CN116734848A

CN116734848A - Vehicle navigation with respect to pedestrians and determining vehicle free space

Info

Publication number: CN116734848A
Application number: CN202310036353.6A
Authority: CN
Inventors: 雅各布·本托利拉; 伊丹·盖勒; 里奥·雷格夫; 托默·巴巴; 阿里尔·贝努; 大卫·阿洛尼; 德米特里·克兹纳; 杰克·卡里奥; 哈纳内尔·毕克
Original assignee: Mobileye Vision Technologies Ltd
Current assignee: Mobileye Vision Technologies Ltd
Priority date: 2020-01-03
Filing date: 2020-12-31
Publication date: 2023-09-12
Also published as: CN116601671A

Abstract

Systems and methods are provided for vehicle navigation. In one implementation, a navigation system for a host vehicle may include at least one processor programmed to: receiving at least one captured image representing an environment of the vehicle from a camera on the host vehicle; detecting a pedestrian represented in the at least one captured image; analyzing the at least one captured image to determine an angular rotation indicator and a tilt angle indicator associated with the head of the pedestrian represented in the at least one captured image; and causing at least one navigational action of the host vehicle based on the angular rotation indicator and the inclination angle indicator associated with the head of a pedestrian.

Description

Vehicle navigation with respect to pedestrians and determining vehicle free space

Cross Reference to Related Applications

The present application claims U.S. provisional application No.62/956,964 (filed 1/3/2020); U.S. provisional application No.62/956,970 (filed 1/3/2020); U.S. provisional application No.62/957,014 (filed 1/3/2020); and U.S. provisional application No.62/957,524 (filed on 1/6 of 2020). The aforementioned application is incorporated herein by reference in its entirety.

Technical Field

The present disclosure relates generally to autonomous vehicle navigation.

Background

As technology continues to advance, the goal of a fully autonomous vehicle capable of navigating on the road is to be achieved. Autonomous vehicles may need to take into account a variety of factors and make appropriate decisions based on those factors to safely and accurately reach the intended destination. For example, autonomous vehicles may need to process and interpret visual information (e.g., information captured from a camera), and may also use information obtained from other sources (e.g., from a GPS device, a speed sensor, an accelerometer, a suspension sensor, etc.). Meanwhile, in order to navigate to a destination, an autonomous vehicle may also need to identify its location within a particular road (e.g., a particular lane within a multi-lane road), navigate alongside other vehicles, avoid obstacles and pedestrians, observe traffic signals and signs, and travel from one road to another at an appropriate intersection or overpass. Utilizing and interpreting the vast amount of information collected by autonomous vehicles as they travel to their destination poses a number of design challenges. The large amount of data (e.g., captured image data, map data, GPS data, sensor data, etc.) that an autonomous vehicle may need to analyze, access, and/or store poses challenges that may actually limit or even adversely affect autonomous navigation. Furthermore, the large amount of data required to store and update maps poses a significant challenge if autonomous vehicles rely on traditional mapping techniques for navigation.

Disclosure of Invention

Embodiments in accordance with the present disclosure provide systems and methods for autonomous vehicle navigation. The disclosed embodiments may use cameras to provide autonomous vehicle navigation features. For example, in accordance with the disclosed embodiments, the disclosed system may include one, two, or more cameras that monitor the environment of the vehicle. The disclosed system may provide a navigational response based on, for example, analysis of images captured by one or more of the cameras.

In one embodiment, a navigation system for a host vehicle may include at least one processor. The at least one processor may be programmed to receive a captured image from a camera on the host vehicle, the captured image comprising a representation of a obscured pedestrian in the environment of the host vehicle. The captured image may lack a representation of the area of the occluded pedestrian contacting the ground surface. The at least one processor may also be programmed to provide the captured image to an analysis module configured to generate an output for the captured image. The generated output includes an indicator of a contact location of the obscured pedestrian with the ground surface. The at least one processor may be further programmed to receive the generated output from the analysis module including the indicator of the contact location of the obscured pedestrian with the ground surface. The at least one processor may be further programmed to cause at least one navigational action of the host vehicle based on the indicator of the contact location of the obscured pedestrian with the ground surface.

In one embodiment, a navigation system for a host vehicle may include at least one processor. The at least one processor may be programmed to receive at least one captured image representing an environment of the vehicle from a camera on the host vehicle. The at least one processor may also be programmed to detect a pedestrian represented in the at least one captured image. The at least one processor may be further programmed to analyze the at least one captured image to determine an angular rotation indicator and a tilt angle indicator associated with the head of the pedestrian represented in the at least one captured image. The at least one processor may also be programmed to cause at least one navigational action of the host vehicle based on the angular rotation indicator and the inclination angle indicator associated with the head of a pedestrian.

In one embodiment, a method for navigating a host vehicle may include: receiving at least one captured image representing an environment of the vehicle from a camera on the host vehicle; detecting a pedestrian represented in the at least one captured image; analyzing the at least one captured image to determine an angular rotation indicator and a tilt angle indicator associated with the head of the pedestrian represented in the at least one captured image; and causing at least one navigational action of the host vehicle based on the angular rotation indicator and the inclination angle indicator associated with the head of a pedestrian.

In one embodiment, a system for detecting free space relative to a host vehicle and for navigating the host vehicle relative to the detected free space is disclosed. The system may include at least one processor. The at least one processor may be programmed to receive a plurality of images from one or more cameras associated with the host vehicle. The plurality of images may represent areas in front of, behind, and to the sides of the host vehicle. The at least one processor may also be programmed to correlate the plurality of images to provide an uninterrupted 360 degree view of the environment surrounding the host vehicle. The at least one processor may be further programmed to analyze the correlated plurality of images to identify a free space region in the environment surrounding the host vehicle. The at least one processor may be further programmed to cause the host vehicle to navigate within the identified free-space region while avoiding navigating within regions outside of the identified free-space region.

In one embodiment, a computer-implemented method for detecting free space relative to a host vehicle and for navigating the host vehicle relative to the detected free space is disclosed. The method may include receiving a plurality of images from one or more cameras associated with the host vehicle. The plurality of images may represent areas in front of, behind, and to the sides of the host vehicle. The method may further include correlating the plurality of images to provide an uninterrupted 360 degree view of the environment surrounding the host vehicle. The method may further include analyzing the correlated plurality of images to identify a free space region in the environment surrounding the host vehicle. The method may further include causing the host vehicle to navigate within the identified free-space region while avoiding navigating within regions outside of the identified free-space region.

According to other disclosed embodiments, a non-transitory computer readable storage medium may store program instructions that are executed by at least one processing device and perform any of the methods described herein.

The foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the claims.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate various disclosed embodiments. In the drawings:

FIG. 1 is a diagrammatic representation of an exemplary system in accordance with a disclosed embodiment.

Fig. 2A is a diagrammatic side view representation of an exemplary vehicle including a system in accordance with a disclosed embodiment.

Fig. 2B is a diagrammatic top view representation of the vehicle and system shown in fig. 2, in accordance with the disclosed embodiments.

Fig. 2C is a diagrammatic top view representation of another embodiment of a vehicle including a system according to the disclosed embodiments.

Fig. 2D is a diagrammatic top view representation of yet another embodiment of a vehicle including a system in accordance with the disclosed embodiments.

Fig. 2E is a diagrammatic top view representation of yet another embodiment of a vehicle including a system in accordance with the disclosed embodiments.

FIG. 2F is a diagrammatical representation of an exemplary vehicle control system in accordance with the disclosed embodiments.

FIG. 3A is a diagrammatic representation of a vehicle interior including a rearview mirror and a user interface for a vehicle imaging system in accordance with a disclosed embodiment.

Fig. 3B is an illustration of an example of a camera mount configured to be positioned behind a rear view mirror and against a vehicle windshield in accordance with the disclosed embodiments.

Fig. 3C is an illustration of the camera mount shown in fig. 3B from a different angle, in accordance with the disclosed embodiments.

Fig. 3D is an illustration of an example of a camera mount configured to be positioned behind a rear view mirror and against a vehicle windshield in accordance with the disclosed embodiments.

FIG. 4 is an exemplary block diagram of a memory configured to store instructions for performing one or more operations in accordance with the disclosed embodiments.

FIG. 5A is a flowchart illustrating an exemplary process for eliciting one or more navigational responses based on monocular image analysis, in accordance with the disclosed embodiments.

FIG. 5B is a flowchart illustrating an exemplary process of detecting one or more vehicles and/or pedestrians in a collection of images, in accordance with the disclosed embodiments.

FIG. 5C is a flowchart illustrating an exemplary process of detecting road markings and/or lane geometry information in a set of images, in accordance with the disclosed embodiments.

FIG. 5D is a flowchart illustrating an exemplary process of detecting traffic lights in a collection of images, in accordance with the disclosed embodiments.

FIG. 5E is a flowchart illustrating an exemplary process of eliciting one or more navigational responses based on vehicle paths in accordance with the disclosed embodiments.

Fig. 5F is a flowchart illustrating an exemplary process of determining whether a forward vehicle (driving vehicle) is changing lanes, in accordance with the disclosed embodiments.

FIG. 6 is a flowchart illustrating an exemplary process for eliciting one or more navigational responses based on stereoscopic image analysis, in accordance with the disclosed embodiments.

FIG. 7 is a flowchart illustrating an exemplary process for eliciting one or more navigational responses based on analysis of three sets of images, in accordance with the disclosed embodiment.

FIG. 8 illustrates a sparse map for providing autonomous vehicle navigation, in accordance with a disclosed embodiment.

Fig. 9A shows a polynomial representation of a portion of a road segment in accordance with a disclosed embodiment.

Fig. 9B illustrates a curve in three-dimensional space representing a target trajectory of a vehicle for a particular road segment contained in a sparse map, in accordance with the disclosed embodiments.

FIG. 10 illustrates example roadmap that may be included in a sparse map, in accordance with the disclosed embodiments.

FIG. 11A illustrates a polynomial representation of a trajectory in accordance with disclosed embodiments.

Fig. 11B and 11C illustrate target trajectories along a multi-lane roadway in accordance with the disclosed embodiments.

FIG. 11D illustrates an example road signature profile in accordance with the disclosed embodiments.

Fig. 12 is a schematic diagram of a system that uses crowd-sourced data (crowd sourcing data) received from a plurality of vehicles for autonomous vehicle navigation in accordance with the disclosed embodiments.

FIG. 13 illustrates an example autonomous vehicle road navigation model represented by a plurality of three-dimensional splines, in accordance with the disclosed embodiments.

Fig. 14 shows a map sketch (map skeleton) generated from combining location information from multiple drives, in accordance with the disclosed embodiments.

Fig. 15 illustrates an example of two drives being longitudinally aligned with an example sign as a road sign, in accordance with the disclosed embodiments.

Fig. 16 illustrates an example of multiple drives being longitudinally aligned with an example sign as a road sign, in accordance with the disclosed embodiments.

FIG. 17 is a schematic diagram of a system for generating driving data using a camera, a vehicle, and a server, in accordance with the disclosed embodiments.

Fig. 18 is a schematic diagram of a system for crowdsourcing sparse maps, in accordance with a disclosed embodiment.

Fig. 19 is a flow chart illustrating an exemplary process of generating a sparse map for autonomous vehicle navigation along a road segment in accordance with the disclosed embodiments.

Fig. 20 shows a block diagram of a server in accordance with the disclosed embodiments.

FIG. 21 shows a block diagram of a memory in accordance with the disclosed embodiments.

FIG. 22 illustrates a process of clustering vehicle trajectories associated with vehicles in accordance with the disclosed embodiments.

FIG. 23 illustrates a navigation system of a vehicle that may be used for autonomous navigation in accordance with the disclosed embodiments.

24A, 24B, 24C and 24D illustrate exemplary lane markings that may be detected in accordance with a disclosed embodiment.

FIG. 24E illustrates an exemplary mapped lane marking in accordance with the disclosed embodiments.

FIG. 24F illustrates an exemplary anomaly associated with detecting lane markings in accordance with a disclosed embodiment.

FIG. 25A illustrates an exemplary image of a vehicle surroundings based on navigation of mapped lane markers in accordance with the disclosed embodiments.

FIG. 25B illustrates lateral positioning correction of a vehicle in a road navigation model based on mapped lane markings, in accordance with a disclosed embodiment.

25C and 25D provide conceptual representations of a positioning technique for positioning a host vehicle along a target trajectory using mapping features included in a sparse map.

FIG. 26A is a flowchart illustrating an exemplary process for mapping lane markings for use in autonomous vehicle navigation in accordance with a disclosed embodiment.

Fig. 26B is a flowchart illustrating an exemplary process for autonomously navigating a host vehicle along a road segment using mapped lane markings in accordance with a disclosed embodiment.

FIG. 27 is an exemplary block diagram of a memory configured to store instructions for performing one or more operations in accordance with the disclosed embodiments.

FIG. 28A provides a diagram of an exemplary captured image including a representation of a partially obscured pedestrian, in accordance with the disclosed embodiments.

FIG. 28B provides a diagram of an exemplary captured image including a representation of a partially obscured pedestrian, in accordance with the disclosed embodiments.

Fig. 28C provides a diagram illustrating an exemplary bounding box encompassing a partially obscured pedestrian in the captured image of fig. 28B, in accordance with a disclosed embodiment.

Fig. 28D provides a diagram illustrating the pixel size and real world size of the bounding box of fig. 28C in accordance with the disclosed embodiments.

FIG. 29 provides an algorithm flow diagram for navigating a vehicle in accordance with the disclosed embodiments.

FIG. 30 is an exemplary block diagram of a memory configured to store instructions for performing one or more operations in accordance with the disclosed embodiments.

Fig. 31A provides a diagram describing the rotation angle and tilt angle of the head pose according to the disclosed embodiments.

Fig. 31B provides a graph of navigation changes based on pedestrian gaze direction in accordance with the disclosed embodiments.

FIG. 32 provides an algorithm flow diagram for navigating a vehicle in accordance with the disclosed embodiments.

FIG. 33 is an exemplary block diagram of a memory configured to store instructions for performing one or more operations in accordance with the disclosed embodiments.

Fig. 34A provides a diagram of a camera system for imaging an environment surrounding a host vehicle in accordance with the disclosed embodiments.

Fig. 34B provides a top view of the environment surrounding the host vehicle in accordance with the disclosed embodiments.

FIG. 35 provides an algorithm flow diagram for navigating a vehicle in accordance with the disclosed embodiments.

Detailed Description

The following detailed description refers to the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the following description to refer to the same or like parts. While several illustrative embodiments are described herein, modifications, adaptations, and other implementations are possible. For example, substitutions, additions or modifications may be made to the components illustrated in the drawings, and the illustrative methods described herein may be modified by substituting, reordering, removing, or adding steps to the disclosed methods. Accordingly, the following detailed description is not limited to the disclosed embodiments and examples. The proper scope is defined by the appended claims.

Autonomous vehicle overview

The term "autonomous vehicle" as used throughout this disclosure refers to a vehicle that is capable of effecting at least one navigational change without driver input. "navigation change" means a change in one or more of steering, braking, or acceleration of the vehicle. To be autonomous, the vehicle need not be fully automated (e.g., full-scale operation without driver or driver input). Autonomous vehicles include those vehicles that are capable of operating under driver control during certain periods of time and without driver control during other periods of time. Autonomous vehicles may also include vehicles that control only some aspects of vehicle navigation, such as steering (e.g., to keep the vehicle route between vehicle lane constraints), while other aspects may be left to the driver (e.g., braking). In some cases, the autonomous vehicle may handle some or all aspects of braking, speed control, and/or steering of the vehicle.

Since human drivers typically rely on visual cues and observations to control vehicles, transportation infrastructure is accordingly built in which lane markings, traffic signs and traffic lights are all designed to provide visual information to the driver. In view of these design characteristics of the transport infrastructure, the autonomous vehicle may include a camera and a processing unit that analyzes visual information captured from the environment of the vehicle. Visual information may include, for example, components of the transport infrastructure (e.g., lane markings, traffic signs, traffic lights, etc.) and other obstructions (e.g., other vehicles, pedestrians, debris, etc.) that are observable by the driver. In addition, the autonomous vehicle may also use stored information, such as information that provides a model of the vehicle environment when navigating. For example, the vehicle may use GPS data, sensor data (e.g., from accelerometers, speed sensors, suspension sensors, etc.), and/or other map data to provide information related to its environment while traveling, and the vehicle (and other vehicles) may use this information to locate itself on the model.

In some embodiments of the present disclosure, the autonomous vehicle may use information obtained (e.g., from a camera, GPS device, accelerometer, speed sensor, suspension sensor, etc.) while navigating. In other embodiments, an autonomous vehicle may use information obtained by the vehicle (or other vehicles) from past navigation while navigating. In still other embodiments, autonomous vehicles may use a combination of information obtained while navigating and information obtained from past navigation. The following section provides an overview of forward imaging systems and methods in accordance with the disclosed embodiments, followed by an overview of the systems. The following section discloses systems and methods for constructing, using, and updating sparse maps for autonomous vehicle navigation.

Overview of the System

FIG. 1 is a block diagram representation of a system 100 according to the disclosed exemplary embodiments. The system 100 may include various components as required by a particular implementation. In some embodiments, the system 100 may include a processing unit 110, an image acquisition unit 120, a position sensor 130, one or more memory units 140 and 150, a map database 160, a user interface 170, and a wireless transceiver 172. The processing unit 110 may include one or more processing devices. In some embodiments, processing unit 110 may include an application processor 180, an image processor 190, or any other suitable processing device. Similarly, the image acquisition unit 120 may include any number of image acquisition devices and components, as desired for a particular application. In some embodiments, the image acquisition unit 120 may include one or more image capturing devices (e.g., cameras), such as image capturing device 122, image capturing device 124, and image capturing device 126. The system 100 may further comprise a data interface 128, which data interface 128 communicatively connects the processing unit 110 to the image acquisition device 120. For example, the data interface 128 may include one or more any wired and/or wireless links for communicating image data acquired by the image acquisition device 120 to the processing unit 110.

The wireless transceiver 172 may include one or more devices configured to exchange transmissions to one or more networks (e.g., cellular, internet, etc.) via an air interface using radio frequencies, infrared frequencies, magnetic fields, or electric fields. The wireless transceiver 172 may use any known standard to transmit and/or receive data (e.g., wi-Fi,Bluetooth Smart, 802.15.4, zigBee, etc.). Such transmissions can include communications from the host vehicle to one or more remotely located servers. Such transmissions may also include communications (unidirectional or bi-directional) between the host vehicle and one or more target vehicles in the host vehicle's environment (e.g., to facilitate coordination of navigation of the host vehicle in view of or along with the target vehicles in the host vehicle's environment) or even broadcast transmissions to unspecified recipients in the vicinity of the transmitting vehicle.

Both the application processor 180 and the image processor 190 may include various types of processing devices. For example, either or both of application processor 180 and image processor 190 may include a microprocessor, a preprocessor (e.g., an image preprocessor), a Graphics Processing Unit (GPU), a Central Processing Unit (CPU), support circuits, digital signal processors, integrated circuits, memory, or any other type of device suitable for running an application and suitable for image processing and analysis. In some embodiments, application processor 180 and/or image processor 190 may comprise any type of single-core or multi-core processor, mobile device microcontroller, central processor, or the like. Various processing means may be used, including, for example, a processing means such as a processing unit Processors available to the manufacturer of etc. or from, for example +.>Available to manufacturers of etcGPUs, and may include various architectures (e.g., x86 processors,/and/or the like)>Etc.).

In some embodiments, application processor 180 and/or image processor 190 may include a slaveAny of the available processor chips of the eye q family. Each of these processor designs includes a plurality of processing units having a local memory and an instruction set therein. Such a processor may include video input for receiving image data from a plurality of image sensors, and may also include video output capabilities. In one example, a->90 nm-micro technology operating at 332Mhz was used.The architecture includes two floating point hyper-threading 32 bit RISC CPUs (+)>Core), five Visual Compute Engines (VCEs), three vector microcode processors +.>Denali 64-bit mobile DDR controller, 128-bit internal Sonics interconnect, dual 16-bit video input and 18-bit video output controller, 16-channel DMA, and several peripherals. MIPS34K CPU manages five VCEs and three VMPs ^TM And DMA, second MIPS34K CPU and multi-channel DMA, among other peripherals. Five VCEs, three +.>And the MIPS34K CPU is capable of performing the intensive visual computations required by the multi-function binding application. In the other of the examples described above, in which the first and second embodiments, Is a third generation processor and is of the ratio +.>Six times stronger, it can be used in the disclosed embodiments. In other examples, a->And/or +.>May be used in the disclosed embodiments. Of course, any newer or future EyeQ handling devices may also be used in conjunction with the disclosed embodiments.

Any of the processing devices disclosed herein may be configured to perform certain functions. Configuring a processing device (e.g., any of the EyeQ processors or another controller or microprocessor) to perform certain functions may include programming computer-executable instructions and making those instructions available to the processing device for execution during operation. In some embodiments, configuring the processing device may include directly programming the processing device with architectural instructions. For example, a processing device (e.g., a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), etc.) may be configured using, for example, one or more Hardware Description Languages (HDLs).

In other embodiments, configuring the processing device may include storing executable instructions on a memory that is accessible to the processing device during operation. For example, the processing device may access memory during operation to obtain and execute stored instructions. In either case, the processing device configured to perform the sensing, image analysis, and/or navigation functions disclosed herein represents a dedicated hardware-based system that controls a plurality of hardware-based components of the host vehicle.

Although fig. 1 shows two separate processing devices contained in the processing unit 110, more or fewer processing devices may be used. For example, in some embodiments, a single processing device may be used to accomplish the tasks of application processor 180 and image processor 190. In other embodiments, these tasks may be performed by more than two processing devices. Further, in some embodiments, the system 100 may include one or more of the processing units 110 without including other components (e.g., the image acquisition unit 120).

The processing unit 110 may comprise various types of devices. For example, the processing unit 110 may include various devices such as a controller, an image processor, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), support circuits, a digital signal processor, an integrated circuit, a memory, or any other type of device for image processing and analysis. The image pre-processor may include a video processor for capturing, digitizing, and processing the imagery from the image sensor. The CPU may include any number of microcontrollers or microprocessors. The GPU may also include any number of microcontrollers or microprocessors. The support circuits may be any number of circuits generally known in the art, including cache, power supplies, clocks, and input-output circuits. The memory may store software that, when executed by the processor, controls the operation of the system. The memory may include a database and image processing software. The memory may include any number of random access memory, read only memory, flash memory, disk drives, optical storage devices, magnetic tape storage devices, removable storage devices, and other types of storage devices. In one example, the memory may be separate from the processing unit 110. In another example, the memory may be integrated into the processing unit 110.

Each memory 140, 150 may include software instructions that, when executed by a processor (e.g., application processor 180 and/or image processor 190), may control the operation of various aspects of the system 100. These memory units may include various databases and image processing software, and trained systems (e.g., neural networks or deep neural networks). The memory unit may include Random Access Memory (RAM), read Only Memory (ROM), flash memory, a disk drive, an optical storage device, a tape storage device, a removable storage device, and/or other types of storage devices. In some embodiments, the memory units 140, 150 may be separate from the application processor 180 and/or the image processor 190. In other embodiments, these memory units may be integrated into the application processor 180 and/or the image processor 190.

The position sensor 130 may comprise any type of device suitable for determining a position associated with at least one component of the system 100. In some embodiments, the location sensor 130 may include a GPS receiver. Such receivers are capable of determining user position and velocity by processing signals broadcast by global positioning system satellites. The location information from the location sensor 130 may be made available to the application processor 180 and/or the image processor 190.

In some embodiments, the system 100 may include components such as a speed sensor (e.g., tachometer, speedometer) for measuring the speed of the vehicle 200 and/or an accelerometer (single or multi-axis) for measuring the acceleration of the vehicle 200.

The user interface 170 may comprise any device suitable for providing information to or receiving input from one or more users of the system 100. In some embodiments, the user interface 170 may include user input devices including, for example, a touch screen, microphone, keyboard, pointer device, track wheel, camera, knob, button, and the like. Through such input devices, a user may be able to provide information input or commands to system 100 by typing instructions or information, providing voice commands, selecting menu options on a screen using buttons, pointers, or eye tracking capabilities, or through any other suitable technique for communicating information to system 100.

The user interface 170 may be equipped with one or more processing devices configured to provide/receive information to/from a user and process that information for use by, for example, the application processor 180. In some embodiments, such processing devices may execute instructions for recognizing and tracking eye movements, receiving and interpreting voice commands, recognizing and interpreting touches and/or gestures made on a touch screen, responding to keyboard inputs or menu selections, and the like. In some embodiments, the user interface 170 may include a display, speakers, haptic devices, and/or any other device for providing output information to a user.

Map database 160 may include any type of database for storing map data useful to system 100. In some embodiments, map database 160 may include data for various items (including roads, water features, geographic features, businesses, points of interest, restaurants, gas stations, etc.) related to locations in a reference coordinate system. The map database 160 may store not only the locations of such items, but also descriptors related to those items, including, for example, names associated with any of the stored features. In some embodiments, map database 160 may be physically located with other components of system 100. Alternatively or additionally, the map database 160, or a portion thereof, may be remotely located relative to other components of the system 100 (e.g., the processing unit 110). In such embodiments, the information from map database 160 may be downloaded via a wired or wireless data connection to the network (e.g., via a cellular network and/or the Internet, etc.). In some cases, the map database 160 may store a sparse data model that contains polynomial representations of certain road features (e.g., lane markers) or target trajectories of the host vehicle. Systems and methods of generating such maps are discussed below with reference to fig. 8-19.

Image capture devices 122, 124, and 126 may each comprise any type of device suitable for capturing at least one image from an environment. Further, any number of image capture devices may be used to acquire images for input to the image processor. Some embodiments may include only a single image capture device, while other embodiments may include two, three, or even four or more image capture devices. The image capture devices 122, 124, and 126 are further described below with reference to fig. 2B-2E.

The system 100, or various components thereof, may be incorporated into a variety of different platforms. In some embodiments, the system 100 may be included on a vehicle 200, as shown in fig. 2A. For example, the vehicle 200 may be equipped with any of the processing unit 110 and other components of the system 100, as described above with respect to fig. 1. Although in some embodiments, the vehicle 200 may be equipped with only a single image capture device (e.g., a camera), in other embodiments, such as the embodiments described in connection with fig. 2B-2E, multiple image capture devices may be used. For example, as shown in fig. 2A, either of the image capturing devices 122 and 124 of the vehicle 200 may be an integral part of an ADAS (advanced driver assistance system) imaging set.

The image capturing devices included on the vehicle 200 as part of the image acquisition unit 120 may be positioned in any suitable location. In some embodiments, as shown in FIGS. 2A-2E and 3A-3C, the image capture device 122 may be located near a rearview mirror. This location may provide a line of sight similar to the driver of the vehicle 200, which may help determine what is visible and invisible to the driver. The image capture device 122 may be positioned anywhere near the rear view mirror, but placing the image capture device 122 on the driver side of the rear view mirror may further help obtain an image representative of the driver's field of view and/or line of sight.

Other locations of the image capturing means of the image capturing unit 120 may also be used. For example, the image capture device 124 may be located on or in a bumper of the vehicle 200. Such a position may be particularly suitable for image capturing devices having a wide field of view. The line of sight of the bumper-positioned image capture device can be different from the driver's line of sight, and thus the bumper image capture device and the driver may not necessarily see the same object. Image capture devices (e.g., image capture devices 122, 124, and 126) may also be located in other locations. For example, the image capture device may be located on or in one or both of the side mirrors of the vehicle 200, on the roof of the vehicle 200, on the hood of the vehicle 200, on the trunk of the vehicle 200, on the side of the vehicle 200, mounted on either of the windows of the vehicle 200, positioned behind or in front of it, in or near the light images mounted on the front and/or back of the vehicle 200, and so forth.

In addition to the image capture device, the vehicle 200 may also include various other components of the system 100. For example, the processing unit 110 may be included on the vehicle 200, integrated with or separate from an Engine Control Unit (ECU) of the vehicle. The vehicle 200 may also be equipped with a location sensor 130 (e.g., a GPS receiver), and may also include a map database 160 and memory units 140 and 150.

As previously described, the wireless transceiver 172 may receive data and/or over one or more networks (e.g., cellular network, internet, etc.). For example, the wireless transceiver 172 may upload data collected by the system 100 to one or more servers and download data from the one or more servers. Via the wireless transceiver 172, the system 100 may receive periodic or on-demand updates to the data stored in the map database 160, memory 140, and/or memory 150. Similarly, wireless transceiver 172 may upload any data (e.g., images captured by image acquisition unit 120, data received by position sensor 130 or other sensors, vehicle control systems, etc.) and/or any data processed by processing unit 110 from system 100 to one or more servers.

The system 100 may set up to upload data to a server (e.g., to the cloud) based on the privacy level. For example, the system 100 may implement privacy level settings to adjust or limit the type of data (including metadata) sent to the server, which may uniquely identify the vehicle and/or the driver/owner of the vehicle. Such settings may be set by a user via, for example, the wireless transceiver 172, initialized by factory default settings, or by data received by the wireless transceiver 172.

In some embodiments, the system 100 may upload data at a "high" privacy level, and under setup settings, the system 100 may transmit data (e.g., route-related location information, captured images, etc.) without any details regarding the particular vehicle and/or driver/owner. For example, when uploading data in a "high" privacy setting, the system 100 may not include a Vehicle Identification Number (VIN) or the name of the driver or owner of the vehicle, but may instead transmit data such as captured images and/or limited location information related to the route.

Other privacy classes are contemplated. For example, the system 100 may transmit data to the server at an "intermediate" privacy level and include additional information not included at a "high" privacy level, such as the brand and/or model number of the vehicle and/or the type of vehicle (e.g., passenger car, sport utility car, truck, etc.). In some embodiments, the system 100 may upload data at a "low" privacy level. Under a "low" privacy level setting, the system 100 may upload data and include information sufficient to uniquely identify a particular vehicle, owner/driver, and/or a portion or all of the route traveled by the vehicle. Such "low" privacy level data may include one or more of the following: such as VIN, driver/owner name, origin of the vehicle before departure, intended destination of the vehicle, brand and/or model of the vehicle, type of vehicle, etc.

FIG. 2A is a diagrammatic side view representation of an exemplary vehicle imaging system in accordance with a disclosed embodiment. Fig. 2B is a diagrammatic top view illustration of the embodiment shown in fig. 2A. As shown in fig. 2B, the disclosed embodiments may include a vehicle 200 including the system 100 in a body of the vehicle 200 with a first image capture device 122 positioned near a rear view mirror and/or near a driver of the vehicle 200, a second image capture device 124 positioned on or in a bumper area (e.g., one of the bumper areas 210) of the vehicle 200, and the processing unit 110.

As shown in fig. 2C, both image capture devices 122 and 124 may be positioned near a rear view mirror and/or near a driver of the vehicle 200. Additionally, while two image capture devices 122 and 124 are shown in fig. 2B and 2C, it should be understood that other embodiments may include more than two image capture devices. For example, in the embodiment shown in fig. 2D and 2E, the first, second, and third image capture devices 122, 124, and 126 are included in the system 100 of the vehicle 200.

As shown in fig. 2D, the image capture device 122 may be positioned near a rear view mirror and/or near a driver of the vehicle 200, and the image capture devices 124 and 126 may be positioned on or in a bumper area of the vehicle 200 (e.g., one of the bumper areas 210). And as shown in fig. 2E, the image capture devices 122, 124, and 126 may be positioned near the rear view mirror and/or near the driver's seat of the vehicle 200. The disclosed embodiments are not limited to any particular number and configuration of image capture devices, and image capture devices may be positioned in any suitable location within and/or on vehicle 200.

It is to be understood that the disclosed embodiments are not limited to vehicles, but may be applied in other contexts. It is also to be understood that the disclosed embodiments are not limited to a particular type of vehicle 200, but may be applicable to all types of vehicles, including automobiles, trucks, trailers, and other types of vehicles.

The first image capture device 122 may comprise any suitable type of image capture device. The image capture device 122 may include an optical axis. In one example, the image capture device 122 may include an aptna M9V024WVGA sensor with a global shutter. In other embodiments, the image capture device 122 may provide a resolution of 1280×960 pixels and may include a rolling shutter. The image capture device 122 may include various optical elements. In some embodiments, one or more lenses may be included, for example, to provide a desired focal length and field of view for the image capture device. In some embodiments, the image capture device 122 may be associated with a 6mm lens or a 12mm lens. In some embodiments, the image capture device 122 may be configured to capture an image having a desired field of view (FOV) 202, as shown in fig. 2D. For example, the image capture device 122 may be configured to have a conventional FOV, including a 46 degree FOV, a 50 degree FOV, a 52 degree FOV, or more, for example, in the range of 40 degrees to 56 degrees. Alternatively, the image capture device 122 may be configured to have a narrow FOV in the range of 23 to 40 degrees, such as a 28 degree FOV or a 36 degree FOV. In addition, the image capture device 122 may be configured to have a wide FOV in the range of 100 to 180 degrees. In some embodiments, the image capture device 122 may include a wide-angle bumper camera or a camera up to a 180 degree FOV. In some embodiments, the image capture device 122 may be a 7.2M pixel image capture device having an aspect ratio of about 2:1 (e.g., hx v=3800 x 1900 pixels), with a horizontal FOV of about 100 degrees. Such an image capture device may be used in place of a three image capture device configuration. Due to the significant lens distortion, the vertical FOV of such an image capture device may be significantly less than 50 degrees in an implementation where the image capture device uses radially symmetric lenses. For example, such lenses may not be radially symmetric, which allows for a vertical FOV of greater than 50 degrees with a 100 degree horizontal FOV.

