EP4721026A1 - Augmenting bird's eye view object proposals with input-dependent perspective object proposals - Google Patents

Abstract

Some systems, methods, and computer program products for augmenting bird's eye view object proposals described include: receiving camera images representing a perspective view of an environment relative to a vehicle; extracting image features from the camera images; generating one or more perspective bounding boxes containing one or more detected objects based on the image features; selecting one or more first object proposals represented in a bird's eye view (BEV) view of the environment based on the one or more perspective bounding boxes; and generating one or more BEV bounding boxes containing the one or more detected objects based on: the image features represented in the perspective view of the environment, the one or more first object proposals represented in the BEV view of the environment, and one or more generic object proposals represented in the BEV view of the environment. Systems and computer program products are also provided.

Description

AUGMENTING BIRD’S EYE VIEW OBJECT PROPOSALS WITH INPUTDEPENDENT PERSPECTIVE OBJECT PROPOSALS

CROSS-REFERENCE TO RELATED APPLICATIONS

[1] This application claims priority to US Provisional Application No. 63/470,125, having a Filing Date of May 31, 2023, the disclosure of which is considered part of the disclosure of this application, and is incorporated by reference in its entirety into this application.

BACKGROUND

[2] Autonomous vehicles (AV) navigate within an environment by detecting physical objects around the AV. Object detection locates instances of objects in captured images or videos of the environment.

BRIEF DESCRIPTION OF THE FIGURES

[3] FIG. 1 is an example environment in which a vehicle including one or more components of an autonomous system can be implemented;

[4] FIG. 2 is a diagram of one or more systems of a vehicle including an autonomous system;

[5] FIG. 3 is a diagram of components of one or more devices and/or one or more systems of

FIGS. 1 and 2;

[6] FIG. 4 is a diagram of certain components of an autonomous system;

[7] FIG. 5 is a diagram of an implementation of a perception system in an autonomous vehicle

(AV) for detecting objects around the AV;

[8] FIG. 6 is a diagram of a neural network that detects objects in camera images;

[9] FIG. 7 is a diagram of an example BEV detection head;

[10] FIG. 8 illustrates an example flow chart of a process for detecting objects around the AV.

DETAILED DESCRIPTION

[11] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure for the purposes of explanation. It will be apparent, however, that the implementations described by the present disclosure can be practiced without these specific details. Tn some instances, well-known structures and devices are illustrated in block diagram form in order to avoid unnecessarily obscuring aspects of the present disclosure.

[12] Specific arrangements or orderings of schematic elements, such as those representing systems, devices, modules, instruction blocks, data elements, and/or the like, are illustrated in the drawings for ease of description. However, it will be understood by those skilled in the art that the specific ordering or arrangement of the schematic elements in the drawings is not meant to imply that a particular order or sequence of processing, or separation of processes, is required unless explicitly described as such. Further, the inclusion of a schematic element in a drawing is not meant to imply that such element is required in all implementations or that the features represented by such element may not be included in or combined with other elements in some implementations unless explicitly described as such.

[13] Further, where connecting elements such as solid or dashed lines or arrows are used in the drawings to illustrate a connection, relationship, or association between or among two or more other schematic elements, the absence of any such connecting elements is not meant to imply that no connection, relationship, or association can exist. In other words, some connections, relationships, or associations between elements are not illustrated in the drawings so as not to obscure the disclosure. In addition, for ease of illustration, a single connecting element can be used to represent multiple connections, relationships, or associations between elements. For example, where a connecting element represents communication of signals, data, or instructions (e g., “software instructions”), it should be understood by those skilled in the art that such element can represent one or multiple signal paths (e.g., a bus), as may be needed, to affect the communication.

[14] Although the terms first, second, third, and/or the like are used to describe various elements, these elements should not be limited by these terms. The terms first, second, third, and/or the like are used only to distinguish one element from another. For example, a first contact could be termed a second contact and, similarly, a second contact could be termed a first contact without departing from the scope of the described implementations. The first contact and the second contact are both contacts, but they are not the same contact.

[15] The terminology used in the description of the various described implementations herein is included for the purpose of describing particular implementations only and is not intended to be limiting. As used in the description of the various described implementations and the appended claims, the singular forms “a,” “an” and “the” are intended to include the plural forms as well and can be used interchangeably with “one or more” or “at least one,” unless the context clearly indicates otherwise. It will also be understood that the term “and/or” as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. It will be further understood that the terms “includes,” “including,” “comprises,” and/or “comprising,” when used in this description specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

[16] As used herein, the terms “communication” and “communicate” refer to at least one of the reception, receipt, transmission, transfer, provision, and/or the like of information (or information represented by, for example, data, signals, messages, instructions, commands, and/or the like). For one unit (e.g., a device, a system, a component of a device or system, combinations thereof, and/or the like) to be in communication with another unit means that the one unit is able to directly or indirectly receive information from and/or send (e.g., transmit) information to the other unit. This may refer to a direct or indirect connection that is wired and/or wireless in nature. Additionally, two units may be in communication with each other even though the information transmitted may be modified, processed, relayed, and/or routed between the first and second unit. For example, a first unit may be in communication with a second unit even though the first unit passively receives information and does not actively transmit information to the second unit. As another example, a first unit may be in communication with a second unit if at least one intermediary unit (e.g., a third unit located between the first unit and the second unit) processes information received from the first unit and transmits the processed information to the second unit. In some implementations, a message may refer to a network packet (e.g., a data packet and/or the like) that includes data.

[17] As used herein, the term “if’ is, optionally, construed to mean “when”, “upon”, “in response to determining,” “in response to detecting,” and/or the like, depending on the context. Similarly, the phrase “if it is determined” or “if [a stated condition or event] is detected” is, optionally, construed to mean “upon determining,” “in response to determining,” “upon detecting [the stated condition or event],” “in response to detecting [the stated condition or event],” and/or the like, depending on the context. Also, as used herein, the terms “has”, “have”, “having”, or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based at least partially on” unless explicitly stated otherwise. [18] Reference will now be made in detail to implementations, examples of which are illustrated in the accompanying drawings. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the various described implementations. However, it will be apparent to one of ordinary skill in the art that the various described implementations can be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail so as not to unnecessarily obscure aspects of the implementations.

[19] General Overview

[20] In some aspects and/or implementations, systems, methods, and computer program products described herein include and/or implement augmenting bird’s eye view (BEV) object proposals with input-dependent perspective object proposals in object detection. In some implementations, a neural network for object detection includes a backbone for converting camera images to image features, and a main detection head (BEV detection head) for generating BEV bounding boxes representative of detected objects based on the image features. The camera images and image features are represented and processed in perspective space (perspective view), while the BEV bounding boxes are represented and processed in BEV space (bird’s eye view). The neural network further includes an auxiliary detection head (perspective detection head) for generating perspective bounding boxes representative of detected objects based on the image features, and a neural network layer (e.g., a feed-forward layer) for converting the perspective bounding boxes to BEV object proposals. Both the BEV object proposals and BEV generic object proposals are used as initial object proposals for the main detection head.

