JP7337121B2 - Roundabout navigation method, apparatus, device and storage medium - Google Patents

Description

The present application relates to the field of vehicle technology, in particular to the field of navigation technology, and proposes a roundabout navigation method, apparatus, device and storage medium.

AR (Augmented Reality, Augmented Reality) navigation combines AR technology and map navigation to provide users with more intuitive navigation guidance. The display serves an intuitive guidance function.

Currently, vehicles have a problem that guide lines cannot be displayed in close contact with the road surface accurately when performing AR navigation in a roundabout scene.

This application is intended to solve at least to some extent one of the technical problems in the related art.

To this end, the present application proposes a roundabout navigation method, apparatus, device and storage medium.

An embodiment of the first aspect of the present application proposes a roundabout navigation method, said method comprising the steps of: obtaining a distance from a roundabout of a vehicle; if below a threshold, obtaining an image of the roundabout; separating road features of the roundabout from the image of the roundabout to obtain distances between the vehicle and the road features; and correcting navigation positioning information based on the distance between the vehicle and the road surface feature.

An embodiment of the second aspect of the present application proposes a roundabout navigation device, said device comprising: a distance obtaining module for obtaining a distance of a vehicle from a roundabout; an image acquisition module for acquiring an image of the roundabout, if lower than a preset threshold, to separate the road surface features of the roundabout from the roundabout image to determine the distance between the vehicle and the road surface features; and a correction module for correcting navigation positioning information based on the distance between the vehicle and the road surface feature.

An embodiment of the third aspect of the present application proposes an electronic device, said electronic device comprising at least one processor and a memory communicatively coupled to said at least one processor, wherein said The memory stores instructions executable by the at least one processor, the instructions for enabling the at least one processor to perform a roundabout navigation method according to an embodiment of the first aspect. Executed by one processor.

An embodiment of the fourth aspect of the present application proposes a non-transitory computer-readable storage medium having computer instructions stored thereon, said computer instructions being stored in said computer as described in an embodiment of the first aspect. Used to drive the intersection navigation method.

An embodiment of the fifth aspect of the present application proposes a computer program, said computer program, when running on a computer, causing said computer to perform the roundabout navigation method according to an embodiment of the first aspect. used for

One embodiment of the above application has the following advantages and beneficial effects. Obtaining the distance of the vehicle from the roundabout, obtaining an image of the roundabout if the distance of the vehicle from the roundabout is lower than a preset threshold, and further obtaining the road surface characteristics of the roundabout from the roundabout image is used to obtain the distance between the vehicle and the road surface feature, and the navigation positioning information is corrected based on the distance between the vehicle and the road surface feature, so the guideline can be drawn accurately. , to further improve the accuracy of AR navigation.

It is understood that nothing contained herein is intended to identify key features or critical features of the embodiments of the present application, nor is it intended to limit the scope of the present application. want to be Other features of the present application will be readily understood from the following specification.

The drawings are used for better understanding of the present technical solution and are not intended to limit the present application.
1 is a schematic flow chart of a roundabout navigation method provided in an embodiment of the present application; Fig. 4 is a schematic flow chart of obtaining the distance between a vehicle and a road surface feature provided in the embodiments of the present application; 3 is a schematic flowchart of another roundabout navigation method provided in an embodiment of the present application; 1 is a schematic structural diagram of a roundabout navigation device provided in an embodiment of the present application; FIG. FIG. 4 is a schematic structural diagram of another roundabout navigation device provided in an embodiment of the present application; 1 is a block diagram of an exemplary electronic device suitable for implementing embodiments of the present application; FIG.

Illustrative embodiments of the present application are described below in conjunction with the drawings, and various details of the embodiments of the present application are included therein for ease of understanding and are merely exemplary. should be regarded as Accordingly, those skilled in the art can make various changes and modifications to the examples described herein without departing from the scope and spirit of this application. Similarly, for the sake of clarity and brevity, the following description omits descriptions of well-known functions and constructions.