The first image capture device 122 may acquire a plurality of first images relative to a scene associated with the vehicle 200. Each of the plurality of first images may be acquired as a series of image scan lines, which may be captured using a rolling shutter. Each scan line may include a plurality of pixels.

The first image capture device 122 may have a scan rate associated with acquisition of each of the first series of image scan lines. The scan rate may represent the rate at which the image sensor is able to acquire image data associated with each pixel contained in a particular scan line.

The image capture devices 122, 124, and 126 may include any suitable type and number of image sensors, including CCD sensors or CMOS sensors, for example. In one embodiment, a CMOS image sensor may be used in conjunction with a rolling shutter such that each pixel in a row is read one at a time, and scanning of the row proceeds row by row until the entire image frame has been captured. In some embodiments, the rows may be captured sequentially from top to bottom with respect to the frame.

In some embodiments, one or more of the image capture devices disclosed herein (e.g., image capture devices 122, 124, and 126) may constitute a high resolution imager and may have a resolution of greater than 5M pixels, 7M pixels, 10M pixels, or more.

The use of rolling shutters may cause pixels in different rows to be exposed and captured at different times, which may cause skew and other image artifacts in the captured image frames. On the other hand, when image capture device 122 is configured to operate with a global or synchronized shutter, all pixels may be exposed for the same amount of time during a common exposure period. Thus, image data in frames collected from a system employing a global shutter represents a snapshot of the entire FOV (e.g., FOV 202) at a particular time. In contrast, in rolling shutter applications, each row in a frame is exposed and data is captured at a different time. Therefore, in an image capturing apparatus having a rolling shutter, a moving object may be distorted. This phenomenon will be described in more detail below.

The second image capturing device 124 and the third image capturing device 126 may be any type of image capturing device. Similar to the first image capturing device 122, each of the image capturing devices 124 and 126 may include an optical axis. In one embodiment, each of image capture devices 124 and 126 may include an aptna M9V024 WVGA sensor with a global shutter. Alternatively, each of image capture devices 124 and 126 may include a rolling shutter. Similar to image capture device 122, image capture devices 124 and 126 may be configured to include various lenses and optical elements. In some embodiments, the lenses of the associated image capture devices 124 and 126 may provide the same or a narrower FOV (e.g., FOVs 204 and 206) as compared to the FOV (e.g., FOV 202) of the associated image capture device 122. For example, the image capture devices 124 and 126 may have FOVs of 40 degrees, 30 degrees, 26 degrees, 23 degrees, 20 degrees, or less.

The image capture devices 124 and 126 may acquire a plurality of second and third images relative to a scene associated with the vehicle 200. Each of the plurality of second and third images may be acquired as second and third series of image scan lines, which may be captured using a rolling shutter. Each scan line or row may have a plurality of pixels. The image capture devices 124 and 126 may have second and third scan rates associated with acquisition of each of the image scan lines included in the second and third series.

Each image capture device 122, 124, and 126 may be positioned in any suitable position and orientation relative to the vehicle 200. The relative positioning of the image capture devices 122, 124, and 126 may be selected to help fuse together information acquired from the image capture devices. For example, in some embodiments, the FOV of the associated image capture device 124 (e.g., FOV 204) may partially or completely overlap with the FOV of the associated image capture device 122 (e.g., FOV 202) and the FOV of the associated image capture device 126 (e.g., FOV 206).

The image capture devices 122, 124, and 126 may be located at the vehicleAny suitable relative height above the tool 200. In one example, there may be a height difference between the image capture devices 122, 124, and 126, which may provide sufficient parallax information to enable stereoscopic analysis. For example, as shown in FIG. 2A, the two image capture devices 122 and 124 are at different heights. There may also be a lateral displacement difference between the image capturing means 122, 124 and 126, for example, giving additional parallax information for the processing unit 110 to perform a stereoscopic analysis. The difference in lateral displacement can be obtained by d _x As shown in fig. 2C and 2D. In some embodiments, a forward or rearward displacement (e.g., a range displacement) may exist between the image capture devices 122, 124, and 126. For example, image capture device 122 may be located 0.5 to 2 meters or more behind image capture device 124 and/or image capture device 126. This type of displacement may enable one of the image capturing devices to overlay the potential blind spot of the other image capturing device(s).

Image capture device 122 may have any suitable resolution capability (e.g., number of pixels associated with the image sensor), and the resolution of the image sensor(s) of associated image capture device 122 may be higher, lower, or the same as the resolution of the image sensor(s) of associated image capture devices 124 and 126. In some embodiments, the image sensor(s) associated with image capture device 122 and/or image capture devices 124 and 126 may have a resolution of 640 x 480, 1024 x 768, 1280 x 960, or any other suitable resolution.

The frame rate (e.g., the rate at which the image capture device acquires a set of pixel data for one image frame before continuing to capture pixel data associated with a next image frame) may be controllable. The frame rate of the associated image capture device 122 may be higher, lower, or the same as the frame rate of the associated image capture devices 124 and 126. The frame rate associated with image capture devices 122, 124, and 126 may depend on a variety of factors, which may affect the timing of the frame rate. For example, one or more of image capture devices 122, 124, and 126 may include a selectable pixel delay period applied before or after acquiring image data associated with one or more pixels of an image sensor in image capture devices 122, 124, and/or 126. In general, image data corresponding to each pixel may be acquired at the clock rate of the device (e.g., one pixel per clock cycle). Additionally, in embodiments that include a rolling shutter, one or more of image capture devices 122, 124, and 126 may include a selectable horizontal blanking period applied before or after acquiring image data associated with a row of pixels of an image sensor in image capture devices 122, 124, and/or 126. Further, one or more of image capture devices 122, 124, and 126 may include an optional vertical blanking period applied before or after image data associated with image frames of image capture devices 122, 124, and/or 126 is acquired.

These timing controls may enable synchronization of the frame rates associated with the image capture devices 122, 124, and 126, even where the line scan rates of each are different. In addition, as will be discussed in more detail below, these selectable timing controls, as well as other factors (e.g., image sensor resolution, maximum line scan rate, etc.), may enable synchronization of image capture from areas where the FOV of image capture device 122 overlaps with one or more FOVs of image capture devices 124 and 126, even where the field of view of image capture device 122 is different from the FOVs of image capture devices 124 and 126.

The frame rate timing in image capture devices 122, 124, and 126 may depend on the resolution of the associated image sensor. For example, assuming similar line scan rates for two devices, if one device includes an image sensor with a resolution of 640 x 480 and the other device includes an image sensor with a resolution of 1280 x 960, more time is required to acquire frames of image data from the sensor with the higher resolution.

Another factor that may affect the timing of image data acquisition in image capture devices 122, 124, and 126 is the maximum line scan rate. For example, acquiring a line of image data from the image sensors included in image capture devices 122, 124, and 126 would require some minimum amount of time. Assuming that the pixel delay period is not increased, this minimum amount of time for acquiring a line of image data will be related to the maximum line scan rate for a particular device. Devices that provide a higher maximum line scan rate have the potential to provide a higher frame rate than devices with a lower maximum line scan rate. In some embodiments, one or more of image capture devices 124 and 126 may have a maximum line scan rate that is higher than the maximum line scan rate of associated image capture device 122. In some embodiments, the maximum line scan rate of image capture devices 124 and/or 126 may be 1.25, 1.5, 1.75, or 2 times or more the maximum line scan rate of image capture device 122.

In another embodiment, image capture devices 122, 124, and 126 may have the same maximum line scan rate, but image capture device 122 may operate at a scan rate less than or equal to its maximum scan rate. The system may be configured such that one or more of image capture devices 124 and 126 operate at a line scan rate equal to the line scan rate of image capture device 122. In other examples, the system may be configured such that the line scan rate of image capture device 124 and/or image capture device 126 may be 1.25, 1.5, 1.75, or 2 times or more the line scan rate of image capture device 122.

In some embodiments, the image capture devices 122, 124, and 126 may be asymmetric. That is, they may include cameras with different fields of view (FOV) and focal lengths. The fields of view of the image capture devices 122, 124, and 126 may include, for example, any desired area relative to the environment of the vehicle 200. In some embodiments, one or more of the image capture devices 122, 124, and 126 may be configured to acquire image data from the environment of the front of the vehicle 200, the rear of the vehicle 200, the sides of the vehicle 200, or a combination thereof.

Further, the focal length associated with each image capturing device 122, 124, and/or 126 may be selectable (e.g., by including appropriate lenses, etc.) such that each device acquires an image of the object at the desired distance range relative to the vehicle 200. For example, in some embodiments, the image capture devices 122, 124, and 126 may acquire images of close-up objects within a few meters from the vehicle. The image capture devices 122, 124, and 126 may also be configured to acquire images of objects at a range of greater distances from the vehicle (e.g., 25m, 50m, 100m, 150m, or more). Further, the focal lengths of image capture devices 122, 124, and 126 may be selected such that one image capture device (e.g., image capture device 122) is capable of capturing images of objects relatively close to the vehicle (e.g., within 10m or within 20 m), while the other image capture devices (e.g., image capture devices 124 and 126) are capable of capturing images of objects farther from the vehicle 200 (e.g., greater than 20m, 50m, 100m, 150m, etc.).

According to some embodiments, the FOV of one or more of the image capture devices 122, 124, and 126 may have a wide angle. For example, it may be advantageous to have a FOV of 140 degrees, particularly for image capture devices 122, 124, and 126 that may be used to capture images of areas in the vicinity of vehicle 200. For example, the image capture device 122 may be used to capture images of an area to the right or left of the vehicle 200, and in such embodiments, it may be desirable for the image capture device 122 to have a wide FOV (e.g., at least 140 degrees).

The field of view associated with each of the image capture devices 122, 124, and 126 may depend on the respective focal length. For example, as the focal length increases, the corresponding field of view decreases.

The image capture devices 122, 124, and 126 may be configured to have any suitable field of view. In one particular example, image capture device 122 may have a horizontal FOV of 46 degrees, image capture device 124 may have a horizontal FOV of 23 degrees, and image capture device 126 may have a horizontal FOV between 23 and 46 degrees. In another example, image capture device 122 may have a horizontal FOV of 52 degrees, image capture device 124 may have a horizontal FOV of 26 degrees, and image capture device 126 may have a horizontal FOV between 26 and 52 degrees. In some embodiments, the ratio of the FOV of image capture device 122 to the FOV of image capture device 124 and/or image capture device 126 may vary from 1.5 to 2.0. In other embodiments, this ratio may vary between 1.25 and 2.25.

The system 100 may be configured such that the field of view of the image capture device 122 at least partially or completely overlaps with the field of view of the image capture device 124 and/or the image capture device 126. In some embodiments, the system 100 may be configured such that the fields of view of the image capture devices 124 and 126 fall within (e.g., are narrower than) the field of view of the image capture device 122 and share a common center with the field of view of the image capture device 122, for example. In other embodiments, image capture devices 122, 124, and 126 may capture adjacent FOVs, or may have partial overlap in their FOVs. In some embodiments, the fields of view of image capture devices 122, 124, and 126 may be aligned such that the center of the narrower FOV image capture devices 124 and/or 126 may be located in the lower half of the field of view of the wider FOV device 122.

FIG. 2F is a diagrammatical representation of an exemplary vehicle control system in accordance with the disclosed embodiments. As shown in fig. 2F, the vehicle 200 may include a throttle system 220, a brake system 230, and a steering system 240. The system 100 may provide inputs (e.g., control signals) to one or more of the throttle system 220, the brake system 230, and the steering system 240 via one or more data links (e.g., one or more any wired and/or wireless links for transmitting data). For example, based on analysis of images acquired by image capture devices 122, 124, and/or 126, system 100 may provide control signals to one or more of throttle system 220, brake system 230, and steering system 240 to navigate vehicle 200 (e.g., by causing accelerations, turns, lane changes, etc.). Further, the system 100 may receive input from one or more of the throttle system 220, the brake system 230, and the steering system 24 indicative of an operating condition of the vehicle 200 (e.g., speed, whether the vehicle 200 is braking and/or turning, etc.). Additional details are provided below in connection with fig. 4-7.

As shown in fig. 3A, the vehicle 200 may also include a user interface 170 for interacting with a driver or passenger of the vehicle 200. For example, the user interface 170 in a vehicle application may include a touch screen 320, a knob 330, buttons 340, and a microphone 350. The driver or passenger of the vehicle 200 may also interact with the system 100 using a handle (e.g., located on or near a steering column of the vehicle 200, including, for example, a turn signal handle), a button (e.g., located on a steering wheel of the vehicle 200), and the like. In some embodiments, microphone 350 may be positioned adjacent to rear view mirror 310. Similarly, in some embodiments, image capture device 122 may be located near rearview mirror 310. In some embodiments, the user interface 170 may also include one or more speakers 360 (e.g., speakers of a vehicle audio system). For example, the system 100 may provide various notifications (e.g., alarms) via the speaker 360.

Fig. 3B-3D are illustrations of an exemplary camera mount 370 in accordance with the disclosed embodiments, the camera mount 370 configured to be positioned against a vehicle windshield behind a rear view mirror (e.g., rear view mirror 310). As shown in fig. 3B, the camera mount 370 may include image capture devices 122, 124, and 126. The image capture devices 124 and 126 may be positioned behind a light shield (glare shield) 380, which light shield 380 may be flush with the vehicle windshield and include components of film and/or anti-reflective material. For example, the light shield 380 may be positioned such that the light shield is aligned with a vehicle windshield having a mating bevel. In some embodiments, each of the image capture devices 122, 124, and 126 may be positioned behind a light shield 380, as shown in fig. 3D. The disclosed embodiments are not limited to any particular configuration of image capture devices 122, 124, and 126, camera mount 370, and light shield 380. Fig. 3C is an illustration of the camera mount 370 of fig. 3B from a front perspective.

Many variations and/or modifications to the above-disclosed embodiments may be made, as will be appreciated by those skilled in the art having the benefit of the present disclosure. For example, not all components may be necessary for operation of the system 100. Further, any component may be located in any suitable portion of the system 100, and the components may be rearranged in a variety of configurations while providing the functionality of the disclosed embodiments. Thus, the above-described configuration is an example, and regardless of the above-described configuration, the system 100 is capable of providing a wide range of functionality to analyze the surroundings of the vehicle 200 and navigate the vehicle 200 in response to the analysis.

As discussed in more detail below and in accordance with various disclosed embodiments, the system 100 may provide a variety of features related to autonomous driving and/or driver assistance techniques. For example, the system 100 may analyze image data, location data (e.g., GPS location information), map data, speed data, and/or data from sensors included in the vehicle 200. The system 100 may collect data for analysis from, for example, the image acquisition unit 120, the position sensor 130, and other sensors. Further, the system 100 can analyze the collected data to determine whether the vehicle 200 should take some action, and then automatically take the determined action without human intervention. For example, the system 100 may automatically control braking, acceleration, and/or steering of the vehicle 200 (e.g., by sending control signals to one or more of the throttle system 220, the brake system 230, and the steering system 240) as the vehicle 200 navigates without human intervention. Further, the system 100 can analyze the collected data and issue a warning and/or alert to a vehicle occupant based on the analysis of the collected data. Additional details regarding the various embodiments provided by the system 100 are provided below.

Forward multiple imaging system

As described above, the system 100 may provide driving assistance functionality using a multiple camera system. The multiple camera system may use one or more cameras facing forward of the vehicle. In other embodiments, the multiple camera system may include one or more cameras facing the side of the vehicle or the rear of the vehicle. In one embodiment, for example, the system 100 may use a two-camera imaging system in which a first camera and a second camera (e.g., image capture devices 122 and 124) may be positioned on the front and/or sides of a vehicle (e.g., vehicle 200). The first camera may have a field of view that is greater than, less than, or partially overlapping the field of view of the second camera. In addition, the first camera may be connected to the first image processor to perform monocular image analysis of the image provided by the first camera, and the second camera may be connected to the second image processor to perform monocular image analysis of the image provided by the second camera. The outputs (e.g., processed information) of the first and second image processors may be combined. In some embodiments, the second image processor may receive images from the first camera and the second camera to perform stereoscopic analysis. In another embodiment, the system 100 may use a three-camera imaging system, where each of the cameras has a different field of view. Thus, such a system may make decisions based on information derived from objects located at varying distances in front of and to the sides of the vehicle. References to monocular image analysis may represent examples of performing image analysis based on images captured from a single point of view (e.g., from a single camera). Stereoscopic image analysis may represent an example of performing image analysis based on two or more images captured with one or more changes in image capture parameters. For example, captured images suitable for performing stereoscopic image analysis may include images captured by: from two or more different positions, from different fields of view, using different focal lengths, along with parallax information, etc.

For example, in one embodiment, the system 100 may implement a three-camera configuration using image capture devices 122, 124, and 126. In such a configuration, image capture device 122 may provide a narrow field of view (e.g., 34 degrees, or other values selected from a range of about 20 to 45 degrees, etc.), image capture device 124 may provide a wide field of view (e.g., 150 degrees, or other values selected from a range of about 100 to about 180 degrees), and image capture device 126 may provide an intermediate field of view (e.g., 46 degrees, or other values selected from a range of about 35 to about 60 degrees). In some embodiments, the image capture device 126 may act as a primary or primary camera. Image capture devices 122, 124, and 126 may be positioned behind rear view mirror 310 and positioned substantially side-by-side (e.g., 6cm apart). Further, in some embodiments, as described above, one or more of the image capture devices 122, 124, and 126 may be mounted behind a light shield 380, the light shield 380 being flush with the windshield of the vehicle 200. Such shielding may act to minimize the effect of any reflections from the interior of the vehicle on the image capture devices 122, 124, and 126.

In another embodiment, as described above in connection with fig. 3B and 3C, a wide field-of-view camera (e.g., image capture device 124 in the above example) may be mounted lower than a narrow and main field-of-view camera (e.g., image devices 122 and 126 in the above example). This configuration may provide a free line of sight from a wide field of view camera. To reduce reflection, a camera may be mounted near the windshield of the vehicle 200 and may include a polarizer on the camera to dampen reflected light.

A three camera system may provide certain performance characteristics. For example, some embodiments may include verifying the ability of one camera to detect an object based on detection results from another camera. In the three-camera configuration described above, the processing unit 110 may include, for example, three processing devices (e.g., three eye q-series processor chips as described above), each dedicated to processing images captured by one or more of the image capturing devices 122, 124, and 126.

In a three-camera system, the first processing device may receive images from the main camera and the narrow field-of-view camera and perform visual processing of the narrow FOV camera, for example to detect other vehicles, pedestrians, lane markers, traffic signs, traffic lights, and other road objects. Further, the first processing device may calculate differences in pixels between images from the main camera and the narrow camera and create a 3D reconstruction of the environment of the vehicle 200. The first processing means may then combine the 3D reconstruction with the 3D map data or with 3D information calculated based on information from another camera.

The second processing device may receive the image from the primary camera and perform visual processing to detect other vehicles, pedestrians, lane markers, traffic signs, traffic lights, and other road objects. In addition, the second processing means may calculate the camera displacement and, based on the displacement, the difference of pixels between successive images and create a 3D reconstruction of the scene (e.g. from moving structures). The second processing device may send the structure from the motion-based 3D reconstruction to the first processing device for combination with the stereoscopic 3D image.

The third processing device may receive the image from the wide FOV camera and process the image to detect vehicles, pedestrians, lane markers, traffic signs, traffic lights, and other road objects. The third processing device may further execute additional processing instructions to analyze the image to identify objects moving in the image, such as vehicles changing lanes, pedestrians, etc.

In some embodiments, having the stream of image-based information captured and processed separately may provide opportunities for redundancy in the system. Such redundancy may include, for example, using a first image capture device and images processed from that device to verify and/or supplement information obtained by capturing and processing image information from at least a second image capture device.

In some embodiments, the system 100 may use two image capture devices (e.g., image capture devices 122 and 124) in providing navigation assistance for the vehicle 200, and a third image capture device (e.g., image capture device 126) to provide redundancy and verify analysis of data received from the other two image capture devices. For example, in such a configuration, image capture devices 122 and 124 may provide images for stereoscopic analysis by system 100 for navigating vehicle 200, while image capture device 126 may provide images for monocular analysis by system 100 to provide redundancy and verification based on information obtained from images captured by image capture device 122 and/or image capture device 124. That is, the image capture device 126 (and corresponding processing device) may be considered to provide a redundant subsystem for providing an inspection of the analysis derived from the image capture devices 122 and 124 (e.g., providing an Automatic Emergency Braking (AEB) system). Further, in some embodiments, redundancy and verification of the received data may be supplemented based on information received from one or more sensors (e.g., radar, lidar, acoustic sensors, information received from one or more transceivers external to the vehicle, etc.).

Those skilled in the art will appreciate that the above-described camera configurations, camera placement, number of cameras, camera positions, etc. are examples only. These components, as well as other components described with respect to the overall system, may be assembled and used in a variety of different configurations without departing from the scope of the disclosed embodiments. Additional details regarding the use of the multiple camera system to provide driver assistance and/or autonomous vehicle functionality are provided below.

FIG. 4 is an exemplary functional block diagram of memory 140 and/or 150 that may be stored/programmed with instructions for performing one or more operations in accordance with the disclosed embodiments. Although reference is made below to memory 140, one skilled in the art will appreciate that instructions may be stored in memory 140 and/or 150.

As shown in fig. 4, the memory 140 may store a monocular image analysis module 402, a stereoscopic image analysis module 404, a velocity and acceleration module 406, and a navigation response module 408. The disclosed embodiments are not limited to any particular configuration of memory 140. Further, application processor 180 and/or image processor 190 may execute instructions stored in any of modules 402, 404, 406, and 408 contained in memory 140. Those skilled in the art will appreciate that references to processing unit 110 in the following discussion may refer to application processor 180 and image processor 190 individually or collectively. Accordingly, the steps of any of the following processes may be performed by one or more processing devices.

In one embodiment, the monocular image analysis module 402 may store instructions (e.g., computer vision software) that, when executed by the processing unit 110, perform monocular image analysis of a set of images acquired by one of the image capturing devices 122, 124, and 126. In some embodiments, processing unit 110 may combine information from the image collection with additional sensed information (e.g., information from radar, lidar, etc.) to perform monocular image analysis. As described below in connection with fig. 5A-5D, monocular image analysis module 402 may include instructions for detecting a feature set within an image set (e.g., lane markers, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, dangerous objects, and any other features associated with the environment of a vehicle). Based on this analysis, the system 100 (e.g., via the processing unit 110) may cause one or more navigational responses in the vehicle 200, such as turns, lane changes, changes in acceleration, etc., as described below in connection with the navigational response module 408.

In one embodiment, the stereoscopic image analysis module 404 may store instructions (e.g., computer vision software) that, when executed by the processing unit 110, perform stereoscopic image analysis of the first and second sets of images acquired from a combination of image capture devices selected from any of the image capture devices 122, 124, and 126. In some embodiments, the processing unit 110 may combine information from the first and second sets of images with additional sensed information (e.g., information from radar) to perform stereoscopic image analysis. For example, the stereoscopic image analysis module 404 may include instructions for performing stereoscopic image analysis based on a first set of images acquired by the image capture device 124 and a second set of images acquired by the image capture device 126. As described below in connection with fig. 6, the stereoscopic image analysis module 404 may include instructions for detecting feature sets (e.g., lane markers, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, dangerous objects, etc.) within the first and second sets of images. Based on the analysis, the processing unit 110 may cause one or more navigational responses in the vehicle 200, such as turns, lane changes, changes in acceleration, etc., as described below in connection with the navigational response module 408. Further, in some embodiments, the stereoscopic image analysis module 404 may implement techniques associated with trained systems (e.g., neural networks or deep neural networks) or untrained systems (e.g., systems that may be configured to detect and/or tag objects in an environment from which sensed information is captured and processed using computer vision algorithms). In one embodiment, the stereoscopic image analysis module 404 and/or other image processing module may be configured to use a combination of trained and untrained systems.

In one embodiment, the speed and acceleration module 406 may store software configured to analyze data received from one or more computing and electromechanical devices in the vehicle 200 configured to cause changes in the speed and/or acceleration of the vehicle 200. For example, the processing unit 110 may execute instructions associated with the speed and acceleration module 406 to calculate a target speed of the vehicle 200 based on data derived from execution of the monocular image analysis module 402 and/or the stereoscopic image analysis module 404. Such data may include, for example, target position, speed, and/or acceleration, position and/or speed of the vehicle 200 relative to nearby vehicles, pedestrians, or road objects, position information of the vehicle 200 relative to lane markings of a road, and the like. In addition, the processing unit 110 may calculate the target speed of the vehicle 200 based on sensed inputs (e.g., information from radar) as well as inputs from other systems of the vehicle 200 (e.g., the throttle system 220, the brake system 230, and/or the steering system 240 of the vehicle 200). Based on the calculated target speed, the processing unit 110 may transmit electronic signals to the throttle system 220, the brake system 230, and/or the steering system 240 of the vehicle 200 to trigger a change in speed and/or acceleration by, for example, physically depressing a brake or releasing an accelerator of the vehicle 200.

In one embodiment, the navigation response module 408 may store software executable by the processing unit 110 to determine an expected navigation response based on data derived from execution of the monocular image analysis module 402 and/or the stereoscopic image analysis module 404. Such data may include location and speed information associated with nearby vehicles, pedestrians, and road objects, target location information for the vehicle 200, and the like. Additionally, in some embodiments, the navigation response may be based (in part or in whole) on map data, a predetermined location of the vehicle 200, and/or a relative velocity or relative acceleration between the vehicle 200 and one or more objects detected from execution of the monocular image analysis module 402 and/or the stereoscopic image analysis module 404. The navigation response module 408 may also determine an expected navigation response based on sensed inputs (e.g., information from radar) as well as inputs from other systems of the vehicle 200 (e.g., the throttle system 220, the brake system 230, and the steering system 240 of the vehicle 200). Based on the desired navigational response, the processing unit 110 may transmit electronic signals to the throttle system 220, the brake system 230, and the steering system 240 of the vehicle 200 to trigger the desired navigational response by, for example, turning the steering wheel of the vehicle 200 to achieve a predetermined angle of rotation. In some embodiments, the processing unit 110 may use the output of the navigation response module 408 (e.g., the expected navigation response) as an input to the execution of the speed and acceleration module 406 for calculating the change in speed of the vehicle 200.

Further, any of the modules disclosed herein (e.g., modules 402, 404, and 406) may implement techniques associated with a trained system (e.g., neural network or deep neural network) or an untrained system.

FIG. 5A is a flowchart illustrating an exemplary process 500A for eliciting one or more navigational responses based on monocular image analysis, in accordance with the disclosed embodiments. At step 510, the processing unit 110 may receive a plurality of images via the data interface 128 between the processing unit 110 and the image acquisition unit 120. For example, a camera contained in the image acquisition unit 120 (e.g., the image capture device 122 having the field of view 202) may capture multiple images of an area in front of the vehicle 200 (or e.g., a side or rear of the vehicle) and communicate them to the processing unit 110 via a data connection (e.g., digital, wired, USB, wireless, bluetooth, etc.). The processing unit 110 may execute the monocular image analysis module 402 at step 520 to analyze the plurality of images, as described in more detail below in connection with fig. 5B-5D. By performing the analysis, the processing unit 110 may detect feature sets within the image set, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, etc.

The processing unit 110 may also execute the monocular image analysis module 402 at step 520 to detect various roadway hazards, such as components of truck tires, dropped roadway markers, loose cargo, small animals, and the like. Road hazards may vary in structure, shape, size, and color, which can make detection of such hazards more tricky. In some embodiments, the processing unit 110 may execute the monocular image analysis module 402 to perform a multi-frame analysis on the plurality of images to detect roadway hazards. For example, the processing unit 110 may estimate camera motion between successive image frames and calculate differences in pixels between the frames to construct a 3D map of the road. The processing unit 110 may then use the 3D map to detect the road surface and the hazards present above the road surface.

At step 530, the processing unit 110 may execute the navigation response module 408 to cause one or more navigation responses in the vehicle 200 based on the analysis performed at step 520 and the techniques described above in connection with fig. 4. The navigational response may include, for example, turns, lane changes, changes in acceleration, etc. In some embodiments, the processing unit 110 may use data derived from the execution of the velocity and acceleration module 406 to elicit one or more navigational responses. Additionally, multiple navigational responses may occur simultaneously, sequentially, or in any combination thereof. For example, the processing unit 110 may cause the vehicle 200 to change lanes and then accelerate by, for example, sequentially transmitting control signals to the steering system 240 and the throttle system 220 of the vehicle 200. Alternatively, the processing unit 110 may brake the vehicle 200 while changing lanes by, for example, simultaneously transmitting control signals to the braking system 230 and the steering system 240 of the vehicle 200.

FIG. 5B is a flowchart illustrating an exemplary process 500B of detecting one or more vehicles and/or pedestrians in a collection of images, in accordance with the disclosed embodiments. The processing unit 110 may execute the monocular image analysis module 402 to implement the process 500B. At step 540, the processing unit 110 may determine a set of candidate objects representing possible vehicles and/or pedestrians. For example, the processing unit 110 may scan one or more images, compare the images to one or more predetermined patterns, and identify possible locations within each image that may contain an object of interest (e.g., a vehicle, a pedestrian, or a portion thereof). The predetermined pattern may be designed in such a way that a high "miss" rate and a low "miss" rate are achieved. For example, the processing unit 110 may identify the candidate object as a possible vehicle or pedestrian using a low threshold of similarity to the predetermined pattern. Doing so may allow the processing unit 110 to reduce the probability of a miss (e.g., unidentified) representing a candidate for a vehicle or pedestrian.

At step 542, processing unit 110 may filter the set of candidate objects based on the classification criteria to exclude certain candidates (e.g., objects that are irrelevant or less relevant). Such criteria may be derived from various properties associated with object types stored in a database (e.g., a database stored in memory 140). Properties may include object shape, size, texture, location (e.g., relative to vehicle 200), and the like. Thus, the processing unit 110 may use one or more sets of criteria to exclude pseudo-candidates from the candidate set.

At step 544, the processing unit 110 may analyze the plurality of frames of the image to determine whether an object in the candidate object set represents a vehicle and/or a pedestrian. For example, the processing unit 110 may track the detected candidate object across successive frames and accumulate frame-by-frame data (e.g., size, position relative to the vehicle 200, etc.) associated with the detected object. In addition, the processing unit 110 may estimate parameters of the detected object and compare frame-by-frame position data of the object with the predicted position.

At step 546, the processing unit 110 may construct a measurement set of the detected objects. Such measurements may include, for example, position, velocity, and acceleration values associated with the detected object (relative to the vehicle 200). In some embodiments, the processing unit 110 may construct the measurements based on estimation techniques using a series of time-based observations, such as Kalman (Kalman) filters or Linear Quadratic Estimations (LQE), and/or based on available modeling data for different object types, such as automobiles, trucks, pedestrians, bicycles, road signs, etc. The kalman filter may be based on a measurement of the scale of the object, where the scale measurement is proportional to the collision time (e.g., the amount of time the vehicle 200 reaches the object). Thus, by performing steps 540-546, the processing unit 110 may identify vehicles and pedestrians that appear within the captured image set, and derive information (e.g., location, speed, size) associated with the vehicles and pedestrians. Based on the identification and the resulting information, the processing unit 110 may cause one or more navigational responses in the vehicle 200, as described above in connection with fig. 5A.

In step 548, the processing unit 110 may perform optical flow analysis of the one or more images to reduce the probability of detecting "false hits" and misses representing candidates for vehicles or pedestrians. Optical flow analysis may represent, for example, analysis of a pattern of motion relative to vehicle 200 in one or more images associated with other vehicles and pedestrians, the pattern of motion being different from road surface motion. The processing unit 110 may calculate the motion of the candidate object by observing different positions of the object across multiple image frames captured at different times. The processing unit 110 may use the position and time values as inputs to a mathematical model for calculating the motion of the candidate object. Accordingly, optical flow analysis may provide another method of detecting vehicles and pedestrians in the vicinity of the vehicle 200. Processing unit 110 may perform optical flow analysis in conjunction with steps 540-546 to provide redundancy in detecting vehicles and pedestrians and to increase the reliability of system 100.

FIG. 5C is a flowchart illustrating an exemplary process 500C of detecting road markings and/or lane geometry information in a set of images, in accordance with the disclosed embodiments. The processing unit 110 may execute the monocular image analysis module 402 to implement the process 500C. At step 550, the processing unit 110 may detect the set of objects by scanning one or more images. To detect lane marker segments, lane geometry information, and other relevant road markers, the processing unit 110 may filter the set of objects to exclude those objects (e.g., small pits, small rocks, etc.) that are determined to be irrelevant. In step 552, the processing unit 110 may group together segments belonging to the same road or lane marking detected in step 550. Based on the groupings, the processing unit 110 may develop a model, such as a mathematical model, that represents the detected segments.

