Generation of corrected future maneuver parameters in a planner

Abstract

Method (600) comprising: Receiving (602), using at least one processor, a first set of parameters associated with one or more previous maneuvers of a vehicle and a second set of parameters associated with a maneuver goal of the vehicle; Generating (604), using the at least one processor and using the first parameter set and the second parameter set, at least one future maneuver parameter corresponding to one or more future maneuvers of the vehicle; Training (606), using the at least one processor, at least one data model by comparing the at least one future maneuver parameter with one or more reference maneuver parameters identified as expert driver maneuver parameters; Generating (608), using the at least one processor and based on the training, at least one corrected future maneuver parameter, wherein the corrected future maneuver parameter comprises at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle based on error correction values indicating a difference between the at least one future maneuver parameter and the one or more reference maneuver parameters; Generating, using the at least one processor, a signal for controlling the vehicle to execute the at least one corrected future maneuver using the at least one corrected future maneuver parameter; and Using the signal, causing the vehicle to operate at least in part based on the corrected future maneuver parameter.

Description

BACKGROUND

[0001] An autonomous vehicle is capable of sensing its environment and navigating without human input. After receiving data representing the environment and / or other parameters, the vehicle processes the data to make movement decisions, such as stopping, moving forward / backward, turning, etc. These decisions are used to navigate the vehicle safely along a selected path, avoiding obstacles, and responding to a variety of scenarios, such as the presence, movement, etc., of other vehicles, pedestrians, and / or other objects.

[0002] US Pat. No. 10,829,149 B1 describes that model-based control of dynamic systems typically requires precise domain-specific knowledge and specifications for system components. In general, the dynamics of the steering actuator can be difficult to model, e.g., due to an integrated power steering control unit, proprietary black-box controls, etc. Furthermore, it is difficult to capture the complex interplay of nonlinear interactions such as power steering, tire forces, etc. with sufficient accuracy. To overcome this limitation, a recurrent neural network can be used to model the steering dynamics of an autonomous vehicle. The resulting model can be used to generate forward steering commands for the integrated controller.Such a neural network model can be automatically generated with less domain-specific knowledge, can predict steering dynamics more accurately, and performs comparably to a high-fidelity first-principal model when used to control the steering system of a self-driving vehicle.

[0003] The publication US 2019 / 0 185 011 A1 describes a method, a system, a medium, and the implementation of lane planning in an autonomous vehicle. Sensor data is received that captures ground images of a road on which the autonomous vehicle is traveling. Based on the sensor data, a current lane of the road that the autonomous vehicle is currently occupying is detected. Lane control for the autonomous vehicle is planned based on the detected current lane characteristics and the parameters of the self-aware capability according to a lane control model. The parameters for the self-aware capability are used to predict the operational capability of the autonomous vehicle with respect to the current location of the autonomous vehicle. The lane control model is generated based on recorded human driving data to achieve human-like lane control behavior in various scenarios.

[0004] US 2018 / 0 120 843 A1 describes a method and system that generates a time-series signal indicative of a change in the environment near the vehicle with respect to a vehicle's movement, and transmits the time-series signal to a neural network to generate a reference trajectory as a function of time that satisfies temporal and spatial constraints on a vehicle's position. The neural network is trained to convert time-series signals into reference trajectories of the vehicle. The trajectory that follows the reference trajectory while the conditions for the vehicle's movement are met is determined, and the vehicle's movement is controlled to follow the trajectory.

[0005] The document DE 10 2020 129 456 A1 describes, among other things, techniques for predicting how an agent (e.g., a vehicle, bicycle, pedestrian, etc.) will move in an environment based on previous movement, the road network, surrounding objects, and / or other relevant environmental factors. One trajectory prediction technique involves generating a probability map for an agent's movement. Another trajectory prediction technique involves generating a trajectory grid for an agent's movement. Furthermore, another trajectory prediction technique involves multimodal regression, where a classifier (e.g., a neural network) is trained to classify the probability from a number of (learned) modes, so that each model generates a trajectory based on the current input.