[21] By virtue of the implementation of systems, methods, and computer program products described herein, some of the advantages of these techniques include using an auxiliary detection head operating in a perspective space (camera images input into the neural network are also in a perspective space) to add supervision to a main detection head operating in a BEV space, which facilitates faster neural network convergence and improves network accuracy. Further, BEV object proposals converted from the perspective bounding boxes are input dependent (e.g., dependent on camera images input into the neural network), which can further facilitate faster neural network convergence and thus save computation costs.

[22] Referring now to FIG. 1, illustrated is example environment 100 in which vehicles that include autonomous systems, as well as vehicles that do not, are operated. As illustrated, environment 100 includes vehicles 102a-102n, objects 104a-104n, routes 106a-106n, area 108, vehicle-to-infrastructure (V2I) device 110, network 112, remote autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118. Vehicles 102a-102n, vehicle-to- infrastructure (V2I) device 110, network 112, autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118 interconnect (e.g., establish a connection to communicate and/or the like) via wired connections, wireless connections, or a combination of wired or wireless connections. In some implementations, objects 104a-104n interconnect with at least one of vehicles 102a-102n, vehicle-to-infrastructure (V2I) device 110, network 112, autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118 via wired connections, wireless connections, or a combination of wired or wireless connections.

[23] Vehicles 102a-102n (referred to individually as vehicle 102 and collectively as vehicles 102) include at least one device configured to transport goods and/or people. In some implementations, vehicles 102 are configured to be in communication with V2I device 110, remote AV system 114, fleet management system 116, and/or V2I system 118 via network 112. In some implementations, vehicles 102 include cars, buses, trucks, trains, and/or the like. In some implementations, vehicles 102 are the same as, or similar to, vehicles 200, described herein (see FIG. 2). In some implementations, a vehicle 200 of a set of vehicles 200 is associated with an autonomous fleet manager. In some implementations, vehicles 102 travel along respective routes 106a-106n (referred to individually as route 106 and collectively as routes 106), as described herein. In some implementations, one or more vehicles 102 include an autonomous system (e.g., an autonomous system that is the same as or similar to autonomous system 202).

[24] Objects 104a-104n (referred to individually as object 104 and collectively as objects 104) include, for example, at least one vehicle, at least one pedestrian, at least one cyclist, at least one structure (e.g., a building, a sign, a fire hydrant, etc.), and/or the like. Each object 104 is stationary (e.g., located at a fixed location for a period of time) or mobile (e.g., having a velocity and associated with at least one trajectory). In some implementations, objects 104 are associated with corresponding locations in area 108.

[25] Routes 106a-106n (referred to individually as route 106 and collectively as routes 106) are each associated with (e.g., prescribe) a sequence of actions (also known as a trajectory) connecting states along which an AV can navigate. Each route 106 starts at an initial state (e.g., a state that corresponds to a first spatiotemporal location, velocity, and/or the like) and ends at a final goal state (e.g., a state that corresponds to a second spatiotemporal location that is different from the first spatiotemporal location) or goal region (e.g., a subspace of acceptable states (e.g., terminal states)). In some implementations, the first state includes a location at which an individual or individuals are to be picked-up by the AV and the second state or region includes a location or locations at which the individual or individuals picked-up by the AV are to be dropped-off. In some implementations, routes 106 include a plurality of acceptable state sequences (e.g., a plurality of spatiotemporal location sequences), the plurality of state sequences associated with (e.g., defining) a plurality of trajectories. In an example, routes 106 include only high level actions or imprecise state locations, such as a series of connected roads dictating turning directions at roadway intersections. Additionally, or alternatively, routes 106 may include more precise actions or states such as, for example, specific target lanes or precise locations within the lane areas and targeted speed at those positions. In an example, routes 106 include a plurality of precise state sequences along the at least one high level action sequence with a limited lookahead horizon to reach intermediate goals, where the combination of successive iterations of limited horizon state sequences cumulatively correspond to a plurality of trajectories that collectively form the high level route to terminate at the final goal state or region.

[26] Area 108 includes a physical area (e.g., a geographic region) within which vehicles 102 can navigate. In an example, area 108 includes at least one state (e.g., a country, a province, an individual state of a plurality of states included in a country, etc.), at least one portion of a state, at least one city, at least one portion of a city, etc. In some implementations, area 108 includes at least one named thoroughfare (referred to herein as a “road”) such as a highway, an interstate highway, a parkway, a city street, etc. Additionally, or alternatively, in some examples area 108 includes at least one unnamed road such as a driveway, a section of a parking lot, a section of a vacant and/or undeveloped lot, a dirt path, etc. In some implementations, a road includes at least one lane (e.g., a portion of the road that can be traversed by vehicles 102). In an example, a road includes at least one lane associated with (e.g., identified based on) at least one lane marking.

[27] Vehicle-to-Infrastructure (V2I) device 110 (sometimes referred to as a Vehicle-to- Infrastructure or Vehicle-to-Everything (V2X) device) includes at least one device configured to be in communication with vehicles 102 and/or V2I infrastructure system 118. In some implementations, V2I device 110 is configured to be in communication with vehicles 102, remote AV system 114, fleet management system 116, and/or V2I system 118 via network 112. In some implementations, V2I device 110 includes a radio frequency identification (RFID) device, signage, cameras (e.g., two-dimensional (2D) and/or three-dimensional (3D) cameras), lane markers, streetlights, parking meters, etc. In some implementations, V2I device 110 is configured to communicate directly with vehicles 102. Additionally, or alternatively, in some implementations V2I device 110 is configured to communicate with vehicles 102, remote AV system 114, and/or fleet management system 116 via V2I system 118. In some implementations, V2I device 110 is configured to communicate with V2I system 118 via network 112.

[28] Network 112 includes one or more wired and/or wireless networks. In an example, network 112 includes a cellular network (e.g., a long term evolution (LTE) network, a third generation (3G) network, a fourth generation (4G) network, a fifth generation (5G) network, a code division multiple access (CDMA) network, etc ), a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., the public switched telephone network (PSTN), a private network, an ad hoc network, an intranet, the Internet, a fiber optic-based network, a cloud computing network, etc., a combination of some or all of these networks, and/or the like.

[29] Remote AV system 114 includes at least one device configured to be in communication with vehicles 102, V2I device 110, network 112, fleet management system 116, and/or V2I system 118 via network 112. In an example, remote AV system 114 includes a server, a group of servers, and/or other like devices. In some implementations, remote AV system 114 is co-located with the fleet management system 116. In some implementations, remote AV system 114 is involved in the installation of some or all of the components of a vehicle, including an autonomous system, an autonomous vehicle compute, software implemented by an autonomous vehicle compute, and/or the like. In some implementations, remote AV system 114 maintains (e.g., updates and/or replaces) such components and/or software during the lifetime of the vehicle.

[30] Fleet management system 116 includes at least one device configured to be in communication with vehicles 102, V2I device 110, remote AV system 114, and/or V2I infrastructure system 118. In an example, fleet management system 116 includes a server, a group of servers, and/or other like devices. In some implementations, fleet management system 116 is associated with a ridesharing company (e.g., an organization that controls operation of multiple vehicles (e.g., vehicles that include autonomous systems and/or vehicles that do not include autonomous systems) and/or the like). [31] In some implementations, V2I system 1 18 includes at least one device configured to be in communication with vehicles 102, V2I device 110, remote AV system 114, and/or fleet management system 116 via network 112. In some examples, V2I system 118 is configured to be in communication with V2I device 110 via a connection different from network 112. In some implementations, V2I system 118 includes a server, a group of servers, and/or other like devices. In some implementations, V2I system 118 is associated with a municipality or a private institution (e.g., a private institution that maintains V2I device 110 and/or the like).