In step 554, the processing unit 110 may construct a set of measurements associated with the detected segment. In some embodiments, the processing unit 110 may create a projection of the detected segment from the image plane to the real world plane. The projection may be characterized using a cubic polynomial having coefficients corresponding to physical properties such as the position, grade, curvature, and curvature derivative of the detected road. In generating the projections, the processing unit 110 may take into account the variations in the road surface as well as the tilt and roll rates associated with the vehicle 200. In addition, the processing unit 110 may model the road elevation by analyzing the locations and motion cues present on the road surface. Further, the processing unit 110 may estimate the tilt and scroll rate associated with the vehicle 200 by tracking a set of feature points in one or more images.

At step 556, processing unit 110 may perform multi-frame analysis by, for example, tracking the detected segments across successive image frames and accumulating frame-by-frame data associated with the detected segments. When the processing unit 110 performs multi-frame analysis, the measurement set constructed at step 554 may become more reliable and associated with higher and higher confidence levels. Thus, by performing steps 550, 552, 554 and 556, the processing unit 110 may identify the road markings appearing within the captured image set and derive lane geometry information. Based on the identification and the resulting information, the processing unit 110 may cause one or more navigational responses in the vehicle 200, as described above in connection with fig. 5A.

In step 558, the processing unit 110 may consider additional information sources to further develop a security model of the vehicle 200 in the context of the surrounding environment. The processing unit 110 may use a security model to define a context in which the system 100 may perform autonomous control of the vehicle 200 in a secure manner. To develop a security model, in some embodiments, the processing unit 110 may consider the locations and movements of other vehicles, detected road edges and barriers, and/or general road shape descriptions extracted from map data (e.g., data from the map database 160). By taking additional information sources into account, the processing unit 110 may provide redundancy in detecting road markings and lane geometries and increase the reliability of the system 100.

Fig. 5D is a flowchart illustrating an exemplary process 500D of detecting traffic lights in a collection of images, in accordance with the disclosed embodiments. The processing unit 110 may execute the monocular image analysis module 402 to implement the process 500D. At step 560, the processing unit 110 may scan the set of images and identify objects that appear in the images that may contain locations of traffic lights. For example, the processing unit 110 may filter the identified objects to construct a set of candidate objects, thereby excluding those objects that are unlikely to correspond to traffic lights. The filtering may be based on various properties associated with the traffic light, such as shape, size, texture, location (e.g., relative to the vehicle 200), and so forth. Such properties may be based on multiple examples of traffic lights and traffic control signals and stored in a database. In some embodiments, the processing unit 110 may perform a multi-frame analysis on a set of candidate objects reflecting possible traffic lights. For example, the processing unit 110 may track candidate objects across successive image frames, estimate the real world locations of the candidate objects, and filter out those objects that are moving (not likely to be traffic lights). In some embodiments, the processing unit 110 may perform color analysis on the candidate object and identify the relative locations of detected colors that may occur inside the traffic light.

In step 562, the processing unit 110 may analyze the geometry of the intersection. The analysis may be based on any combination of the following: (i) the number of lanes detected on either side of the vehicle 200, (ii) markers detected on the road (e.g., arrow markers), and (iii) descriptions of intersections extracted from map data (e.g., data from the map database 160). The processing unit 110 may use information derived from the execution of the monocular analysis module 402 for analysis. In addition, the processing unit 110 may determine a correspondence between the traffic light detected at step 560 and a lane that appears near the vehicle 200.

As the vehicle 200 approaches the intersection, the processing unit 110 may update the confidence level associated with the analyzed intersection geometry and the detected traffic lights at step 564. For example, the estimated number of intersections may affect the confidence level as compared to the number of traffic lights actually present at the intersection. Thus, based on the confidence level, the processing unit 110 may delegate control to the driver of the vehicle 200 in order to improve the safety conditions. By performing steps 560, 562, and 564, the processing unit 110 may identify traffic lights that appear within the captured image set and analyze intersection geometry information. Based on the identification and analysis, the processing unit 110 may cause one or more navigational responses in the vehicle 200, as described above in connection with fig. 5A.

FIG. 5E is a flowchart illustrating an exemplary process 500E for eliciting one or more navigational responses in the vehicle 200 based on the vehicle path, in accordance with the disclosed embodiments. At step 570, the processing unit 110 may construct an initial vehicle path associated with the vehicle 200. The vehicle path may be represented using a set of points expressed in terms of coordinates (x, z), and the distance d between two points in the set of points _i May fall in the range of 1 to 5 meters. In one embodiment, the processing unit 110 may use two polynomials (e.g., left and right road polynomials) to construct the initial vehicle path. The processing unit 110 may be in a computational geometry between two polynomialsThe points, and if present, each point included in the generated vehicle path is offset by a predetermined offset (e.g., a smart lane offset) (a zero offset may correspond to travel in the middle of a lane). The offset may be in a direction perpendicular to the segment between any two points in the vehicle path. In another embodiment, the processing unit 110 may use one polynomial and the estimated lane width to offset each point in the vehicle path by half the estimated lane width plus a predetermined offset (e.g., a smart lane offset).

In step 572, the processing unit 110 may update the vehicle path constructed in step 570. The processing unit 110 may reconstruct the vehicle path constructed at step 570 using a higher resolution such that the distance d between two points in the set of points representing the vehicle path _k Less than the distance d _i . For example, distance d _k May fall in the range of 0.1 to 0.3 meters. The processing unit 110 may reconstruct the vehicle path using a parabolic spline algorithm that may generate a cumulative distance vector S corresponding to the total length of the vehicle path (i.e., based on the set of points representing the vehicle path).

At step 574, the processing unit 110 may determine a front viewpoint (expressed in terms of coordinates as (x) _l ,z _l )). The processing unit 110 may extract a front view from the accumulated distance vector S, and the front view may be associated with a front view distance and a front view time. The forward looking distance may have a lower limit ranging from 10 to 20 meters, which may be calculated as the product of the speed of the vehicle 200 and the forward looking time. For example, as the speed of the vehicle 200 decreases, the forward looking distance may also decrease (e.g., until it reaches a lower limit). The forward looking time may range from 0.5 to 1.5 seconds and may be inversely proportional to the gain associated with one or more control loops (e.g., heading error tracking control loops) that cause a navigational response in the vehicle 200. For example, the gain of the heading error tracking control loop may depend on the bandwidth of the yaw rate loop, the steering actuator loop, the lateral dynamics of the car, etc. Thus, the higher the gain of the heading error tracking control loop, the shorter the forward time.

In step 576, the processing unit 110 may determine a heading error and a yaw-rate command based on the front viewpoint determined in step 574. The processing unit 110 may calculate the inverse tangent of the front view (e.g., arctan (x _l /z _l ) A heading error is determined. The processing unit 110 may determine the yaw rate command as a product of the heading error and the advanced control gain. The advanced control gain may be equal to: (2/forward looking time) if the forward looking distance is not at the lower limit. Otherwise, the advanced control gain may be equal to: (2 x speed/forward looking distance of vehicle 200).

FIG. 5F is a flowchart illustrating an exemplary process 500F of determining whether a forward-moving vehicle is changing lanes, in accordance with the disclosed embodiments. At step 580, the processing unit 110 may determine navigation information associated with the traveling vehicle (e.g., the vehicle traveling in front of the vehicle 200). For example, the processing unit 110 may determine the position, velocity (e.g., direction and speed), and/or acceleration of the forward-moving vehicle using the techniques described above in connection with fig. 5A and 5B. The processing unit 110 may also determine one or more road polynomials, forward viewpoints (associated with the vehicle 200), and/or snail trails (e.g., a set of points describing the path taken by the advancing vehicle) using the techniques described above in connection with fig. 5E.

In step 582, the processing unit 110 may analyze the navigation information determined in step 580. In one embodiment, the processing unit 110 may calculate the distance between the snail trail and the road polynomial (e.g., along the trail). If this distance along the trail varies by more than a predetermined threshold (e.g., 0.1 to 0.2 meters on straight roads, 0.3 to 0.4 meters on medium curved roads, and 0.5 to 0.6 meters on roads with sharp bends), the processing unit 110 may determine that the forward moving vehicle may be changing lanes. In the event that multiple vehicles are detected traveling in front of the vehicle 200, the processing unit 110 may compare the snail trail associated with each vehicle. Based on the comparison, the processing unit 110 may determine that a vehicle whose snail trail does not match the snail trail of other vehicles may be changing lanes. The processing unit 110 may also compare the curvature of the snail trail (associated with the advancing vehicle) to the expected curvature of the road segment the advancing vehicle is traveling on. The predicted curvature may be extracted from map data (e.g., data from map database 160), from road polynomials, from snail trails of other vehicles, from a priori knowledge about the road, etc. If the difference between the curvature of the snail trail and the predicted curvature of the road segment exceeds a predetermined threshold, the processing unit 110 may determine that the lead vehicle may be changing lanes.

In another embodiment, the processing unit 110 may compare the instantaneous location of the forward moving vehicle to the forward viewpoint (associated with the vehicle 200) for a particular period of time (e.g., 0.5 to 1.5 seconds). If the distance between the instantaneous position of the forward moving vehicle and the forward viewpoint changes during a particular period of time and the accumulated sum of the changes exceeds a predetermined threshold (e.g., 0.3 to 0.4 meters on straight roads, 0.7 to 0.8 meters on medium curved roads, and 1.3 to 1.7 meters on roads with sharp bends), the processing unit 110 may determine that the forward moving vehicle may be changing lanes. In another embodiment, the processing unit 110 may analyze the geometry of the snail trail by comparing the lateral distance traveled along the trail with the expected curvature of the snail trail. The predicted radius of curvature may be determined as follows: (delta) _z ² +δ _x ² )/2/(δ _x ) Wherein delta _x Represents the lateral distance travelled and delta _z Representing the longitudinal distance travelled. If the difference between the lateral distance travelled and the expected curvature exceeds a predetermined threshold (e.g., 500 to 700 meters), the processing unit 110 may determine that the lead vehicle may be changing lanes. In another embodiment, the processing unit 110 may analyze the location of the forward moving vehicle. If the location of the lead vehicle obscures the road polynomial (e.g., the lead vehicle is overlaid on top of the road polynomial), the processing unit 110 may determine that the lead vehicle may be changing lanes. In the event that the position of the forward moving vehicle is such that the other vehicle is detected in front of the forward moving vehicle and the snail trails of the two vehicles are not parallel, the processing unit 110 may determine that the (closer) forward moving vehicle may be changing lanes.

In step 584, the processing unit 110 may determine whether the lead vehicle 200 is changing lanes based on the analysis performed in step 582. For example, the processing unit 110 may make the determination based on a weighted average of the individual analyses performed at step 582. Under such a scenario, for example, a determination by the processing unit 110 that a preceding vehicle may be changing lanes based on a particular type of analysis may be assigned a value of "1" (while "0" indicates that a lane change is not possible with respect to the preceding vehicle). The different analyses performed at step 582 may be assigned different weights, and the disclosed embodiments are not limited to any particular combination of analyses and weights.

FIG. 6 is a flow chart illustrating an exemplary process 600 for eliciting one or more navigational responses based on stereoscopic image analysis in accordance with the disclosed embodiments. At step 610, the processing unit 110 may receive the first and second plurality of images via the data interface 128. For example, cameras (e.g., image capture devices 122 and 124 having fields of view 202 and 204) included in the image acquisition unit 120 may capture first and second pluralities of images of an area in front of the vehicle 200 and communicate them to the processing unit 110 via a digital connection (e.g., USB, wireless, bluetooth, etc.). In some embodiments, the processing unit 110 may receive the first and second pluralities of images via two or more data interfaces. The disclosed embodiments are not limited to any particular data interface configuration or protocol.

In step 620, the processing unit 110 may execute the stereoscopic image analysis module 404 to perform stereoscopic image analysis of the first and second plurality of images to create a 3D map of the road in front of the vehicle and detect features within the image, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, road hazards, and the like. Stereoscopic image analysis may be performed in a similar manner to the steps described above in connection with fig. 5A-5D. For example, the processing unit 110 may execute the stereo image analysis module 404 to detect candidate objects (e.g., vehicles, pedestrians, road signs, traffic lights, road hazards, etc.) within the first and second plurality of images, filter out a subset of the candidate objects based on various criteria, and perform multi-frame analysis, construct a measurement, and determine a confidence level for the remaining candidate objects. In performing the above steps, the processing unit 110 may consider information from the first and second plurality of images rather than information from one set of images alone. For example, the processing unit 110 may analyze differences in pixel-level data (or other data subsets from among two captured image streams) of candidate objects appearing in the first and second plurality of images. As another example, the processing unit 110 may estimate the location and/or velocity of the candidate object (e.g., relative to the vehicle 200) by observing that the object appears in one image of the plurality of images instead of another image, or relative to other discrepancies (which may exist relative to objects appearing in both image streams). For example, the position, velocity, and/or acceleration relative to the vehicle 200 may be determined based on the trajectory, position, movement characteristics, etc. of features associated with objects appearing in one or both of the image streams.

At step 630, the processing unit 110 may execute the navigation response module 408 to cause one or more navigation responses in the vehicle 200 based on the analysis performed at step 620 and the techniques described above in connection with fig. 4. The navigational response may include, for example, turns, lane changes, changes in acceleration, changes in speed, braking, etc. In some embodiments, the processing unit 110 may use data derived from the execution of the velocity and acceleration module 406 to elicit one or more navigational responses. Additionally, multiple navigational responses may occur simultaneously, sequentially, or in any combination thereof.

FIG. 7 is a flowchart illustrating an exemplary process 700 for eliciting one or more navigational responses based on analysis of three sets of images, in accordance with the disclosed embodiment. At step 710, the processing unit 110 may receive the first, second, and third pluralities of images via the data interface 128. For example, cameras (e.g., image capture devices 122, 124, and 126 having fields of view 202, 204, and 206) included in the image acquisition unit 120 may capture first, second, and third pluralities of images of areas in front of and/or to the sides of the vehicle 200 and communicate them to the processing unit 110 via a digital connection (e.g., USB, wireless, bluetooth, etc.). In some embodiments, the processing unit 110 may receive the first, second, and third pluralities of images via three or more data interfaces. For example, each of the image capture devices 122, 124, 126 may have an associated data interface for communicating data to the processing unit 110. The disclosed embodiments are not limited to any particular data interface configuration or protocol.

At step 720, the processing unit 110 may analyze the first, second, and third plurality of images to detect features within the images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, road hazards, and the like. The analysis may be performed in a similar manner to the steps described above in connection with fig. 5A-5D and fig. 6. For example, the processing unit 110 may perform monocular image analysis (e.g., via execution of the monocular image analysis module 402 and based on the steps described above in connection with fig. 5A-5D) on each of the first, second, and third pluralities of images. Alternatively, the processing unit 110 may perform stereoscopic image analysis on the first and second plurality of images, the second and third plurality of images, and/or the first and third plurality of images (e.g., via execution of the stereoscopic image analysis module 404 and based on the steps described above in connection with fig. 6). The processed information corresponding to the analysis of the first, second and/or third plurality of images may be combined. In some embodiments, the processing unit 110 may perform a combination of monocular and stereoscopic image analysis. For example, the processing unit 110 may perform monocular image analysis on the first plurality of images (e.g., via execution of the monocular image analysis module 402) and stereoscopic image analysis on the second and third plurality of images (e.g., via execution of the stereoscopic image analysis module 404). The configuration of the image capture devices 122, 124, and 126, including their respective locations and fields of view 202, 204, and 206, may affect the type of analysis performed on the first, second, and third pluralities of images. The disclosed embodiments are not limited to a particular configuration of image capture devices 122, 124, and 126 or to the type of analysis performed on the first, second, and third pluralities of images.

In some embodiments, processing unit 110 may perform a test on system 100 based on the images acquired and analyzed at steps 710 and 720. Such testing may provide an indicator of the overall performance of the system 100 for certain configurations of the image capture devices 122, 124, and 126. For example, the processing unit 110 may determine the proportion of "misses" (e.g., where the system 100 incorrectly determines the presence of a vehicle or pedestrian) and "misses.

At step 730, the processing unit 110 may cause one or more navigational responses in the vehicle 200 based on information derived from two of the first, second, and third pluralities of images. The selection of two of the first, second, and third pluralities of images may depend on various factors, such as the number, type, and size of objects detected in each of the plurality of images. The processing unit 110 may also select based on image quality and resolution, the effective field of view reflected in the image, the number of frames captured, the degree to which one or more objects of interest are actually present in the frames (e.g., the percentage of frames in which the objects are present, the proportion of objects present in each such frame, etc.), and so forth.

In some embodiments, the processing unit 110 may select information derived from two of the first, second, and third pluralities of images by determining a degree to which information derived from one image source is consistent with information derived from the other image source. For example, the processing unit 110 may combine the processed information derived from each of the image capture devices 122, 124, and 126 (whether by monocular analysis, stereo analysis, or any combination of the two) and determine that the captured images across each of the image capture devices 122, 124, and 126 are consistent visual indicators (e.g., lane markings, detected vehicles and their locations and/or paths, detected traffic lights, etc.). The processing unit 110 may also exclude information that is inconsistent across the captured images (e.g., vehicle changing lanes, lane models indicating vehicles too close to the vehicle 200, etc.). Thus, the processing unit 110 may select information derived from two of the first, second, and third pluralities of images based on the determination of the consistent and inconsistent information.

The navigational response may include, for example, turns, lane changes, changes in acceleration, etc. The processing unit 110 may cause one or more navigational responses based on the analysis performed at step 720 and the techniques as described above in connection with fig. 4. The processing unit 110 may also use data derived from the execution of the velocity and acceleration module 406 to elicit one or more navigational responses. In some embodiments, the processing unit 110 may cause one or more navigational responses based on the relative position, relative velocity, and/or relative acceleration between the vehicle 200 and the detected object within any of the first, second, and third pluralities of images. Multiple navigational responses may occur simultaneously, sequentially, or in any combination thereof.

Sparse road model for autonomous vehicle navigation

In some embodiments, the disclosed systems and methods may use sparse maps for autonomous vehicle navigation. In particular, sparse maps may be used for autonomous vehicle navigation along road segments. For example, a sparse map may provide sufficient information for navigating an autonomous vehicle without storing and/or updating large amounts of data. As discussed in more detail below, the autonomous vehicle may use the sparse map to navigate one or more roads based on one or more stored trajectories.

Sparse map for autonomous vehicle navigation

In some embodiments, the disclosed systems and methods may generate a sparse map of autonomous vehicle navigation. For example, a sparse map may provide sufficient information for navigation without requiring excessive data storage or data transmission rates. As described in more detail below, a vehicle (which may be an autonomous vehicle) may use a sparse map to navigate one or more roads. For example, in some embodiments, the sparse map may include data related to roads and potentially roadways along the roads that may be sufficient for vehicle navigation, but that also presents a small data footprint. For example, sparse data maps, described in detail below, may require significantly less storage space and data transmission bandwidth as compared to digital maps that include detailed map information (e.g., image data collected along roads).

For example, rather than storing a detailed representation of the road segments, the sparse data map may store a three-dimensional polynomial representation of the preferred vehicle path along the road. These paths may require little data storage space. Additionally, in the sparse data map, roadways may be identified and included in the sparse map road model to aid navigation. These landmarks may be positioned at any spacing suitable for achieving vehicle navigation, but in some cases, such landmarks need not be identified at high density and short spacing and contained in a model. Instead, in some cases, navigation may be possible based on roadsigns that are spaced apart by at least 50 meters, at least 100 meters, at least 500 meters, at least 1 kilometer, or at least 2 kilometers. As will be discussed in more detail elsewhere, the sparse map may be generated based on data collected or measured by vehicles equipped with various sensors and devices (e.g., image capture devices, global positioning system sensors, motion sensors, etc.) while traveling along a road. In some cases, the sparse map may be generated based on data collected during multiple passes of one or more vehicles along a particular road. Generating a sparse map using multiple passes of one or more vehicles may be referred to as "crowd-sourced" sparse maps.

In accordance with the disclosed embodiments, an autonomous vehicle system may use a sparse map for navigation. For example, the disclosed systems and methods may distribute sparse maps for generating road navigation models of autonomous vehicles, and may navigate autonomous vehicles along a road segment using the sparse maps and/or the generated road navigation models. A sparse map in accordance with the present disclosure may include one or more three-dimensional contours that may represent predetermined trajectories that autonomous vehicles may traverse while moving along an associated road segment.

Sparse maps in accordance with the present disclosure may also include data representing one or more road features. Such road features may include identified road signs, road signature profiles, and any other road-related features useful in navigating a vehicle. Sparse maps in accordance with the present disclosure may enable autonomous navigation of a vehicle based on a small amount of data contained in the sparse map. For example, rather than including detailed representations of roads (e.g., road edges, road curvature, images associated with road segments, or data detailing other physical characteristics associated with road segments), the disclosed embodiments of sparse maps may require relatively little storage space (and relatively little bandwidth when portions of the sparse map are transmitted to the vehicle), but may still adequately provide autonomous vehicle navigation. The small data footprint of the disclosed sparse map, discussed in more detail below, may be implemented in some embodiments by storing representations of road-related elements that require a small amount of data, but still enable autonomous navigation.

For example, rather than storing detailed representations of various aspects of the road, the disclosed sparse map may store a polynomial representation of one or more trajectories along which the vehicle may progress. Thus, rather than storing (or having to transmit) details regarding the physical properties of the road to enable navigation along the road, using the disclosed sparse map, the vehicle may instead be navigated along a particular road segment without needing to interpret the physical aspects of the road in some cases, but rather by aligning its travel path with a trajectory (e.g., a polynomial spline) along the particular road segment. In this way, the vehicle may navigate based primarily on stored trajectories (e.g., polynomial splines), which may require significantly less storage space than the manner in which storage of road images, road parameters, road layout, etc., is involved.

In addition to the stored polynomial representation of the trajectory along the road segment, the disclosed sparse map may also include small data objects that may represent road features. In some embodiments, the small data object may include a digital signature derived from a digital image (or digital signal) obtained by a sensor (e.g., a camera or another sensor, such as a suspension sensor) on a vehicle traveling along the roadway section. The digital signature may have a reduced size relative to the signal acquired by the sensor. In some embodiments, the digital signature may be created to be compatible with a classifier function configured to detect and identify road features, for example, from signals acquired by sensors during subsequent driving. In some embodiments, the digital signature may be created such that the digital signature has as little footprint as possible while maintaining the ability to correlate or match road features with stored signatures based on images of road features captured at a later time by cameras on vehicles traveling along the same road segment (or digital signals generated by sensors if the stored signature is not based on images and/or includes other data).

In some embodiments, the size of the data object may be further associated with the uniqueness of the road feature. For example, for road features detectable by a camera on the vehicle, and where the camera system on the vehicle is coupled to a classifier that is capable of distinguishing image data corresponding to that road feature as being associated with a particular type of road feature (e.g., a road sign), and where such road sign is locally unique in that area (e.g., the same road sign or the same type of road sign is not present nearby), it may be sufficient to store data indicative of the type of road feature and its location.

As will be discussed in more detail below, road features (e.g., roadmarks along a road segment) may be stored as small data objects that may represent road features in fewer bytes while providing sufficient information to identify and use such features for navigation. In one example, the road sign may be identified as an identified road sign that underlies navigation of the vehicle. The representation of the road sign may be stored in a sparse map to include, for example, several bytes of data indicating the type of the sign (e.g., stop sign) and several bytes of data indicating the location (e.g., coordinates) of the sign. Navigation based on such data-light representations of roadmap (e.g., using representations sufficient for locating, identifying, and navigating based on roadmap) may provide an expected level of navigation functionality associated with sparse maps without significantly increasing the data overhead associated with sparse maps. This lean representation (lean representation) of the road sign (and other road features) may utilize sensors and processors included on such vehicles that are configured to detect, identify, and/or classify certain road features.

When, for example, a marker or even a specific type of marker is locally unique in a given area (e.g., when another marker is not present or another marker of the same type is not present), the sparse map may use data indicative of the type of marker (marker or specific type of marker) and during navigation (e.g., autonomous navigation) when a camera on the autonomous vehicle captures an image of an area including the marker (or specific type of marker), the processor may process the image, detect the marker (if actually present in the image), classify the image as a marker (or as a specific type of marker), and correlate the location of the image with the location of the marker as stored in the sparse map.

The sparse map may include any suitable representation of objects identified along the road segment. In some cases, an object may be referred to as a semantic object or a non-semantic object. The semantic objects may include, for example, objects associated with a predetermined type of classification. Such type classification may be useful for reducing the amount of data required to describe semantic objects identified in the environment, which may be beneficial both during the acquisition phase (e.g., reducing costs associated with bandwidth for transmitting driving information from multiple acquisition vehicles to a server) and during the navigation phase (e.g., the reduction of map data may speed up the transmission of map tiles from a server to a navigation vehicle, and may also reduce costs associated with bandwidth for such transmission). Semantic object classification types may be assigned to any type of object or feature that is desired to be encountered along a roadway.

Semantic objects may also be divided into two or more logical groups. For example, in some cases, a set of semantic object types may be associated with a predetermined dimension. Such semantic objects may include certain speed limit signs, let-down signs, merge signs, stop signs, traffic lights, directional arrows on roads, manholes covers, or any other type of object that may be associated with a standardized size. One benefit provided by such semantic objects is that very little data may be required to represent/fully define the object. For example, if the normalized size of the speed limit size is known, the collecting vehicle may only need to identify (by analysis of the captured image) the presence of the speed limit sign (identified type) along with an indication of the location of the detected speed limit sign (e.g., the 2D location of the center of the sign or some angle of the sign in the captured image (or, alternatively, the 3D location in real world coordinates)) to provide sufficient information for map generation on the server side. In the case of transmitting the 2D image location to the server, the location associated with the captured image of the detected sign may also be transmitted so that the server may determine the real world location of the sign (e.g., by using structure in motion technology from multiple captured images of one or more acquisition vehicles). Even with such limited information (only a few bytes are needed to define each detected object), the server can construct a map comprising the fully represented speed limit signs based on the type classification (representing speed limit signs) received from the one or more collecting vehicles along with the location information of the detected signs.

Semantic objects may also include other identified object or feature types that are not associated with certain standardized characteristics. Such objects or features may include pits, asphalt joints, lamp posts, non-standardized signs, curbs, trees, branches, or any other type of identified object type having one or more variable characteristics (e.g., variable dimensions). In this case, the collecting vehicle may transmit an indication of the size of the object or feature in addition to transmitting an indication of the type of object or feature detected (e.g., pothole, pole, etc.) and location information of the object or feature detected to the server. The size may be expressed in a 2D image dimension (e.g., having a bounding box or one or more dimension values) or a real world dimension (determined by structure in motion calculations, based on LIDAR or RADAR system outputs, based on trained neural network outputs, etc.).

Non-semantic objects or features may include any detectable object or feature that falls outside of the identified category or type but may still provide valuable information in map generation. In some cases, such non-semantic features may include detected building corners or detected building window corners, unique stones or objects near the road, concrete splatters in the road shoulders, or any other detectable object or feature. Upon detection of such an object or feature, one or more acquisition vehicles may communicate to a map generation server the location of one or more points (2D image points or 3D real world points) associated with the detected object/feature. In addition, compressed or reduced image segments (e.g., image hashes) may be generated for areas of the captured image that include the detected object or feature. The image hash may be calculated based on a predetermined image processing algorithm and may form a valid signature of the detected non-semantic object or feature. Such signatures are useful for navigation relative to sparse maps that include non-semantic features or objects, as vehicles traversing roads can apply algorithms similar to those used to generate image hashes in order to confirm/verify the presence of mapped non-semantic features or objects in the captured image. Using this technique, the non-semantic features may increase the richness of the sparse map (e.g., to enhance its usefulness in navigation) without increasing significant data overhead.

As described above, the target track may be stored in a sparse map. These target trajectories (e.g., 3D splines) may represent a preferred or recommended path for each available lane of the road, each effective path through the intersection for merging and exiting, etc. In addition to the target trajectory, other road features may be detected, collected and combined in the form of representational splines in the sparse map. Such features may include, for example, road edges, lane markings, curbs, guardrails, or any other object or feature extending along a road or road segment.

Generating sparse maps

In some embodiments, the sparse map may include at least one line representation of road surface features extending along a road segment and a plurality of road signs associated with the road segment. In certain aspects, the sparse map may be generated via "crowdsourcing," for example, through image analysis of a plurality of images acquired as one or more vehicles traverse a road segment.

Fig. 8 illustrates a sparse map 800 accessible to one or more vehicles, such as vehicle 200 (which may be autonomous vehicles), for providing autonomous vehicle navigation. The sparse map 800 may be stored in a memory (e.g., memory 140 or 150). Such memory devices may include any type of non-transitory storage device or computer-readable medium. For example, in some embodiments, memory 140 or 150 may include a hard disk drive, a compact disc, a flash memory, a magnetic-based memory device, an optical-based memory device, and the like. In some embodiments, the sparse map 800 may be stored in a database (e.g., map database 160) that may be stored in memory 140 or 150 or other type of storage.

In some embodiments, the sparse map 800 may be stored on a storage device provided on the vehicle 200 or on a non-transitory computer readable medium (e.g., a storage device included in a navigation system on the vehicle 200). A processor (e.g., processing unit 110) provided on the vehicle 200 may access a sparse map 800 stored in a storage device or computer readable medium provided on the vehicle 200 to generate navigation instructions for guiding the autonomous vehicle 200 as the vehicle traverses a road segment.

But the sparse map 800 need not be stored locally with respect to the vehicle. In some embodiments, the sparse map 800 may be stored on a storage device or computer readable medium provided on a remote server in communication with the vehicle 200 or a device associated with the vehicle 200. A processor (e.g., processing unit 110) provided on the vehicle 200 may receive data contained in the sparse map 800 from a remote server and may execute the data for guiding autonomous driving of the vehicle 200. In such embodiments, the remote server may store the entire sparse map 800 or only a portion thereof. Accordingly, a storage device or computer readable medium on the vehicle 200 and/or on one or more additional vehicles may store the remaining portion(s) of the sparse map 800.

Further, in such embodiments, the sparse map 800 may be made accessible to multiple vehicles (e.g., tens, hundreds, thousands, millions, or millions of vehicles, etc.) traversing various road segments. It should also be noted that the sparse map 800 may include multiple sub-maps. For example, in some embodiments, the sparse map 800 may include hundreds, thousands, millions, or more sub-maps (e.g., map tiles) that may be used in a navigation vehicle. Such sub-maps may be referred to as local maps or map tiles, and a vehicle traveling along a road may access any number of local maps related to the location where the vehicle is traveling. The local map portion of the sparse map 800 may be stored with Global Navigation Satellite System (GNSS) keys as an index to a database of the sparse map 800. Thus, while the calculation of steering angles for navigating the host vehicle in the present system may be performed without reliance on the GNSS location, road characteristics, or roadmap of the host vehicle, such GNSS information may be used to retrieve relevant local maps.

In general, the sparse map 800 may be generated based on data (e.g., driving information) collected from one or more vehicles while traveling along a road. For example, using one or more on-vehicle sensors (e.g., cameras, speedometers, GPS, accelerometers, etc.), the trajectory of one or more vehicles traveling along the road may be recorded, and a polynomial representation of a preferred trajectory for subsequent travel of the vehicle along the road may be determined based on the collected trajectory of one or more vehicles traveling. Similarly, data collected by one or more vehicles may help identify potential roadways along a particular roadway. The data collected from passing vehicles may also be used to identify road profile information, such as road width profile, road roughness profile, traffic line spacing profile, road conditions, and the like. Using the collected information, the sparse map 800 may be generated and distributed (e.g., for local storage or via instant data transmission) for use in navigating one or more autonomous vehicles. However, in some embodiments, map generation may not end at the time of initial generation of the map. As will be discussed in more detail below, the sparse map 800 may be updated continuously or periodically based on data collected from vehicles as those vehicles continue through roads contained in the sparse map 800.

The data recorded in the sparse map 800 may include location information based on Global Positioning System (GPS) data. For example, location information may be included in the sparse map 800 for various map elements, including, for example, road sign locations, road profile locations, and the like. The locations of map elements included in the sparse map 800 may be obtained using GPS data collected from vehicles traversing the road. For example, a vehicle passing the identified roadmap may determine the location of the identified roadmap using GPS location information associated with the vehicle and a determination of the location of the identified roadmap relative to the vehicle (e.g., based on image analysis of data collected from one or more cameras on the vehicle). Such a location determination of the identified roadmap (or any other feature contained in the sparse map 800) may be repeated as additional vehicles pass the location of the identified roadmap. Some or all of the additional location determinations may be used to refine the location information stored in the sparse map 800 relative to the identified roadmap. For example, in some embodiments, multiple location measurements relative to a particular feature stored in the sparse map 800 may be averaged together. However, any other mathematical operation may also be used to refine the stored location of the map element based on the plurality of determined locations of the map element.

In a particular example, the collection vehicle may traverse a particular road segment. Each acquisition vehicle captures an image of its respective environment. The images may be collected at any suitable frame capture rate (e.g., 9Hz, etc.). The image analysis processor(s) on each acquisition vehicle analyze the captured images to detect the presence of semantic and/or non-semantic features/objects. At a high level, the collecting vehicle communicates to the mapping server detection indications of semantic and/or non-semantic objects/features and locations associated with those objects/features. In more detail, a type indicator, a size indicator, etc. may be transmitted together with the location information. The location information may include any suitable information for enabling the mapping server to aggregate the detected objects/features into a sparse map useful in navigation. In some cases, the location information may include one or more 2D image locations (e.g., X-Y pixel locations) in the captured image where the semantic or non-semantic feature/object was detected. Such image locations may correspond to the center, corners, etc. of the feature/object. In this case, to help the mapping server reconstruct the driving information and align the driving information from the plurality of collecting vehicles, each collecting vehicle may also provide the server with the location (e.g., GPS location) where each image was captured.