FIG. 1 is a schematic flowchart of a roundabout navigation method provided in an embodiment of the present application, and as shown in FIG. 1, the method includes steps 101-104.

Step 101, obtain the distance of the vehicle from the roundabout.

Step 102, if the distance of the vehicle from the roundabout is lower than a preset threshold, obtain an image of the roundabout.

The roundabout navigation method of the embodiments of the present application can be applied to AR (Augmented Reality) navigation of vehicles.

In this embodiment, the distance of the vehicle from the roundabout can be obtained first. Optionally, the position of the roundabout can be pre-obtained and the real-time position of the vehicle can be obtained via a positioning system, and the distance from the roundabout of the vehicle can be determined based on the real-time position of the vehicle and the position of the roundabout. calculate.

In this embodiment, if the vehicle's distance from the roundabout is less than a preset threshold, an image of the roundabout is captured. Here, the image of the roundabout can be obtained by photographing with an image collecting device installed in the vehicle. For example, the image of the roundabout is obtained by photographing the road environment around the vehicle with an in-vehicle camera.

For example, obtain the position of the roundabout in advance, detect the position of the vehicle in real time while the vehicle is traveling, calculate the real-time distance of the vehicle from the roundabout based on the position information of the roundabout and the position of the vehicle, Further, if the distance of the vehicle from the roundabout is less than a preset threshold, it is determined that the vehicle is traveling toward the roundabout, in which case an image of the roundabout is acquired by the on-board image acquisition device.

Here, the preset threshold can be set as required, and can also be determined according to a large amount of experimental data, for example, the preset threshold is 20 meters.

Step 103, separating the road feature of the roundabout from the image of the roundabout to obtain the distance between the vehicle and the road feature.

In this embodiment, after obtaining the image of the roundabout, semantic segmentation processing can be performed on the image of the roundabout by semantic segmentation technology, and by separating the road surface features of the roundabout from the image of the roundabout, Obtaining the area of the road surface feature in the image of the roundabout is realized. Here, road surface features include, for example, road edges, road surfaces, corners, and the like.

As an example, a sample image of a roundabout road may be pre-collected, where the sample image includes marked areas and types of road surface features corresponding to the marked areas, for example, the sample image includes a road edge Areas, corner areas, road surface areas, etc. are marked. Moreover, the sample image can be input to the preset model to obtain the prediction result, and the image semantic segmentation model can be trained according to the prediction result and the marked result. Furthermore, after obtaining the image of the roundabout, the image semantic segmentation model is used to perform semantic segmentation processing on the image of the roundabout to obtain the road surface features such as road edges, road surfaces, and corners in the image of the roundabout.

In this embodiment, obtaining the distance between the vehicle and the road surface feature may include, for example, obtaining the distance between the vehicle and the corner of the roundabout, e.g. may further include obtaining a distance between

Step 104, correct navigation positioning information according to the distance between the vehicle and the road surface feature.

In this embodiment, navigation positioning information is acquired via a positioning system, and the navigation positioning information is corrected based on the distance between the vehicle and the road surface feature. Here, the positioning system includes GPS (Global Positioning System), and the navigation positioning information is, for example, the GPS position. GPS positioning usually has an error of 5-10 meters, so depending on the actual distance from the road surface features such as the edge/corner of the vehicle, along with the current road trajectory, the error of the positioning information of the navigation , thereby obtaining an accurate vehicle position.

For example, after obtaining the navigation positioning information through the positioning system, the road on which the vehicle is currently located can be determined based on the navigation positioning information and the pre-stored map information; and the road surface feature and the road on which the vehicle is currently located, determine the actual position of the vehicle on the current road, and replace the positioning information of the navigation with the actual position.

As another example, the road on which the vehicle is currently located can be determined based on navigation positioning information and pre-stored map information, and furthermore the distance between the vehicle and the road surface feature and the distance between the vehicle and the current road surface. Based on the road on which it is located, determine the vehicle's actual position on the current road. Further, the actual position and the navigation positioning information are compared to obtain an error between them, and if the error is greater than a preset threshold, the navigation positioning information is modified according to the actual position.