[0006] US 2020 / 0 156 632 A1 describes methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying high-priority agents in the vicinity of a vehicle. In one aspect, a method comprises processing an input characterizing a trajectory of the vehicle in an environment, using an importance scoring model to generate an output defining a respective importance score for each of the various agents in the environment near the vehicle. The importance score of an agent indicates the estimated impact of the agent on planning decisions generated by a planning system of the vehicle that plans a future trajectory of the vehicle. The high-priority agents are identified as an appropriate subset of the plurality of agents with the highest importance scores. BRIEF DESCRIPTION OF THE CHARACTERS Fig. 1 is an exemplary environment in which a vehicle including one or more components of an autonomous system may be implemented; Fig. 2 is a diagram of one or more systems of a vehicle including an autonomous system; Fig. 3 is a diagram of components of one or more devices and / or one or more systems of Fig. 1 and Fig. 2; Fig. 4A is a diagram of certain components of an autonomous system; Fig. Figure 4B is a diagram of a neural network implementation; Fig. 4C and Fig. 4D are diagrams illustrating an example operation of a CNN; Fig. 5A illustrates an example of a system implementing a self-correcting scheduler, according to some embodiments of the present subject matter; Fig. Figure 5B illustrates additional details of the Fig. 5A as executed by its controller in accordance with some embodiments of the present subject matter; Fig. Figure 5C illustrates a training network used by the autonomous vehicle system (e.g., as shown in Fig. 5A and B) may be used during training, according to some embodiments of the current subject matter; Fig. 5D illustrates a process for generating correction(s) to input signals according to some embodiments of the present subject matter; Fig. Figure 5E illustrates an example of an experimental implementation of the self-correcting planner system (as in Fig. 5A) according to some embodiments of the current subject matter; Fig. 5F illustrates examples of feature vectors that may be generated by the feature extractor and rasterization components, according to some embodiments of the present subject matter; Fig. 5G illustrates another example of an experimental implementation of the self-correcting scheduler system (as in Fig. 5A) according to some embodiments of the present subject matter; and Fig. 6 illustrates an example process for generating a self-correction in a planner component of a vehicle according to some embodiments of the present subject matter. DETAILED DESCRIPTION

[0007] In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it should be understood that the embodiments described by the present disclosure may be practiced without these specific details. In some instances, well-known structures and devices are presented in block diagram form to avoid unnecessarily obscuring aspects of the present disclosure.

[0008] 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, one of ordinary skill in the art will understand that the specific order or arrangement of the schematic elements in the drawings is not intended to imply that a particular order or sequence of processing or a separation of processes is required unless expressly described. Furthermore, the inclusion of a schematic element in a drawing is not intended to imply that that element is required in all embodiments or that the features represented by that element cannot be incorporated into or combined with other elements in some embodiments unless expressly described.

[0009] Furthermore, in drawings in which connecting elements such as solid or dashed lines or arrows are used 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 intended to imply that no connection, relationship, or association can exist. In other words, some connections, relationships, or associations between elements are not shown in the drawings in order not to obscure the disclosure. In addition, for ease of illustration, a single connecting element may be used to represent multiple connections, relationships, or associations between elements. For example, if a connecting element represents communication of signals, data, or instructions (e.g., "software instructions"), one skilled in the art should understand that such an element may represent one or more signal paths (e.g.,a bus), depending on what is required to effect communication.

[0010] Although the terms "first," "second," "third," and / or the like are used to describe various elements, these elements are not intended to 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 referred to as a second contact, and similarly, a second contact could be referred to as a first contact, without departing from the scope of the described embodiments. Both the first contact and the second contact are contacts, but not the same contact.

[0011] The terminology used herein in the description of the various described embodiments is included only for the purpose of describing particular embodiments and is not intended to be limiting. The singular forms "a" and "an" as used in the description of the various described embodiments and the claims are intended to include the plural forms and may be used synonymously with "one or more" or "at least one," unless the context obviously indicates otherwise. Furthermore, it is understood that the term "and / or" as used herein refers to and includes all possible combinations of one or more of the associated listed elements.It is further understood that the terms "include / have", "including / having", "comprises" and / or "comprising", as used in this specification, indicate the presence of stated features, numbers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more further features, numbers, steps, operations, elements, components and / or groups thereof.