In other cases, the acquisition vehicle may provide one or more 3D real world points associated with the detected objects/features to a server. Such 3D points may be related to a predetermined origin (e.g., the origin of the driver's leg) and may be determined by any suitable technique. In some cases, structures in motion techniques may be used to determine the 3D real world location of the detected object/feature. For example, an object such as a particular speed limit sign may be detected in two or more captured images. Using information such as the known self-motion of the acquiring vehicle between the captured images (speed, trajectory, GPS position, etc.), along with observed changes in the speed limit markers in the captured images (changes in X-Y pixel positions, changes in size, etc.), the real world position of one or more points associated with the speed limit markers can be determined and communicated to the mapping server. This approach is optional because it requires more computation on the part of the acquisition vehicle system. Sparse maps of the disclosed embodiments may use a smaller amount of stored data to enable autonomous navigation of a vehicle. In some embodiments, the sparse map 800 may have a data density (e.g., including data representing target trajectories, roadmarks, and any other stored road characteristics) of less than 2Mb per kilometer road, less than 1Mb per kilometer road, less than 500kB per kilometer road, or less than 100kB per kilometer road. In some embodiments, the data density of the sparse map 800 may be less than 10kB per kilometer road or even less than 2kB per kilometer road (e.g., 1.6kB per kilometer), or no more than 10kB per kilometer road or no more than 20kB per kilometer road. In some embodiments, a majority (if not all) of the U.S. roads may be autonomously navigated using a sparse map with a total of 4GB or less of data. These data density values may represent an average number for the entire sparse map 800, for local maps within the sparse map 800, and/or for particular road segments within the sparse map 800.

As described, the sparse map 800 may include representations of multiple target trajectories 810 for guiding autonomous driving or navigation along a road segment. Such target trajectories may be stored as three-dimensional splines. The target trajectory stored in the sparse map 800 may be determined, for example, based on two or more reconstructed trajectories of vehicles previously traversed along a particular road segment. The road segment may be associated with a single target track or multiple target tracks. For example, on a two-lane road, a first target trajectory may be stored to represent an expected travel path along the road in a first direction, and a second target trajectory may be stored to represent an expected travel path along the road in another direction (e.g., opposite the first direction). Additional target trajectories may be stored with respect to a particular road segment. For example, on a multi-lane road, one or more target trajectories may be stored representing an expected travel path for a vehicle in one or more lanes associated with the multi-lane road. In some embodiments, each lane of a multi-lane road may be associated with its own target track. In other embodiments, there may be fewer stored target trajectories than lanes present on a multi-lane road. In such a case, a vehicle traveling on a multi-lane road may guide its navigation using any one of the stored target trajectories by taking into account the lane offset from the lane for which the target trajectory is stored (for example, if the vehicle travels in the leftmost lane of a three-lane road and stores the target trajectory only for the middle lane of the road, the vehicle may navigate using the target trajectory of the middle lane by taking into account the lane offset between the middle lane and the leftmost lane when generating the navigation instruction).

In some embodiments, the target trajectory may represent an ideal path that the vehicle should take while traveling. The target track may be located, for example, in the approximate center of the driving lane. In other cases, the target trajectory may be located at other positions relative to the road segment. For example, the target trajectory may be substantially coincident with a center of the road, an edge of the road, or an edge of the lane, or the like. In such cases, the navigation based on the target track may include the determined offset amount by which the position relative to the target track is to be maintained. Further, in some embodiments, the determined offset by which the position relative to the target track is to be maintained may vary based on the type of vehicle (e.g., a passenger vehicle including two axles may have a different offset along at least a portion of the target track than a truck including more than two axles).

The sparse map 800 may also include data related to a plurality of predetermined landmarks 820, the landmarks 820 being associated with a particular road segment, local map, etc. As discussed in more detail below, these landmarks may be used in the navigation of autonomous vehicles. For example, in some embodiments, roadmarks may be used to determine the current position of the vehicle relative to a stored target trajectory. With this location information, the autonomous vehicle may be able to adjust the heading direction to match the direction of the target track at the determined location.

The plurality of landmarks 820 may be identified at any suitable spacing and stored in the sparse map 800. In some embodiments, roadmarks may be stored at a higher density (e.g., every few meters or more). However, in some embodiments, significantly larger road marking pitch values may be employed. For example, in the sparse map 800, the identified (or recognized) roadways may be spaced apart by 10 meters, 20 meters, 50 meters, 100 meters, 1 kilometer, or 2 kilometers. In some cases, the identified roadways may be located at distances even more than 2 kilometers apart.

Between landmarks, and thus between determinations of vehicle position relative to the target trajectory, the vehicle may navigate based on dead reckoning, where the vehicle uses sensors to determine its own motion and estimate its position relative to the target trajectory. Because errors may accumulate during navigation through dead reckoning, position determination relative to the target trajectory may become less and less accurate over time. The vehicle may use the landmarks (and their known locations) that appear in the sparse map 800 to remove dead reckoning induced errors in position determination. In this way, the identified roadmap contained in the sparse map 800 may be used as a navigation anchor from which the precise location of the vehicle relative to the target trajectory may be determined. Because a certain amount of error may be acceptable in position location, the identified roadmap need not always be available to autonomous vehicles. Proper navigation, as described above, may be possible even based on road marking spacings of 10 meters, 20 meters, 50 meters, 100 meters, 500 meters, 1km, 2 km or more. In some embodiments, a density of 1 identified roadmap per 1km road may be sufficient to maintain the longitudinal position determination accuracy within 1 m. Thus, not every potential landmark that appears along a road segment needs to be stored in the sparse map 800.

Further, in some embodiments, lane markings may be used for positioning of the vehicle during road marking intervals. By using the roadway markers during the road-marking intervals, error accumulation during navigation through dead reckoning can be minimized.

In addition to the target trajectory and the identified roadmap, the sparse map 800 may also include information related to various other road features. For example, fig. 9A shows a representation of a curve along a particular road segment that may be stored in the sparse map 800. In some embodiments, a single lane of a road may be modeled by a three-dimensional polynomial description of the left and right sides of the road. Such polynomials representing the left and right sides of a single lane are shown in fig. 9A. The road can be represented using a polynomial in a similar manner to that shown in fig. 9A, regardless of how many lanes the road has. For example, the left and right sides of a multi-lane road may be represented by polynomials similar to those shown in fig. 9A, and intermediate lane markings (e.g., a dotted line marking representing lane boundaries, huang Shixian representing boundaries between lanes traveling in different directions, etc.) included on the multi-lane road may also be represented using polynomials such as those shown in fig. 9A.

As shown in fig. 9A, a polynomial (e.g., a polynomial of first order, second order, third order, or any suitable order) may be used to represent the lane 900. For ease of illustration, the lane 900 is shown as a two-dimensional lane, and the polynomial is shown as a two-dimensional polynomial. As shown in fig. 9A, the lane 900 includes a left side 910 and a right side 920. In some embodiments, more than one polynomial may be used to represent the location of each side of a road or lane boundary. For example, each of the left side 910 and the right side 920 may be represented by a plurality of polynomials of any suitable length. In some cases, the polynomial may have a length of approximately 100m, although other lengths greater or less than 100m may be used. In addition, the polynomials can overlap one another to facilitate seamless transitions in navigation based on polynomials that are subsequently encountered when the host vehicle is traveling along a road. For example, each of the left side 910 and the right side 920 may be represented by a plurality of third-order polynomials which are divided into segments of about 100 meters long (an example of a first predetermined range) and overlap each other by about 50 meters. The polynomials representing left side 910 and right side 920 may or may not have the same order. For example, in some embodiments, some polynomials may be second order polynomials, some may be third order polynomials, and some may be fourth order polynomials.

In the example shown in fig. 9A, the left side 910 of the lane 900 is represented by two sets of third order polynomials. The first group includes polynomial segments 911, 912 and 913. The second set includes polynomial segments 914, 915 and 916. The two groupings, while substantially parallel to each other, follow the position of the respective sides of the road. Polynomial segments 911, 912, 913, 914, 915 and 916 have a length of approximately 100 meters and adjacent segments in the overlapping series are approximately 50 meters. However, as previously described, polynomials of different lengths and different amounts of overlap may also be used. For example, the polynomials may have a length of 500m, 1km or more, and the overlap amount may vary from 0 to 50m, 50m to 100m, or more than 100 m. In addition, while fig. 9A is shown as representing polynomials extending in 2D space (e.g., at the surface of a sheet of paper), it is to be understood that these polynomials may represent curves extending in three dimensions (e.g., including height components) to represent elevation changes of road segments in addition to X-Y curvatures. In the example shown in fig. 9A, the right side 920 of the lane 900 is further represented by a first set of polynomial segments 921, 922 and 923 and a second set of polynomial segments 924, 925 and 926.

Returning to the target trajectory of the sparse map 800, fig. 9B shows a three-dimensional polynomial representing the target trajectory of a vehicle traveling along a particular road segment. The target trajectory represents not only the X-Y path that the host vehicle should travel along a particular road segment, but also the elevation changes that the host vehicle will encounter while traveling along that road segment. Thus, each target trajectory in the sparse map 800 may be represented by one or more three-dimensional polynomials (e.g., three-dimensional polynomial 950 shown in fig. 9B). The sparse map 800 may include multiple trajectories (e.g., millions or billions or more to represent trajectories of vehicles along various segments of roads throughout the world). In some embodiments, each target trajectory may correspond to a spline connecting three-dimensional polynomial segments.

Regarding the data footprint of the polynomials stored in the sparse map 800, in some embodiments, each cubic polynomial may be represented by four parameters each requiring four bytes of data. A polynomial of degree three requiring approximately 192 bytes of data per 100m may be employed to obtain the appropriate representation. This may translate into approximately 200kB per hour in data usage/transmission requirements for a host vehicle traveling at approximately 100 km/hr.

The sparse map 800 may describe a network of lanes using a combination of geometric descriptors and metadata. The geometry may be described by a polynomial or spline as described above. The metadata may describe the number of lanes, special characteristics (e.g., common lanes for cars), and possibly other sparse tags. The total footprint of such indicators may be negligible.

Accordingly, a sparse map according to embodiments of the present disclosure may include at least one line representation of road surface features extending along a road segment, each line representation being to represent a path along the road segment that substantially corresponds to the road surface features. In some embodiments, as described above, the at least one line representation of the road surface feature may comprise a spline, a polynomial representation, or a curve. Further, in some embodiments, the road surface features may include at least one of road edges or lane markings. Further, as described below with respect to "crowd sourcing," road surface features may be identified by image analysis of a plurality of images acquired as one or more vehicles traverse a road segment.

As previously described, the sparse map 800 may include a plurality of predetermined roadways associated with road segments. Rather than storing actual images of landmarks and relying on, for example, image recognition analysis based on captured images and stored images, each landmark in the sparse map 800 may be represented and identified using less data than would otherwise be required by the stored actual images. The data representing the roadmap may still include sufficient information to describe or identify the roadmap along the road. Storing data describing characteristics of the landmark, rather than an actual image of the landmark, may reduce the size of the sparse map 800.

Fig. 10 shows an example of the types of roadways that may be represented in the sparse map 800. A road sign may include any visible and identifiable object along a road segment. The landmarks may be selected such that they are fixed and do not change frequently with respect to their location and/or content. The roadmap included in the sparse map 800 may be useful in determining the location of the vehicle 200 through a particular road segment relative to the target trajectory. Examples of road signs may include traffic signs, direction signs, general signs (e.g., rectangular signs), roadside fixtures (e.g., lampposts, reflectors, etc.), and any other suitable category. In some embodiments, lane markings on the road may also be included as roadmarks in the sparse map 800.

Examples of road signs shown in fig. 10 include traffic signs, direction signs, roadside fixtures, and general signs. Traffic signs may include, for example, speed limit signs (e.g., speed limit sign 1000), let-off signs (e.g., let-off sign 1005), route number signs (e.g., route number sign 1010), traffic light signs (e.g., traffic light sign 1015), stop signs (e.g., stop sign 1020). The direction indicator may comprise an indicator comprising one or more arrows indicating one or more directions to different locations. For example, the direction flag may include: road signs 1025 with arrows for guiding the vehicle to different roads or places; an exit sign 1030 having an arrow for guiding the vehicle off the road; etc. Accordingly, at least one of the plurality of road signs may comprise a road sign.

The general sign may be traffic independent. For example, a general sign may include a billboard for advertising or a welcome board adjacent to a boundary between two countries, states, counties, cities or towns. Fig. 10 shows a general logo 1040 ("Joe's restaurant"). While the general logo 1040 may have a rectangular shape as shown in fig. 10, the general logo 1040 may have other shapes, such as square, circular, triangular, etc.

The road sign may also include a roadside fixture. Roadside fixtures may not be the subject of a sign and may not be traffic or direction related. For example, the roadside fixture may include a lamppost (e.g., lamppost 1035), a wire pole, a traffic lamppost, and the like.

The roadmap may also include beacons that may be specifically designed for use in an autonomous vehicle navigation system. For example, such beacons may include independent structures placed at predetermined intervals to aid in navigating the host vehicle. Such beacons may also include visual/graphical information (e.g., icons, badges, bar codes, etc.) added to existing road signs, which may be identified or recognized by vehicles traveling along the road segment. Such beacons may also include electronic components. In such embodiments, an electronic beacon (e.g., an RFID tag, etc.) may be used to transmit non-visual information to the host vehicle. Such information may include, for example, landmark identification and/or landmark location information that the host vehicle may use in determining the location along the target trajectory.

In some embodiments, the roadmap contained in the sparse map 800 may be represented by a data object of a predetermined size. The data representing a landmark may include any suitable parameters for identifying a particular landmark. For example, in some embodiments, the roadmap stored in the sparse map 800 may include parameters such as physical size of the roadmap (e.g., to support estimating distance to the roadmap based on a known size/scale), distance to a previous roadmap, lateral offset, altitude, type code (e.g., roadmap type-which type of direction sign, traffic sign, etc.), GPS coordinates (e.g., to support global positioning), and any other suitable parameters. Each parameter may be associated with a data size. For example, a landmark size may be stored using 8 bytes of data. 12 bytes of data may be used to specify distance, lateral offset, and height to the previous landmark. The type code associated with a road sign (e.g., a direction sign or traffic sign) may require approximately 2 bytes of data. For a general token, a 50 byte data store may be used to store an image signature that implements the identity of the general token. The landmark GPS location may be associated with a 16 byte data store. These data sizes for each parameter are examples only, and other data sizes may be used. Representing landmarks in sparse map 800 in this manner may provide a thin solution for effectively representing landmarks in a database. In some embodiments, an object may be referred to as a standard semantic object or a non-standard semantic object. Standard semantic objects may include any class of objects for which a standardized set of characteristics exist (e.g., speed limit signs, warning signs, direction signs, traffic lights, etc., of known size or other characteristics). Non-standard semantic objects may include any object that is not associated with a standardized set of characteristics (e.g., universal advertising signs, signs identifying businesses, pits, trees, etc., which may have variable sizes). Each nonstandard semantic object may be represented with 38 bytes of data (e.g., 8 bytes for size; 12 bytes for distance from previous flags, lateral offset and height; 2 bytes for type code; and 16 bytes for location coordinates). Standard semantic objects may be represented using even less data because the mapping server may not need size information to fully represent objects in a sparse map.

The sparse map 800 may use a label system to represent landmark types. In some cases, each traffic sign or direction sign may be associated with its own tag, which may be stored in a database as part of the landmark identification. For example, the database may include approximately 1000 different labels representing various traffic signs and approximately 10000 different labels representing direction signs. Of course, any suitable number of tags may be used, and additional tags may be created as desired. In some embodiments less than about 100 bytes may be used to represent the universal flag (e.g., about 86 bytes, including: 8 bytes for size, 12 bytes for distance to previous signpost, lateral offset and height, 50 bytes for image signature, and 16 bytes for GPS coordinates).

Thus, for semantic road signs that do not require image signing, the data density impact on the sparse map 800 may be about 760 bytes per kilometer even at a higher landmark density of about 1 per 50m (e.g., 20 landmarks per kilometer x 38 bytes per landmark = 760 bytes). Even for a general signature including an image signature component, the data density impact is about 1.72kB per kilometer (20 signposts per kilometer x 86 bytes per signpost = 1720 bytes). For semantic road signs this corresponds to a data usage of about 76kB per hour for a vehicle travelling at 100 km/hr. For the universal sign this corresponds to about 170kB per hour for a vehicle traveling at 100 km/hr. It should be noted that in some environments (e.g., urban environments), there may be a much higher detected object density (possibly more than one per meter) available for inclusion in the sparse map. In some embodiments, generally rectangular objects (e.g., rectangular markers) may be represented in the sparse map 800 by no more than 100 bytes of data. The representation of the generally rectangular object (e.g., the general marker 1040) in the sparse map 800 may include a compressed image signature or image hash (e.g., the compressed image signature 1045) associated with the generally rectangular object. This compressed image signature/image hash may be determined using any suitable image hashing algorithm and may be used, for example, to help identify a generic marker, such as an identified marker. Such a compressed image signature (e.g., image information derived from actual image data representing the object) may obviate the need to store an actual image of the object or to perform a comparative image analysis on the actual image in order to identify landmarks.

Referring to fig. 10, the sparse map 800 may include or store a compressed image signature 1045 associated with a general marker 1040 instead of an actual image of the general marker 1040. For example, after an image capture device (e.g., image capture device 122, 124, or 126) captures an image of the general signature 1040, a processor (e.g., image processor 190 or any other processor capable of processing an image on or remotely located relative to the host vehicle) may perform image analysis to extract/create a compressed image signature 1045, the compressed image signature 1045 including a unique signature or pattern associated with the general signature 1040. In one embodiment, the compressed image signature 1045 may include a shape, a color pattern, a brightness pattern, or any other feature that may be extracted from the image of the general logo 1040 for describing the general logo 1040.

For example, in fig. 10, circles, triangles, and stars shown in the compressed image signature 1045 may represent areas of different colors. The pattern represented by circles, triangles, and stars may be stored in the sparse map 800, for example within 50 bytes specified to include an image signature. Notably, circles, triangles, and stars are not necessarily intended to indicate that such shapes are stored as part of an image signature. These shapes are instead intended to conceptually represent identifiable regions with discernable color differences, text regions, graphic regions, or other variations in characteristics that may be associated with a universal sign. Such compressed image signatures can be used to identify landmarks in the form of general landmarks. For example, the compressed image signature can be used to perform a same-not-same (same-not-same) analysis based on a comparison of the stored compressed image signature with image data captured, for example, using a camera on an autonomous vehicle.

Accordingly, the plurality of roadways may be identified by image analysis of a plurality of images acquired as one or more vehicles traverse the road segment. As described below with respect to "crowdsourcing," in some embodiments, image analysis identifying a plurality of landmarks may include accepting potential landmarks when the ratio of images in which landmarks appear to images in which landmarks do not appear exceeds a threshold value. Further, in some embodiments, the image analysis identifying the plurality of landmarks may include rejecting potential landmarks when a ratio of images in which landmarks do not appear to images in which landmarks appear exceeds a threshold value.

Returning to the target trajectories that the host vehicle may use to navigate a particular road segment, fig. 11A shows a polynomial-representative trajectory captured during the process of building or maintaining the sparse map 800. The polynomial representation of the target trajectory included in the sparse map 800 may be determined based on two or more reconstructed trajectories of vehicles previously traversed along the same road segment. In some embodiments, the polynomial representation of the target trajectory contained in the sparse map 800 may be an aggregation of two or more reconstructed trajectories of a previous traversal of the vehicle along the same road segment. In some embodiments, the polynomial representation of the target trajectory contained in the sparse map 800 may be an average of two or more reconstructed trajectories of vehicles previously traversed along the same road segment. Other mathematical operations may also be used to construct a target trajectory along a road path based on reconstructed trajectories collected from vehicles traveling along the road segment.

As shown in fig. 11A, a road segment 1100 may be traveled by multiple vehicles 200 at different times. Each vehicle 200 may collect data related to the path taken by the vehicle along the road segment. The path traveled by a particular vehicle may be determined based on camera data, accelerometer information, speed sensor information, and/or GPS information, among other potential sources. Such data may be used to reconstruct trajectories of vehicles traveling along the road segment, and based on these reconstructed trajectories, a target trajectory (or multiple target trajectories) may be determined for the particular road segment. Such target trajectories may represent preferred paths of the host vehicle (e.g., guided by an autonomous navigation system) while traveling along the road segment.

In the example shown in fig. 11A, the first reconstructed track 1101 may be determined based on data received from a first vehicle traversing the road segment 1100 for a first period of time (e.g., day 1), the second reconstructed track 1102 may be obtained from a second vehicle traversing the road segment 1100 for a second period of time (e.g., day 2), and the third reconstructed track 1103 may be obtained from a third vehicle traversing the road segment 1100 for a third period of time (e.g., day 3). Each of the trajectories 1101, 1102, and 1103 may be represented by a polynomial (e.g., a three-dimensional polynomial). It should be noted that in some embodiments, any of the reconstructed trajectories may be assembled on a vehicle traversing the road segment 1100.

Additionally or alternatively, such a reconstructed trajectory may be determined on the server side based on information received from vehicles traversing the road segment 1100. For example, in some embodiments, the vehicle 200 may transmit data (e.g., steering angle, heading, time, location, speed, sensed road geometry, and/or sensed roadmap, etc.) related to their movement along the road segment 1100 to one or more servers. The server may reconstruct the trajectory of the vehicle 200 based on the received data. The server may also generate a target trajectory based on the first, second, and third trajectories 1101, 1102, and 1103 for guiding navigation of an autonomous vehicle that will later travel along the same road segment 1100. While the target trajectories may be associated with a single previous pass of the road segment, in some embodiments, each target trajectory included in the sparse map 800 may be determined based on two or more reconstructed trajectories of vehicles passing through the same road segment. In fig. 11A, a target track is represented by 1110. In some embodiments, the target track 1110 may be generated based on an average of the first, second, and third tracks 1101, 1102, and 1103. In some embodiments, the target trajectory 1110 included in the sparse map 800 may be an aggregation (e.g., a weighted combination) of two or more reconstructed trajectories.

At the mapping server, the server may receive actual trajectories for a particular road segment from a plurality of collection vehicles traversing the road segment. In order to generate a target trajectory for each effective path along the road segment (e.g., each lane, each driving direction, each path through the intersection, etc.), the received actual trajectories may be aligned. The alignment process may include correlating actual, acquired trajectories with each other using the identified detected objects/features along the road segment along with the acquisition locations of those detected objects/features. Once aligned, an average or "best fit" target track for each available lane, etc., may be determined based on the aggregated, correlated/aligned actual tracks.

Fig. 11B and 11C further illustrate the concept of a target track associated with a road segment existing within the geographic area 1111. As shown in fig. 11B, a first segment 1120 within the geographic area 1111 may include a multi-lane road including two lanes 1122 designated for vehicle travel in a first direction and two additional lanes 1124 designated for vehicle travel in a second direction opposite the first direction. Lanes 1122 and 1124 may be separated by a double yellow line 1123. The geographic area 1111 may also include a branch road segment 1130 intersecting the road segment 1120. The road segment 1130 may include a two-lane road, each lane being designated for a different direction of travel. The geographic region 1111 may also include other road features such as stop line 1132, stop flag 1134, speed limit flag 1136, and hazard flag 1138.

As shown in fig. 11C, the sparse map 800 may include a local map 1140 that includes a road model for assisting autonomous navigation of vehicles within the geographic area 1111. For example, the local map 1140 may include target trajectories for one or more lanes associated with road segments 1120 and/or 1130 within the geographic area 1111. For example, the local map 1140 may include target trajectories 1141 and/or 1142 that an autonomous vehicle may access or rely on when traversing the lane 1122. Similarly, the local map 1140 may include target trajectories 1143 and/or 1144 that an autonomous vehicle may access or rely on when traversing lane 1124. Further, the local map 1140 may include target trajectories 1145 and/or 1146 that the autonomous vehicle may access or rely on when traversing the road segment 1130. The target trajectory 1147 represents a preferred path that the autonomous vehicle should follow when transitioning from the lane 1120 (and in particular with respect to the target trajectory 1141 associated with the rightmost lane of the lane 1120) to the road segment 1130 (and in particular with respect to the target trajectory 1145 associated with the first side of the road segment 1130). Similarly, the target trajectory 1148 represents a preferred path that the autonomous vehicle should follow when transitioning from the road segment 1130 (and specifically with respect to the target trajectory 1146) to a portion of the road segment 1124 (and specifically with respect to the target trajectory 1143 associated with the left lane of the lane 1124 as shown).

The sparse map 800 may also include representations of other road related features associated with the geographic area 1111. For example, the sparse map 800 may also include a representation of one or more roadmarks identified in the geographic area 1111. Such landmarks may include a first landmark 1150 associated with stop line 1132, a second landmark 1152 associated with stop sign 1134, a third landmark associated with speed limit sign 1154, and a fourth landmark 1156 associated with hazard sign 1138. Such landmarks may be used, for example, to assist an autonomous vehicle in determining its current position relative to any of the illustrated target trajectories so that the vehicle may adjust its heading to match the direction of the target trajectory at the determined position.

In some embodiments, the sparse map 800 may also include a road signature profile. Such a road signature profile may be associated with any discernable/measurable change in at least one parameter of the associated road. For example, in some cases, such profiles may be associated with changes in road surface information (e.g., changes in surface roughness of a particular road segment, changes in road width for a particular road segment, changes in distance between dashed lines drawn along a particular road segment, changes in road curvature along a particular road segment, etc.). Fig. 11D shows an example of a road sign profile 1160. While the profile 1160 may represent any of the parameters described above, etc., in one example, the profile 1160 may represent a measure of road surface roughness as obtained by monitoring one or more sensors that provide an output indicative of the amount of suspension displacement when the vehicle is traveling a particular road segment.

Alternatively or concurrently, profile 1160 may represent a change in road width as determined based on image data obtained via cameras on a vehicle traveling a particular road segment. Such profiles may be useful, for example, in determining a particular location of an autonomous vehicle relative to a particular target trajectory. That is, as an autonomous vehicle traverses a road segment, it may measure a profile associated with one or more parameters associated with the road segment. If the measured profile can be correlated/matched with a predetermined profile that maps a parameter change relative to a location along the road segment, the measured and predetermined profiles (e.g., by overlapping corresponding portions of the measured and predetermined profiles) can be used to determine the current location along the road segment and to determine the current location relative to the target trajectory of the road segment.

In some embodiments, the sparse map 800 may include different trajectories based on different characteristics associated with a user of the autonomous vehicle, environmental conditions, and/or other parameters related to driving. For example, in some embodiments, different trajectories may be generated based on different user preferences and/or profiles. The sparse map 800 including such different trajectories may be provided to different autonomous vehicles of different users. For example, some users may prefer to avoid toll roads, while other users may prefer to take the shortest or fastest route, regardless of whether there are toll roads on the line. The disclosed system may generate different sparse maps with different trajectories based on such different user preferences or profiles. As another example, some users may prefer to travel on a fast moving lane, while other users may prefer to maintain a location in a center lane at all times.

Different trajectories may be generated based on different environmental conditions (e.g., day and night, snow, rain, fog, etc.) and included in the sparse map 800. Autonomous vehicles that are driven under different environmental conditions may be provided with a sparse map 800 generated based on such different environmental conditions. In some embodiments, a camera provided on the autonomous vehicle may detect environmental conditions and in turn may provide such information to a server that generates and provides a sparse map. For example, the server may generate or update the sparse map 800 that has been generated to include trajectories that may be more suitable or safe for autonomous driving under the detected environmental conditions. Updating the sparse map 800 based on environmental conditions may be performed dynamically as the autonomous vehicle travels along a road.

Other different parameters related to driving may also be used as a basis for generating different sparse maps and providing to different autonomous vehicles. For example, when an autonomous vehicle is traveling at high speed, the turn may be tighter. A track associated with a particular lane, rather than a road, may be included in the sparse map 800 such that an autonomous vehicle may remain within the particular lane while following the particular track. When an image captured by a camera on an autonomous vehicle indicates that the vehicle has drifted outside of a lane (e.g., passed through a lane marker), an action may be triggered within the vehicle to bring the vehicle back to the specified lane in a particular trajectory.

Crowdsourcing sparse map

The disclosed sparse map may be efficiently (and passively) generated by crowdsourcing capabilities. For example, any private or commercial vehicle equipped with a camera (e.g., a simple low resolution camera typically included as OEM equipment on today's vehicles) and a suitable image analysis processor may be used as the acquisition vehicle. No special equipment (e.g., high definition imaging and/or positioning systems) is required. As a result of the disclosed crowdsourcing technique, the generated sparse map may be extremely accurate and may include extremely fine positional information (allowing for navigation error limits of 10cm or less) without requiring any specialized imaging or sensing devices as input to the map generation process. Crowd sourcing also enables much faster (and cheaper) updates to the generated map, as new driving information from any road that is minimally equipped to also serve as a collection vehicle for private or commercial vehicles to traverse is continually available to the mapping server system. There is no need for a given vehicle equipped with high definition imaging and mapping sensors. Thus, the costs associated with building such a dedicated vehicle can be avoided. Furthermore, the update of the presently disclosed sparse map may be much faster than systems that rely on dedicated, specialized mapping vehicles (which are typically limited to a number of specialized vehicle fleets that are far lower than the number of private or commercial vehicles that are already available to perform the disclosed acquisition techniques due to their cost and specialized equipment).

The disclosed sparse maps generated by crowdsourcing can be extremely accurate in that they can be generated based on many inputs from multiple (tens, hundreds, millions, etc.) acquisition vehicles that have collected driving information along a particular road segment. For example, each acquisition vehicle traveling along a particular road segment may record its actual trajectory and may determine location information relative to detected objects/features along the road segment. The information is communicated from the plurality of collection vehicles to the server. The actual trajectories are aggregated to generate refined target trajectories for each effective driving path along the road segment. In addition, the location information collected from multiple collection vehicles for each detected object/feature (semantic or non-semantic) along the road segment may also be aggregated. As a result, the mapped location of each detected object/feature may constitute an average of hundreds, thousands, or millions of individually determined locations for each detected object/feature. Such techniques may be generated for detected objects/features and their exact mapped locations.

In some embodiments, the disclosed systems and methods may generate a sparse map of autonomous vehicle navigation. For example, the disclosed systems and methods may use crowd-sourced data to generate a sparse map that one or more autonomous vehicles may use to navigate along a road system. As used herein, "crowd sourcing" means that data is received from various vehicles (e.g., autonomous vehicles) traveling on a road segment at different times, and such data is used to generate and/or update a road model, including sparse map tiles. Any of the models or sparse map tiles thereof may in turn be transmitted to vehicles or other vehicles that later travel along the road segment for assisting autonomous vehicle navigation. The road model may include a plurality of target trajectories representing preferred trajectories that the autonomous vehicle should follow when traversing the road segment. The target trajectory may be the same as the reconstructed actual trajectory collected from the vehicles traversing the road segment, which may be transmitted from the vehicles to the server. In some embodiments, the target trajectory may be different from the actual trajectory taken by one or more vehicles when previously traversing the road segment. The target trajectory may be generated based on the actual trajectory (e.g., by averaging or any other suitable operation).

The vehicle trajectory data that the vehicle may upload to the server may correspond to an actual reconstructed trajectory of the vehicle, or may correspond to a recommended trajectory that may be based on or related to the actual reconstructed trajectory of the vehicle, but may be different from the actual reconstructed trajectory. For example, the vehicle may modify its actual reconstructed trajectory and submit (e.g., recommend) the modified actual trajectory to the server. The road model may use the recommended modified trajectory as a target trajectory for autonomous navigation of other vehicles.

In addition to trajectory information, other information that may be used in constructing the sparse data map 800 may also include information related to potential landmark candidates. For example, through crowdsourcing of information, the disclosed systems and methods can identify potential roadways in an environment and refine roadway locations. The roadmap may be used by the navigation system of the autonomous vehicle to determine and/or adjust the position of the vehicle along the target trajectory.

The reconstructed trajectory that the vehicle can generate while traveling along the road may be obtained by any suitable method. In some embodiments, the reconstructed trajectory may be formed by stitching together the motion segments of the vehicle using, for example, self-motion estimation (e.g., three-dimensional translation and three-dimensional rotation of the camera and thus the body of the vehicle). The rotation and translation estimates may be determined based on analysis of images captured by one or more image capture devices along with information from other sensors or devices (e.g., inertial and speed sensors). For example, the inertial sensor may include an accelerometer or other suitable sensor configured to measure changes in translation and/or rotation of the vehicle body. The vehicle may include a speed sensor that measures the speed of the vehicle.

In some embodiments, the self-motion of the camera (and thus the vehicle body) may be estimated based on optical flow analysis of the captured images. Optical flow analysis of the image sequence identifies movement of pixels from the image sequence and determines movement of the vehicle based on the identified movement. The self-movement may integrate over time and along the road segment to reconstruct a trajectory associated with the road segment that the vehicle has followed.