A roundabout belongs to a special type of road, and in related technology, when a vehicle performs AR (Augmented Reality, augmented reality) navigation in a roundabout scene, a drawn guideline may be displayed in close contact with the road surface. Not possible, the guideline will not be on the road surface, affecting the accuracy of AR navigation.

A roundabout navigation method in an embodiment of the present application obtains a distance from a roundabout of a vehicle, and obtains an image of the roundabout if the distance of the vehicle from the roundabout is lower than a preset threshold. . Further, the road feature of the roundabout is separated from the image of the roundabout to obtain the distance between the vehicle and the road feature. Correct navigation positioning information based on the distance between the vehicle and road surface features. As a result, by navigating with the corrected navigation positioning information, it is possible to accurately draw the guidelines, and as a result, the guidelines are displayed in close contact with the road surface. By correcting the positioning information of , the guideline does not adhere to the road surface in the roundabout scene, which solves the problem and further improves the accuracy of AR navigation.

Based on the above embodiment, an implementation mode for obtaining the distance between the vehicle and the road surface feature will be described below.

FIG. 2 is a schematic flowchart of obtaining the distance between the vehicle and the road surface feature provided in the embodiment of the present application, as shown in FIG. 2, obtaining the distance between the vehicle and the road surface feature The steps include steps 201-204.

Step 201, obtaining pixel values of a plurality of pixels corresponding to road features.

Step 202, depth estimation is performed on the pixel values of the plurality of pixels based on a deep learning algorithm to generate depth vector values of the plurality of pixels for the vehicle's camera.

In this embodiment, the road surface features in the image of the roundabout have already been obtained by semantic segmentation, i.e., the target area corresponding to the road surface features in the image is determined, and the pixel values of a plurality of pixels in the target area are determined. to get Further, depth estimation is performed on the pixel values of the plurality of pixels based on a deep learning algorithm to generate a plurality of pixel depth vector values for the camera within the vehicle. Here, the depth vector value corresponding to each pixel is (0 to 1), for example.

Step 203, obtain the position of the camera in the vehicle.

In this embodiment, the vehicle is equipped with a camera for collecting images of the roundabout. The camera may be installed at any position within the vehicle, for example, the camera may be installed at the front of the vehicle, or the camera may be installed at the rear of the vehicle, here It is not specifically limited.

Step 204, generate the distance between the road surface feature and the vehicle based on the camera position and the depth vector values of the plurality of pixels for the vehicle's camera.

In this embodiment, the position of the camera in the vehicle can be obtained, and the internal and external parameters of the camera can be obtained, where the camera internal parameters are parameters related to the characteristics of the camera itself; For example, it refers to the focal length of the camera, the pixel size, etc., and the camera extrinsic parameters refer to parameters in the world coordinate system, such as the rotation direction of the camera. Further, based on the intrinsic and extrinsic parameters of the camera and the depth vector values of the plurality of pixels for the camera of the vehicle, the absolute depth is calculated, and the distance between the road surface feature and the vehicle is determined based on the absolute depth.

Embodiments of the present application can determine the distance between the vehicle and the road surface feature based on the depth estimation and the camera position, and accurately obtain the distance between the vehicle and the road surface feature; It is used to correct the navigation positioning information to improve the accuracy of the navigation positioning information.

After correcting the navigation positioning information based on the distance between the vehicle and the road surface feature according to the above embodiment, the method further includes the step of navigating based on the corrected navigation positioning information.

In an embodiment of the present application, the step of navigating based on the corrected navigation positioning information includes obtaining roundabout route data and obtaining branch lines of the current route based on the roundabout route data; obtaining a trajectory of a branch line of the current route, and generating a guideline based on the trajectory and the corrected navigation positioning information. Here, the route data of the roundabout is obtained based on the pre-stored map information in the related map application, the branch line of the current route on which the vehicle is located is obtained from the route data of the roundabout, and the branch line of the current route is obtained from the pre-stored map information. A guideline is generated based on the trajectory of the branch line of the current route in the map information and the navigation positioning information after correction.