[32] The number and arrangement of elements illustrated in FIG. 1 are provided as an example. There can be additional elements, fewer elements, different elements, and/or differently arranged elements, than those illustrated in FIG. 1. Additionally, or alternatively, at least one element of environment 100 can perform one or more functions described as being performed by at least one different element of FIG. 1. Additionally, or alternatively, at least one set of elements of environment 100 can perform one or more functions described as being performed by at least one different set of elements of environment 100.

[33] Referring now to FIG. 2, vehicle 200 (which may be the same as, or similar to vehicle 102 of FIG. 1) includes or is associated with autonomous system 202, powertrain control system 204, steering control system 206, and brake system 208. In some implementations, vehicle 200 is the same as or similar to vehicle 102 (see FIG. 1). In some implementations, autonomous system 202 is configured to confer vehicle 200 autonomous driving capability (e.g., implement at least one driving automation or maneuver-based function, feature, device, and/or the like that enable vehicle 200 to be partially or fully operated without human intervention including, without limitation, fully autonomous vehicles (e.g., vehicles that forego reliance on human intervention such as Level 5 ADS-operated vehicles), highly autonomous vehicles (e.g., vehicles that forego reliance on human intervention in certain situations such as Level 4 ADS-operated vehicles), conditional autonomous vehicles (e.g., vehicles that forego reliance on human intervention in limited situations such as Level 3 ADS-operated vehicles) and/or the like. In one implementation, autonomous system 202 includes operational or tactical functionality required to operate vehicle 200 in on-road traffic and perform part or all of Dynamic Driving Task (DDT) on a sustained basis. In another implementation, autonomous system 202 includes an Advanced Driver Assistance System (ADAS) that includes driver support features. Autonomous system 202 supports various levels of driving automation, ranging from no driving automation (e.g., Level 0) to full driving automation (e.g., Level 5). For a detailed description of fully autonomous vehicles and highly autonomous vehicles, reference may be made to SAE International’s standard J3016: Taxonomy and Definitions for Terms Related to On-Road Motor Vehicle Automated Driving Systems, which is incorporated by reference in its entirety. In some implementations, vehicle 200 is associated with an autonomous fleet manager and/or a ridesharing company.

[34] Autonomous system 202 includes a sensor suite that includes one or more devices such as cameras 202a, LiDAR sensors 202b, radar sensors 202c, and microphones 202d. In some implementations, autonomous system 202 can include more or fewer devices and/or different devices (e.g., ultrasonic sensors, inertial sensors, GPS receivers (discussed below), odometry sensors that generate data associated with an indication of a distance that vehicle 200 has traveled, and/or the like). In some implementations, autonomous system 202 uses the one or more devices included in autonomous system 202 to generate data associated with environment 100, described herein. The data generated by the one or more devices of autonomous system 202 can be used by one or more systems described herein to observe the environment (e.g., environment 100) in which vehicle 200 is located. In some implementations, autonomous system 202 includes communication device 202e, autonomous vehicle compute 202f, drive-by-wire (DBW) system 202h, and safety controller 202g.

[35] Cameras 202a include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and/or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Cameras 202a include at least one camera (e.g., a digital camera using a light sensor such as a Charge-Coupled Device (CCD), a thermal camera, an infrared (IR) camera, an event camera, and/or the like) to capture images including physical objects (e.g., cars, buses, curbs, people, and/or the like). In some implementations, camera 202a generates camera data as output. In some examples, camera 202a generates camera data that includes image data associated with an image. In this example, the image data may specify at least one parameter (e.g., image characteristics such as exposure, brightness, etc., an image timestamp, and/or the like) corresponding to the image. In such an example, the image may be in a format (e.g., RAW, JPEG, PNG, and/or the like). In some implementations, camera 202a includes a plurality of independent cameras configured on (e.g., positioned on) a vehicle to capture images for the purpose of stereopsis (stereo vision). In some examples, camera 202a includes a plurality of cameras that generate image data and transmit the image data to autonomous vehicle compute 202f and/or a fleet management system (e.g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1). In such an example, autonomous vehicle compute 202f determines depth to one or more objects in a field of view of at least two cameras of the plurality of cameras based on the image data from the at least two cameras. In some implementations, cameras 202a is configured to capture images of objects within a distance from cameras 202a (e.g., up to 100 meters, up to a kilometer, and/or the like). Accordingly, cameras 202a include features such as sensors and lenses that are optimized for perceiving objects that are at one or more distances from cameras 202a.

[36] In an implementation, camera 202a includes at least one camera configured to capture one or more images associated with one or more traffic lights, street signs and/or other physical objects that provide visual navigation information. In some implementations, camera 202a generates traffic light data associated with one or more images. In some examples, camera 202a generates TLD (Traffic Light Detection) data associated with one or more images that include a format (e.g., RAW, JPEG, PNG, and/or the like). In some implementations, camera 202a that generates TLD data differs from other systems described herein incorporating cameras in that camera 202a can include one or more cameras with a wide field of view (e.g., a wide-angle lens, a fish-eye lens, a lens having a viewing angle of approximately 120 degrees or more, and/or the like) to generate images about as many physical objects as possible.

[37] Light Detection and Ranging (LiDAR) sensors 202b include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and/or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). LiDAR sensors 202b include a system configured to transmit light from a light emitter (e.g., a laser transmitter). Light emitted by LiDAR sensors 202b include light (e.g., infrared light and/or the like) that is outside of the visible spectrum. In some implementations, during operation, light emitted by LiDAR sensors 202b encounters a physical object (e.g., a vehicle) and is reflected back to LiDAR sensors 202b. In some implementations, the light emitted by LiDAR sensors 202b does not penetrate the physical objects that the light encounters. LiDAR sensors 202b also include at least one light detector which detects the light that was emitted from the light emitter after the light encounters a physical object. In some implementations, at least one data processing system associated with LiDAR sensors 202b generates an image (e.g., a point cloud, a combined point cloud, and/or the like) representing the objects included in a field of view of LiDAR sensors 202b. In some examples, the at least one data processing system associated with LiDAR sensor 202b generates an image that represents the boundaries of a physical object, the surfaces (e.g., the topology of the surfaces) of the physical object, and/or the like. In such an example, the image is used to determine the boundaries of physical objects in the field of view of LiDAR sensors 202b.

[38] Radio Detection and Ranging (radar) sensors 202c include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and/or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Radar sensors 202c include a system configured to transmit radio waves (either pulsed or continuously). The radio waves transmitted by radar sensors 202c include radio waves that are within a predetermined spectrum. In some implementations, during operation, radio waves transmitted by radar sensors 202c encounter a physical object and are reflected back to radar sensors 202c. In some implementations, the radio waves transmitted by radar sensors 202c are not reflected by some objects. In some implementations, at least one data processing system associated with radar sensors 202c generates signals representing the objects included in a field of view of radar sensors 202c. For example, the at least one data processing system associated with radar sensor 202c generates an image that represents the boundaries of a physical object, the surfaces (e.g., the topology of the surfaces) of the physical object, and/or the like. In some examples, the image is used to determine the boundaries of physical objects in the field of view of radar sensors 202c.