[0012] As used herein, the terms “communication” and “communicate” refer to the reception and / or transmission and / or sharing and / or provision and / or the like of information (or information represented, for example, by data, signals, messages, instructions, commands, and / or the like). When one entity (e.g., a device, a system, a component of a device or system, combinations thereof, and / or the like) is in communication with another entity, this means that one entity is capable of directly or indirectly receiving information from the other entity and / or sending (e.g., transmitting) information to the other entity. This can refer to a direct or indirect connection that is wired and / or wireless.Additionally, two entities may be in communication with each other even if the transmitted information is modified, processed, forwarded, and / or routed between the first and second entities. For example, a first entity may be in communication with a second entity even if the first entity is passively receiving information and not actively transmitting information to the second entity. As another example, a first entity may be in communication with a second entity if at least one intermediate entity (e.g., a third entity located between the first and second entities) processes information received from the first entity and forwards the processed information to the second entity. In some embodiments, a message may refer to a network packet (e.g., a data packet and / or the like) that contains data.

[0013] The term "if" is to be interpreted optionally as "when," "upon / after," or "in response to determining," "in response to detecting," and / or the like, depending on the context. Similarly, the phrase "if determined" or "if [a stated condition or event] is detected" is to be interpreted optionally as "upon / after determining," "in response to determining," "upon / after detecting [the stated condition or event]," "in response to detecting [the stated condition or event]," and / or the like, depending on the context. The phrases "has / has," "have / have," "having," or the like, as used herein, are also to be understood as open-ended expressions. Furthermore, the phrase "based on" is intended to mean "at least in part based on," unless explicitly stated otherwise.

[0014] Reference will now be made in detail to embodiments, 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 embodiments. However, one skilled in the art will understand that the various described embodiments may be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail to avoid unnecessarily obscuring aspects of the embodiments. General overview

[0015] A vehicle (e.g., an autonomous vehicle) contains sensors that monitor various parameters associated with the vehicle. For example, some sensors (e.g., cameras, LIDAR sensors, RADAR sensors, SONAR sensors, etc.) monitor / detect changes occurring in the vehicle's environment (e.g., actions and / or presence of other vehicles, pedestrians, street lighting, etc.). The information / data received from the sensors is used by the vehicle's control unit (or another processing component), which includes a planner component ("Planner"), to determine the path, direction, speed, and / or other movement parameters.

[0016] To reduce the impact of errors, a planner self-corrects during training by comparing its future outputs with a reference output (which may be a predetermined set of vehicle maneuver values defined by an "expert driver"). A planner is a causal, recursive training system whose output depends on past and / or current inputs. For example, a vehicle's current heading is used by the planner to determine (or otherwise influence) its future heading. Because errors occur when determining a future output of the planner (e.g., a vehicle's heading should have been 2 degrees to the left of a curb), and because the planner's input includes future outputs and associated errors, the effects of errors can be significantly amplified if left uncorrected.

[0017] The difference between the future output and the reference output corresponds to an error correction value. The future output and the error correction values are provided as input to the planner during a training phase to reduce the accumulating effect of errors. To perform the comparison, one or more features related to the vehicle and / or the environment (e.g., route centerline proximity, course agreement, ego shifts, etc.) are extracted from the future output, and a vector containing values of these features is generated and rasterized. The planner then compares the generated / rasterized vector with a vector containing reference values to determine an error correction vector, which is fed back along with a future maneuver output in a loop to repeat the process described above. The process can be repeated continuously or at specific times.

[0018] In some embodiments, one or more processors (e.g., an arbitration unit of the ego vehicle, a control unit, etc.) receive at least a first set of parameters (e.g., current, past maneuvers of the ego vehicle, etc.) associated with one or more previous maneuvers of a vehicle (e.g., the ego vehicle), and at least a second set of parameters (e.g., a maneuver goal of the vehicle) associated with a maneuver goal of the vehicle. Using the first and second set of parameters, the processor(s) generate(s) a future maneuver parameter corresponding to a future maneuver of the vehicle (e.g., the vehicle's control unit determines future maneuvers of the vehicle, e.g., the ego vehicle's entry into an adjacent lane, etc.). The processor(s) (e.g., a planner component of the processor(s)) train(s) at least one data model (e.g.,a motion data model) by comparing the generated future maneuver parameter with one or more reference maneuver parameters (e.g., one or more reference maneuvers determined by an "expert driver"). Based on the training, the processor(s) generate a corrected future maneuver parameter. The corrected future maneuver parameter includes a future maneuver of the vehicle and a correction to the future maneuver of the vehicle. To provide an end-to-end training system that enables correction of past parameters, in some embodiments the first parameter set may include the generated corrected future maneuver parameters. The generated corrected future maneuver parameter is used to correct at least one first parameter in the first parameter set (e.g., correction of past parameters, maneuvers, etc.).The processor(s) can then train the data model using the corrected first parameter.