In an embodiment of the present application, the step of navigating based on the corrected navigation positioning information includes obtaining roundabout route data and obtaining branch lines of the current route based on the roundabout route data; obtaining a fitting trajectory based on the branches of the current route, generating a guideline based on the fitting trajectory, and adjusting the guideline based on the corrected navigation positioning information.

The step of performing navigation based on the corrected navigation positioning information will be further described below.

FIG. 3 is a schematic flowchart of another roundabout navigation method provided in an embodiment of the present application, and as shown in FIG. 3, the method includes steps 301-305.

Step 301, obtaining roundabout route data.

Step 302, obtaining the branch line of the current route according to the roundabout route data;

In this embodiment, the roundabout route data in the pre-stored map information is acquired, where the roundabout route data includes current route branch lines, route branch line nodes, and route shape points. , but not limited to.

Step 303, obtain a fitting trajectory according to the branch line of the current route.

In this example, each branch of the route can correspond to one fitting trajectory.

In one embodiment of the present application, the fitting trajectory is obtained by obtaining a bounding rectangle of a roundabout, obtaining a plurality of sample trajectories in which a plurality of sample vehicles enter the bounding rectangle, and fitting a plurality of sample trajectories of a plurality of sample vehicles. , obtained by the step of generating the fitting trajectory.

Here, the circumscribed rectangle of the roundabout is used to cover the entire area of the roundabout. Specifically, it may be set as required. good too. As an example, if the roundabout is a circular area with a diameter of 200m, then construct a bounding rectangle of 200m*200m so that the bounding rectangle contains the entire area of the roundabout.

In this embodiment, a plurality of sample vehicles drive into a roundabout, sample trajectory refers to the trajectory of a vehicle driving through a roundabout, and each sample vehicle can correspond to one sample trajectory. . A plurality of sample trajectories of a plurality of sample vehicles are statistically fitted to fit the plurality of sample trajectories to generate a fitted trajectory.

In one embodiment of the present application, fitting a plurality of sample trajectories of a plurality of sample vehicles to generate a fitted trajectory includes clustering the plurality of sample trajectories to generate a plurality of trip types; fitting the sample trajectories corresponding to , to generate a fitted trajectory corresponding to each run type.

Here, the driving type includes, for example, a type entering a roundabout, a type leaving a roundabout, and the like. As one possible embodiment, the type of travel can be determined by the position of each route shape point in the road network, the course angle and the topological shape of the traveled trajectory, different positions, course angles and topological shapes being of different types. handle.

In this embodiment, the driving type of each sample trajectory is determined, and a plurality of sample trajectories are clustered to generate a plurality of driving types. For each travel type, when the number of sample trajectories corresponding to the travel type reaches a preset number, the sample trajectories corresponding to the travel type are fitted to generate a fitting trajectory.

Here, the embodiment of fitting the sample trajectory corresponding to each driving type and generating the fitting trajectory corresponding to each driving type can be selected as needed, and one possible embodiment is the Bezier fitting algorithm to fit the sample trajectory corresponding to each driving type to generate a fitting trajectory corresponding to each driving type.

This realizes that the roundabout fitting trajectory is generated based on the actual driving data of the vehicle, and the branch line of the determined current route corresponds to the branch line of the current route in the roundabout fitting trajectory. A fitting trajectory can be obtained.

Step 304, generate a guideline based on the fitting trajectory.

Step 305, adjust the guidelines according to the corrected navigation positioning information;

In this embodiment, by substituting the fitting trajectory for the roundabout route data in the map, a guideline is drawn based on the fitting trajectory, and the guideline is adjusted based on the corrected navigation positioning information. As an example, if it is determined that the vehicle is currently traveling straight based on the fitting trajectory and the corrected navigation positioning information, a straight guideline is drawn, and the vehicle is guided based on the fitting trajectory and the corrected navigation positioning information. If we determine that we are currently curving to the left, then we draw a guide line that turns to the left.