[39] Microphones 202d includes at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and/or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Microphones 202d include one or more microphones (e.g., array microphones, external microphones, and/or the like) that capture audio signals and generate data associated with (e.g., representing) the audio signals. In some examples, microphones 202d include transducer devices and/or like devices. In some implementations, one or more systems described herein can receive the data generated by microphones 202d and determine a position of an object relative to vehicle 200 (e.g., a distance and/or the like) based on the audio signals associated with the data.

[40] Communication device 202e includes at least one device configured to be in communication with cameras 202a, LiDAR sensors 202b, radar sensors 202c, microphones 202d, autonomous vehicle compute 202f, safety controller 202g, and/or DBW (Drive-By-Wire) system 202h. For example, communication device 202e may include a device that is the same as or similar to communication interface 314 of FIG. 3. In some implementations, communication device 202e includes a vehicle-to-vehicle (V2V) communication device (e.g., a device that enables wireless communication of data between vehicles).

[41] Autonomous vehicle compute 202f include at least one device configured to be in communication with cameras 202a, LiDAR sensors 202b, radar sensors 202c, microphones 202d, communication device 202e, safety controller 202g, and/or DBW system 202h. In some examples, autonomous vehicle compute 202f includes a device such as a client device, a mobile device (e.g., a cellular telephone, a tablet, and/or the like), a server (e.g., a computing device including one or more central processing units, graphical processing units, and/or the like), and/or the like. In some implementations, autonomous vehicle compute 202f is the same as or similar to autonomous vehicle (AV) compute 400, described herein. Additionally, or alternatively, in some implementations, autonomous vehicle compute 202f is configured to be in communication with an autonomous vehicle system (e.g., an autonomous vehicle system that is the same as or similar to remote AV system 114 of FIG. 1), a fleet management system (e.g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1), a V2I device (e.g., a V2I device that is the same as or similar to V2I device 110 of FIG. 1), and/or a V2I system (e.g., a V2I system that is the same as or similar to V2I system 118 of FIG. 1).

[42] Safety controller 202g includes at least one device configured to be in communication with cameras 202a, LiDAR sensors 202b, radar sensors 202c, microphones 202d, communication device 202e, autonomous vehicle computer 202f, and/or DBW system 202h. In some examples, safety controller 202g includes one or more controllers (electrical controllers, electromechanical controllers, and/or the like) that are configured to generate and/or transmit control signals to operate one or more devices of vehicle 200 (e.g., powertrain control system 204, steering control system 206, brake system 208, and/or the like). In some implementations, safety controller 202g is configured to generate control signals that take precedence over (e.g., overrides) control signals generated and/or transmitted by autonomous vehicle compute 202f.

[43] DBW system 202h includes at least one device configured to be in communication with communication device 202e and/or autonomous vehicle compute 202f. In some examples, DBW system 202h includes one or more controllers (e.g., electrical controllers, electromechanical controllers, and/or the like) that are configured to generate and/or transmit control signals to operate one or more devices of vehicle 200 (e.g., powertrain control system 204, steering control system 206, brake system 208, and/or the like). Additionally, or alternatively, the one or more controllers of DBW system 202h are configured to generate and/or transmit control signals to operate at least one different device (e.g., a turn signal, headlights, door locks, windshield wipers, and/or the like) of vehicle 200.

[44] Powertrain control system 204 includes at least one device configured to be in communication with DBW system 202h. In some examples, powertrain control system 204 includes at least one controller, actuator, and/or the like. In some implementations, powertrain control system 204 receives control signals from DBW system 202h, and powertrain control system 204 causes vehicle 200 to make longitudinal vehicle motion, such as start moving forward, stop moving forward, start moving backward, stop moving backward, accelerate in a direction, decelerate in a direction or to make lateral vehicle motion such as performing a left turn, performing a right turn, and/or the like. In an example, powertrain control system 204 causes the energy (e.g., fuel, electricity, and/or the like) provided to a motor of the vehicle to increase, remain the same, or decrease, thereby causing at least one wheel of vehicle 200 to rotate or not rotate.

[45] Steering control system 206 includes at least one device configured to rotate one or more wheels of vehicle 200. In some examples, steering control system 206 includes at least one controller, actuator, and/or the like. In some implementations, steering control system 206 causes the front two wheels and/or the rear two wheels of vehicle 200 to rotate to the left or right to cause vehicle 200 to turn to the left or right. In other words, steering control system 206 causes activities necessary for the regulation of the y-axis component of vehicle motion.

[46] Brake system 208 includes at least one device configured to actuate one or more brakes to cause vehicle 200 to reduce speed and/or remain stationary. In some examples, brake system 208 includes at least one controller and/or actuator that is configured to cause one or more calipers associated with one or more wheels of vehicle 200 to close on a corresponding rotor of vehicle 200. Additionally, or alternatively, in some examples brake system 208 includes an automatic emergency braking (AEB) system, a regenerative braking system, and/or the like.

[47] In some implementations, vehicle 200 includes at least one platform sensor (not explicitly illustrated) that measures or infers properties of a state or a condition of vehicle 200. In some examples, vehicle 200 includes platform sensors such as a global positioning system (GPS) receiver, an inertial measurement unit (IMU), a wheel speed sensor, a wheel brake pressure sensor, a wheel torque sensor, an engine torque sensor, a steering angle sensor, and/or the like. Although brake system 208 is illustrated to be located on the near side of vehicle 200 in FIG. 2, brake system 208 may be located anywhere in vehicle 200.

[48] Referring now to FIG. 3, illustrated is a schematic diagram of a device 300. As illustrated, device 300 includes processor 304, memory 306, storage component 308, input interface 310, output interface 312, communication interface 314, and bus 302. In some implementations, device 300 corresponds to at least one device of vehicles 102 (e.g., at least one device of a system of vehicles 102), and/or one or more devices of network 112 (e.g., one or more devices of a system of network 112). In some implementations, one or more devices of vehicles 102 (e.g., one or more devices of a system of vehicles 102), and/or one or more devices of network 112 (e.g., one or more devices of a system of network 112) include at least one device 300 and/or at least one component of device 300. As shown in FIG. 3, device 300 includes bus 302, processor 304, memory 306, storage component 308, input interface 310, output interface 312, and communication interface 314.

[49] Bus 302 includes a component that permits communication among the components of device 300. In some cases, processor 304 includes a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), and/or the like), a microphone, a digital signal processor (DSP), and/or any processing component (e.g., a field- programmable gate array (FPGA), an application specific integrated circuit (ASIC), and/or the like) that can be programmed to perform at least one function. Memory 306 includes random access memory (RAM), read-only memory (ROM), and/or another type of dynamic and/or static storage device (e.g., flash memory, magnetic memory, optical memory, and/or the like) that stores data and/or instructions for use by processor 304.

[50] Storage component 308 stores data and/or software related to the operation and use of device 300. In some examples, storage component 308 includes a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, a solid state disk, and/or the like), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, a CD-ROM, RAM, PROM, EPROM, FLASH-EPROM, NV-RAM, and/or another type of computer readable medium, along with a corresponding drive.