Data collected by multiple vehicles driving along a road segment at different times (e.g., reconstructing trajectories) may be used to construct a road model (e.g., including target trajectories, etc.) contained in the sparse map 800. Data collected by multiple vehicles driving along a road segment at different times may also be averaged to increase the accuracy of the model. In some embodiments, data relating to road geometry and/or road sign may be received from multiple vehicles traveling through a common road segment at different times. Such data received from different vehicles may be combined to generate a road model and/or update the road model.

The geometry of the reconstructed trajectory (and also the target trajectory) along the road segment may be represented by a curve in three-dimensional space, which may be a spline connecting three-dimensional polynomials. The reconstructed trajectory curve may be determined from an analysis of a video stream or a plurality of images captured by a camera mounted on the vehicle. In some embodiments, the location is identified in each frame or image a few meters in front of the current location of the vehicle. This position is the position to which the vehicle is expected to travel for a predetermined period of time. This operation may be repeated frame by frame and at the same time the vehicle may calculate the camera's own motion (rotation and translation). At each frame or image, a short-range model of the expected path is generated by the vehicle in a frame of reference attached to the camera. The short-range models may be stitched together to obtain a three-dimensional model of the road in a coordinate system, which may be any or a predetermined coordinate system. The three-dimensional model of the road may then be fitted by a spline, which may include or connect one or more polynomials of appropriate order.

To infer the short-range road model at each frame, one or more detection modules may be used. For example, a bottom-up (bottom-up) lane detection module may be used. The bottom-up lane detection module may be useful when drawing lane markings on a road. This module may look for edges in the image and assemble them together to form lane markings. The second module may be used in conjunction with a bottom-up lane detection module. The second module is an end-to-end depth neural network that can be trained to predict the correct short-range path from the input image. In both modules, the road model may be detected in an image coordinate system and transformed into a three-dimensional space, which may be virtually attached to the camera.

While the reconstruction trajectory modeling approach may introduce a build-up of errors (which may include noise components) due to integration of self-motion over long periods of time, such errors may be insignificant because the generated model may provide sufficient accuracy for navigation of the local scale. In addition, it is possible to eliminate integration errors by using external information sources (e.g., satellite images or geodetic measurements). For example, the disclosed systems and methods may use a GNSS receiver to eliminate accumulated errors. However, GNSS positioning signals may not necessarily be available and accurate. The disclosed systems and methods may implement steering applications that are weakly dependent on availability and accuracy of GNSS positioning. In such systems, the use of GNSS signals may be limited. For example, in some embodiments, the disclosed systems may use GNSS signals for purposes of facilitating database indexing only.

In some embodiments, the range scale (e.g., local scale) associated with the autonomous vehicle navigation steering application may be about 50 meters, 100 meters, 200 meters, 300 meters, etc. Such distances can be used because the geometric road model is mainly used for two purposes: planning a forward trajectory, and positioning the vehicle on the road model. In some embodiments, the planning task may use the model for a typical range of 40 meters ahead (or any other suitable distance ahead, e.g., 20 meters, 30 meters, 50 meters) when the control algorithm maneuvers the vehicle at the target point that is located 1.3 seconds ahead (or any other time, e.g., 1.5 seconds, 1.7 seconds, 2 seconds, etc.). The localization task uses the road model for a typical range of 60 meters (or any other suitable distance, e.g., 50 meters, 100 meters, 150 meters, etc.) behind the car in accordance with another method called "tail alignment" described in more detail in another section. The disclosed systems and methods may generate a geometric model with sufficient accuracy for a particular range (e.g., 100 meters) that the planned trajectory will not deviate more than, for example, 30cm from the lane center.

As described above, a three-dimensional road model may be constructed by detecting short-range portions and stitching them together. Stitching may be accomplished by calculating a six degree self-motion model using video and/or images captured by the cameras, data from inertial sensors reflecting the motion of the vehicle, and the master vehicle speed signal. The accumulated error may be small enough for a certain local range scale, for example about 100 meters. This aspect may all be accomplished in a single drive through a particular road segment.

In some embodiments, multiple driving may be used to average the generated model and further increase its accuracy. The same car may travel the same route multiple times or multiple cars may send their collected model data to a central server. In any event, a matching process may be performed to identify overlapping models and to implement averaging to generate target trajectories. Once the convergence criterion is met, the constructed model (e.g., including the target trajectory) is available for manipulation. Subsequent driving may be used for other model improvements and to accommodate infrastructure changes.

Sharing of driving experiences (e.g., sensed data) between multiple automobiles becomes feasible when they are connected to a central server. Each vehicle client may store a partial copy of the generic road model, which may be correlated to its current location. The bi-directional update process between the vehicle and the server may be performed by the vehicle and the server. The small footprint concepts described above enable the disclosed systems and methods to perform bi-directional updates using very little bandwidth.

Information related to the potential roadmap may also be determined and forwarded to the central server. For example, the disclosed systems and methods may determine one or more physical properties of a potential landmark based on one or more images including the landmark. Physical properties may include the physical size (e.g., height, width) of the landmark, the distance from the vehicle to the landmark, the distance between the landmark and the previous landmark, the lateral position of the landmark (e.g., the position of the landmark relative to the driving lane), the GPS coordinates of the landmark, the type of landmark, the identification of text on the landmark, etc. For example, the vehicle may analyze one or more images captured by the camera to detect potential roadsigns (e.g., speed limit signs).

The vehicle may determine a distance from the vehicle to the road sign or a location associated with the road sign (e.g., any semantic or non-semantic objects or features along the road segment) based on the analysis of the one or more images. In some embodiments, the distance may be determined based on an analysis of the image of the landmark using a suitable image analysis method (e.g., a scaling method and/or an optical flow method). As previously described, the location of the object/feature may include the 2D image location of one or more points associated with the object/feature (e.g., X-Y pixel locations in one or more captured images), or may include the 3D real world location of one or more points (e.g., determined by structure in motion/optical flow techniques, LIDAR or RADAR information, etc.). In some embodiments, the disclosed systems and methods may be configured to determine the type or class of potential roadways. In the event that the vehicle determines that a potential landmark corresponds to a predetermined type or class stored in the sparse map, it may be sufficient for the vehicle to transmit an indication of the type or class of landmark to the server along with its location. The server may store such an indication. At a later time, during navigation, the navigation vehicle may capture an image including a representation of the landmark, process the image (e.g., using a classifier), and compare the resulting landmark to confirm detection of the mapped landmark, and use the mapped landmark in positioning the navigation vehicle relative to the sparse map.

In some embodiments, multiple autonomous vehicles traveling on a road segment may communicate with a server. The vehicle (or customer) may generate a curve describing its driving in any coordinate system (e.g., through self-motion integration). The vehicle may detect the landmarks and locate them in the same frame. The vehicle may upload the curves and roadmarks to the server. The server may collect data from the vehicle for multiple drives and generate a unified road model. For example, as described below with respect to fig. 19, the server may use the uploaded curves and roadmarks to generate a sparse map with a unified road model.

The server may also distribute the model to clients (e.g., vehicles). For example, the server may distribute the sparse map to one or more vehicles. The server may continuously or periodically update the model as new data is received from the vehicle. For example, the server may process the new data to evaluate whether the data includes information that should trigger an update or create new data on the server. The server may distribute the updated model or updates to the vehicle for providing autonomous vehicle navigation.

The server may use one or more criteria to determine whether new data received from the vehicle should trigger an update to the model or trigger creation of new data. For example, when the new data indicates that a previously identified landmark at a particular location is no longer present or is replaced with another landmark, the server may determine that the new data should trigger an update to the model. As another example, when the new data indicates that the road segment has been closed, and when this is verified by data from other vehicles, the server may determine that the new data should trigger an update to the model.

The server may distribute the updated model (or updated portion of the model) to one or more vehicles traveling on the road segment associated with the update of the model. The server may also distribute the updated model to vehicles that are to travel on the road segment with which the update of the model is associated or vehicles whose planned journey includes the road segment. For example, while an autonomous vehicle travels along another road segment before reaching the road segment with which the update is associated, the server may distribute the update or updated model to the autonomous vehicle before reaching the road segment.

In some embodiments, the remote server may collect trajectories and roadways from multiple clients (e.g., vehicles traveling along a common road segment). The server may use the roadmap to match the curve and create an average road model based on the trajectories collected from the plurality of vehicles. The server may also calculate a graph of the road and the most likely path at each node or junction of the road segment. For example, the remote server may align the trajectories to generate a crowdsourcing sparse map from the collected trajectories.

The server may average the road marking properties received from a plurality of vehicles traveling along a common road segment, such as the distance measured by the plurality of vehicles from one road marking to another (e.g., a preceding road marking along the road segment) to determine arc length parameters and support positioning along the path and speed calibration for each customer vehicle. The server may average physical dimensions of road signs measured by a plurality of vehicles traveling along a common road segment and identifying the same road sign. The average physical size may be used to support distance estimation, such as distance from a vehicle to a road sign. The server may average lateral positions of road signs (e.g., positions from lanes where vehicles are driving to road signs) as measured by multiple vehicles driving along a common road segment and identifying the same road sign. The average lateral position may be used to support lane assignment. The server may average the GPS coordinates of road signs measured by a plurality of vehicles traveling along the same road segment and identifying the same road sign. The average GPS coordinates of the roadmap may be used to support global localization or positioning of the roadmap in the road model.

In some embodiments, the server may identify model changes, such as construction, detouring, new signs, removal of signs, etc., based on data received from the vehicle. The server may continuously or periodically or instantaneously update the model as new data is received from the vehicle. The server may distribute updates to the model or updated models to the vehicles for providing autonomous navigation. For example, as discussed further below, the server may use crowd-sourced data to filter out "false" road signs detected by the vehicle.

In some embodiments, the server may analyze driver intervention during autonomous driving. The server may analyze data received from the vehicle at the time and location of the intervention and/or data received prior to the intervention. The server may identify certain portions of data that cause or are closely related to the intervention, such as data indicating temporary lane closure establishment, data indicating pedestrians in the road. The server may update the model based on the identified data. For example, the server may modify one or more tracks stored in the model.

FIG. 12 is a schematic diagram of a system that uses crowdsourcing to generate sparse maps (and uses crowdsourcing sparse maps for distribution and navigation). Fig. 12 shows a road segment 1200 that includes one or more lanes. Multiple vehicles 1205, 1210, 1215, 1220, and 1225 may travel on road segment 1200 at the same time or at different times (although shown in fig. 12 as occurring on road segment 1200 at the same time). At least one of vehicles 1205, 1210, 1215, 1220, and 1225 may be an autonomous vehicle. For simplicity of this example, all vehicles 1205, 1210, 1215, 1220, and 1225 are assumed to be autonomous vehicles.

Each vehicle may be similar to the vehicles disclosed in other embodiments (e.g., vehicle 200) and may include components or devices included in or associated with the vehicles disclosed in other embodiments. Each vehicle may be equipped with an image capture device or camera (e.g., image capture device 122 or camera 122). Each vehicle may communicate with remote server 1230 via one or more networks (e.g., through a cellular network and/or the internet, etc.) via wireless communication path 1235 (shown in phantom). Each vehicle may transmit data to the server 1230 and receive data from the server 1230. For example, the server 1230 may collect data from a plurality of vehicles traveling on the road segment 1200 at different times and may process the collected data to generate an autonomous vehicle road navigation model or an update to the model. The server 1230 may transmit the autonomous vehicle road navigation model or updates to the model to the vehicle (which transmits data to the server 1230). The server 1230 may transmit the autonomous vehicle road navigation model or updates to the model to other vehicles that later travel on the road segment 1200.

As vehicles 1205, 1210, 1215, 1220, and 1225 travel on road segment 1200, the navigation information collected (e.g., detected, sensed, or measured) by vehicles 1205, 1210, 1215, 1220, and 1225 may be transmitted to server 1230. In some embodiments, the navigation information may be associated with the common road segment 1200. The navigation information may include trajectories associated with each of the vehicles 1205, 1210, 1215, 1220, and 1225 as each travels through the road segment 1200. In some embodiments, the trajectory may be reconstructed based on data sensed by various sensors and devices provided on the vehicle 1205. For example, the trajectory may be reconstructed based on at least one of accelerometer data, speed data, roadmap data, road geometry or profile data, vehicle positioning data, and self-movement data. In some embodiments, the trajectory may be reconstructed based on data from inertial sensors (e.g., accelerometers) and the speed of the vehicle 1205 sensed by the speed sensor. Additionally, in some embodiments, the trajectory may be determined by a processor on each of vehicles 1205, 1210, 1215, 1220, and 1225 based on the sensed self-motion of the camera, which may indicate three-dimensional translation and/or three-dimensional rotation (or rotational motion). The self-motion of the camera (and thus the body) may be determined from analysis of one or more images captured by the camera.

In some embodiments, the trajectory of vehicle 1205 may be determined by a processor provided on vehicle 1205 and transmitted to server 1230. In other embodiments, server 1230 may receive data sensed by various sensors and devices provided in vehicle 1205 and determine trajectories based on the data received from vehicle 1205.

In some embodiments, the navigation information transmitted from vehicles 1205, 1210, 1215, 1220, and 1225 to server 1230 may include data related to a road surface, road geometry, or road profile. The geometry of the road segment 1200 may include a lane structure and/or a road sign. The lane structure may include a total number of lanes of the road segment 1200, a type of lane (e.g., one-way lane, two-way lane, roadway, overpass, etc.), a marking on the lane, a width of the lane, etc. In some embodiments, the navigation information may include lane assignments, such as in which lane of the plurality of lanes the vehicle is traveling. For example, a lane assignment may be associated with a value of "3" that indicates that the vehicle is traveling on a third lane from the left or right. As another example, the lane assignment may be associated with a text value "center lane" that indicates that the vehicle is traveling on the center lane.

The server 1230 may store navigation information on non-transitory computer readable media (e.g., hard disk drives, compact discs, magnetic tapes, memory, etc.). The server 1230 may generate (e.g., by a processor included in the server 1230) at least a portion of an autonomous vehicle road navigation model for the common road segment 1200 based on navigation information received from the plurality of vehicles 1205, 1210, 1215, 1220, and 1225, and may store the model as part of a sparse map. The server 1230 may determine a trajectory associated with each lane based on crowd-sourced data (e.g., navigation information) received from multiple vehicles (e.g., 1205, 1210, 1215, 1220, and 1225) traveling on lanes of a road segment at different times. The server 1230 may generate an autonomous vehicle road navigation model or a portion of the model (e.g., an updated portion) based on a plurality of trajectories determined from crowd-sourced navigation data. The server 1230 may transmit the model or an updated portion of the model to one or more of the autonomous vehicles 1205, 1210, 1215, 1220, and 1225 traveling on the road segment 1200 or any other autonomous vehicles that later travel on the road segment for updating the existing autonomous vehicle road navigation model provided in the navigation system of the vehicle. The autonomous vehicle road navigation model may be used by the autonomous vehicle in autonomous navigation along the common road segment 1200.

As described above, the autonomous vehicle road navigation model may be included in a sparse map (e.g., sparse map 800 shown in fig. 8). The sparse map 800 may include sparse records of data related to road geometry and/or roadways along roads that may provide sufficient information to guide autonomous navigation of an autonomous vehicle, but do not require excessive data storage. In some embodiments, the autonomous vehicle road navigation model may be stored separately from the sparse map 800, and map data from the sparse map 800 may be used when the model is executed for navigation. In some embodiments, the autonomous vehicle road navigation model may use map data contained in the sparse map 800 for determining a target trajectory along the road segment 1200 for guiding autonomous navigation of autonomous vehicles 1205, 1210, 1215, 1220, and 1225 or other vehicles that later travel along the road segment 1200. For example, when the autonomous vehicle road navigation model is executed by a processor included in the navigation system of the vehicle 1205, the model may cause the processor to compare a trajectory determined based on navigation information received from the vehicle 1205 with a predetermined trajectory included in the sparse map 800 to verify and/or correct the current travel route of the vehicle 1205.

In an autonomous vehicle road navigation model, the geometry of a road feature or target trajectory may be encoded by a curve in three-dimensional space. In one embodiment, the curve may be a three-dimensional spline, including one or more splines connecting three-dimensional polynomials. As will be appreciated by those skilled in the art, a spline may be a numerical function that is defined piecewise by a series of polynomials for fitting data. The spline used to fit the three-dimensional geometric data of the road may comprise a linear spline (first order), a quadratic spline (second order), a cubic spline (third order), or any other spline (other order), or a combination thereof. A spline may comprise one or more three-dimensional polynomials of different orders connecting (e.g., fitting) data points of three-dimensional geometric data of a road. In some embodiments, the autonomous vehicle road navigation model may include a three-dimensional spline corresponding to a target trajectory along a common road segment (e.g., road segment 1200) or a lane of road segment 1200.

As described above, the autonomous vehicle road navigation model contained in the sparse map may include other information, such as an identification of at least one road sign along the road segment 1200. The roadmap is visible within the field of view of a camera (e.g., camera 122) mounted on each of vehicles 1205, 1210, 1215, 1220, and 1225. In some embodiments, the camera 122 may capture images of the roadmap. A processor (e.g., processor 180, 190 or processing unit 110) provided on the vehicle 1205 may process the images of the roadmap to extract identification information of the roadmap. The landmark identification information may be stored in the sparse map 800 instead of the actual image of the landmark. The landmark identification information may require much less memory than the actual image. Other sensors or systems (e.g., GPS systems) may also provide some identifying information of the landmark (e.g., the location of the landmark). The road sign may comprise at least one of a traffic sign, an arrow sign, a lane sign, a dashed lane sign, a traffic light, a stop line, a direction sign (e.g. a highway exit sign with an indicated direction, a highway sign with an arrow pointing in a different direction or location), a road sign beacon or a light pole. Road marking beacons represent devices (e.g., RFID devices) installed along road segments that transmit or reflect signals to vehicle-mounted receivers such that when a vehicle passes through the device, the beacons received by the vehicle and the location of the device (e.g., determined from the GPS location of the device) may be used as road markings to be included in the autonomous vehicle road navigation model and/or the sparse map 800.

The identification of the at least one landmark may include a location of the at least one landmark. The location of the landmark may be determined based on location measurements performed using sensor systems (e.g., global positioning system, inertial-based positioning system, landmark beacons, etc.) associated with the plurality of vehicles 1205, 1210, 1215, 1220, and 1225. In some embodiments, the location of the roadmap may be determined by averaging the location measurements detected, collected, or received over multiple drives for the sensor systems on different vehicles 1205, 1210, 1215, 1220, and 1225. For example, vehicles 1205, 1210, 1215, 1220, and 1225 may transmit location measurement data to server 1230, and server 1230 may average the location measurements and use the average location measurements as the location of the roadmap. The location of the roadmap may be continuously refined by measurements received from the vehicle in subsequent driving.

The identification of the landmark may include the size of the landmark. A processor provided on the vehicle (e.g., 1205) may estimate the physical size of the roadmap based on the analysis of the image. The server 1230 may receive multiple estimates of the physical size of the same roadmap from different vehicles through different drives. The server 1230 may average the different estimates to derive the physical size of the roadmap and store that roadmap size in the road model. The physical size estimate may be used to further determine or estimate the distance from the vehicle to the road sign. The distance to the roadmap may be estimated based on the current speed of the vehicle and an extended scale from the position of the roadmap in the image relative to the extended focal point of the camera. For example, the distance to a landmark can be estimated by z=v×dt×r/D, where V is the speed of the vehicle, R is the distance from the landmark at time t1 to the extended focus in the image, and D is the change in distance from the landmark at t1 to t2 in the image. dt represents (t 2-t 1). For example, the distance to a landmark can be estimated by z=v×dt×r/D, where V is the speed of the vehicle, R is the distance between the image landmark and the extended focus, dt is the time interval, and D is the image displacement of the landmark along the epipolar line. Other equations equivalent to the above equation (e.g., z=v×ω/Δω) may be used to estimate the distance to the road sign. Where V is the vehicle speed, ω is the image length (e.g., object width), and Δω is the change in that image length per unit time.

When the physical size of the landmark is known, the distance to the landmark may also be determined based on the following equation: z=f×w/ω, where f is the focal length, W is the size (e.g., height or width) of the landmark, and ω is the number of pixels when the landmark leaves the image. From the above formula, Δz=f×w×Δω/ω can be used ² The change in distance Z is calculated by +f×Δw/ω, where Δw decays to zero by averaging, and where Δω is the number of pixels representing the bounding box accuracy in the image. The value of the estimated physical size of the roadmap may be calculated by averaging multiple observations at the server side. The resulting error in the distance estimation may be small. There are two sources of error that can occur using the above formula, Δw and Δω. Their contribution to the distance error is determined by Δz=f×w×Δω/ω ² +f.DELTA.W/ω. However, Δw decays to zero by averaging; Δz is thus determined by Δω (e.g., inaccuracy of the bounding box in the image).

For landmarks of unknown size, the distance to the landmark can be estimated by tracking feature points on the landmark between successive frames. For example, certain features that appear on the speed limit markers may be tracked between two or more image frames. Based on these tracked features, a distance distribution per feature point may be generated. The distance estimate may be extracted from the distance distribution. For example, the most frequent distance that occurs in the distance distribution may be used as the distance estimate. As another example, an average of the distance distribution may be used as the distance estimate.

FIG. 13 illustrates an example autonomous vehicle road navigation model represented by a plurality of three-dimensional splines 1301, 1302 and 1303. Curves 1301, 1302 and 1303 are shown in fig. 13 for ease of illustration only. Each spline may include one or more three-dimensional polynomials connecting the plurality of data points 1310. Each polynomial may be a first order polynomial, a second order polynomial, a third order polynomial, or a combination of any suitable polynomials having different orders. Each data point 1310 may be associated with navigation information received from vehicles 1205, 1210, 1215, 1220, and 1225. In some embodiments, each data point 1310 may be associated with data related to a road sign (e.g., size, location, and identification information of the road sign) and/or a road signature profile (e.g., road geometry, road roughness profile, road curvature profile, road width profile). In some embodiments, some data points 1310 may be associated with data related to road signs, while other data points may be associated with data related to road signature profiles.

Fig. 14 shows raw location data 1410 (e.g., GPS data) received from five separate drives. One drive may be separated from another drive when passed by a separate vehicle at the same time, by the same vehicle at a separate time, or by a separate vehicle at a separate time. To account for errors in the location data 1410 and different locations of vehicles within the same lane (e.g., one vehicle may be driving closer to the left side of the lane than another vehicle), the server 1230 may use one or more statistical techniques to generate the map profile 1420 to determine whether the change in the raw location data 1410 represents an actual discrepancy or statistical error. Each path within the sketch 1420 may in turn be linked to the original data 1410 that forms the path. For example, the path between A and B within the sketch 1420 is linked to raw data 1410 from drives 2, 3, 4, and 5, but not from drive 1. The sketch 1420 may not be sufficiently detailed to be used to navigate a vehicle (e.g., because it combines driving from multiple lanes on the same road, unlike the splines described above), but may provide useful topology information and may be used to define intersections.

Fig. 15 shows an example by which additional details may be generated for sparse maps within map sketch segments (e.g., segments a through B within sketch 1420). As shown in fig. 15, the data (e.g., self-movement data, road marking data, etc.) may be shown as a location S (or S) along the drive ₁ Or S ₂ ) Is a function of (2). The server 1230 may identify the roadmap of the sparse map by identifying unique matches between roadmaps 1501, 1503, and 1505 of the drive 1510 and roadmaps 1507 and 1509 of the drive 1520. Such a matching algorithm may produce identifications of landmarks 1511, 1513, and 1515. Those skilled in the art will appreciate that other matching algorithms may be used. For example, probability optimization may be used in place of or in combination with unique matching. The server 1230 may be driven in longitudinal alignment to align matching roadmarks. For exampleThe server 1230 may select one drive (e.g., drive 1520) as the reference drive and then offset and/or stretch the other drive(s) (e.g., drive 1510) elastically for alignment.

Fig. 16 shows an example of aligned landmark data for use in a sparse map. In the example of fig. 16, the road sign 1610 includes a road sign. The example of fig. 16 further illustrates data from multiple drives 1601, 1603, 1605, 1607, 1609, 1611, and 1613. In the example of fig. 16, the data from the drive 1613 consists of "virtual" roadways, and the server 1230 may identify it as no drive 1601, 1603, 1605, 1607, 1609, and 1611 includes identification of roadways in the vicinity of the identified roadways in the drive 1613. Accordingly, the server 1230 may accept the potential roadmap when the ratio of the image in which the roadmap appears to the image in which the roadmap does not appear exceeds a threshold value and/or may reject the potential roadmap when the ratio of the image in which the roadmap does not appear to the image in which the roadmap appears exceeds a threshold value.

Fig. 17 illustrates a system 1700 for generating driving data that may be used to crowd source sparse maps. As shown in fig. 17, the system 1700 may include a camera 1701 and a positioning device 1703 (e.g., a GPS locator). The camera 1701 and positioning device 1703 may be mounted on a vehicle (e.g., one of vehicles 1205, 1210, 1215, 1220, and 1225). The camera 1701 may generate a plurality of data of various types, such as self-movement data, traffic sign data, road data, and the like. The camera data and the position data may be segmented into driver segments 1705. For example, the driver segments 1705 may each have camera data and position data from a drive of less than 1 km.

In some embodiments, system 1700 may remove redundancy in driver 1705. For example, if a landmark appears in multiple images from the camera 1701, the system 1700 may strip the redundant data such that the driver section 1705 contains only one copy of the location of the landmark and any metadata related to the landmark. As another example, if a lane marker appears in multiple images from the camera 1701, the system 1700 may strip the redundant data such that the driver's leg 1705 contains only one copy of the location of the lane marker and any metadata related to the lane marker.

The system 1700 also includes a server (e.g., server 1230). The server 1230 may receive the driver segments 1705 from the vehicle and reassemble the driver segments 1705 into a single drive 1707. Such an arrangement may allow for reduced bandwidth requirements when transmitting data between the vehicle and the server, while also allowing the server to store data related to the overall driving.

Fig. 18 shows the system 1700 of fig. 17 further configured for crowdsourcing sparse maps. As in fig. 17, the system 1700 includes a vehicle 1810 that captures driving data using, for example, a camera (which generates, for example, self-movement data, traffic sign data, road data, etc.) and a positioning device (e.g., a GPS locator). As in fig. 17, the vehicle 1810 segments the collected data into driving segments (shown in fig. 18 as "DS1 1", "DS2 1", "DSN 1"). The server 1230 then receives the driving segments, and reconstructs driving (shown as "driving 1" in fig. 18) from the received segments.

As further shown in fig. 18, the system 1700 also receives data from additional vehicles. The vehicle 1820 also captures driving data using, for example, a video camera (which generates, for example, self-movement data, traffic sign data, road data, etc.) and a positioning device (e.g., a GPS locator). Similar to vehicle 1810, vehicle 1820 segments the collected data into driving segments (shown in fig. 18 as "DS1 2", "DS2 2", "DSN 2"). The server 1230 then receives the driving segments, and reconstructs driving (shown as "driving 2" in fig. 18) from the received segments. Any number of additional vehicles may be used. For example, fig. 18 also includes "car N" which captures driving data, segments it into driving segments (shown as "DS 1N", "DS 2N", "DSN" in fig. 18), and sends it to server 1230 for reconstruction into driving (shown as "driving N" in fig. 18).

As shown in fig. 18, server 1230 may construct a sparse map using reconstructed drives (e.g., "drive 1", "drive 2", and "drive N") collected from multiple vehicles (e.g., "car 1" (again labeled 1810), "car 2" (again labeled car 1820), and "car N").

FIG. 19 is a flowchart illustrating an example process 1900 of generating a sparse map for autonomous vehicle navigation along a road segment. Process 1900 may be performed by one or more processing devices included in server 1230.

Process 1900 may include receiving a plurality of images acquired as one or more vehicles traverse the road segment (step 1905). The server 1230 may receive images from cameras contained within one or more of the vehicles 1205, 1210, 1215, 1220, and 1225. For example, camera 122 may capture one or more images of the environment surrounding vehicle 1205 as vehicle 1205 travels along road segment 1200. In some embodiments, server 1230 may also receive stripped-down image data that is redundancy-removed by a processor on vehicle 1205, as described above with respect to fig. 17.

The process 1900 may further include identifying at least one line representation of a road surface feature extending along the road segment based on the plurality of images (step 1910). Each line representation may represent a path along a road segment that substantially corresponds to a road surface feature. For example, the server 1230 may analyze the environmental image received from the camera 122 to identify a road edge or lane marker and determine a travel trajectory along the road segment 1200 associated with the road edge or lane marker. In some embodiments, the trajectory (or line representation) may include a spline, a polynomial representation, or a curve. The server 1230 may determine the travel track of the vehicle 1205 based on the camera self-motions (e.g., three-dimensional translational and/or three-dimensional rotational motions) received at step 1905.

The process 1900 may also include identifying a plurality of road signs associated with the road segment based on the plurality of images (step 1910). For example, server 1230 may analyze the environmental image received from camera 122 to identify one or more roadmarks (e.g., road signs along road segment 1200). The server 1230 may identify roadsigns using analysis of multiple images acquired as one or more vehicles traverse the road segment. To enable crowdsourcing, the analysis may include rules related to accepting and rejecting possible roadways for the associated road segments. For example, the analysis may include accepting the potential roadmap when the ratio of the image in which the roadmap appears to the image in which the roadmap does not appear exceeds a threshold value, and/or rejecting the potential roadmap when the ratio of the image in which the roadmap does not appear to the image in which the roadmap appears exceeds a threshold value.

Process 1900 may include other operations or steps performed by server 1230. For example, the navigation information may include target trajectories for vehicles traveling along the road segment, and the process 1900 may include clustering, by the server 1230, vehicle trajectories related to a plurality of vehicles traveling on the road segment, and determining the target trajectories based on the clustered vehicle trajectories, as discussed in more detail below. Clustering the vehicle trajectories may include clustering, by the server 1230, a plurality of trajectories related to vehicles traveling on the road segment into a plurality of clusters based on at least one of an absolute heading of the vehicle or a lane assignment of the vehicle. Generating the target trajectory may include averaging, by the server, the clustered trajectories. As another example, process 1900 may include aligning the data received at step 1905. As described above, other processes or steps performed by the server 1230 may also be included in the process 1900.

The disclosed systems and methods may include other features. For example, the disclosed system may use local coordinates instead of global coordinates. For autonomous driving, some systems may present data in world coordinates. For example, longitude and latitude coordinates of the earth's surface may be used. To use the map for maneuvers, the host vehicle may determine its position and orientation relative to the map. It seems natural to use an onboard GPS device to locate vehicles on a map and to find rotational transformations (e.g. north, east and down) between the body reference frame and the world reference frame. Once the vehicle body reference frame is aligned with the map reference frame, the intended route may be expressed in the vehicle body reference frame and the steering command may be calculated or generated.

The disclosed systems and methods may employ a low footprint model to enable autonomous vehicle navigation (e.g., steering control), which may be collected by the autonomous vehicle itself without resorting to expensive survey equipment. To support autonomous navigation (e.g., steering applications), the road model may include a sparse map having the geometry of the road, its lane structure, and road signs, which may be used to determine the location or position of the vehicle along the trajectories contained in the model. As described above, the generation of the sparse map may be performed by a remote server that communicates with and receives data from vehicles traveling on roads. The data may include sensed data, a reconstructed track based on the sensed data, and/or a recommended track that may represent a modified reconstructed track. As described below, the server may transmit the model back to the vehicle or other vehicles that later travel on the road to aid autonomous navigation.

Fig. 20 shows a block diagram of a server 1230. The server 1230 may include a communication unit 2005, which communication unit 2005 may include hardware components (e.g., communication control circuitry, switches, and antennas) and software components (e.g., communication protocols, computer code). For example, the communication unit 2005 may include at least one network interface. The server 1230 may communicate with vehicles 1205, 1210, 1215, 1220, and 1225 through communication unit 2005. For example, server 1230 may receive navigation information transmitted from vehicles 1205, 1210, 1215, 1220, and 1225 through communication unit 2005. The server 1230 may distribute the autonomous vehicle road navigation model to one or more autonomous vehicles via the communication unit 2005.

The server 1230 may include at least one non-transitory storage medium 2010, such as a hard disk drive, compact disc, magnetic tape, or the like. The storage 1410 may be configured to store data, such as navigation information received from vehicles 1205, 1210, 1215, 1220, and 1225 and/or autonomous vehicle road navigation models generated by the server 1230 based on the navigation information. Storage 2010 may be configured to store any other information, such as a sparse map (e.g., sparse map 800 described above with respect to fig. 8).

In addition to, or in lieu of, storage 2010, server 1230 may include memory 2015. Memory 2015 may be similar to or different from memory 140 or 150. Memory 2015 may be non-transitory memory such as flash memory, random access memory, or the like. The memory 2015 may be configured to store data, such as computer code or instructions executable by a processor (e.g., the processor 2020), map data (e.g., data of the sparse map 800), autonomous vehicle road navigation models, and/or navigation information received from the vehicles 1205, 1210, 1215, 1220, and 1225.