In one embodiment of the present application, the distance between the vehicle and the road surface feature is obtained based on a preset period, and the navigation positioning information is corrected based on the distance between the vehicle and the road surface feature. do. As an example, every k seconds, an image of the roundabout is acquired and the road surface features of the roundabout are separated from the roundabout image, where k is 1, for example. When the road surface feature is acquired, the distance between the vehicle and the road surface feature is further acquired, and the navigation positioning information is corrected based on the distance. In addition, after correcting the navigation positioning information each time, the guideline is adjusted according to the corrected navigation positioning information, so that the guideline is centered in the process of turning at a roundabout. can be done.

In practical application, there is a deviation between the map route data of the roundabout and the actual driving route, so in order to further improve the accuracy of the AR navigation, the roundabout navigation method of the embodiment of the present application adopts the real route. To solve the problem that the roundabout map route data is not accurate by forming a fitting trajectory by fitting the running trajectory of a vehicle, and replacing the map roundabout route data with the fitting trajectory, thereby , ensures that the drawn guideline can be accurately adhered to the road, further improving the accuracy of AR navigation.

In order to implement the above embodiments, the present application further provides a roundabout navigation device.

FIG. 4 is a schematic structural diagram of a roundabout navigation device provided in an embodiment of the present application, as shown in FIG. , including a correction module 40 .

Here, the distance acquisition module 10 is used to acquire the distance of the vehicle from the roundabout.

The image acquisition module 20 is used to acquire an image of the roundabout when the vehicle's distance from the roundabout is below a preset threshold.

A processing module 30 is used to separate the road surface features of the roundabout from the roundabout image to obtain the distance between the vehicle and the road surface features.

A correction module 40 is used to correct navigation positioning information based on the distance between the vehicle and the road surface feature.

In addition to FIG. 4, the roundabout navigation device shown in FIG. 5 further includes a navigation module 50, a trajectory acquisition module 60 and a generation module .

Here, the navigation module 50 is used for navigation based on the corrected navigation positioning information.

In one embodiment of the present application, the navigation module 50 specifically obtains the route data of the roundabout, obtains the branch line of the current route based on the route data of the roundabout, and follows the branch line of the current route. to obtain a fitting trajectory based on, generate a guideline based on the fitting trajectory, and adjust the guideline based on the corrected navigation positioning information. Here, the trajectory acquisition module 60 is used to acquire a bounding rectangle of a roundabout, and to acquire a plurality of sample trajectories in which a plurality of sample vehicles enter the bounding rectangle.

A generation module 70 is used to fit a plurality of sample trajectories of a plurality of sample vehicles to generate a fitted trajectory.

In one embodiment of the present application, the generation module 70 includes a clustering unit for clustering the plurality of sample trajectories and generating a plurality of run types, and fitting a sample trajectory corresponding to each run type, each run type and a generation unit for generating a fitting trajectory corresponding to .

Further, the generating unit is specifically used for fitting the sample trajectory corresponding to each driving type with a Bezier fitting algorithm to generate said fitting trajectory corresponding to each driving type.

In one embodiment of the present application, the processing module 30 specifically obtains pixel values of a plurality of pixels corresponding to the road surface feature, and based on a deep learning algorithm, for the pixel values of the plurality of pixels: performing depth estimation to generate depth vector values of the plurality of pixels for a camera of the vehicle, obtaining a position of the camera at the vehicle, and determining the position of the camera and the plurality of pixels for the vehicle camera is used to generate the distance between the road surface feature and the vehicle based on the depth vector value of .

In one embodiment of the present application, the distance between the vehicle and the road surface feature is obtained based on a preset period, and navigation positioning information is obtained based on the distance between the vehicle and the road surface feature. correct.

The interpretation and description of the roundabout navigation method of the previous embodiment are equally applicable to the roundabout navigation apparatus of the present embodiment, and detailed description thereof will be omitted here.