[51] Input interface 310 includes a component that permits device 300 to receive information, such as via user input (e.g., a touchscreen display, a keyboard, a keypad, a mouse, a button, a switch, a microphone, a camera, and/or the like). Additionally or alternatively, in some implementations input interface 310 includes a sensor that senses information (e.g., a global positioning system (GPS) receiver, an accelerometer, a gyroscope, an actuator, and/or the like). Output interface 312 includes a component that provides output information from device 300 (e.g., a display, a speaker, one or more light-emitting diodes (LEDs), and/or the like).

[52] In some implementations, communication interface 314 includes a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, and/or the like) that permits device 300 to communicate with other devices via a wired connection, a wireless connection, or a combination of wired and wireless connections. In some examples, communication interface 314 permits device 300 to receive information from another device and/or provide information to another device. In some examples, communication interface 314 includes an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a cellular network interface, and/or the like.

[53] In some implementations, device 300 performs one or more processes described herein. Device 300 performs these processes based on processor 304 executing software instructions stored by a computer-readable medium, such as memory 305 and/or storage component 308. A computer-readable medium (e.g., a non-transitory computer readable medium) is defined herein as a non-transitory memory device. A non-transitory memory device includes memory space located inside a single physical storage device or memory space spread across multiple physical storage devices.

[54] In some implementations, software instructions are read into memory 306 and/or storage component 308 from another computer-readable medium or from another device via communication interface 314. When executed, software instructions stored in memory 306 and/or storage component 308 cause processor 304 to perform one or more processes described herein. Additionally or alternatively, hardwired circuitry is used in place of or in combination with software instructions to perform one or more processes described herein. Thus, implementations described herein are not limited to any specific combination of hardware circuitry and software unless explicitly stated otherwise.

[55] Memory 306 and/or storage component 308 includes data storage or at least one data structure (e.g., a database and/or the like). Device 300 is capable of receiving information from, storing information in, communicating information to, or searching information stored in the data storage or the at least one data structure in memory 306 or storage component 308. In some examples, the information includes network data, input data, output data, or any combination thereof.

[56] In some implementations, device 300 is configured to execute software instructions that are either stored in memory 306 and/or in the memory of another device (e.g., another device that is the same as or similar to device 300). As used herein, the term “module” refers to at least one instruction stored in memory 306 and/or in the memory of another device that, when executed by processor 304 and/or by a processor of another device (e.g., another device that is the same as or similar to device 300) cause device 300 (e.g., at least one component of device 300) to perform one or more processes described herein. In some implementations, a module is implemented in software, firmware, hardware, and/or the like.

[57] The number and arrangement of components illustrated in FIG. 3 are provided as an example. In some implementations, device 300 can include additional components, fewer components, different components, or differently arranged components than those illustrated in FIG. 3. Additionally or alternatively, a set of components (e.g., one or more components) of device 300 can perform one or more functions described as being performed by another component or another set of components of device 300.

[58] Referring now to FIG. 4, illustrated is an example block diagram of an autonomous vehicle compute 400 (sometimes referred to as an “AV stack”). As illustrated, autonomous vehicle compute 400 includes perception system 402 (sometimes referred to as a perception module), planning system 404 (sometimes referred to as a planning module), localization system 406 (sometimes referred to as a localization module), control system 408 (sometimes referred to as a control module), and database 410. In some implementations, perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included and/or implemented in an autonomous navigation system of a vehicle (e.g., autonomous vehicle compute 202f of vehicle 200). Additionally, or alternatively, in some implementations, perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included in one or more standalone systems (e.g., one or more systems that are the same as or similar to autonomous vehicle compute 400 and/or the like). In some examples, perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included in one or more standalone systems that are located in a vehicle and/or at least one remote system as described herein. In some implementations, any and/or all of the systems included in autonomous vehicle compute 400 are implemented in software (e.g., in software instructions stored in memory), computer hardware (e.g., by microprocessors, microcontrollers, application-specific integrated circuits (ASICs), Field Programmable Gate Arrays (FPGAs), and/or the like), or combinations of computer software and computer hardware. It will also be understood that, in some implementations, autonomous vehicle compute 400 is configured to be in communication with a remote system (e g., an autonomous vehicle system that is the same as or similar to remote AV system 114, a fleet management system 116 that is the same as or similar to fleet management system 116, a V2I system that is the same as or similar to V2I system 118, and/or the like).

[59] In some implementations, perception system 402 receives data associated with at least one physical object (e.g., data that is used by perception system 402 to detect the at least one physical object) in an environment and classifies the at least one physical object. In some examples, perception system 402 receives image data captured by at least one camera (e.g., cameras 202a), the image associated with (e.g., representing) one or more physical objects within a field of view of the at least one camera. In such an example, perception system 402 classifies at least one physical object based on one or more groupings of physical objects (e.g., bicycles, vehicles, traffic signs, pedestrians, and/or the like). In some implementations, perception system 402 transmits data associated with the classification of the physical objects to planning system 404 based on perception system 402 classifying the physical objects.

[60] In some implementations, planning system 404 receives data associated with a destination and generates data associated with at least one route (e.g., routes 106 of FIG. 1) along which a vehicle (e.g., vehicles 102) can travel along toward a destination. In some implementations, planning system 404 periodically or continuously receives data from perception system 402 (e.g., data associated with the classification of physical objects, described above) and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by perception system 402. In other words, planning system 404 may perform tactical function-related tasks that are required to operate vehicle 102 in on-road traffic. Tactical efforts involve maneuvering the vehicle in traffic during a trip, including but not limited to deciding whether and when to overtake another vehicle, change lanes, or selecting an appropriate speed, acceleration, deacceleration, etc. In some implementations, planning system 404 receives data associated with an updated position of a vehicle (e.g., vehicles 102) from localization system 406 and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by localization system 406.

[61] In some implementations, localization system 406 receives data associated with (e.g., representing) a location of a vehicle (e.g., vehicles 102) in an area. In some examples, localization system 406 receives LiDAR data associated with at least one point cloud generated by at least one LiDAR sensor (e.g., LiDAR sensors 202b). In certain examples, localization system 406 receives data associated with at least one point cloud from multiple LiDAR sensors and localization system 406 generates a combined point cloud based on each of the point clouds. In these examples, localization system 406 compares the at least one point cloud or the combined point cloud to two- dimensional (2D) and/or a three-dimensional (3D) map of the area stored in database 410. Localization system 406 then determines the position of the vehicle in the area based on localization system 406 comparing the at least one point cloud or the combined point cloud to the map. In some implementations, the map includes a combined point cloud of the area generated prior to navigation of the vehicle. In some implementations, maps include, without limitation, high-precision maps of the roadway geometric properties, maps describing road network connectivity properties, maps describing roadway physical properties (such as traffic speed, traffic volume, the number of vehicular and cyclist traffic lanes, lane width, lane traffic directions, or lane marker types and locations, or combinations thereof), and maps describing the spatial locations of road features such as crosswalks, traffic signs or other travel signals of various types. In some implementations, the map is generated in real-time based on the data received by the perception system.