The server 1230 may include at least one processing device 2020, the processing device 2020 being configured to execute computer code or instructions stored in the memory 2015 to perform various functions. For example, the processing device 2020 may analyze the navigation information received from the vehicles 1205, 1210, 1215, 1220, and 1225 and generate an autonomous vehicle road navigation model based on the analysis. The processing device 2020 may control the communication unit 1405 to distribute the autonomous vehicle road navigation model to one or more autonomous vehicles (e.g., one or more of the vehicles 1205, 1210, 1215, 1220, and 1225 or any vehicle that is later traveling on the road segment 1200). The processing device 2020 may be similar to or different from the processors 180, 190 or the processing unit 110.

Fig. 21 shows a block diagram of a memory 2015, which memory 2015 may store computer code or instructions for performing one or more operations of generating a road navigation model for use in autonomous vehicle navigation. As shown in fig. 21, the memory 2015 may store one or more modules for performing operations to process vehicle navigation information. For example, the memory 2015 may include a model generation module 2105 and a model distribution module 2110. The processor 2020 may execute instructions stored in either of the modules 2105 and 2110 contained in the memory 2015.

The model generation module 2105 can store instructions which, when executed by the processor 2020, can generate at least a portion of an autonomous vehicle road navigation model for a common road segment (e.g., road segment 1200) based on navigation information received from vehicles 1205, 1210, 1215, 1220, and 1225. For example, in generating the autonomous vehicle road navigation model, the processor 2020 may cluster vehicle trajectories along the common road segment 1200 into different clusters. The processor 2020 may determine a target trajectory along the common road segment 1200 based on the clustered vehicle trajectories of each of the different clusters. Such operations may include finding a mean or average track of the clustered vehicle tracks in each cluster (e.g., by averaging data representing the clustered vehicle tracks). In some embodiments, the target trajectory may be associated with a single lane of the common road segment 1200.

The road model and/or sparse map may store trajectories associated with the road segments. These trajectories may be referred to as target trajectories that are provided to an autonomous vehicle for autonomous navigation. The target trajectory may be received from a plurality of vehicles or may be generated based on actual or recommended trajectories (with some modified actual trajectories) received from the plurality of vehicles. New trajectories received from other vehicles may be employed to continuously update (or average) target trajectories contained in a road model or sparse map.

Vehicles traveling on road segments may collect data via various sensors. The data may include road signs, road signature profiles, vehicle movements (e.g., accelerometer data, speed data), vehicle positions (e.g., GPS data), and the actual trajectory itself may be reconstructed or the data transmitted to a server that will reconstruct the actual trajectory of the vehicle. In some embodiments, the vehicle may transmit track-related data (e.g., curves in any frame of reference), road marking data, and lane alignment along the travel path to the server 1230. Various vehicles traveling along the same road segment while driving multiple times may have different trajectories. The server 1230 may identify a route or track associated with each lane from tracks received from vehicles through a clustering process.

Fig. 22 illustrates a process of clustering vehicle trajectories associated with vehicles 1205, 1210, 1215, 1220, and 1225 for determining a target trajectory for a common road segment (e.g., road segment 1200). The target track or tracks determined from the clustering process may be included in an autonomous vehicle road navigation model or sparse map 800. In some embodiments, vehicles 1205, 1210, 1215, 1220, and 1225 traveling along segment 1200 may communicate multiple trajectories 2200 to server 1230. In some embodiments, server 1230 may generate trajectories based on road signs, road geometries, and vehicle motion information received from vehicles 1205, 1210, 1215, 1220, and 1225. To generate an autonomous vehicle road navigation model, the server 1230 may cluster the vehicle trajectories 1600 into a plurality of clusters 2205, 2210, 2215, 2220, and 2230, as shown in fig. 22.

Clustering may be performed using various criteria. In some embodiments, the absolute heading of all driving pairs in the cluster along the road segment 1200 may be similar. Absolute heading may be obtained from GPS signals received by vehicles 1205, 1210, 1215, 1220, and 1225. In some embodiments, dead reckoning may be used to obtain absolute heading. As will be appreciated by those skilled in the art, dead reckoning may be used to determine the current location, and thus the heading of vehicles 1205, 1210, 1215, 1220, and 1225, by using previously determined locations, estimated speeds, and the like. Trajectories clustered by absolute heading may be useful for identifying routes along roads.

In some embodiments, lane assignments (e.g., in the same lane before and after the intersection) for all driving in the cluster versus driving along the road segment 1200 may be similar. The clustered trajectories by lane assignment may be useful for identifying lanes along a road. In some embodiments, both criteria (e.g., absolute heading and lane assignment) may be used for clustering.

In each cluster 2205, 2210, 2215, 2220, 2225, and 2230, the tracks may be averaged to obtain a target track associated with the particular cluster. For example, trajectories from multiple drives associated with the same lane cluster may be averaged. The average track may be a target track associated with a particular lane. To average the trajectory clusters, the server 1230 may select a reference frame for any trajectory C0. For all other tracks (C1..cn), server 1230 may look up a rigid transformation mapping Ci to C0, where i=1, 2..cn, n, where n is a positive integer, corresponding to the total number of tracks contained in the cluster. The server 1230 may calculate an average curve or trajectory in the C0 reference frame.

In some embodiments, the roadmap may define an arc length match between different maneuvers, which may be used for alignment of the track with the lane. In some embodiments, lane markings before and after the intersection may be used for alignment of the track with the lane.

To assemble lanes from the track, the server 1230 may select a frame of reference for any lane. The server 1230 may map the partially overlapping lanes to the selected frame of reference. The server 1230 may continue mapping until all lanes are in the same frame of reference. Lanes adjacent to each other may be aligned as if they were the same lane, and they may later be laterally offset.

Road signs identified along road segments may be mapped to a common reference frame, first at the lane level and then at the intersection level. For example, the same roadmap may be identified multiple times by multiple vehicles in multiple drives. The data relating to the same roadmap received in different driving may be slightly different. Such data may be averaged and mapped to the same reference frame, e.g., the C0 reference frame. Additionally or alternatively, changes in data of the same roadmap received during multiple driving may be calculated.

In some embodiments, each lane of the road segment 120 may be associated with a target trajectory and certain roadways. The target trajectory or a plurality of such target trajectories may be included in an autonomous vehicle road navigation model that may later be used by other autonomous vehicles traveling along the same road segment 1200. Road signs identified by vehicles 1205, 1210, 1215, 1220, and 1225 while traveling along road segment 1200 may be recorded in association with the target trajectory. The data of the target trajectory and roadmap may be updated continuously or periodically with new data received from other vehicles in subsequent driving.

For autonomous vehicle positioning, the disclosed systems and methods may use an extended kalman filter. The position of the vehicle may be determined based on three-dimensional position data and/or three-dimensional orientation data by self-moving integration of predictions of future positions in front of the current position of the vehicle. The positioning of the vehicle may be corrected or adjusted by image viewing of the roadmap. For example, when the vehicle detects a landmark within the image captured by the camera, the landmark may be compared to known landmarks stored in the road model or sparse map 800. The known roadmap may have a known location (e.g., GPS data) along the road model and/or target trajectory stored in the sparse map 800. Based on the current speed and the image of the landmark, a distance from the vehicle to the landmark may be estimated. The position along the target trajectory may be adjusted based on the distance to the roadmap and the known position of the roadmap (stored in the road model or sparse map 800). The road model and/or the location/positioning data (e.g., average from multiple drives) of the roadmap stored in the sparse map 800 may be assumed to be accurate.

In some embodiments, the disclosed system may form a closed loop subsystem in which an estimate of a six degree of freedom position (e.g., three-dimensional position data plus three-dimensional orientation data) of a vehicle may be used to navigate an autonomous vehicle (e.g., steering its steering wheel) to reach a desired point (e.g., 1.3 seconds in front of being stored). The measured data from the steering and actual navigation can in turn be used to estimate the six degree of freedom position.

In some embodiments, poles along the road (e.g., lampposts and power or cable poles) may be used as road signs for locating vehicles. Other landmarks (e.g., traffic signs, traffic lights, arrows on roads, stop lines) may also be used as landmarks for locating vehicles, as may static features or signatures of objects along road segments. When the pole is used for positioning, an x-view of the pole (i.e., from the perspective of the vehicle) may be used instead of a y-view (i.e., distance to the pole) because the bottom of the pole may be obscured and sometimes they are not on the road plane.

Fig. 23 illustrates a navigation system of a vehicle that may be used for autonomous navigation using crowd-sourced sparse maps. For ease of illustration, the vehicle is referred to as vehicle 1205. The vehicle shown in fig. 23 may be any other vehicle disclosed herein, including, for example, vehicles 1210, 1215, 1220, and 1225, as well as vehicle 200 shown in other embodiments. As shown in fig. 12, vehicle 1205 may communicate with a server 1230. The vehicle 1205 may include an image capture device 122 (e.g., a camera 122). The vehicle 1205 may include a navigation system 2300 configured to provide navigation guidance for the vehicle 1205 to travel on a road (e.g., road segment 1200). The vehicle 1205 may also include other sensors such as a speed sensor 2320 and an accelerometer 2325. The speed sensor 2320 may be configured to detect a speed of the vehicle 1205. The accelerometer 2325 may be configured to detect acceleration or deceleration of the vehicle 1205. The vehicle 1205 shown in fig. 23 may be an autonomous vehicle, and the navigation system 2300 may be used to provide navigation guidance for autonomous driving. Alternatively, the vehicle 1205 may also be a non-autonomous, manually controlled vehicle, and the navigation system 2300 may still be used to provide navigation guidance.

The navigation system 2300 can include a communication unit 2305, the communication unit 2305 configured to communicate with a server 1230 via a communication path 1235. The navigation system 2300 can also include a GPS unit 2310, the GPS unit 2310 configured to receive and process GPS signals. The navigation system 2300 can further include at least one processor 2315, the processor 2315 configured to process data, such as GPS signals, map data from a sparse map 800 (which sparse map 800 can be stored on a storage device provided on the vehicle 1205 and/or received from the server 1230), road geometry sensed by the road profile sensor 2330, images captured by the camera 122, and/or autonomous vehicle road navigation models received from the server 1230. The road profile sensor 2330 may include different types of devices for measuring different types of road profiles (e.g., road roughness, road width, road elevation, road curvature, etc.). For example, the road profile sensor 2330 may include a device that measures movement of the suspension of the vehicle 2305 to derive a road roughness profile. In some embodiments, the road profile sensor 2330 may include a radar sensor to measure the distance from the vehicle 1205 to both sides of the road (e.g., the barrier on both sides of the road), thereby measuring the width of the road. In some embodiments, the road profile sensor 2330 may include a device configured to measure the elevation of a road. In some embodiments, the road profile sensor 2330 may include a device configured to measure road curvature. For example, a camera (e.g., camera 122 or another camera) may be used to capture an image of the road showing the curvature of the road. Vehicle 1205 may use such images to detect road curvature.

The at least one processor 2315 may be programmed to receive at least one environmental image associated with the vehicle 1205 from the camera 122. The at least one processor 2315 may analyze the at least one environmental image to determine navigational information related to the vehicle 1205. The navigation information may include a trajectory related to travel of the vehicle 1205 along the road segment 1200. The at least one processor 2315 may determine a trajectory based on the motion (e.g., three-dimensional translational and three-dimensional rotational motion) of the camera 122 (and thus the vehicle). In some embodiments, the at least one processor 2315 may determine translational and rotational movement of the camera 122 based on analysis of a plurality of images acquired by the camera 122. In some embodiments, the navigation information may include lane assignment information (e.g., in which lane the vehicle 1205 is traveling along the road segment 1200). The navigation information transmitted from the vehicle 1205 to the server 1230 may be used by the server to generate and/or update an autonomous vehicle road navigation model that may be transmitted back from the server 1230 to the vehicle 1205 for providing autonomous navigation guidance for the vehicle 1205.

The at least one processor 2315 may also be programmed to communicate navigation information from the vehicle 1205 to the server 1230. In some embodiments, navigation information may be transmitted to the server 1230 along with road information. The road location information may include at least one of a GPS signal, road sign information, road geometry, lane information, etc., received by the GPS unit 2310. The at least one processor 2315 may receive from the server 1230 an autonomous vehicle road navigation model or a portion of the model. The autonomous vehicle road navigation model received from the server 1230 may include at least one update based on navigation information transmitted from the vehicle 1205 to the server 1230. The portion of the model transmitted from server 1230 to vehicle 1205 may include an updated portion of the model. The at least one processor 2315 may cause at least one navigational maneuver (e.g., turn, brake, accelerate, pass another vehicle, etc.) by the vehicle 1205 based on the received autonomous vehicle road navigation model or an updated portion of the model.

The at least one processor 2315 may be configured to communicate with various sensors and components included in the vehicle 1205, including the communication unit 1705, the GPS unit 2315, the camera 122, the speed sensor 2320, the accelerometer 2325, and the road profile sensor 2330. The at least one processor 2315 may collect information or data from the various sensors and components and transmit the information or data to the server 1230 through the communication unit 2305. Alternatively or in addition, various sensors or components of the vehicle 1205 may also communicate with the server 1230 and communicate data or information collected by the sensors or components to the server 1230.

In some embodiments, vehicles 1205, 1210, 1215, 1220, and 1225 may communicate with each other and may share navigational information with each other such that at least one of vehicles 1205, 1210, 1215, 1220, and 1225 may use crowd sourcing to generate an autonomous vehicle road navigation model, for example, based on information shared by other vehicles. In some embodiments, vehicles 1205, 1210, 1215, 1220, and 1225 may share navigation information with each other, and each vehicle may update its own autonomous vehicle road navigation model provided in the vehicle. In some embodiments, at least one of vehicles 1205, 1210, 1215, 1220, and 1225 (e.g., vehicle 1205) may be used as a hub vehicle. The at least one processor 2315 of the hub vehicle (e.g., vehicle 1205) may perform some or all of the functions performed by the server 1230. For example, the at least one processor 2315 of the hub vehicle may communicate with other vehicles and receive navigational information from the other vehicles. The at least one processor 2315 of the hub vehicle may generate an autonomous vehicle road navigation model or update to the model based on shared information received from other vehicles. The at least one processor 2315 of the hub vehicle may transmit the autonomous vehicle road navigation model or updates to the model to other vehicles for providing autonomous navigation guidance.

Sparse map-based guideNavigation system

As previously described, the autonomous vehicle road navigation model including the sparse map 800 may include a plurality of mapped lane markers and a plurality of mapped objects/features associated with the road segment. As discussed in more detail below, these mapped lane markers, objects, and features may be used when navigating an autonomous vehicle. For example, in some embodiments, the map objects and features may be used to locate the host vehicle relative to the map (e.g., relative to the map target track). The mapped lane markings may be used (e.g., as an inspection) to determine a lateral position and/or orientation relative to the planned or target trajectory. With this location information, the autonomous vehicle may be able to adjust the heading direction to match the direction of the target track at the determined location.

The vehicle 200 may be configured to detect lane markings in a given road segment. A road segment may include any indicia on a road for guiding vehicle traffic on the road. For example, the lane markings may be continuous or broken lines distinguishing edges of the driving lane. The lane markings may also include double lines, such as double continuous lines, double dashed lines, or a combination of continuous and dashed lines, indicating, for example, whether an overtaking is permitted in an adjacent lane. The lane markings may also include highway entrance and exit markings indicating a decelerating lane, such as an exit ramp, or a dashed line indicating that the lane is turning only or that the lane is ending. The markings may further indicate work areas, temporary lane changes, travel paths through intersections, intermediate belts, special lanes (e.g., bicycle lanes, HOV lanes, etc.), or other various markings (e.g., sidewalks, speed bumps, railroad crossings, stop lines, etc.).

The vehicle 200 may use cameras (e.g., image capture devices 122 and 124 included in the image acquisition unit 120) to capture images of surrounding lane markings. The vehicle 200 may analyze the images to detect a point location associated with the lane marker based on features identified within one or more of the captured images. These point locations may be uploaded to a server to represent lane markings in the sparse map 800. Depending on the position and field of view of the camera, lane markings may be detected simultaneously on both sides of the vehicle from a single image. In other embodiments, different cameras may be used to capture images of multiple sides of the vehicle. Instead of uploading an actual image of the lane markers, the markers may be stored in the sparse map 800 as splines or a series of points, thus reducing the size of the sparse map 800 and/or the data that must be uploaded remotely by the vehicle.

24A-24D illustrate exemplary point locations that may be detected by the vehicle 200 to represent particular lane markings. Similar to the roadmap described above, the vehicle 200 may use various image recognition algorithms or software to identify the point locations within the captured image. For example, the vehicle 200 may identify a series of edge points, corner points, or various other point locations associated with a particular lane marker. Fig. 24A shows a continuous lane marker 2410 that may be detected by the vehicle 200. The lane markings 2410 may represent the outside edge of the road, as represented by continuous white lines. As shown in fig. 24A, the vehicle 200 may be configured to detect a plurality of edge location points 2411 along lane markers. The location points 2411 may be collected at any interval sufficient to create mapped lane markings in a sparse map to represent lane markings. For example, lane markings may be represented by one point per meter of detected edge, one point per five meters of detected edge, or at other suitable spacing. In some embodiments, the spacing may be determined by other factors other than at set intervals, such as the point based on the highest confidence rating of the vehicle 200 having the location of the detection point. Although fig. 24A shows edge position points on the inner edge of the lane markings 2410, the points may be collected at the outer edge of the line or along both edges. Further, although a single line is shown in fig. 24A, similar edge points may be detected for a bicontinuous line. For example, detection point 2411 may be detected along an edge of one or both of the continuous lines.

The vehicle 200 may also represent lane markings in different ways depending on the type or shape of the lane markings. Fig. 24B shows an exemplary dashed lane marker 2420 that may be detected by the vehicle 200. Instead of identifying edge points as in fig. 24A, the vehicle may detect a series of corner points 2421 representing the corners of the lane dashed line to define the full boundary of the dashed line. Although fig. 24B shows each corner marked with a given dashed line being located, the vehicle 200 may detect or upload a subset of the points shown in the figure. For example, the vehicle 200 may detect the leading edge or angle of a given dashed mark, or may detect the two corner points closest to the interior of the lane. Further, not every dashed mark may be captured, e.g., vehicle 200 may capture and/or record points marked by a dashed line representing a sample of the dashed mark (e.g., every other, every third, every fifth, etc.) or at a predefined pitch (e.g., every meter, every five meters, every ten meters, etc.). Corner points may also be detected for similar lane markings, such as markings showing lanes for exit ramps, specific lane ending, or other various lane markings that may have detectable corner points. Corner points may also be detected for lane markings consisting of double dashed lines or a combination of continuous and dashed lines.

In some embodiments, points uploaded to the server to generate mapped lane markings may represent points other than detected edge points or corner points. FIG. 24C illustrates a series of points that may represent the centerline of a given lane marker. For example, the continuous lane 2410 may be represented by a centerline point 2441 along the centerline 2440 of the lane marker. In some embodiments, the vehicle 200 may be configured to detect these center points using various image recognition techniques, such as Convolutional Neural Networks (CNNs), scale-invariant feature transforms (SIFTs), oriented gradient Histogram (HOG) features, or other techniques. Alternatively, the vehicle 200 may detect other points (e.g., edge points 2411 shown in fig. 24A), and may calculate the centerline points 2441, for example, by detecting points along each edge and determining midpoints between the edge points. Similarly, the dashed lane marker 2420 may be represented by a centerline point 2451 along the centerline 2450 of the lane marker. The centerline points may be located at the edges of the dashed lines as shown in fig. 24C, or at various other locations along the centerline. For example, each dashed line may be represented by a single point in the geometric center of the dashed line. The points may also be spaced apart at predetermined intervals along the centerline (e.g., every meter, 5 meters, 10 meters, etc.). The centerline point 2451 may be directly detected by the vehicle 200 or may be calculated based on other detected reference points (e.g., the corner point 2421, as shown in fig. 24B). The center line may also be used to represent other lane marker types (e.g., double lines) using similar techniques as described above.

In some embodiments, the vehicle 200 may identify points representing other features, such as vertices between two intersecting lane markers. Fig. 24D shows an exemplary point representing the intersection between two lane markings 2460 and 2465. The vehicle 200 may calculate a vertex 2466 representing the intersection between the two lane markings. For example, one of the lane markings 2460 or 2465 may represent a train passing area or another passing area in the road segment. Although lane markings 2460 and 2465 are shown intersecting perpendicularly to each other, various other configurations may be detected. For example, lane markings 2460 and 2465 may intersect at other angles, or one or both of the lane markings may terminate at vertex 2466. Similar techniques may also be applied to intersections between dashed lines or other lane marker types. In addition to the vertex 2466, various other points 2467 may also be detected, providing other information about the orientation of the lane markings 2460 and 2465.

The vehicle 200 may associate real world coordinates with each detected point of the lane marker. For example, a location identifier may be generated that includes coordinates of each point to upload to a server for mapping lane markings. The location identifier may further include other identifying information about the point, including whether the point represents a corner point, an edge point, a center point, etc. Accordingly, the vehicle 200 may be configured to determine the real world location of each point based on the analysis of the images. For example, the vehicle 200 may detect other features in the image (e.g., the various roadmarks described above) to locate the real world location of the lane markers. This may involve determining the position of the lane marker in the image relative to the detected landmark, or determining the position of the vehicle based on the detected landmark and then determining the distance from the vehicle (or the target trajectory of the vehicle) to the lane marker. When the roadmap is not available, the location of the lane marking point may be determined relative to the location of the vehicle determined based on dead reckoning. The real world coordinates contained in the location identifier may be represented as absolute coordinates (e.g., latitude/longitude coordinates) or may be relative to other features, e.g., based on a longitudinal location along the target track and a lateral distance from the target track. The location identifiers may then be uploaded to a server for use in generating mapped lane markings in a navigation model (e.g., sparse map 800). In some embodiments, the server may construct a spline representing the lane markings of the road segments. Alternatively, the vehicle 200 may generate a spline and upload it to a server for recording in the navigation model.

Fig. 24E shows an exemplary navigation model or sparse map including corresponding road segments that map lane markings. The sparse map may include target trajectories 2475 to be followed by vehicles along the road segment. As described above, the target trajectory 2475 may represent an ideal path taken by the vehicle while traveling on a corresponding road segment, or may be located elsewhere on the road (e.g., the centerline of the road, etc.). The target trajectory 2475 may be calculated by various methods described above, such as based on an aggregation (e.g., weighted combination) of two or more reconstructed trajectories of vehicles traversing the same road segment.

In some embodiments, the target trajectory may be generated equally for all vehicle types and for all roads, vehicles, and/or environmental conditions. However, in other embodiments, various other factors or variables may also be considered in generating the target trajectory. Different target trajectories may be generated for different types of vehicles (e.g., private cars, light trucks, and full trailers). For example, a target trajectory with a smaller turning radius may be generated for a small private car as compared to a larger semi-trailer. In some embodiments, roads, vehicles, and environmental conditions are also contemplated. For example, different target trajectories may be generated for different road conditions (e.g., wet, ice, dry, etc.), vehicle conditions (e.g., tire conditions or estimated tire conditions, braking conditions or estimated braking conditions, amount of fuel remaining, etc.), or environmental factors (e.g., time of day, visibility, weather, etc.). The target trajectory may also depend on one or more aspects or characteristics of the particular road segment (e.g., speed limit, frequency and size of turns, grade, etc.). In some embodiments, various user settings may also be used to determine the target trajectory, such as setting a driving mode (e.g., anticipated driving aggressiveness, economy mode, etc.).

The sparse map may also include mapped lane markings 2470 and 2480 representing lane markings along the road segment. The mapped lane markings may be represented by a plurality of location identifiers 2471 and 2481. As described above, the location identifier may include a location in real world coordinates of a point associated with the detected lane marker. Similar to the target trajectory in the model, the lane markings may also include elevation data and may be represented as curves in three-dimensional space. For example, the curve may be a spline connecting a three-dimensional polynomial of appropriate order, and the curve may be calculated based on the location identifier. The mapped lane markings may also include other information or metadata related to the lane markings, such as an identifier of the type of lane marking (e.g., between two lanes with the same direction of travel, between two lanes with opposite directions of travel, edges of a road, etc.) and/or other characteristics of the lane marking (e.g., continuous, broken line, single line, double line, yellow, white, etc.). In some embodiments, the mapped lane markers may be continuously updated within the model, for example, using crowdsourcing techniques. The same vehicle may upload the location identifier during multiple occasions when traveling on the same road segment, or the data may be selected from multiple vehicles (e.g., 1205, 1210, 1215, 1220, and 1225) traveling on the road segment at different times. The sparse map 800 may then be updated or refined based on subsequent location identifiers received from the vehicle and stored in the system. When the mapped lane markers are updated and refined, the updated road navigation model and/or the sparse map may be distributed to multiple autonomous vehicles.

Generating mapped lane markings in the sparse map may also include detecting and/or mitigating errors based on anomalies in the image or in the actual lane markings themselves. Fig. 24F shows an exemplary anomaly 2495 associated with detecting lane markings 2490. Anomalies 2495 may occur in images captured by the vehicle 200, such as objects from camera views obstructing lane markings, debris on lenses, and the like. In some cases, the anomalies may be due to the lane markings themselves, which may be damaged or worn or partially covered by, for example, dust, debris, water, snow, or other materials on the road. Anomaly 2495 can cause false point 2491 to be detected by vehicle 200. The sparse map 800 may provide correct mapped lane markings while excluding errors. In some embodiments, the vehicle 200 may detect the error point 2491, for example, by detecting an anomaly 2495 in the image or by identifying errors based on detected lane marker points before and after the anomaly. Based on detecting the anomaly, the vehicle may ignore the point 2491 or may adjust it to coincide with other detected points. In other embodiments, errors may be corrected after points have been uploaded, for example, by determining that points are outside of an expected threshold based on other points uploaded during the same journey or based on an aggregation of data from previous journeys along the same road segment.

The navigation model and/or mapped lane markings in the sparse map may also be used for navigation by autonomous vehicles traveling on the corresponding road. For example, a vehicle navigating along a target track may periodically align itself with the target track using mapped lane markings in a sparse map. As described above, between roadways, a vehicle may navigate based on dead reckoning, where the vehicle uses sensors to determine its own motion and estimate its position relative to a target trajectory. Errors can accumulate over time and vehicle position determination relative to the target trajectory can become increasingly inaccurate. Accordingly, the vehicle may use lane markers (and their known locations) that appear in the sparse map 800 to reduce dead reckoning induced errors in position determination. In this way, the identified lane markers contained in the sparse map 800 may be used as navigation anchor points from which the precise location of the vehicle relative to the target trajectory may be determined.

FIG. 25A illustrates an exemplary image 2500 of a vehicle surroundings that can be used for map lane marker-based navigation. The image 2500 may be captured, for example, by the vehicle 200 through image capture devices 122 and 124 included in the image acquisition unit 120. The image 2500 may include an image of at least one lane marker 2510, as shown in fig. 25A. The image 2500 may also include one or more roadways 2521 for navigation as described above, such as road signs. Some elements shown in fig. 25A are also shown for reference, such as elements 2511, 2530, and 2520 that are not present in captured image 2500 but are detected and/or determined by vehicle 200.

Using the various techniques described above with respect to fig. 24A-D and 24F, the vehicle may analyze the image 2500 to identify lane markings 2510. Various points 2511 may be detected as features corresponding to lane markings in the image. For example, the point 2511 may correspond to an edge of a lane marker, an angle of a lane marker, a midpoint of a lane marker, a vertex between two intersecting lane markers, or various other features or locations. The point 2511 is detected as a position corresponding to a point stored in the navigation model received from the server. For example, if a sparse map is received that contains points representing the centerline of the mapped lane marker, the point 2511 may also be detected based on the centerline of the lane marker 2510.

The vehicle may also determine a longitudinal position represented by element 2520 and located along the target track. The longitudinal position 2520 may be determined from the image 2500, for example, by detecting landmarks 2521 within the image 2500 and comparing the measured position to known landmark positions stored in the road model or sparse map 800. The location along the target trajectory may then be determined based on the distance to the landmark and the known location of the landmark. The longitudinal position 2520 may also be determined from images other than the image used to determine the position of the lane marker. For example, longitudinal position 2520 may be determined by detecting landmarks in images from other cameras within image acquisition unit 120 taken at or near the same time as image 2500. In some cases, the vehicle may not be proximate any road sign or other reference point for determining the longitudinal position 2520. In such cases, the vehicle may navigate based on dead reckoning, and thus may use sensors to determine its own motion and estimate a longitudinal position 2520 relative to the target trajectory. The vehicle may also determine a distance 2530 that represents the actual distance between the vehicle and the observed lane marker 2510 in the captured image(s). Camera angle, speed of the vehicle, width of the vehicle, or various other factors may be considered in determining distance 2530.

Fig. 25B illustrates lateral positioning correction of a vehicle based on mapped lane markings in a road navigation model. As described above, the vehicle 200 may use one or more images captured by the vehicle 200 to determine the distance 2530 between the vehicle 200 and the lane markings 2510. The vehicle 200 may also have access to a road navigation model (e.g., sparse map 800) that may include mapped lane markers 2550 and target trajectories 2555. The mapped lane markers 2550 may be modeled using the techniques described above, for example, using crowd-sourced position identifiers captured by multiple vehicles. The target trajectory 2555 may also be generated using various techniques previously described. The vehicle 200 may also determine or estimate a longitudinal position 2520 along the target track 2555 as described above with respect to fig. 25A. The vehicle 200 may then determine the expected distance 2540 based on the lateral distance between the target trajectory 2555 and the mapped lane markings corresponding to the longitudinal position 2520. The lateral positioning of the vehicle 200 may be corrected or adjusted by comparing the actual distance 2530 measured using the captured image(s) to the expected distance 2540 from the model.

Fig. 25C and 25D provide illustrations associated with another example for locating a host vehicle during navigation based on mapped roadmap/objects/features in a sparse map. Fig. 25C conceptually illustrates a series of images captured from a vehicle navigating along a road segment 2560. In this example, road segment 2560 includes a straight segment of an expressway separated by two lanes depicted by road edges 2561 and 2562 and a center lane marker 2563. As shown, the host vehicle is navigating along a lane 2564 associated with a mapped target trajectory 2565. Thus, in an ideal situation (and without an influencer, e.g., the presence of a target vehicle or object in the road, etc.), the host vehicle should closely track the mapped target trajectory 2565 as the host vehicle navigates along the lane 2564 of the road segment 2560. In fact, the host vehicle may experience drift as it navigates along the mapped target trajectory 2565. For efficient and safe navigation, the drift should be maintained within acceptable limits (e.g., lateral displacement of +/-10cm from the target trajectory 2565 or any other suitable threshold). To periodically account for drift and make any required route corrections to ensure that the host vehicle follows the target trajectory 2565, the disclosed navigation system may be able to locate the host vehicle along the target trajectory 2565 (e.g., determine the lateral and longitudinal position of the host vehicle relative to the target trajectory 2565) using one or more mapped features/objects included in the sparse map.

As a simple example, fig. 25C shows a speed limit marker 2566, which speed limit marker 2566 may appear in five different, sequentially captured images as the host vehicle navigates along road segment 2560. For example, at a first time t ₀ A marker 2566 may appear in the captured image near the horizon. When the host vehicle approaches the flag 2566, at time t ₁ 、t ₂ 、t ₃ And t ₄ The marker 2566 will appear at a different 2D X-Y pixel location in the captured image. For example, in the captured image space, the marker 2566 will move down and to the right along a curve 2567 (e.g., a curve extending through the center of the marker in each of the five captured image frames). As the marker 2566 is approached by the host vehicle, the marker 2566 will also appear to increase in size (i.e., it will occupy a large number of pixels in the subsequently captured image).

These changes in the image space representation of the object, such as marker 2566, may be utilized to determine the local position of the host vehicle along the target trajectory. For example, as described in the present disclosure, any detectable object or feature (e.g., a semantic feature or a detectable non-semantic feature like a logo 2566) may be identified by one or more collection vehicles previously traversing a road segment (e.g., road segment 2560). The mapping server may collect collected driving information from a plurality of vehicles, aggregate and correlate the information, and generate a sparse map including, for example, target trajectories 2565 for lanes 2564 of road segments 2560. The sparse map may also store the location of the markers 2566 (along with type information, etc.). During navigation (e.g., prior to entering road segment 2560), a map tile including a sparse map of road segment 2560 may be provided to the host vehicle. To navigate in the lane 2564 of the road segment 2560, the host vehicle may follow the mapped target trajectory 2565.

The mapped representation of the marker 2566 can be used by the host vehicle to locate itself relative to the target trajectory. For example, a camera on the host vehicle will capture an image 2570 of the environment of the host vehicle, and the captured image 2570 may include an image representation of a logo 2566 having a certain size and a certain X-Y image position, as shown in fig. 25D. The size and X-Y image position may be used to determine the position of the host vehicle relative to the target track 2565. For example, based on a sparse map including representations of markers 2566, the navigation processor of the host vehicle may determine that the representations of markers 2566 should appear in the captured image in response to the host vehicle traveling along target trajectory 2565 such that the center of markers 2566 will move along line 2567 (in image space). If a captured image, such as image 2570, shows a center (or other reference point) off line 2567 (e.g., an expected image space trajectory), the host vehicle navigation system can determine that it is not located on the target trajectory 2565 at the time of the captured image. From the image, however, the navigation processor may determine an appropriate navigation correction to return the host vehicle to the target track 2565. For example, if the analysis shows that the image position of the marker 2566 is shifted in the image to the left of the expected image space position on line 2567 by a distance 2572, the navigation processor may cause the orientation of the host vehicle to change (e.g., change the steering angle of the wheels) to move the host vehicle to the left by a distance 2573 in this manner, each captured image may be used as part of the feedback loop process so that the difference between the observed image position of the marker 2566 and the expected image trajectory 2567 may be minimized to ensure that the host vehicle continues along the target trajectory 2565 with little to no deviation.