The roundabout navigation device of the embodiment of the present application obtains the distance of the vehicle from the roundabout, and obtains the image of the roundabout when the distance of the vehicle from the roundabout is lower than a preset threshold. . Further, the road feature of the roundabout is separated from the image of the roundabout to obtain the distance between the vehicle and the road feature. The navigation positioning information is corrected based on the distance between the vehicle and the road surface feature, and navigation is performed based on the corrected navigation positioning information. As a result, the guideline can be accurately drawn and displayed in close contact with the road surface, and by correcting navigation positioning information with the distance from the vehicle's road surface characteristics, the guideline adheres closely to the road surface in roundabout scenes. It solves the problem of not doing so and further improves the accuracy of AR navigation.

In order to implement the above embodiments, the present application further proposes a computer program product which, when instructions in the computer program product are executed by a processor, performs a roundabout navigation method according to any of the preceding embodiments. is carried out.

According to embodiments of the present application, the present application further provides an electronic device and a readable storage medium.

Referring to FIG. 6, it is a block diagram of electronics of a roundabout navigation method according to an embodiment of the present application. Electronic equipment is intended to represent various forms of digital computers such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronics can also represent various forms of mobile devices such as personal digital processors, mobile phones, smart phones, wearable devices, and other similar computing devices. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the description and/or required implementation of the application herein.

As shown in FIG. 6, the electronic device includes one or more processors 601, memory 602, and interfaces for connecting components, including high speed and low speed interfaces. Each component is interconnected by a different bus and can be mounted on a common motherboard or otherwise mounted as desired. The processor is capable of processing instructions executed within the electronic device, which instructions are stored in memory for displaying graphical information of the GUI on an external input/output device (eg, a display device coupled to the interface, etc.). contains instructions stored in or in memory. In other embodiments, multiple processors and/or multiple buses may be used along with multiple memories, if desired. Similarly, multiple electronic devices can be connected, and each electronic device can provide some required operation (eg, a server array, a group of blade servers, or a multi-processor system). . In FIG. 6, one processor 601 is taken as an example.

Memory 602 is a non-transitory computer-readable storage medium provided by the present application. Therein, stored in the memory are instructions executable by at least one processor, such that the at least one processor performs the roundabout navigation method provided by the present application. A non-transitory computer-readable storage medium of the present application stores computer instructions for causing a computer to perform the roundabout navigation method provided by the present application.

The memory 602, as a non-transitory computer-readable storage medium, stores program instructions/modules (for example, the distance acquisition module 10, the image acquisition module 20, etc. shown in FIG. 4) corresponding to the roundabout navigation method in the embodiments of the present application. , processing module 30, correction module 40, navigation module 50), non-transitory computer-executable programs and modules. Processor 601 performs the various functional applications and data processing of the server by executing non-transitory software programs, instructions and modules stored in memory 602, namely the roundabout in the above method embodiment. Implement a navigation method.

The memory 602 can include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for at least one function, and the data storage area can store , data created through the use of electronic devices, and the like. Memory 602 may also include high speed random access memory and may further include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state memory device. It is a state storage device. In some embodiments, memory 602 can optionally include memory configured remotely to processor 601, and these remote memories can be connected to the electronic device via a network. . Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The roundabout navigation method electronics may further include an input device 603 and an output device 604 . The processor 601, the memory 602, the input device 603, and the output device 604 can be connected via a bus or other manner, and the connection via the bus is taken as an example in FIG.

The input device 603 can receive input numeric or character information and generate key signal inputs for user settings and functional control of electronic devices, such as touch screens, keypads, mice, trackpads, touchpads. , a pointing stick, one or more mouse buttons, a trackball, a joystick, or the like. Output devices 604 may include display devices, supplemental lighting devices (eg, LEDs), tactile feedback devices (eg, vibration motors), and the like. Such display devices can include, but are not limited to, liquid crystal displays (LCD), light emitting diode (LED) displays, and plasma displays. In some embodiments, the display device may be a touchscreen.