[62] In another example, localization system 406 receives Global Navigation Satellite System (GNSS) data generated by a global positioning system (GPS) receiver. In some examples, localization system 406 receives GNSS data associated with the location of the vehicle in the area and localization system 406 determines a latitude and longitude of the vehicle in the area. In such an example, localization system 406 determines the position of the vehicle in the area based on the latitude and longitude of the vehicle. In some implementations, localization system 406 generates data associated with the position of the vehicle. In some examples, localization system 406 generates data associated with the position of the vehicle based on localization system 406 determining the position of the vehicle. In such an example, the data associated with the position of the vehicle includes data associated with one or more semantic properties corresponding to the position of the vehicle.

[63] In some implementations, control system 408 receives data associated with at least one trajectory from planning system 404 and control system 408 controls operation of the vehicle. In some examples, control system 408 receives data associated with at least one trajectory from planning system 404 and control system 408 controls operation of the vehicle by generating and transmitting control signals to cause a powertrain control system (e.g., DBW system 202h, powertrain control system 204, and/or the like), a steering control system (e.g., steering control system 206), and/or a brake system (e.g., brake system 208) to operate. For example, control system 408 is configured to perform operational functions such as a lateral vehicle motion control or a longitudinal vehicle motion control. The lateral vehicle motion control causes activities necessary for the regulation of the y-axis component of vehicle motion. The longitudinal vehicle motion control causes activities necessary for the regulation of the x-axis component of vehicle motion. In an example, where a trajectory includes a left turn, control system 408 transmits a control signal to cause steering control system 206 to adjust a steering angle of vehicle 200, thereby causing vehicle 200 to turn left. Additionally, or alternatively, control system 408 generates and transmits control signals to cause other devices (e.g., headlights, turn signal, door locks, windshield wipers, and/or the like) of vehicle 200 to change states.

[64] In some implementations, perception system 402, planning system 404, localization system 406, and/or control system 408 implement at least one machine learning model (e.g., at least one multilayer perceptron (MLP), at least one convolutional neural network (CNN), at least one recurrent neural network (RNN), at least one autoencoder, at least one transformer, and/or the like). In some examples, perception system 402, planning system 404, localization system 406, and/or control system 408 implement at least one machine learning model alone or in combination with one or more of the above-noted systems. In some examples, perception system 402, planning system 404, localization system 406, and/or control system 408 implement at least one machine learning model as part of a pipeline (e.g., a pipeline for identifying one or more objects located in an environment and/or the like).

[65] Database 410 stores data that is transmitted to, received from, and/or updated by perception system 402, planning system 404, localization system 406, and/or control system 408. In some examples, database 410 includes a storage component (e.g., a storage component that is the same as or similar to storage component 308 of FIG. 3) that stores data and/or software related to the operation and uses at least one system of autonomous vehicle compute 400. In some implementations, database 410 stores data associated with 2D and/or 3D maps of at least one area. In some examples, database 410 stores data associated with 2D and/or 3D maps of a portion of a city, multiple portions of multiple cities, multiple cities, a county, a state, a State (e.g., a country), and/or the like). In such an example, a vehicle (e.g., a vehicle that is the same as or similar to vehicles 102 and/or vehicle 200) can drive along one or more drivable regions (e.g., single-lane roads, multi-lane roads, highways, back roads, off road trails, and/or the like) and cause at least one LiDAR sensor (e.g., a LiDAR sensor that is the same as or similar to LiDAR sensors 202b) to generate data associated with an image representing the objects included in a field of view of the at least one LiDAR sensor.

[66] In some implementations, database 410 can be implemented across a plurality of devices. In some examples, database 410 is included in a vehicle (e.g., a vehicle that is the same as or similar to vehicles 102 and/or vehicle 200), an autonomous vehicle system (e.g., an autonomous vehicle system that is the same as or similar to remote AV system 114, a fleet management system (e g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1, a V2I system (e.g., a V2I system that is the same as or similar to V2I system 118 of FIG. 1) and/or the like.

[67] FIG. 5 is a diagram of an implementation 500 of a perception system 402 in an autonomous vehicle (AV) for detecting objects around the AV. In some implementations, implementation 500 includes a perception system 402 (e.g., perception system 402 of FIG. 4). The perception system 402 receives camera images 504 of physical objects around the AV captured by cameras (cameras 202a of FIG. 2) and outputs detected objects 512 contained in the camera images 504. In some implementations, the perception system 402 includes a neural network 502, which includes backbone 506 for extracting image features 508 from camera images 504 and detection head 510 for generating BEV bounding boxes representative of detected objects 512 based on the image features 508.

[68] In examples, the terms backbone and head refer to the structural components of a neural network (e.g., neural network 502). In examples, a backbone (e.g., backbone 506) extracts image features from camera image data (raw pixel values), and one or more heads perform a predetermined task using the image features. For example, the detection head 510 detects a center of objects based on image features extracted from the camera images and determines respective attributes of the objects. The attributes include size, orientation, velocity, and the like. In examples, the output of the detection head is one or more bounding boxes associated with objects in the environment and classification scores for the presence of object class instances (e.g., vehicles, pedestrians, cyclists) within the bounding boxes. The higher the classification score, the more likely the corresponding object class instance is present in a bounding box. Image features are deep learning information extracted from raw pixel values. Image features are the pixel values with only the relevant information extracted by the backbone. The relevant information is hidden features that help detection head 510 to perform a task, e.g., detect objects 512. In examples, the detection head includes a classification head and a bounding box head.

[69] FIG. 6 is a diagram of a neural network 502 that detects objects in camera images. In the example of FIG. 6, the camera images are captured by cameras (cameras 202a of FIG. 2) of an AV. As shown in FIGS. 5 and 6, the detection head 510 includes BEV detection head 602 for generating BEV bounding boxes 612 (in a BEV space) representative of detected objects 512 based on the image features 508, and perspective detection head 604 for generating perspective bounding boxes (in a perspective space) 606 representative of detected objects 512 based on the image features 508.

[70] The detection head 510 refers to the top or final layer(s) of the neural network 502 that outputs predictions about the presence or absence of specific objects within an input camera image 504. The detection head 510 obtains input that includes the image features 508 extracted from the input camera image 504 by the lower layers (e.g., backbone 506) of a neural network 502 and produces the final predictions. In some implementations, the final predictions include the class labels (e.g., pedestrians, buildings, other vehicles, traffic lights, etc.) for classifying the detected objects 512 and bounding box coordinates (coordinates of perspective bounding boxes 606 and BEV bounding boxes 612).

[71] The perspective space refers to a space as viewed from a camera position, where the objects in the scene are represented in terms of their positions and orientations relative to the camera. In examples, with respect to a vehicle coordinate frame of reference, the perspective space captures images in the XZ or YZ plane, with a fixed or substantially fixed range of Z-values. The BEV space refers to a top-down or overhead view of the 3D space, where the objects in the scene are projected onto a 2D plane from a bird’s eye view perspective. In the BEV space, the 3D positions and orientations of objects are represented in terms of their 2D coordinates on the plane. In examples, with respect to a vehicle coordinate frame of reference, the BEV space is a view of the XY plane, with zero Z-values.

[72] The neural network 502 further includes a neural network layer, such as a feed-forward layer 608, for converting the perspective bounding boxes 606 to BEV object proposals 610, in addition to BEV generic object proposals 605. The feed-forward layer 608, also known as a dense layer or a fully connected layer, includes a set of neurons or nodes, where each node is connected to every node in the previous layer by a weighted connection. The input to each node is the output of the previous layer, which is multiplied by the weight of the connection and summed with a bias term. The resulting sum is then passed through an activation function to produce the output of the node, which becomes the input to the next layer. The weights and biases of a feed-forward layer are learned during the training process through backpropagation.