The above procedure may be used to detect the lateral orientation or displacement of the host vehicle relative to the target trajectory. Positioning the host vehicle relative to the target track 2565 may also include determining a longitudinal position of the target vehicle along the target track. For example, the captured image 2570 includes a representation of a logo 2566 having a certain image size (e.g., 2D X-Y pixel area). As the mapped marker 2566 travels through image space along line 2567 (e.g., as the size of the marker gradually increases, as shown in fig. 25C), the size may be compared to the expected image size of the mapped marker 2566. Based on the image size of the markers 2566 in the image 2570, and based on the expected size progression in the image space relative to the mapped target trajectory 2565, the host vehicle can determine its longitudinal position relative to the target trajectory 2565 (when the image 2570 is captured). As described above, this longitudinal position coupled with any lateral displacement relative to target track 2565 allows for complete positioning of the host vehicle relative to target track 2565 as the host vehicle navigates along road 2560.

Fig. 25C and 25D provide only one example of the disclosed positioning technique using a single map object and a single target track. In other examples, there may be many more target tracks (e.g., one target track for each viable lane of a multi-lane highway, city street, complex intersection, etc.), and many more mappings may be available for positioning. For example, a sparse map representing a city environment may include many objects available for positioning per meter.

FIG. 26A is a flowchart illustrating an exemplary process 2600A of mapping lane markings for use in autonomous vehicle navigation, in accordance with a disclosed embodiment. At step 2610, process 2600A may include receiving two or more location identifiers associated with the detected lane marker. For example, step 2610 may be performed by server 1230 or one or more processors associated with the server. The location identifier may include a location in real world coordinates of a point associated with the detected lane marker, as described above with respect to fig. 24E. In some embodiments, the location identifier may also contain other data, such as additional information related to road segments or lane markings. Additional data may also be received during step 2610, such as accelerometer data, speed data, roadmap data, road geometry or profile data, vehicle positioning data, self-movement data, or various other forms of data as described above. The location identifier may be generated by a vehicle (e.g., vehicles 1205, 1210, 1215, 1220, and 1225) based on images captured by the vehicle. For example, the identifier may be determined based on acquisition of at least one image representing an environment of the host vehicle from a camera associated with the host vehicle, analysis of the at least one image to detect lane markers in the environment of the host vehicle, and analysis of the at least one image to determine a location of the detected lane markers relative to a location associated with the host vehicle. As described above, the lane markings may include a variety of different marking types, and the location identifiers may correspond to a variety of points relative to the lane markings. For example, where the detected lane marker is part of a dashed line marking the lane boundary, the point may correspond to the detected angle of the lane marker. In the case where the detected lane marker is part of a continuous line marking the lane boundary, the points may correspond to the detected edges of the lane marker, with various pitches as described above. In some embodiments, the points may correspond to the center line of the detected lane marker as shown in fig. 24C, or may correspond to the vertices between two intersecting lane markers and at least two other points of the associated intersecting lane marker as shown in fig. 24D.

At step 2612, the process 2600A may include associating the detected lane marker with the corresponding road segment. For example, the server 1230 may analyze the real world coordinates or other information received during step 2610 and compare the coordinates or other information with the location information stored in the autonomous vehicle road navigation model. The server 1230 may determine a road segment in the model corresponding to the real-world road segment that detected the lane marker.

At step 2614, the process 2600A may include updating an autonomous vehicle road navigation model relative to the corresponding road segment based on the two or more location identifiers associated with the detected lane markers. For example, the autonomous road navigation model may be a sparse map 800, and the server 1230 may update the sparse map to include or adjust mapped lane markings in the model. The server 1230 may update the model based on various methods or processes described above with respect to fig. 24E. In some embodiments, updating the autonomous vehicle road navigation model may include storing one or more indicators of the location in real world coordinates of the detected lane marker. The autonomous vehicle road navigation model may also include at least one target trajectory that the vehicle is to follow along the corresponding road segment, as shown in fig. 24E.

At step 2616, process 2600A may include distributing the updated autonomous vehicle road navigation model to a plurality of autonomous vehicles. For example, server 1230 may distribute updated autonomous vehicle road navigation models to vehicles 1205, 1210, 1215, 1220, and 1225, which may use the models for navigation. The autonomous vehicle road navigation model may be distributed via one or more networks (e.g., over a cellular network and/or the internet, etc.) via a wireless communication path 1235 as shown in fig. 12.

In some embodiments, lane markings may be mapped using data received from multiple vehicles, for example, by crowdsourcing techniques as described above for fig. 24E. For example, the process 2600A may include receiving a first communication from a first host vehicle including a location identifier associated with a detected lane marker and receiving a second communication from a second host vehicle including an additional location identifier associated with the detected lane marker. For example, the second communication may be received from a subsequent vehicle traveling on the same road segment or from the same vehicle along a subsequent trip of the same road segment. The process 2600A may further include refining the determination of the at least one location associated with the detected lane marker based on the location identifier received in the first communication and based on the additional location identifier received in the second communication. This may include using an average of the plurality of location identifiers, and/or filtering out "virtual" identifiers that may not reflect the real world location of the lane markers.

FIG. 26B is a flowchart illustrating an exemplary process 2600B for autonomously navigating a host vehicle along a road segment using mapped lane markings. Process 2600B may be performed, for example, by processing unit 110 of autonomous vehicle 200. At step 2620, process 2600B may include receiving an autonomous vehicle road navigation model from a server-based system. In some embodiments, the autonomous vehicle road navigation model may include a target trajectory of a host vehicle along a road segment and a location identifier associated with one or more lane markings of the associated road segment. For example, the vehicle 200 may receive the sparse map 800 or another road navigation model developed using the process 2600A. In some embodiments, the target trajectory may be represented as a three-dimensional spline, such as shown in fig. 9B. As described above with respect to fig. 24A-F, the location identifier may include a location in real world coordinates of a point associated with the lane marker (e.g., a corner point of a dashed lane marker, an edge point of a continuous lane marker, a vertex between two intersecting lane markers and other points associated with intersecting lane markers, a centerline associated with a lane marker, etc.).

At step 2621, process 2600B may include receiving at least one image representative of an environment of a vehicle. The image may be received from an image capture device of the vehicle, for example, by image capture devices 122 and 124 included in the image acquisition unit 120. The image may include an image of one or more lane markings similar to image 2500 described above.

At step 2622, process 2600B may include determining a longitudinal position of the host vehicle along the target trajectory. As described above with respect to fig. 25A, this may be based on other information in the captured image (e.g., landmarks, etc.) or dead reckoning by the vehicle between detected landmarks.

At step 2623, the process 2600B may include determining an expected lateral distance to the lane marker based on the determined longitudinal position of the host vehicle along the target trajectory and based on two or more location identifiers associated with the at least one lane marker. For example, the vehicle 200 may use the sparse map 800 to determine an expected lateral distance to lane markers. As shown in fig. 25B, a longitudinal position 2520 along target track 2555 may be determined at step 2622. Using the sparse map 800, the vehicle 200 may determine an expected distance 2540 to a mapped lane marker 2550 corresponding to the longitudinal position 2520.

At step 2624, the process 2600B may include analyzing the at least one image to identify at least one lane marker. For example, the vehicle 200 may use various image recognition techniques or algorithms to identify lane markings within the image, as described above. For example, the lane marker 2510 may be detected by image analysis of the image 2500, as shown in fig. 25A.

At step 2625, the process 2600B may include determining an actual lateral distance to the at least one lane marker based on the analysis of the at least one image. For example, the vehicle may determine a distance 2530 representing the actual distance between the vehicle and the lane marker 2510 as shown in fig. 25A. The camera angle, the speed of the vehicle, the width of the vehicle, the position of the camera relative to the vehicle, or various other factors may be considered in determining the distance 2530.

At step 2626, the process 2600B may include determining an autonomous steering action of the host vehicle based on a difference between the predicted lateral distance to the at least one lane marker and the determined actual lateral distance to the at least one lane marker. For example, as described above with respect to fig. 25B, the vehicle 200 may compare the actual distance 2530 to the predicted distance 2540. The difference between the actual and the expected distance may be indicative of the error (and its magnitude) between the actual position of the vehicle and the target trajectory to be followed by the vehicle. Accordingly, the vehicle may determine an autonomous steering action or another autonomous action based on the difference. For example, if the actual distance 2530 is less than the predicted distance 2540 as shown in fig. 25B, the vehicle may determine an autonomous steering action to direct the vehicle left off of the lane marker 2510. Thus, the position of the vehicle relative to the target trajectory can be corrected. Process 2600B may be used, for example, to improve navigation of a vehicle between roadmarks.

The processes 2600A and 2600B provide only examples of techniques that may be used to navigate a host vehicle using the disclosed sparse map. In other examples, processes consistent with those described with respect to fig. 25C and 25D may also be employed.

Navigation based on partially obscured pedestrians

The disclosed systems and methods may allow for navigation of a host vehicle based on partially obscured pedestrians represented in a captured image. For example, navigation of the host vehicle may be based on the location where the foot of the pedestrian contacts the ground surface (referred to herein as the "ground contact location"). However, if the pedestrian is partially occluded in the captured image such that his or her foot and ground contact locations are occluded in the captured image, the disclosed systems and methods may use the trained model to predict the ground contact locations. In particular, before analyzing the image comprising the representation of the occluded pedestrian, the model may be trained based on training data (including, for example, training images, each training image comprising a ground contact location and a representation of at least one pedestrian) and/or depth information corresponding to the training image (e.g., from Lidar-based measurements or obtained by any other depth sensing device). The trained model may then analyze the image and output a bounding box for the partially obscured pedestrian. Using the bounding box, the disclosed systems and methods can accurately determine the position of a partially obscured pedestrian relative to a host vehicle and maneuver the host vehicle to avoid collisions with the partially obscured pedestrian.

FIG. 27 is an exemplary functional block diagram of memory 140 and/or 150 that may be stored/programmed with instructions for performing one or more operations in accordance with the disclosed embodiments. Although reference is made below to memory 140, one skilled in the art will appreciate that instructions may be stored in memory 140 and/or 150.

As shown in fig. 27, the memory 140 may store an image receiving module 2702, a mask analyzing module 2704, an indicator module 2708, and a navigation action module 2710. The disclosed embodiments are not limited to any particular configuration of memory 140. Further, application processor 180 and/or image processor 190 may execute instructions stored in any of modules 2702, 2704, 2708, and 2710 contained in memory 140. Those skilled in the art will appreciate that references to processing unit 110 in the following discussion may refer to application processor 180 and image processor 190 individually or collectively. Accordingly, the steps of any of the following processes may be performed by one or more processing devices.

In one embodiment, the image receiving module 2702 may store instructions (such as computer vision software) that, when executed by the processing unit 110, cause the processing unit 110 to receive one or more captured images representing an environment of a host vehicle (e.g., the host vehicle 200) from a camera on the host vehicle. As described above, the camera may include at least one of an image capture device 122, an image capture device 124, and an image capture device 126. Alternatively and/or additionally, image receiving module 2702 may store instructions that, when executed by processing unit 110, cause processing unit 110 to receive images from cameras (e.g., dedicated pedestrian detection cameras) other than image capture device 122, image capture device 124, and image capture device 126.

According to the disclosed embodiments, at least one of the received images may include a partially obscured pedestrian in the environment of the host vehicle. Some portions of the pedestrian (e.g., the foot) that are partially obscured are not visible from the received image. In the following description, unless otherwise explained, "a partially-occluded pedestrian" is used to mean that at least the foot of the pedestrian is occluded by another object, and thus includes an image of "a partially-occluded pedestrian" that lacks at least a representation of the area where the foot of the pedestrian contacts the ground surface. The image receiving module 2702 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to perform image analysis of the received image to detect pedestrians that are partially obscured. In particular, in some embodiments, the image analysis may include monocular image analysis as described above in connection with fig. 5A-5D, in which case the image receiving module 2702 may include instructions for detecting feature sets within the image, such as anatomical portions, sidewalks, trees, mailboxes, and other features typically associated with pedestrians. Certain features may be identified as reference points for detecting motion or determining pedestrian speed. Alternatively and/or additionally, in some embodiments, image receiving module 2702 may include stereoscopic image analysis as described above in connection with fig. 6, in which case image receiving module 2702 may include instructions for detecting feature sets within a first set of images (e.g., acquired by image capturing device 124) and a second set of images (e.g., acquired by image capturing device 126). Based on features detected in the monocular image analysis and/or the stereoscopic image analysis, the processing unit 110 may detect a representation of the partially occluded pedestrian from the received image.

In some embodiments, a portion or portion of the pedestrian that is partially obscured is not visible in the received image. For example, the feet of a partially obscured pedestrian may be blocked by an object, and thus the received image may lack a representation of the area where the partially obscured pedestrian contacts the ground surface, which may be, for example, a road surface or a sidewalk. Fig. 28A shows an exemplary image 2801 captured by a camera on the host vehicle. The image 2801 includes a representation of a pedestrian 2803 that is partially obscured, with its lower body (i.e., foot) not visible within the image 2801. In this example, the camera view of the pedestrian's foot is blocked by a vehicle 2805 that is different from the host vehicle.

Fig. 28B shows another example in which an image 2811 captured by an onboard camera of a host vehicle includes a representation of a pedestrian 2813 that is partially obscured. In this example, the pedestrian's foot and its camera view of contact with the ground surface are obscured by the main vehicle's own hood 2815. In this example, the partially obscured pedestrian 2905 may be proximate to the host vehicle.

Referring back to fig. 27, upon detecting a received image comprising a representation of a partially occluded pedestrian, image receiving module 2702 may provide the received image to occlusion analysis module 2704 for further analysis. In one embodiment, the image receiving module 2702 may provide the image or a portion of the image to the mask analysis module 2704 after the pedestrian or pedestrian candidate is detected in the image. Pedestrians, including partially obscured pedestrians, may be detected in the image using any of several known methods including computer vision (e.g., pattern recognition) methods and machine learning based methods. In one embodiment, pedestrian or pedestrian candidate detection may also be implemented as part of the occlusion analysis performed in occlusion analysis module 2704.

The mask analysis module 2704 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to generate an indicator of a contact location (i.e., ground contact location) of the partially masked pedestrian with the ground surface. The indicator may be a bounding box surrounding a representation of the partially obscured pedestrian in the received image. The processing unit 100 may determine the size of the bounding box to encompass the area where the foot of the partially obscured pedestrian contacts the ground surface (if such area is already represented in the received image). For example, fig. 28C shows a rectangular shaped bounding box 2817 generated for a partially obscured pedestrian 2813 in an image 2811. As shown, in the captured image 2811, a portion of the host vehicle (i.e., the hood 2815) occludes the area where the partially occluded pedestrian 2813 contacts the ground surface. The bounding box 2817 includes an area of the captured image 2811 representing a portion of the hood 2815, which corresponds to the foot of a pedestrian (shown in phantom line drawing in fig. 28C) and its contact location with the ground surface.

According to the disclosed embodiments, the bounding box may be defined by a pixel size and a real world size. For example, as shown in fig. 28D, bounding box 2817 (fig. 28C) may be formed of at least a horizontal length ("L") of pixels associated with captured image 2811 _p ") and the height (" H ") of the pixel associated with the captured image 2811 _p ") are defined. L (L) _p And H _p May be measured relative to a reference pixel associated with the captured image 2811 (e.g., a pixel at one of the four corners of the rectangular-shaped bounding box 2817) and may be used to describe the pixel size of a partially obscured pedestrian in the captured image 2811. Furthermore, bounding box 2817 may be formed of at least a real world length ("L") _w ") and real world height (" H _w ") are defined. L (L) _w And H _w Corresponding to the actual size of a partially obscured pedestrian in the real world. As described above in connection with fig. 12, the true size of the partially obscured pedestrian may be estimated based on the change in distance between the partially obscured pedestrian and the host vehicle and the change in pixel size of the bounding box. Thus, the real world size of bounding box 2817 may be estimated in the same manner. Alternatively or additionally, if the distance between the partially obscured pedestrian and the host vehicle is known, thenThe pixel size L of bounding box 2817 based on the distance between the partially obscured pedestrian and the host vehicle _p 、H _p And estimating the real world dimension L of the bounding box 2817 by the focal length of a camera capturing an image comprising a representation of the partially obscured pedestrian _w 、H _w 。

In some embodiments, a predictive model may be used to generate an indicator (e.g., bounding box) of the contact location of the partially obscured pedestrian with the ground surface. In particular, referring back to fig. 27, the mask analysis module 2704 may also store instructions that, when executed by the processing unit 110, cause the processing unit 110 to execute a machine learning algorithm for training the predictive model 2706. For example, the predictive model 2706 may be trained as a neural network, such as a Convolutional Neural Network (CNN) or a Deep Neural Network (DNN). After training, the processing unit 110 may provide the received image including the representation of the partially occluded pedestrian to the trained predictive model 2706 for generating an indicator (e.g., bounding box) of the contact location of the partially occluded pedestrian with the ground surface.

In one embodiment, the processing unit 110 may train the predictive model 2706 based on training data including one or more training images, each training image including a representation of at least one training pedestrian and showing an area where the foot of the training pedestrian contacts the ground surface. The training pedestrian(s) may be the same or different person as the partially obscured pedestrian detected by image receiving module 2702. In particular, processing unit 110 may use the one or more training images to create a first training set and train predictive model 2706 using the first training set. During training, processing unit 110 may configure prediction model 2706 to extract features from the first training set and classify (or cluster) features according to their correspondence with the training pedestrians. For example, features corresponding to the foot of a training pedestrian and its contact location on the ground surface may be identified and marked. Similarly, features corresponding to other portions of a training pedestrian (e.g., head, hand, upper body, etc.) may be identified and marked. Further, features corresponding to different portions of a trained pedestrian may be associated and classified (or clustered) to represent the same pedestrian. The processing unit 110 may also modify the one or more training images to mask areas of the ground surface where the feet of the training pedestrian contact the ground surface, thereby creating a second training set. The processing unit 110 may then train the predictive model 2706 using the second training set. For example, the processing unit 110 may analyze the second training set using the predictive model 2706 to predict the position of the feet of the training pedestrian and their contact location with the ground surface based at least in part on the visible portion (e.g., head, hand, upper body, etc.) of the training pedestrian in the second training set. The processing unit 110 may then compare the predicted position with the real position of the foot in the first training set. Based on the comparison, the processing unit 110 may determine a confidence value associated with the predicted location and adjust parameters of the predictive model 2706 to minimize the difference between the predicted location and the real location.

In one embodiment, processing unit 110 may combine the image data with additional sensory information (e.g., information from radar, lidar, acoustic sensors, etc.) to train predictive model 2706. For example, the training data may include one or more training images and lidar data associated with the training images. Each of the one or more training images may be captured by a camera on the host vehicle and include a representation of at least one training pedestrian, the foot of which may or may not be visible in the training image. The training pedestrian may be the same or a different person than the partially obscured pedestrian detected by image receiving module 2702. If the foot is visible, the processing unit 110 may modify the one or more training images to mask the area where the foot contacts the ground surface. The processing unit 110 may feed one or more training images with occluded foot regions to the predictive model 2706 to generate a bounding box around the training pedestrian. The predicted bounding box may extend from the head of the training pedestrian to his or her foot. Based on the predicted height of the bounding box and the focal length of the camera capturing the training image, the processing unit 110 may determine a predicted distance from the training pedestrian to the host vehicle. The lidar data may include depth information corresponding to each of the one or more training images. Thus, the lidar depth information may indicate a measured distance from the training pedestrian to the host vehicle. The processing unit 110 may compare the predicted distance to lidar depth information. Based on the comparison, processing unit 110 may determine a confidence value associated with the predicted bounding box and adjust parameters of prediction model 2706 to minimize the difference between the predicted distance and the lidar depth information.

In accordance with the disclosed embodiment, the predictive model 2706 may be trained prior to invoking the mask analysis module 2704 to analyze the partially masked pedestrians identified by the image receiving module 2702. Thus, training data may be acquired before the partially occluded pedestrians are analyzed by the occlusion analysis module 2704. In one embodiment, processing unit 110 may maintain a buffer in memory 140 or 150 to store training data. As an example, the buffer may store up to 20 image frames. Each of the buffered image frames may include a representation of a trained pedestrian and/or a representation of an area of the trained pedestrian contacting the ground surface. Further, the buffer may store lidar depth data corresponding to the image frames. By implementing instructions stored in the mask analysis module 2704, the processing unit 110 may provide buffered training data to the mask analysis module 2704 to train the predictive model 2706. In fact, not all of the buffered training data is required to train the predictive model 2706. For example, even though the buffer may contain 20 training image frames, only 15 of them may be used to train the predictive model 2706. In one embodiment, the processing unit 110 may cease training after the confidence value associated with the predictive model 2706 reaches a predetermined threshold.

As described above, according to the disclosed embodiments, the occlusion analysis module 2704 may store instructions that, when executed by the processing unit 100, cause the processing unit to feed images provided by the image receiving module 2702 (i.e., images including representations of partially occluded pedestrians and representations of areas lacking the partially occluded pedestrians' feet contacting the ground surface) to the trained predictive model 2706 to generate bounding boxes surrounding the partially occluded pedestrians. The generated bounding box may include the feet of the partially obscured pedestrian and the area where the feet contact the ground surface. If the prediction model 2706 is trained using only the training image(s), the processing unit 100 may use the trained prediction model 2706 to determine a predicted location of the occluded foot region and then generate a bounding box based on the predicted location. If the prediction model 2706 is trained using both the training image(s) and the corresponding lidar depth information, the processing unit 100 may directly generate a bounding box using the trained prediction model 2706.

Still referring to fig. 27, the indicator module 2708 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to receive an indicator (e.g., a bounding box) of a contact location (i.e., a ground contact location) of the partially obscured pedestrian with the ground surface from the obscured analysis module 2704 and analyze the indicator to determine a distance between the partially obscured pedestrian and a portion of the host vehicle and/or to determine an estimated time to collision between the partially obscured pedestrian and the host vehicle. In one embodiment, the processing unit 110 may determine the estimated distance between the partially obscured pedestrian and the host vehicle based on the pixel size of the bounding box in the captured image, the real world size of the bounding box, the focal length of the camera capturing the image including the representation of the partially obscured pedestrian, and/or the position of the camera relative to the ground surface. In one embodiment, the processing unit 110 may track the bounding box and monitor its changes in the subsequently captured images. Based on this change, the processing unit 110 may determine whether and how quickly the host vehicle is approaching a partially obscured pedestrian. For example, if the size of the bounding box is becoming larger in the subsequently captured image, the processing unit 110 may determine that the host vehicle is approaching a partially obscured pedestrian. Conversely, if the size of the bounding box is becoming smaller in the subsequently captured image, the processing unit 110 may determine that the host vehicle is leaving a partially obscured pedestrian. Further, based on the rate at which the bounding box is becoming larger, the processing unit 110 may determine how quickly the host vehicle may collide with the pedestrian if no action is taken, or when the (planned or predicted) trajectory of the host vehicle will intersect the predicted trajectory of the pedestrian.

Still referring to fig. 27, the navigation action module 2710 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to determine a navigation action of the host vehicle based on the ground contact location and cause at least one adjustment of a navigation actuator of the host vehicle in response to the determined navigation action of the host vehicle. For example, if the bounding box surrounding the partially-occluded pedestrian indicates that the host vehicle is approaching the partially-occluded pedestrian, the processing unit 110 may determine at least one navigational change to avoid a collision with the partially-occluded pedestrian (e.g., by decelerating and/or braking, by rotating the host vehicle away from the partially-occluded pedestrian, by steering the host vehicle to a different direction and/or accelerating in a new direction, etc.). As another example, if the bounding box indicates that the partially obscured pedestrian is moving away from the host vehicle, the processing unit 110 may determine that the host vehicle may maintain the same speed or direction, or accelerate in the same or a different direction.

Based on the determined navigational actions, the processing unit 110 may transmit electronic signals to cause control of at least one navigational actuator of the host vehicle according to the determined navigational actions of the host vehicle. The navigation actuator may include at least one of a steering mechanism, a brake, or an accelerator. In some embodiments, the processing unit 110 may transmit one or more signals to one or more of the throttle system 220, the brake system 230, and the steering system 240 of the vehicle 200 to trigger the determined navigational action, e.g., turn the steering wheel of the vehicle 200 to achieve a predetermined angle of rotation.

According to the disclosed embodiments, any of the modules disclosed herein (e.g., modules 2702, 2704, 2708, and 2710) may be stored remotely in a server in communication with the host vehicle. The captured image may initially be transmitted to a server, which may include one or more processors similar to the processing unit 110 described above. The one or more processors at the server may execute the instructions in modules 2702, 2704, 2708, and/or 2710 in a manner similar to that described above in connection with processing unit 110. The server may then transmit the results to the host vehicle 110. For example, in one embodiment, the mask analysis module 2704 and the predictive model 2706 may be stored in a server. The server may train the predictive model 2706 based on the training data and then transmit the trained model to the host vehicle for further use.

In accordance with the disclosed embodiments, the occlusion analysis module 2704 in the host vehicle or server can track partially occluded pedestrians in subsequently captured images and retrain the predictive model 2706 if the subsequently captured images indicate that the predicted bounding box or predicted ground contact location is inaccurate. For example, if the pedestrian becomes fully visible in the subsequent image, the occlusion analysis module 2704 may compare the actual ground contact position to the predicted ground contact position to determine if there are any discrepancies. Similarly, if a characteristic of the pedestrian is detected in a subsequent image (e.g., the pedestrian is pushing a cart or stepping on a skateboard), the occlusion analysis module 2704 may estimate the ground contact position of the pedestrian based on the particular characteristic (e.g., the position of the cart or skateboard) and compare the estimated actual ground contact position to the predicted ground contact position to determine if there are any discrepancies. In both examples, if the difference exceeds a predetermined limit or ratio, the mask analysis module 2704 may retrain the predictive model 2706. According to one embodiment, if training and retraining is performed by a server, the processing unit 110 in the host vehicle may maintain a buffer in memory 140 or 150 to store the newly captured image for a certain number of frames (e.g., 20 frames). When it is determined that the newly captured image indicates that the predicted bounding box or ground contact location is inaccurate, the processing unit 110 may send the newly captured image to a server for preserving the prediction model 2706.

FIG. 29 is a flowchart illustrating an exemplary process 2900 for navigating a host vehicle based on a detected partially obscured pedestrian, in accordance with the disclosed embodiments. At step 2902, the processing unit 110 may train a predictive model (e.g., predictive model 2706 in fig. 27) of an indicator of a contact location of the partially obscured pedestrian with the ground surface based on the training data. The predictive model may include a neural network. The training data may include one or more training images, each training image including at least one training pedestrian and a representation of an area of the training pedestrian contacting a ground surface (e.g., a road surface or a pavement surface). In addition, processing unit 110 may train the predictive model using lidar depth information corresponding to the training image(s).

At step 2904, the processing unit 110 may receive one or more images captured by a camera on the host vehicle. Each of the one or more captured images may represent an environment of the host vehicle and may be received via the data interface 128. For example, cameras included in the image acquisition unit 120, such as the image capture devices 122, 124, and 126 having fields of view 202, 204, and 206, may capture at least one image of an area in front of and/or to the side of the host vehicle and communicate them to the processing unit 110 via a digital connection (e.g., USB, wireless, bluetooth, etc.).

The processing unit 110 may analyze one or more received images and detect at least one image including a representation of a partially obscured pedestrian in the environment of the host vehicle. The foot of the at least partially occluded pedestrian may be occluded (e.g., blocked by another object) in the detected image, and thus the detected image lacks a representation of the area of the partially occluded pedestrian contacting the ground surface. For example, the ground surface may be a road surface or a pavement surface. In one embodiment, the processing unit 110 may identify the partially obscured pedestrian by detecting one or more attributes or features of the partially obscured pedestrian. The attributes may include a facial region, a core/body region, or one or more attachments associated with a partially obscured pedestrian. The processing unit 110 may also determine that an attribute or feature associated with the foot is lost in the detected image.

At step 2906, the processing unit 110 may feed an image including the partially occluded pedestrian into the trained predictive model to generate an indicator of the contact location of the partially occluded pedestrian with the ground surface (i.e., the ground contact location). The indicator may be a bounding box surrounding the partially obscured pedestrian, including the obscured ground contact location.

At step 2908, the processing unit 110 may determine a ground contact location of the partially obscured pedestrian based on the generated indicator. For example, the lower end of the bounding box corresponds to a ground contact location, which may be used by the processing unit 110 to determine an estimated distance between the partially obscured pedestrian and the host vehicle, and/or to determine an estimated time to collision between the partially obscured pedestrian and the host vehicle.

In one embodiment, the processing unit 110 may also determine the ground contact location of the obscured pedestrian based on other factors, such as information regarding the elevation of the ground surface (including road surface, sidewalk, soft shoulder, etc.) relative to the location of the host vehicle. The ground surface elevation information may be obtained via measurements generated based on sensor data from sensors on the host vehicle, and/or the elevation information may be obtained from a map (such as a sparse map generated using data downloaded into the host vehicle from a remote server). In some embodiments, ground elevation information may be used as input to a prediction module to allow accurate prediction of the contact point of the foot of a partially obscured pedestrian with the ground given the relative elevation (positive or negative) of the ground surface relative to the position of the host vehicle.

At step 2910, the processing unit 110 may cause a navigation action of the host vehicle based on the generated indicator. For example, if it is determined that the distance between the partially obscured pedestrian and the host vehicle is within a predetermined distance, the processing unit 110 may slow, stop, turn away from the partially obscured pedestrian. The processing unit 110 may initiate the navigational action by sending one or more control signals to a navigational actuator, such as a brake, steering wheel, accelerator, etc.

According to the disclosed embodiments, any steps of the process 2900 disclosed herein (e.g., steps 2902, 2904, 2906, 2908, and 2910) may be performed remotely in a server in communication with the host vehicle. For example, the server may perform step 2902 and then transmit the trained model to the host vehicle for further use.

Navigation based on detected head pose of pedestrian

The disclosed systems and methods may allow for navigation of a host vehicle based on a head pose (i.e., head orientation) of a pedestrian detected from image data. The head pose may be represented by a rotation angle (also referred to as a yaw angle) in a horizontal plane parallel to the ground surface and an inclination angle in a vertical plane extending from the nose of the pedestrian and the back of the head. The gaze direction of the pedestrian may be estimated based on the rotation angle and the tilt angle. By tracking the gaze direction of the pedestrian, the host vehicle may be preemptively maneuvered to allow for more efficient navigation and safer driving. For example, when it is determined that the gaze direction of the pedestrian is away from the host vehicle, the host vehicle may navigate by using a greater safety margin than that used when the pedestrian is gazing at or looking in the direction of the host vehicle.

FIG. 30 is an exemplary functional block diagram of memory 140 and/or 150 that may be stored/programmed with instructions for performing one or more operations in accordance with the disclosed embodiments. Although reference is made below to memory 140, one skilled in the art will appreciate that instructions may be stored in memory 140 and/or 150.

As shown in fig. 30, the memory 140 may store an image receiving module 3002, a pedestrian detection module 3004, a head pose analysis module 3006, and a navigation action module 3010. The disclosed embodiments are not limited to any particular configuration of memory 140. Further, the application processor 180 and/or the image processor 190 may execute instructions stored in any one of the modules 3002, 3004, 3006, and 3010 contained in the memory 140. Those skilled in the art will appreciate that references to processing unit 110 in the following discussion may refer to application processor 180 and image processor 190 individually or collectively. Accordingly, the steps of any of the following processes may be performed by one or more processing devices.

In one embodiment, the image receiving module 3002 may store instructions (such as computer vision software) that, when executed by the processing unit 110, cause the processing unit 110 to receive one or more images from a camera on the host vehicle that represent the environment of the host vehicle. As described above, the camera may include at least one of an image capture device 122, an image capture device 124, and an image capture device 126. Alternatively and/or additionally, the image receiving module 3002 may store instructions that, when executed by the processing unit 110, receive images from cameras (e.g., dedicated sidewalk cameras) other than the image capturing device 122, the image capturing device 124, and the image capturing device 126.

In one embodiment, the pedestrian detection module 3004 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to perform image analysis of one or more received images to detect at least one pedestrian in the environment of the host vehicle. In some embodiments, the image analysis may include monocular image analysis as described above in connection with fig. 5A-5D, in which case the pedestrian detection module 3004 may include instructions for detecting one or more features within the image, such as anatomical portions, accessories, clothing, sidewalks, walking sticks, strollers, and other features typically associated with pedestrians. Alternatively and/or additionally, in some embodiments, the pedestrian detection module 3004 may include stereoscopic image analysis as described above in connection with fig. 6, in which case the pedestrian detection module 3004 may include instructions for detecting one or more features within the first set of images (e.g., acquired by the image capture device 124) and the second set of images (e.g., acquired by the image capture device 126). Based on the monocular image analysis and/or the stereoscopic image analysis, the processing unit 110 may detect at least one pedestrian represented in one or more of the received images.