Various embodiments of the systems and techniques described herein may be digital electronic circuit systems, integrated circuit systems, application specific integrated circuits (ASICs), computer hardware, firmware, software, and/or can be realized by a combination of These various embodiments are embodied in one or more computer programs, which can be executed and/or interpreted by a programmable system including at least one programmable processor. , the programmable processor, which may be a dedicated or general-purpose programmable processor, receives data and instructions from the storage system, at least one input device, and at least one output device, and transmits data and instructions to the storage system, the It can be transmitted to at least one input device and to the at least one output device.

These computing programs (also called programs, software, software applications, or code) contain programmable processor machine instructions and are written in high-level process and/or object oriented programming languages and/or assembly/machine language. A computing program can be implemented. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus for providing machine instructions and/or data to a programmable processor. , and/or apparatus (eg, magnetic disk, optical disk, memory, programmable logic device (PLD)), including a machine-readable medium for receiving machine instructions, which are machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and/or data to a programmable processor.

To provide interaction with a user, the systems and techniques described herein can be implemented on a computer that includes a display device (e.g., cathode ray tube (CRT)) for displaying information to the user. ) or LCD (liquid crystal display) monitor), and a keyboard and pointing device (e.g., mouse or trackball) through which a user can provide input to the computer. Other types of devices can also provide interaction with a user, e.g., the feedback provided to the user can be any form of sensing feedback (e.g., visual, auditory, or tactile feedback). may receive input from the user in any form (including acoustic, speech, and tactile input).

The systems and techniques described herein may be computing systems that include back-end components (e.g., data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include front-end components. system (e.g., a user computer having a graphical user interface or web browser through which a user interacts with embodiments of the systems and techniques described herein), or such a background It can be implemented on a computing system including any combination of end components, middleware components and front end components. The components of the system can be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include local area networks (LAN), wide area networks (WAN), and the Internet.

The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server is created by computer programs running on corresponding computers and having a client-server relationship to each other.

Steps may be rearranged, added, or deleted using the various forms of flow shown above. For example, each step described in this application may be performed in parallel, sequentially, or in a different order, but the technology disclosed in this application There is no limitation herein as long as the desired result of the scheme can be achieved.

The above specific embodiments do not limit the protection scope of this application. One skilled in the art can make various modifications, combinations, subcombinations, and substitutions based on design requirements and other factors. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall all fall within the protection scope of this application.

Claims (15)