[73] The feed-forward layer 608 converts the perspective bounding boxes 606 in the perspective space to the BEV object proposals 610 in the BEV space. The perspective bounding boxes 606 are generated by the perspective detection head 604 based on the image features 508, which are extracted from input camera images 504. Thus, the BEV object proposals 610 are dependent on input camera images 504, and are not randomly initialized object proposals. BEV generic object proposals 605 are a diverse and representative set of candidate object regions in the camera images 504, without relying on any prior knowledge or assumptions about the objects to be detected. BEV generic object proposals 605 are randomly initialized object proposals, as a “starting point” for the BEV detection head 602. Since the BEV generic object proposals 605 are randomly initialized, the BEV generic object proposals 605 are not dependent on input camera images 504. During training of the neural network 502, the initialized object proposals are gradually updated to become final object proposals (e.g., BEV bounding boxes 612), so as to minimize the total loss of the neural network 502.

[74] The BEV object proposals 610 and the BEV generic object proposals 605 are both input into the BEV detection head 602 as a “starting point” (initial object proposals), as opposed to only inputting BEV generic object proposals 605 into the BEV detection head 602 as a “starting point”. BEV object proposals 610 are input dependent, as opposed to being randomly initialized. The input dependent BEV object proposals 610 can facilitate faster neural network convergence. BEV bounding boxes 612 are final predictions (final object proposals) output from the BEV detection head 602. The BEV bounding boxes 612 containing detected objects 512 are used by the AV compute (e.g., AV compute 400 of FIG. 4) for trajectory planning based on the detected objects 512.

[75] A total loss of the neural network 502 is calculated to assess the neural network 502. The total loss is x * a loss of the BEV detection head 602 + (1-x) * a loss of the perspective detection head 604, and x is in a range of [0, 1], The neural network 502 is trained to minimize the total loss.

[76] FIG. 7 is a diagram of an example BEV detection head 602. The BEV detection head 602 includes a position encoder 702 for encoding the positional information of each element in the input BEV object proposals 610 and BEV generic object proposals 605; a transformer encoder 704 for processing the input image features 508 and generating a sequence of hidden representations or features that capture the important information in the input image features 508; a transformer decoder 706 for generating a sequence of output data based on the output of transformer encoder 704 and a partial sequence of the output data; a first feed-forward neural network (FFN) 708 for generating class labels 710 (indicating classification) of the detected objects 512; and a second FFN 712 for generating BEV bounding boxes 612 containing the detected objects 512.

[77] The position encoder 702 includes a fixed, learnable embedding matrix that maps each position or time step in the input sequence (e.g., input BEV object proposals 610 and BEV generic object proposals 605) to a corresponding embedding vector. The embedding vectors are then added to the input embeddings of each element in the input sequence, allowing the transformer decoder 706 to distinguish between different positions or time steps in the input sequence.

[78] The transformer encoder 704 includes multiple layers of self-attention and feed-forward neural networks. Each layer of the transformer encoder 704 takes in a sequence of input embeddings, which represent the elements of the input sequence, and computes a sequence of output embeddings that capture the relationships and dependencies between different elements of the input sequence.

[79] The transformer decoder 706 includes multiple layers of self-attention (also known as intraattention), encoder-decoder attention (also known as inter-attention), and feed-forward neural networks. The self-attention is used to capture the relationships between different positions in the input sequence. The encoder-decoder attention is used to attend to the output representations of the transformer encoder 704. This allows the transformer decoder 706 to align the input and output sequences and capture the relevant context from the transformer decoder 706’ s output representations. Encoder-decoder attention helps the decoder to incorporate information from the input sequence and generate output representations that are conditioned on the input. The feedforward neural networks are used to process the self-attention and encoder-decoder attention outputs and generate the final output representations.

[80] An FFN includes multiple layers of interconnected neurons, with each layer responsible for performing a specific type of computation on the input data. The input to a feed-forward neural network is fed through the network in a forward direction, with each layer transforming the input data into a higher-level representation. The output of the final layer of the network represents the network’s prediction or classification of the input data. The first FFN 708 is used to generate class labels 710 of the detected objects 512, and the second FFN 712 is used to generate BEV bounding boxes 612 containing the detected objects 512.

[81] FIG. 8 illustrates an example flow chart of a process 800 for detecting objects around the AV. The example process 800 is implemented (e.g., completely, partially, and/or the like) on AV compute 400 of FIG. 4 (e.g., perception system 402 of FIG. 4), neural network 502 of FIGS. 5 and 6, or device 300 of FIG. 3. In some implementations, the example process 800 shown in FIG. 8 can be modified or reconfigured to include additional, fewer, or different operations, which can be performed in the order shown or in a different order. In some instances, one or more of the operations can be repeated or iterated, for example, until a terminating condition is reached. In some implementations, one or more of the individual operations shown in FIG. 8 can be executed as multiple separate operations, or one or more subsets of the operations shown in FIG. 8 can be combined and executed as a single operation. The objects can be vehicles, pedestrians, bicycles, trees, building, or any other objects around the AV.

[82] At 802, the neural network (e.g., neural network 502 of FIGS. 5 and 6) or processor (processor 304 of FIG. 3) receives camera images (e.g., camera images 504 of FIGS. 5 and 6) from cameras (e.g., cameras 202a of FIG. 2). The camera images are represented in a perspective view of the environment, as captured from a camera position.

[83] At 804, the neural network or processor extracts image features (e.g., image features 508 of FIGS. 5-7) from the camera images. The image features are represented in the perspective view of the environment, as viewed from a camera position. [84] At 806, the neural network or processor generates one or more perspective bounding boxes (e.g., perspective bounding boxes of FIG. 6) containing one or more detected objects (e.g., detected objects of FIG. 5) based on the image features. The one or more perspective bounding boxes bound the one or more detected objects when represented in the perspective view of the environment.

[85] At 808, the neural network or processor converts the one or more perspective bounding boxes into one or more first object proposals (e.g., BEV object proposals 640 of FIGS. 6 and 7) represented in a bird’s eye view (BEV) of the environment (a top-down view, looking directly down onto the ground plane, as if from above).

[86] At 810, the neural network or processor generates one or more BEV bounding boxes (e.g., BEV bounding boxes 612 of FIGS. 6 and 7) containing the one or more detected objects, based on the image features represented in the perspective view of the environment, the one or more first object proposals represented in the BEV of the environment, and one or more generic object proposals (e.g., BEV generic object proposals 605 of FIGS. 6 and 7) represented in the BEV of the environment. The one or more BEV bounding boxes bound the one or more detected objects when represented in the BEV view of the environment.