[0019] In some embodiments, the present subject matter may include one or more of the following optional features. The processor(s) execute the generated corrected future maneuver of the vehicle. For example, a drive-by-wire (DBW) component of the vehicle may execute the corrected future maneuver after completion of model training.

[0020] In some embodiments, receiving the first and second sets of parameters includes continuously receiving the generated corrected future maneuver parameters following the execution to correct one or more previously received first parameters. The vehicle's planner executes a closed-loop process in which all generated future maneuver parameters are incorporated into the planner process as past maneuver parameters, i.e., as first parameters. As already mentioned, this enables correction of past parameters.

[0021] In some embodiments, the processor(s) further receive at least one third parameter associated with a state of the vehicle. The third parameter corresponds, for example, to an environment and / or scene input (e.g., pedestrian on the sidewalk, vehicle in an oncoming lane, etc.). Training the data model includes training the data model using the future maneuver parameter, the third parameter (e.g., related to the environment and / or scene), and one or more reference maneuver parameters. Training the data model thus includes using environment / scene data as inputs for training purposes.

[0022] In some embodiments, the above process further includes repeating the receiving of the first and second sets of parameters, comparing the generated future maneuver parameter with the reference maneuver parameter(s), training the data model, and generating the corrected future maneuver parameter. Thus, the planning process is repetitive, and corrected parameters are continuously fed to the planner.

[0023] In some embodiments, at least one of the first and second parameters corresponds to at least one feature associated with at least one maneuver of the vehicle. During the process, the processor(s) extract one or more features associated with the maneuvers for training purposes. The processor(s) then rasterize the extracted feature(s) and backpropagate them for training. Training the data model further includes training the data model using at least one rasterized feature that can be associated with a movement of the vehicle.

[0024] In some embodiments, executing the corrected further maneuver includes executing the corrected future maneuver of the vehicle while the vehicle is operating. This enables the generation of movements during travel time.

[0025] In some embodiments, the first and second sets of parameters include at least one of the following: a speed, a position, an acceleration, a direction of movement, and any combination of these, of the vehicle.

[0026] In some embodiments, at least one received first parameter is differentiable from at least one of the following parameters: at least one previously received first parameter and at least one corrected first parameter. Alternatively or additionally, at least one received first parameter is differentiable from at least one first anchor parameter in the first parameter set. The first anchor parameter can be associated with at least one of the following: at least one reference feature associated with at least one maneuver of the vehicle and at least one reference maneuver parameter.

[0027] The implementation of the systems, methods and computer program products described herein provides methods for enabling self-correction and / or elimination of errors that may be associated with the planning and / or execution of maneuvers by a control unit of a vehicle, which in turn may avoid undesirable consequences (e.g., accidents).

[0028] In Fig. 1 shows an exemplary environment 100 in which both vehicles with autonomous systems and vehicles without such systems are operated. As illustrated, the environment 100 includes vehicles 102a-102n, objects 104a-104n, routes 106a-106n, an area 108, a vehicle-to-infrastructure (V21) device 110, a network 112, a remote AV (autonomous vehicle) system 114, a fleet management system 116, and a V2I system 118. The vehicles 102a-102n, the vehicle-to-infrastructure (V21) device 110, the network 112, the AV system 114, the fleet management system 116, and the V2I system 118 are interconnected via wired connections, wireless connections, or a combination of wired or wireless connections (e.g., they connect for communication and / or the like).In some embodiments, objects 104a-104n are connected to vehicles 102a-102n and / or vehicle-to-infrastructure (V2I) device 110 and / or network 112 and / or AV system 114 and / or fleet management system 116 and / or V2I system 118 via wired connections, wireless connections, or a combination of wired or wireless connections.