In one embodiment, the head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to analyze one or more received images to determine a head pose of at least one detected pedestrian. Specifically, the processing unit 110 may first identify the head of the detected pedestrian. The processing unit 110 may then determine the angular rotation indicator and the tilt angle indicator associated with the identified head. Based on the rotation angle and the tilt angle, the processing unit 110 may determine a gaze direction of the detected pedestrian.

In one embodiment, the head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to identify a head by analyzing one or more received images to identify features associated with the head. For example, the processing unit 110 may identify an area in the image having a certain shape. The processing unit 110 may perform shape analysis on the identified regions to determine whether some of them correspond to a portion of the head. Alternatively or additionally, the processing unit 110 may perform a motion analysis on the identified region to detect a contour around a portion of the head.

After identifying the head of the pedestrian from the one or more received images, the processing unit 110 may continue to determine the pose of the head. As shown in fig. 31A, the head pose may be defined by a rotation angle Φ and an inclination angle θ. The rotation angle phi may describe the angular rotation of the head in a horizontal plane parallel to the ground surface. In other words, the rotational angle φ may indicate an angular movement of the head about a vertical axis passing through the center of the head. Further, the tilt angle θ may describe the angular movement of the head in a vertical plane extending from the nose of the pedestrian to the rear of the head. In other words, the tilt angle θ may be indicative of the angular movement of the head about an axis passing through the ear of the pedestrian.

Referring back to fig. 30, the processing unit 110 may feed one or more received images to the model 3008 stored in the head pose analysis module 3006 to determine rotational angles and tilt angles associated with the head. Model 3008 may include a neural network (e.g., CNN or DNN) for classifying or clustering image features associated with different head poses. Prior to using model 3008 to determine the head pose, processing unit 110 may train model 3008 based on a plurality of training images including representations of pedestrians exhibiting various head poses and gaze directions. The processing unit 110 may train the model 3008 to classify or cluster image portions and/or image features in the training image according to their corresponding head orientations or gaze directions.

After training is complete, processing unit 110 may analyze one or more received images using trained model 3008 to determine a rotation angle and a tilt angle associated with the head and determine a gaze direction associated with the head based on the rotation angle and the tilt angle. In one embodiment, the head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to determine that a pedestrian is gazing toward a host vehicle when a gaze direction intersects the host vehicle. For example, the processing unit 110 may calculate a cone that includes the host vehicle (e.g., at an apex within the cone or at another location) and has an axis along the axis of the host vehicle (e.g., from the hood to the trunk). The taper may have an angle of 90 degrees or less, 45 degrees or less, etc. The processing unit 110 may determine that a pedestrian is looking toward the host vehicle when the gaze direction intersects the side of the cone at an angle less than a threshold (e.g., less than 90 degrees, less than 45 degrees, etc.). Alternatively, the processing unit 110 may calculate a cone that includes the pedestrian (e.g., at an apex within the cone or at another location within the cone) and has an axis from the pedestrian to the host vehicle. The taper may have an angle of 90 degrees or less, 45 degrees or less, etc. The processing unit 110 may determine that the pedestrian is looking toward the host vehicle when the gaze direction falls within the cone.

Using a similar approach, the processing unit 110 may also determine whether a pedestrian is looking toward other vehicles in the primary vehicle's environment. For example, the head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to identify a target vehicle in one or more received images and determine whether a gaze direction associated with the head intersects the target vehicle.

In one embodiment, the head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to determine that a pedestrian is looking away from the host vehicle when the gaze direction is not facing the host vehicle. The processing unit 110 may use any of the above-described cones to determine that the pedestrian is looking away from the host vehicle. Alternatively, if the processing unit 110 is unable to identify certain facial features (e.g., eyes, nose, mouth) associated with the head, the processing unit 110 may determine that the pedestrian is looking away from the host vehicle. For example, the pedestrian may face away from the host vehicle such that only the sides or back of the pedestrian's head are visible in one or more of the received images.

In one embodiment, the one or more received images may include a plurality of images captured by the camera over time. The head pose analysis module 3006 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to track changes in head pose over time based on a plurality of captured images. For example, the processing unit 110 may determine that pedestrians shown in the plurality of captured images are initially gazing toward the host vehicle and are turning her head away from the host vehicle. As another example, the processing unit 110 may determine that the plurality of captured images show an increased tilt angle of the head, which means that the pedestrian is tilting up or down her head.

Still referring to fig. 30, the navigation action module 3010 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to determine a navigation action of the host vehicle based on the head pose of the pedestrian. In one embodiment, if the head pose of the pedestrian indicates that she is looking away from the host vehicle, the processing unit 110 may determine a first navigational action of the host vehicle. The first navigational action may include slowing or stopping the host vehicle or moving within the travel lane in a direction away from the pedestrian. The first navigational action allows the host vehicle and pedestrians more reaction time.

In one embodiment, if the head pose of the pedestrian indicates that he or she is looking toward the host vehicle, the processing unit 110 may determine a second navigational action of the host vehicle. The second navigational action may include maintaining an original speed and/or heading direction. This is because pedestrians looking toward the host vehicle are aware of and alerted by the approaching host vehicle. Similarly, if the head pose of the pedestrian indicates that he or she is looking at a target vehicle traveling in the same direction and forward of the host vehicle, the processing unit 110 may determine a second navigational action of the host vehicle. This is because pedestrians can more easily see traffic traveling in the same direction. Further, if the head pose of the pedestrian indicates that he or she is looking at a target vehicle located in front of the host vehicle, but traveling in a direction opposite to the host vehicle, the processing unit 110 may determine a second navigation action of the host vehicle. This is because pedestrians who see the target vehicle traveling in the upcoming lane can also see the main vehicle.

As described above in connection with the head pose analysis module 3006, the processing unit 110 may track changes in head pose over time across multiple captured images. The navigation action module 3010 may also store instructions that, when executed by the processing unit 110, cause the processing unit 110 to determine a navigation action of the host vehicle based on a change in head pose over time. For example, if it is determined that the pedestrian was initially gazing toward the host vehicle, but later moves his or her gaze direction away from the host vehicle, the processing unit 110 may determine a first navigational action of the host vehicle when determining movement of the gaze direction.

The navigation action module 3010 may also store instructions that, when executed by the processing unit 110, cause the processing unit 110 to transmit electronic signals to cause control of at least one navigation actuator of the host vehicle in accordance with the determined navigation action of the host vehicle. The navigation actuator may include at least one of a steering mechanism, a brake, or an accelerator. In some embodiments, the processing unit 110 may transmit one or more signals to one or more of the throttle system 220, the brake system 230, and the steering system 240 of the vehicle 200 to trigger the determined navigational action, e.g., turn the steering wheel of the vehicle 200 to achieve a predetermined angle of rotation.

FIG. 31B illustrates an example of an implementation of the disclosed system for navigating a vehicle based on a detected head pose of a pedestrian. The scenario shown in fig. 31B is an example of one of the images that may be captured from the environment of the host vehicle 3105 (which may include the host vehicle 200). As explained above, by analyzing the image, the processing unit 110 can determine the head pose of the pedestrian 3103 in the image. The processing unit 110 may make this determination by determining a rotation angle and a tilt angle associated with the head of the pedestrian 3103. Based on the rotation angle and the tilt angle, the processing unit 110 may also determine the gaze direction of the pedestrian 3103 with respect to the main vehicle 3105 by constructing a cone with the main vehicle 3105 or the pedestrian 3103 as a vertex. If the processing unit 110 determines that the gaze direction of the pedestrian 3103 intersects or is within the cone of view at an angle that is less than the threshold, respectively, the processing unit 110 may determine that the pedestrian 3103 is gazing toward the host vehicle 3105. Thus, the processing unit 110 may determine a navigational action of the main vehicle 3105, such as slowing down, or braking, or switching lanes. In some embodiments, the host vehicle 3105 may switch lanes and/or may slow to a stop. It will be appreciated that a cone is used herein as an example of a geometry, which may be used to convert the gaze direction of a pedestrian into a geometry. It will also be appreciated that other 3-D shapes in real space may be used.

In one embodiment, the navigation action module 3010 may cause the host vehicle to perform a negotiation maneuver, where the host vehicle may perform a relatively subtle or fine maneuver to signal intent and expected reactions from pedestrians during one or more action expected periods. For example, the negotiation operation may be used when a first constraint (e.g., an initial/default safety margin) is reached that indicates that the host vehicle needs to slow down or stop and yield to the pedestrian, but in this case the host vehicle may be allowed (programmed or configured) to proceed under a second constraint (a smaller safety margin) only if the pedestrian's intent is verified and meets the second constraint. The navigation action module 3010 can determine the intent of the pedestrian based on the head pose of the pedestrian or based on detecting a change in the head pose of the pedestrian in response to a signal (e.g., flashing light, horn sounding) of the host vehicle.

According to the disclosed embodiments, any of the modules disclosed herein (e.g., modules 3002, 3004, 3006, and 3010) may be stored remotely in a server in communication with the host vehicle. The captured image may initially be transmitted to a server, which may include one or more processors similar to the processing unit 110 described above. The one or more processors at the server may execute the instructions in modules 3002, 3004, 3006, and/or 3010 in a manner similar to that described above in connection with processing unit 110. The server may then transmit the results to the host vehicle 110. For example, in one embodiment, the head pose analysis module 3006 and model 3008 may be stored in a server. The server may train the model 3008 based on the training data and then transmit the trained model to the host vehicle for further use.

In accordance with the disclosed embodiments, the head pose analysis module 3006 in the host vehicle or server may track the gaze direction or movement of the pedestrian in the subsequently captured image and retrain the model 3008 if the subsequently captured image indicates that the head pose results generated by the model 3008 are inaccurate. For example, if the predicted head pose indicates that the pedestrian is looking toward the host vehicle, but tracking of the pedestrian's eyes in the subsequently captured image indicates that he or she is looking away from the host vehicle, the head pose analysis module 3006 may retain the model 3008. According to one embodiment, if training and retraining is performed by a server, the processing unit 110 in the host vehicle may maintain a buffer in memory 140 or 150 to store the newly captured image for a certain number of frames (e.g., 20 frames). When it is determined that the newly captured image indicates that the head pose predicted by module 3006 is inaccurate, processing unit 110 may send the newly captured image to a server for use in retaining prediction module 3006.

Fig. 32 is a flowchart illustrating an exemplary process 3200 for navigating a host vehicle based on a detected pedestrian head pose in accordance with the disclosed embodiments. At step 3202, the processing unit 110 may receive one or more images representing an environment of the host vehicle via the data interface 128. For example, cameras included in the image acquisition unit 120, such as the image capture devices 122, 124, and 126 having fields of view 202, 204, and 206, may capture at least one image of an area in front of and/or to the side of the host vehicle and communicate them to the processing unit 110 via a digital connection (e.g., USB, wireless, bluetooth, etc.).

At step 3204, the processing unit 110 may analyze the one or more received images to detect at least one pedestrian in the environment of the host vehicle. For example, detecting at least one pedestrian may include detecting one or more attributes or features of the pedestrian. The attributes may include a facial region, a core/body region, or one or more attachments associated with a pedestrian.

At step 3206, processing unit 110 may analyze one or more received images to determine a head pose of a detected pedestrian. For example, the processing unit 110 may use a trained model (e.g., model 3008 in fig. 30) to classify (or cluster) features in one or more received images and determine a rotation angle and a tilt angle associated with the head of the pedestrian based on the classification (or cluster). Based on the rotation angle and the tilt angle, the processing unit 110 may determine a gaze direction of the pedestrian. In some embodiments, the one or more received images may include a plurality of images captured over time, and the processing unit 110 may track changes in head pose and movement of gaze direction over time.

At step 3208, the processing unit 110 may cause control of at least one navigation actuator of the host vehicle based on the determined head pose. For example, if the head pose indicates that the pedestrian is looking toward the host vehicle, the processing unit 110 may slow down and/or brake the host vehicle, or turn the host vehicle to a new heading direction and/or accelerate the host vehicle in the new heading direction. As another example, if the head pose indicates that the pedestrian is looking away from the host vehicle, the processing unit 110 may cause the host vehicle to maintain the original speed and/or heading direction, or accelerate in the original heading direction. The processing unit 110 may initiate the navigational action by sending one or more control signals to a navigational actuator, such as a brake, steering wheel, accelerator, etc.

According to the disclosed embodiments, any of the steps of process 3200 disclosed herein (e.g., steps 3202, 3204, 3206, and 3208) may be performed remotely in a server in communication with the host vehicle. For example, the server may perform steps 3202-3206 and communicate the determined head pose to the host vehicle, which may then determine a navigation action based on the determined head pose.

Navigation based on detected pedestrian gestures

The disclosed systems and methods may allow for navigation of a host vehicle based on detected pedestrian gestures. The gesture may indicate how the pedestrian is moving relative to the host vehicle. The gesture may provide an indicator of the intent of the pedestrian, for example, allowing the host vehicle to advance or signaling the host vehicle to stop. Further, the gesture may indicate some state or activity of the pedestrian, such as making a telephone call or gazing at a mobile device, etc. By detecting and recognizing these gestures, the disclosed systems and methods may cause the host vehicle to respond based at least in part on the gestures. For example, if it is determined that a pedestrian is making a gesture that signals a stop of the host vehicle, the disclosed system may stop or slow down the host vehicle. As another example, if it is determined that a pedestrian is making a telephone call, the disclosed system may stop, slow down, turn away from, or send an alert to the pedestrian.

For example, similar to other embodiments provided by the present disclosure, memory 140 and/or 150 may store computer instructions executable by one or more processors to perform a method for navigating a host vehicle based on detected pedestrian gestures. In particular, the one or more processors may be programmed to receive a plurality of images from a camera associated with the host vehicle. The plurality of images may represent an environment of the host vehicle. The one or more processors may detect a pedestrian represented in the plurality of images and analyze the plurality of images to detect an identified gesture made by the pedestrian. Based on the detected recognized gesture made by the pedestrian, the one or more processors may cause at least one navigational action of the host vehicle. For example, the navigational action may include at least one of accelerating, braking, or rotating the host vehicle.

In one embodiment, the one or more processors may detect the recognized gesture based on an orientation of a hand of the pedestrian. Alternatively or additionally, the one or more processors may detect the recognized gesture based on movement of a hand of the pedestrian represented across the plurality of images.

In one embodiment, the one or more processors may analyze the plurality of images by using at least one trained model configured to detect the recognized gesture based on training data comprising a plurality of training image sets. Each training image set may include a plurality of training images representing at least one pedestrian performing one or more gestures. In accordance with the present disclosure, the trained model may be any suitable machine learning algorithm, such as a neural network.

In one embodiment, the one or more processors may analyze the plurality of images to detect the recognized gesture associated with the pedestrian further based on identifying at least one skeletal anchor point of the pedestrian and identifying movement of one or more skeletal sections relative to the at least one skeletal anchor point. The skeletal anchorage points may be, for example, elbows, shoulders, wrists, knuckles, necks, and the like. The skeletal section may be, for example, a hand, forearm, finger section, upper arm, head, or the like.

In one embodiment, the recognized gesture may indicate a command to move the host vehicle forward. Thus, based on the recognized gesture, the one or more processors may cause the host vehicle to move forward at the selected speed.

In one embodiment, the recognized gesture may indicate a command to stop the host vehicle. Thus, based on the recognized gesture, the one or more processors may cause the host vehicle to stop, slow down, or change heading direction. For example, the one or more processors may actuate a brake of the host vehicle to stop or slow the host vehicle. As another example, the one or more processors may steer the primary vehicle away from its original direction of movement to yield the primary vehicle to the pedestrian.

In one embodiment, the recognized gesture may indicate a command to pass the primary vehicle through a pedestrian. Thus, based on the recognized gesture, the one or more processors may cause the host vehicle to perform at least one of: maintaining the current speed of the host vehicle or maintaining the current heading direction of the host vehicle.

In one embodiment, the recognized gesture may indicate that the pedestrian is making a telephone call. Thus, based on the recognized gesture, the one or more processors may maintain the host vehicle a distance from the pedestrian, increase safety margin/constraints with the pedestrian, or send an alert to the pedestrian to prevent any accident. For example, the one or more processors may cause the vehicle to stop, slow, or move away from the pedestrian. As another example, the one or more processors may sound a horn of the host vehicle to alert pedestrians.

In one embodiment, the recognized gesture may indicate that a pedestrian is gazing toward the display device. Thus, based on the recognized gesture, the one or more processors may cause the host vehicle to perform at least one of: slowing down the host vehicle or changing the heading direction of the host vehicle.

In one embodiment, the recognized gesture may indicate that the pedestrian is looking toward the host vehicle. Thus, based on the recognized gesture, the one or more processors may cause the host vehicle to perform at least one of: maintaining the current speed of the host vehicle or maintaining the current heading direction of the host vehicle.

In one embodiment, the navigational action may include at least one of accelerating, braking, or rotating the host vehicle.

In one embodiment, the recognized gesture may include at least one of: head nodding, head rocking, hand rocking, bending of the wrist with the palm facing away from the host vehicle, bending of the wrist with the palm facing toward the host vehicle, bending of the elbow with the palm facing away from the host vehicle, bending of the elbow with the palm facing toward the host vehicle, bending of the fingers with the palm facing away from the host vehicle. The one or more processors maneuver the host vehicle based on the recognized gesture. For example, when a bend in the wrist, elbow, or finger is identified with the palm facing away from the host vehicle, the one or more processors may advance the host vehicle in its original direction, as these gestures are typically used to signal that the host vehicle is allowed to pass or approach a pedestrian. As another example, when a bend of the wrist or elbow with the palm facing the host vehicle is identified, the one or more processors may cause the host vehicle to stop, as these gestures are typically used to signal the host vehicle that a stop is required.

According to the disclosed embodiments, any of the steps and functions disclosed herein may be performed remotely in a server in communication with the host vehicle. For example, the server may train the model based on the training data and transmit the trained model to the host vehicle for use.

Determination of free space around a vehicle

The disclosed systems and methods may enable a host vehicle to determine a free space region in the host vehicle's surroundings. In particular, the disclosed systems and methods may capture multiple images showing various areas of an environment, and synthesize the captured images to form an uninterrupted 360 degree view of the environment. The uninterrupted 360 degree view may indicate a free space region where the host vehicle may safely navigate. For example, the uninterrupted 360 degree view may include a top view of the primary vehicle surroundings. The top view may indicate free space and non-free space. The main vehicle may be controlled to navigate within free space while avoiding navigating into non-free space.

FIG. 33 is an exemplary functional block diagram of memory 140 and/or 150 that may be stored/programmed with instructions for performing one or more operations in accordance with the disclosed embodiments. Although reference is made below to memory 140, one skilled in the art will appreciate that instructions may be stored in memory 140 and/or 150.

As shown in fig. 33, the memory 140 may store an image receiving module 3302, an image synthesizing module 3304, an image analyzing module 3306, and a navigation action module 3308. The disclosed embodiments are not limited to any particular configuration of memory 140. Further, the application processor 180 and/or the image processor 190 may execute instructions stored in any one of the modules 3302, 3304, 3306, 3308 included in the memory 140. Those skilled in the art will appreciate that references to processing unit 110 in the following discussion may refer to application processor 180 and image processor 190 individually or collectively. Accordingly, the steps of any of the following processes may be performed by one or more processing devices.

In one embodiment, the image receiving module 3302 may store instructions (such as computer vision software) that, when executed by the processing unit 110, cause the processing unit 110 to receive a plurality of images from one or more cameras on the host vehicle that represent areas in front of, behind, and to the sides of the host vehicle. As described above, the camera may include one or more of image capture device 122, image capture device 124, and image capture device 126. Fig. 34A shows an example of a plurality of cameras 3403 on a main vehicle 3401. As shown, each of the plurality of cameras 3403 may have a field of view ("FOV", marked by dashed lines in fig. 34A) that covers a portion of the surrounding environment of the main vehicle 3401. The plurality of cameras 3403 may have overlapping FOVs such that the combined FOV may cover an uninterrupted 360 degree view of the surrounding of the main vehicle 3401. In some embodiments, one or more of the plurality of cameras 3403 may include a wide angle lens to increase their FOV. It is contemplated that the present disclosure is not limited by the example shown in fig. 34A. For example, in some embodiments, instead of using multiple cameras mounted at different locations, the multiple images may be captured by a single camera that rotates continuously about a vertical axis extending through the roof of the main vehicle 3401.

Referring back to fig. 33, in one embodiment, the image composition module 3304 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to correlate the plurality of images to form an uninterrupted 360 degree view of the host vehicle surroundings. In particular, the processing unit 110 may correlate the plurality of images by detecting real world points represented in two or more of the plurality of images and using the detected real world points to align images of adjacent areas in the host vehicle surroundings. The real world point may correspond to some object in an overlapping region of the camera FOV (e.g., sun, a portion of a building structure, a portion of a pedestrian, a portion of another vehicle, etc.) and have pixel values that are distinguishable from the background. Based on the correlated images, the processing unit 110 may generate an uninterrupted 360 degree view of the environment.

In the disclosed embodiments, the uninterrupted 360 degree view may be represented in various ways. In one embodiment, the uninterrupted 360 degree view may be represented as a 360 degree panoramic image, which may be generated by stitching the related plurality of images. In one embodiment, the uninterrupted 360 degree view may be represented as a top view of the primary vehicle surroundings, which may be generated by simulating a top view of the surroundings using the relevant plurality of images. For example, the processing unit 110 may generate the top view by generating a sparse map representing the surrounding of the host vehicle and inserting objects in the environment into the sparse map according to their locations in the associated plurality of images. An example of such a top view is shown in fig. 34B, where the main vehicle 3401 may travel on a road 3421. As shown, the top view may provide a simulated view from a predetermined height above the roof surface of the main vehicle 3401. The top view shows one or more obstacles that the main vehicle 3401 may not be able to travel or may be expected to avoid, such as a pit 3411, building structures (or houses) 3412, 3413, pedestrians 3414, and walls (or fences) 3415. The top view also shows one or more locations where the main vehicle 3401 may travel or park, such as roads 3411, traffic lanes 3422, curbs 3423, and parking lots 3424.

Referring back to fig. 33, in one embodiment, the image analysis module 3306 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to analyze the relevant plurality of images to identify a free-space region in the host vehicle surroundings. In particular, the processing unit 110 may perform a feature analysis of the related plurality of images to identify one or more locations associated with the free space region. Exemplary free space regions may include one or more of roads, driveways, parking lots, or sidewalks, etc., or combinations thereof. Thus, the features associated with the free-space region may include one or more points or lines associated with the boundary of the free-space region, a landmark, one or more points or lines associated with the landmark, etc., or a combination thereof. Referring to the example shown in fig. 34B, processing unit 110 may determine one or more locations associated with the free space region. For example, the processing unit 110 may determine one side (i.e., location identifier) of the roadway 3422 based on analysis of the related plurality of images. As another example, the processing unit 110 may determine one or more points or lines along the curb 3423 of the road 3421 and the boundary of the curb 3423. As yet another example, the processing unit 110 may determine the presence of the building structure 3413 (i.e., road sign) adjacent the roadway 3422 and the parking lot 3424 based on the related plurality of images.

Similarly, the processing unit 110 may perform a feature analysis of the relevant plurality of images to identify one or more locations outside the free space region (i.e., non-free space). Still referring to the example in fig. 34B, the non-free space may include detected obstructions such as walls 3415, areas 3414 including pedestrians, areas 3411 including potholes.

The processing unit 110 may also determine a free space region based on the determined location. For example, still referring to fig. 34B, the processing unit 110 may identify a continuous free space region formed by the road 3411, the roadway 3422, the curb 3423, and the parking lot 3424. Each of the road 3411, the roadway 3422, the curb 3423, and the parking lot 3424 may form a free space sub-region in which the main vehicle 3401 freely moves. The free space sub-regions may be separated from each other by non-free spaces. For example, the roadway 3422 is separated from the parking lot 3424 by the building structure 3413.

Referring back to fig. 33, in one embodiment, the navigation action module 3308 may store instructions that, when executed by the processing unit 110, cause the processing unit 110 to cause the host vehicle to navigate within the identified free-space region while avoiding navigating within regions outside of the identified free-space region. In particular, the processing unit 110 may determine a navigation action to allow the host vehicle to safely navigate in the identified free space region. Referring to the example in fig. 34B, the free space region may include a lane 3422, and the processing unit 110 may determine a series of braking, accelerating, and steering actions that allow the main vehicle 3401 to enter or leave the lane 3422 without colliding with the boundaries of the lane 3422. In addition, a pedestrian 3414 may be present on the traffic lane 3422. To avoid a potential collision, the processing unit 110 may determine a steering action that causes the main vehicle 3401 to move away from the pedestrian 3414. Similarly, the processing unit 110 may determine a series of braking, accelerating, and steering actions that allow the main vehicle 3401 to enter or leave the parking lot 3424.

The processing unit 110 may cause one or more actuator systems associated with the host vehicle to perform the determined navigational action of the host vehicle. Exemplary navigational actions may include braking actions, coasting actions, accelerating or steering actions, and the like, or combinations thereof. For example, referring to fig. 33, an autonomous system associated with the vehicle 3302 may cause one or more actuator systems associated with the vehicle 3302 to navigate into the lane 3351 to reduce speed and turn to the left side lane 3351 by taking braking action.

In accordance with the disclosed embodiments, any of the modules disclosed herein (e.g., modules 3302, 3304, 3306, and 3308) may implement techniques associated with a trained system (such as a neural network or deep neural network) or an untrained system (such as a system that may be configured to detect and/or tag objects in an environment from which sensed information is captured and processed using computer vision algorithms). In one embodiment, the image analysis module 3306 and/or other image processing modules may be configured to use a combination of trained and untrained systems.

According to the disclosed embodiments, any of the modules disclosed herein (e.g., modules 3302, 3304, 3306, and 3308) may be stored remotely in a server in communication with the host vehicle. The captured image may initially be transmitted to a server, which may include one or more processors similar to the processing unit 110 described above. The one or more processors at the server may execute the instructions in modules 3302, 3304, 3306, and/or 3308 in a manner similar to that described above in connection with processing unit 110. The server may then transmit the results to the host vehicle 110. For example, in one embodiment, the image analysis module 3006 may be stored in a server. The server may execute instructions of the image analysis module 3306 to identify the free-space region and then transmit information about the free-space region to the host vehicle for further use.

Fig. 35 is a flow chart illustrating an exemplary process 3500 for navigating a host vehicle based on a determined free space region around the host vehicle in accordance with the disclosed embodiments. At step 3502, the processing unit 110 may receive a plurality of images representing an environment of the host vehicle. The plurality of images may be received via the data interface 128. For example, cameras included in the image acquisition unit 120, such as the image capture devices 122, 124, and 126 having fields of view 202, 204, and 206, may capture images representing areas in front of, behind, and to the sides of the host vehicle and transmit them to the processing unit 110 via a digital connection (e.g., USB, wireless, bluetooth, etc.).

At step 3504, the processing unit 110 may correlate the plurality of images to provide an uninterrupted 360 degree view of the host vehicle surroundings. In one embodiment, the processing unit 110 may render the uninterrupted 360 degree view as a top view of the host vehicle surroundings as if viewed from a predetermined height above the roof surface of the host vehicle.

At step 3506, the processing unit 110 can analyze the correlated plurality of images to identify free-space regions in the host vehicle surroundings. For example, the free space region may be part of a city environment, and the processing unit may analyze the correlated plurality of images to identify features representing free space, such as lanes, parking lots, roads, or curbs, which may collectively form a continuous free space region. The processing unit 110 may also analyze the associated plurality of images to identify features representing non-free space, such as obstacles, pedestrians, building structures, or potholes.

At step 3508, the processing unit 110 may cause the host vehicle to navigate within the identified free-space region while avoiding navigating within regions outside of the identified free-space region. In particular, the processing unit 110 may determine one or more navigational actions to allow the host vehicle to move within the free-space region without colliding with boundaries of the free-space region. The processing unit 110 may then initiate one or more navigational actions by sending one or more control signals to a navigational actuator, such as a brake, steering wheel, accelerator, etc.

According to the disclosed embodiments, any steps of the process 3500 disclosed herein (e.g., steps 3502, 3504, 3506, and 3508) can be performed remotely in a server in communication with the host vehicle. For example, the server may perform steps 3502-3506 and communicate the identified free-space region to the host vehicle, which may then determine a navigational action based on the free-space region.

The foregoing description is provided for convenience of explanation. It is not intended to be exhaustive or to be limited to the precise form or embodiment disclosed. Modifications and adaptations may become apparent to those skilled in the art from a consideration of the present specification and practice of the disclosed embodiments. Additionally, while aspects of the disclosed embodiments are described as being stored in memory, those skilled in the art will appreciate that these aspects can also be stored on other types of computer-readable media, such as secondary storage devices, e.g., hard disk or CD ROM, or other forms of RAM or ROM, USB media, DVD, blu-ray, 4Kultra HD Blu-ray, or other optical drive media.

Computer programs based on the present written description and the disclosed methods are within the skill of an experienced developer. The various programs or program modules can be created using any of the techniques known to those skilled in the art, or can be designed in conjunction with existing software. For example, program segments or program modules can be designed by or with the aid of Net Framework, net Compact Framework (and related languages such as Visual Basic, C, etc.), java, C++, objective-C, HTML, HTML/AJAX combinations, XML, or HTML with Java applets included.

Moreover, although illustrative embodiments have been described herein, the scope of any and all embodiments has equivalent elements, modifications, omissions, combinations (e.g., of aspects across various embodiments), adaptations and/or alterations as would be appreciated by those in the art based on the present disclosure. Limitations in the claims are to be interpreted broadly based on the language employed in the claims and not limited to examples described in the present specification or during the prosecution of the present application. Examples are to be understood as non-exclusive. Furthermore, the steps of the disclosed methods may be modified in any manner, including by reordering steps and/or inserting or deleting steps. It is intended, therefore, that the specification and examples be considered as exemplary only, with a true scope and spirit being indicated by the following claims and their full scope of equivalents.

Claims

1. A navigation system for a host vehicle, the navigation system comprising:

at least one processor programmed to:

receiving a captured image from a camera on the host vehicle, the captured image comprising a representation of a obscured pedestrian in the environment of the host vehicle, wherein the captured image lacks a representation of an area of the obscured pedestrian contacting a ground surface;

providing the captured image to an analysis module configured to generate an output for the captured image, wherein the generated output includes an indicator of a contact location of the obscured pedestrian with the ground surface;

receiving, from the analysis module, the generated output including the indicator of the contact location of the obscured pedestrian with the ground surface; and

causing at least one navigational action of the host vehicle based on the indicator of the contact location of the obscured pedestrian with the ground surface,

wherein the analysis module comprises at least one trained model trained based on training data comprising a plurality of images, each image comprising:

At least one representation of a training pedestrian, an

The area where the training pedestrian contacts the ground surface, and

wherein the training data further comprises the plurality of images, each image modified to mask the area of the training pedestrian contacting the ground surface.

2. The navigation system of claim 1, wherein the analysis module includes at least one trained model trained based on training data, comprising:

a plurality of images, each image including a representation of at least one training pedestrian, and

LIDAR depth information corresponding to the plurality of images.

3. The navigation system of claim 1 or 2, wherein the at least one trained model comprises a neural network.

4. The navigation system of claim 1, wherein the indicator of the contact location of the obscured pedestrian with the ground surface comprises a bounding box.

5. The navigation system of claim 4, wherein the bounding box is used to determine an estimated time to collision between the host vehicle and the pedestrian.

6. The navigation system of claim 4, wherein the bounding box is used to determine an estimated distance between the pedestrian and at least a portion of the host vehicle.

7. The navigation system of claim 4, wherein the bounding box encloses the representation of the at least one occluded pedestrian in the captured image and is sized to enclose such an area if the area where the occluded pedestrian contacts a ground surface has been represented in the captured image.

8. The navigation system of claim 4, wherein the bounding box is defined by at least two pixel sizes relative to a reference pixel associated with the captured image.

9. The navigation system of claim 4, wherein the bounding box is defined by at least two real world dimensions.

10. The navigation system of claim 4, wherein the area of the obscured pedestrian contacting the ground surface is obscured by at least a portion of the host vehicle from view in the captured image, and wherein the bounding box includes an area of the captured image that includes a representation of the at least a portion of the host vehicle.

11. The navigation system of claim 10, wherein the at least a portion of the host vehicle comprises at least a portion of a hood of the host vehicle.

12. The navigation system of claim 1, wherein the ground surface comprises at least one of: road surface or pavement surface.

13. The navigation system of claim 1, wherein the at least one navigational action includes at least one of: accelerating, braking, or rotating the host vehicle.

14. The navigation system of claim 1, wherein the at least one processor is further programmed to:

a buffer is maintained that includes a plurality of images acquired prior to the captured image, wherein one or more of the plurality of images includes a representation of the occluded pedestrian of the captured image.

15. The navigation system of claim 14, wherein one or more of the plurality of images includes a representation of the area of the obscured pedestrian contacting the ground surface.

16. The navigation system of claim 14, wherein the plurality of images includes up to 20 acquired image frames.

17. The navigation system of claim 14,

wherein the at least one processor is programmed to provide one or more of the plurality of images to the analysis module, and

Wherein the indicator of the contact location of the obscured pedestrian with the ground surface is determined based at least in part on the one or more of the plurality of images.