[87] According to some non-limiting implementations or examples, provided is a method, including: receiving, by a processor, one or more camera images representing a perspective view of an environment relative to a vehicle; extracting, by the processor, image features from the camera images; generating, by the processor, one or more perspective bounding boxes containing one or more detected objects based on the image features, wherein the one or more perspective bounding boxes bound the one or more detected objects that are represented in the perspective view of the environment; selecting, by the processor, one or more first obj ect proposals represented in a bird’s eye view (BEV) view of the environment based on the one or more perspective bounding boxes; and generating, by the processor, one or more BEV bounding boxes containing the one or more detected objects based on: the image features represented in the perspective view of the environment, the one or more first object proposals represented in the BEV view of the environment, and one or more generic object proposals represented in the BEV view of the environment, wherein the one or more BEV bounding boxes bound the one or more detected objects when represented in the BEV view of the environment. [88] According to some non-limiting implementations or examples, provided is a system, including at least one processor; and a memory storing instructions thereon that, when executed by the at least one processor, cause the at least one processor to perform the above method.

[89] According to some non-limiting implementations or examples, provided is a non-transitory, computer-readable storage medium having instructions stored thereon, which when executed by at least one processor, cause the at least one processor to perform the above method.

[90] Clause 1 : A method, including: receiving, by a processor, one or more camera images representing a perspective view of an environment relative to a vehicle; extracting, by the processor, image features from the camera images; generating, by the processor, one or more perspective bounding boxes containing one or more detected objects based on the image features, wherein the one or more perspective bounding boxes bound the one or more detected objects that are represented in the perspective view of the environment; selecting, by the processor, one or more first object proposals represented in a bird’s eye view (BEV) view of the environment based on the one or more perspective bounding boxes; and generating, by the processor, one or more BEV bounding boxes containing the one or more detected objects based on: the image features represented in the perspective view of the environment, the one or more first object proposals represented in the BEV view of the environment, and one or more generic object proposals represented in the BEV view of the environment, wherein the one or more BEV bounding boxes bound the one or more detected objects when represented in the BEV view of the environment.

[91] Clause 2: The method of clause 1, wherein generating the one or more perspective bounding boxes comprises providing the image features to a perspective detection head of a neural network to cause the perspective detection head to generate the one or more perspective bounding boxes, and generating the one or more BEV bounding boxes comprises providing the image features, the one or more first object proposals and the one or more generic object proposals to a BEV detection head of the neural network to cause the BEV detection head to generate the one or more BEV bounding boxes.

[92] Clause 3 : The method of clause 2, wherein the neural network is a transformer model.

[93] Clause 4: The method of clause 2, further comprising: calculating a total loss of the neural network, wherein the total loss is x * a loss of the BEV detection head + (1-x) * a loss of the perspective detection head, wherein x is in a range of [0, 1], [94] Clause 5: The method of clause 2, wherein converting the one or more perspective bounding boxes into the one or more first object proposals comprises providing the one or more perspective bounding boxes to a feed-forward layer of the neural network to cause the feed-forward layer to convert the one or more perspective bounding boxes into the one or more first object proposals.

[95] Clause 6: The method of any one of clauses 1-5, further comprising: classifying, by a perception system, the one or more detected objects contained in the generated one or more BEV bounding boxes.

[96] Clause 7: The method of any one of clauses 1-6, further comprising: performing, by a planning system, tactical function related tasks to operate the vehicle in the environment based on the generated one or more BEV bounding boxes.

[97] Clause 8: A non-transitory, computer-readable storage medium having instructions stored thereon, which when executed by at least one processor, cause the at least one processor to perform the method of any one of clauses 1-7.

[98] Clause 9: A system for detecting objects around a vehicle by a neural network, comprising: at least one processor; and a memory storing instructions thereon that, when executed by the at least one processor, cause the at least one processor to perform the method of any one of clauses 1-7.

[99] Clause 10: The system of clause 9, wherein the neural network includes a perspective detection head for generating the one or more perspective bounding boxes and a BEV detection head for generating the one or more BEV bounding boxes, wherein the at least one processor is caused to further perform: calculating a total loss of the neural network, wherein the total loss is x * a loss of the BEV detection head + (1-x) * a loss of the perspective detection head, wherein x is in a range of [0, 1],

[100] In the foregoing description, implementations of the invention have been described with reference to numerous specific details that may vary from implementation to implementation. The description and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense. The sole and exclusive indicator of the scope of the invention, and what is intended by the applicants to be the scope of the invention, is the literal and equivalent scope of the set of claims that issue from this application, in the specific form in which such claims issue, including any subsequent correction. Any definitions expressly set forth herein for terms contained in such claims shall govern the meaning of such terms as used in the claims. In addition, the term “further comprising” is used in the foregoing description or following claims, what follows this phrase can be an additional step or entity, or a sub-step/sub-entity of a previously-recited step or entity.

Claims

WHAT IS CLAIMED IS:

1. A method, comprising: receiving, by a processor, one or more camera images representing a perspective view of an environment relative to a vehicle; extracting, by the processor, image features from the camera images; generating, by the processor, one or more perspective bounding boxes containing one or more detected objects based on the image features, wherein the one or more perspective bounding boxes bound the one or more detected objects that are represented in the perspective view of the environment; selecting, by the processor, one or more first object proposals represented in a bird’s eye view (BE\ view of the environment based on the one or more perspective bounding boxes; and generating, by the processor, one or more BEV bounding boxes containing the one or more detectec objects based on: the image features represented in the perspective view of the environment, the one or more first object proposals represented in the BEV view of the environment, and one or more generic object proposals represented in the BEV view of the environment, wherein the one or more BEV bounding boxes bound the one or more detected objects when represented in the BEV view of the environment.

2. The method of claim 1, wherein generating the one or more perspective bounding boxes comprises providing the image features to a perspective detection head of a neural network to cause the perspective detection head to generate the one or more perspective bounding boxes, and generating the one or more BEV bounding boxes comprises providing the image features, the one or more first object proposals and the one or more generic object proposals to a BEV detection head of the neural network to cause the BEV detection head to generate the one or more BEV bounding boxes.

3. The method of claim 2, wherein the neural network is a transformer model.

4. The method of claim 2, further comprising: calculating a total loss of the neural network, wherein the total loss is x * a loss of the BEV detection head + (1-x) * a loss of the perspective detection head, wherein x is in a range of [0, 1].

5. The method of claim 2, wherein converting the one or more perspective bounding boxes into the one or more first object proposals comprises providing the one or more perspective bounding boxes to a feed-forward layer of the neural network to cause the feed-forward layer to convert the one or more perspective bounding boxes into the one or more first object proposals.

6. The method of any one of claims 1-5, further comprising: classifying, by a perception system, the one or more detected objects contained in the generated one or more BEV bounding boxes.

7. The method of any one of claims 1-6, further comprising: performing, by a planning system, tactical function related tasks to operate the vehicle in the environment based on the generated one or more BEV bounding boxes.

8. A non-transitory, computer-readable storage medium having instructions stored thereon, which when executed by at least one processor, cause the at least one processor to perform the method of any one of claims 1-7.

9. A system for detecting objects around a vehicle by a neural network, comprising: at least one processor; and a memory storing instructions thereon that, when executed by the at least one processor, cause the at least one processor to perform the method of any one of claims 1-7.

10. The system of claim 9, wherein the neural network includes a perspective detection head for generating the one or more perspective bounding boxes and a BEV detection head for generating the one or more BEV bounding boxes, wherein the at least one processor is caused to further perform: calculating a total loss of the neural network, wherein the total loss is x * a loss of the BEV detection head + (1-x) * a loss of the perspective detection head, wherein x is in a range of [0, 1].