[0029] The vehicles 102a-102n (individually referred to as vehicle 102 and collectively as vehicles 102) include at least one device configured to transport goods and / or people. In some embodiments, the vehicles 102 are configured to communicate with the V2I device 110, the remote AV system 114, the fleet management system 116, and / or the V2I system 118 via the network 112. In some embodiments, the vehicles 102 include cars, buses, trucks, trains, and / or the like. In some embodiments, the vehicles 102 are the same as or similar to the presently described vehicles 200 (see Fig. 2). In some embodiments, a vehicle 200 of a set of vehicles 200 is associated with an autonomous fleet manager. In some embodiments, the vehicles 102 travel along respective routes 106a-106n (individually referred to as route 106 and collectively as routes 106), as described herein. In some embodiments, 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).

[0030] Objects 104a-104n (individually referred to 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., at a fixed location for a specific time) or mobile (e.g., having a velocity and at least one trajectory). In some embodiments, objects 104 are associated with corresponding locations in area 108.

[0031] Routes 106a-106n (individually referred to as route 106 and collectively as routes 106) are each associated with (e.g., they specify) a sequence of actions (also referred to as a trajectory) that connect states along which an AV can navigate. Each route 106 begins with an initial state (e.g., a state corresponding to a first spatiotemporal location, a velocity vector, and / or the like) and a final destination state (e.g., a state corresponding to a second spatiotemporal location that is different from the first spatiotemporal location) or a destination region (e.g., a subspace of acceptable states (e.g., final states)).In some embodiments, 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 embodiments, the routes 106 include a plurality of acceptable state sequences (e.g., a plurality of spatiotemporal location sequences), where the plurality of state sequences are associated with (e.g., define) a plurality of trajectories. In one example, the routes 106 include only high-level actions or imprecise state locations, such as a series of connected roads that specify turn directions at road intersections.Additionally or alternatively, the routes 106 may include more precise actions or states, such as, for example, specific target lanes or precise locations within the lane regions and a target speed at those positions. In one example, the routes 106 include a plurality of precise state sequences along the at least one sequence of higher-level actions with a limited look-ahead horizon to reach intermediate goals, wherein the combination of successive iterations of limited-horizon state sequences cumulatively corresponds to a plurality of trajectories that together form the higher-level route to arrive at the final target state or region.

[0032] The area 108 includes a physical area (e.g., a geographic region) in which the vehicles 102 can navigate. In one example, the area 108 includes at least one state (e.g., a country, a province, a single state of a plurality of states comprising a country, etc.), at least a portion of a state, at least one city, at least a portion of a town, etc. In some embodiments, the area 108 includes at least one named thoroughfare (hereinafter referred to as a "road"), such as a highway, a trunk road, an avenue, a city street, etc. Additionally or alternatively, in some examples, the area 108 includes at least one unnamed road, such as a driveway, a portion of a parking lot, a portion of vacant and / or undeveloped land, a dirt road, etc. In some embodiments, a road includes at least one travel lane (e.g.,a portion of the road that can be traveled by vehicles 102). In one example, a road includes at least one lane associated with (e.g., identified based on) at least one lane marking.

[0033] The vehicle-to-infrastructure (V2I) device 110 (sometimes referred to as a vehicle-to-infrastructure (V2X) device) includes at least one device configured to communicate with vehicles 102 and / or the V2I infrastructure system 118. In some embodiments, the V2I device 110 is configured to communicate with the vehicles 102, the remote AV system 114, the fleet management system 116, and / or the V2I system 118 via the network 112. In some embodiments, the V2I device 110 includes a radio frequency identification (RFID) device, traffic signs, cameras (e.g., two-dimensional (2D) and / or three-dimensional (3D) cameras), lane markings, streetlights, parking meters, etc. In some embodiments, the V2I device 110 is configured to communicate directly with the vehicles 102.Additionally or alternatively, in some embodiments, the V2I device 110 is configured to communicate with the vehicles 102, the remote AV system 114, and / or the fleet management system 116 via the V2I system 118. In some embodiments, the V2I device 110 is configured to communicate with the V2I system 118 via the network 112.

[0034] The network 112 includes one or more wired and / or wireless networks. In one example, the network 112 includes a cellular network (e.g., an LTE (Long Term Evolution) network, a 3G (third generation) network, a 4G (fourth generation) network, a 5G (fifth generation) network, a CDMA (code division multiple access) 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 network, a cloud computing network, etc., a combination of some or all of these networks, and / or the like.