A method of navigating a roundabout, comprising:
obtaining the vehicle's distance from the roundabout;
obtaining an image of the roundabout if the vehicle's distance from the roundabout is below a preset threshold;
After obtaining the image of the roundabout, performing a semantic segmentation process on the image of the roundabout by semantic segmentation to extract road surface features including road edges, road surfaces and corners of the roundabout from the image of the roundabout . separating to obtain a distance between the vehicle and the road surface feature;
correcting positioning information for navigation based on the distance between the vehicle and the road surface feature;
navigating based on the corrected navigation positioning information;
The step of navigating based on the corrected navigation positioning information includes:
obtaining route data for the roundabout;
obtaining branches of the current route based on the roundabout route data;
obtaining a fitting trajectory based on branches of the current route;
generating guidelines based on the fitting trajectory;
adjusting the guideline based on the corrected navigation positioning information;
and providing intuitive guidance by displaying the guideline on the road ahead via an AR device .
How to navigate roundabouts. The fitting trajectory is
obtaining a bounding rectangle of the roundabout;
obtaining a plurality of sample trajectories in which a plurality of sample vehicles enter the bounding rectangle;
fitting a plurality of sample trajectories of the plurality of sample vehicles to generate the fitted trajectory;
The roundabout navigation method according to claim 1, characterized in that: fitting a plurality of sample trajectories of the plurality of sample vehicles to generate the fitted trajectories,
clustering the plurality of sample trajectories to generate a plurality of run types;
fitting the sample trajectory corresponding to each run type to generate a fitted trajectory corresponding to each run type;
3. The roundabout navigation method according to claim 2, wherein: Fitting a sample trajectory corresponding to each run type and generating a fitted trajectory corresponding to each run type includes:
fitting a sample trajectory corresponding to each run type with a Bezier fitting algorithm to generate the fitted trajectory corresponding to each run type;
4. The roundabout navigation method according to claim 3, wherein: Obtaining a distance between the vehicle and the road surface feature comprises:
obtaining pixel values of a plurality of pixels corresponding to the road surface feature;
performing depth estimation on pixel values of the plurality of pixels based on a deep learning algorithm to generate depth vector values of the plurality of pixels for a camera of the vehicle;
obtaining a position of the camera on the vehicle;
generating a distance between the road surface feature and the vehicle based on the position of the camera and the depth vector values of the plurality of pixels for the vehicle's camera;
The roundabout navigation method according to claim 1, characterized in that: acquiring a distance between the vehicle and the road surface feature based on a preset cycle, and correcting navigation positioning information based on the distance between the vehicle and the road surface feature;
The roundabout navigation method according to claim 1, characterized in that: A roundabout navigation device comprising:
a distance acquisition module for obtaining the distance of the vehicle from the roundabout;
an image acquisition module for acquiring an image of the roundabout if the vehicle's distance from the roundabout is below a preset threshold;
After obtaining the image of the roundabout, performing a semantic segmentation process on the image of the roundabout by semantic segmentation to extract road surface features including road edges, road surfaces and corners of the roundabout from the image of the roundabout . a processing module for separately obtaining a distance between the vehicle and the road surface feature;
a correction module for correcting navigation positioning information based on the distance between the vehicle and the road surface feature;
a navigation module for navigating based on the corrected navigation positioning information;
The navigation module includes:
obtaining route data for the roundabout;
obtaining branch lines of the current route based on the roundabout route data;
obtaining a fitting trajectory based on branch lines of the current route;
Intuitive guidance is provided by generating a guideline based on the fitting trajectory, adjusting the guideline based on the corrected navigation positioning information , and displaying the guideline on the road ahead via an AR device. used to do
Roundabout navigation device. obtaining a bounding rectangle of the roundabout;
a trajectory acquisition module for acquiring a plurality of sample trajectories in which a plurality of sample vehicles enter the bounding rectangle;
a generation module for fitting a plurality of sample trajectories of the plurality of sample vehicles to generate the fitted trajectories;
The roundabout navigation device according to claim 7, characterized in that: The generation module is
a clustering unit for clustering the plurality of sample trajectories to generate a plurality of run types;
a generation unit for fitting the sample trajectory corresponding to each run type and generating a fitted trajectory corresponding to each run type;
The roundabout navigation device according to claim 8, characterized in that: The generating unit is
used to fit a sample trajectory corresponding to each run type with a Bezier fitting algorithm to generate the fitted trajectory corresponding to each run type;
10. The roundabout navigation device according to claim 9, characterized in that: The processing module is
obtaining pixel values of a plurality of pixels corresponding to the road surface feature;
performing depth estimation on pixel values of the plurality of pixels based on a deep learning algorithm to generate depth vector values of the plurality of pixels for a camera of the vehicle;
obtain a position of the camera on the vehicle; and
used to generate a distance between the road surface feature and the vehicle based on the position of the camera and the depth vector values of the plurality of pixels relative to the vehicle's camera;
The roundabout navigation device according to claim 7, characterized in that: acquiring a distance between the vehicle and the road surface feature based on a preset cycle, and correcting navigation positioning information based on the distance between the vehicle and the road surface feature;
The roundabout navigation device according to claim 7, characterized in that: an electronic device,
at least one processor;
a memory communicatively coupled to the at least one processor;
The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is any one of claims 1 to 6. can perform the roundabout navigation methods of
Electronics. A non-transitory computer-readable storage medium storing computer instructions, said computer instructions for causing said computer to perform the roundabout navigation method according to any one of claims 1 to 6. used,
A non-transitory computer-readable storage medium. causing the computer to execute the roundabout navigation method according to any one of claims 1 to 6 when operating on the computer;
A computer program characterized by:

Images