[0035] The remote AV system 114 includes at least one device configured to communicate with the vehicles 102, the V2I device 110, the network 112, the remote AV system 114, the fleet management system 116, and / or the V2I system 118 via the network 112. In one example, the remote AV system 114 includes a server, a group of servers, and / or other similar devices. In some embodiments, the remote AV system 114 is co-located with the fleet management system 116. In some embodiments, the remote AV system 114 is involved in the installation of some or all of the components of a vehicle, including an autonomous system, an AV computer, software implemented by an AV computer, and / or the like. In some embodiments, the remote AV system 114 maintains (e.g.,updates and / or replaces) the remote AV system 114 such components and / or software during the life of the vehicle.

[0036] The fleet management system 116 includes at least one device configured to be in communication with the vehicles 102, the V2I device 110, the remote AV system 114, and / or the V2I infrastructure system 118. In one example, the fleet management system 116 includes a server, a group of servers, and / or other similar devices. In some embodiments, the fleet management system 116 is associated with a ride-sharing company (e.g., an organization that manages the operation of multiple vehicles (e.g., vehicles that include autonomous systems and / or vehicles that do not include autonomous systems), and / or the like).

[0037] In some embodiments, the V2I system 118 includes at least one device configured to communicate with the vehicles 102, the V2I device 110, the remote AV system 114, and / or the fleet management system 116 via the network 112. In some examples, the V2I system 118 is configured to communicate with the V2I device 110 via a connection other than the network 112. In some embodiments, the V2I system 118 includes a server, a group of servers, and / or other similar devices. In some embodiments, the V2I system 118 is associated with a municipality or a private institution (e.g., a private institution that manages the V2I device 110 and / or the like).

[0038] The number and arrangement of the Fig. 1 is merely exemplary. Additional elements, fewer elements, different elements and / or differently arranged elements than those shown in Fig. 1. Additionally or alternatively, at least one element of the environment 100 may perform one or more functions that are considered to be performed by at least one other element in Fig. 1. Additionally or alternatively, at least one set of elements of the environment 100 may perform one or more functions described as being performed by at least one other set of elements of the environment 100.

[0039] According to Fig. 2, the vehicle 200 includes an autonomous system 202, a powertrain control system 204, a steering control system 206, and a braking system 208. In some embodiments, the vehicle 200 is the same as or similar to the vehicle 102 (see Fig. 1). In some embodiments, the vehicle 102 has autonomous capabilities (e.g., implements at least one function, feature, device, and / or the like that enables the vehicle 200 to operate partially or entirely without human intervention, including, without limitation, fully autonomous vehicles (e.g., vehicles that forgo human intervention), highly autonomous vehicles (e.g., vehicles that forgo human intervention in certain situations), and / or the like). For a detailed description of fully autonomous vehicles and highly autonomous vehicles, reference is made to SAE International's Standard J3016: Taxonomy and Definitions for Terms Related to On-Road Motor Vehicle Automated Driving Systems, which is hereby incorporated in its entirety.In some embodiments, vehicle 200 is associated with an autonomous fleet manager and / or a ride-sharing company.

[0040] The 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 embodiments, the autonomous system 202 may include more or fewer devices and / or different devices (e.g., ultrasonic sensors, inertial sensors, GPS receivers (see below), odometry sensors that generate data associated with an indication of a distance traveled by the vehicle 200, and / or the like). In some embodiments, the autonomous system 202 uses the one or more devices included in the autonomous system 202 to generate data associated with the environment 100 as described herein. The data generated by the one or more devices of the autonomous system 202 may be used by one or more systems described herein to monitor the environment (e.g.,the environment 100) in which the vehicle 200 is located. In some embodiments, the autonomous system 202 includes a communication device 202e, an AV computer 202f, and a drive-by-wire (DBW) system 202h.

[0041] The cameras 202a include at least one device configured to communicate via a bus (e.g., a bus that is similar to the bus 302 in Fig. 3) is in communication with the communication device 202e, the AV computer 202f, and / or the security control unit 202g. The cameras 202a include at least one camera (e.g., a digital camera with a light sensor such as a charge-coupled device (CCD), a thermal imaging camera, an infrared (IR) camera, an event camera, and / or the like) for capturing images with physical objects (e.g., cars, buses, curbs, people, and / or the like). In some embodiments, the camera 202a generates camera data as output. In some examples, the camera 202a generates camera data that includes image data associated with an image. In this example, the image data may indicate at least one parameter corresponding to the image (e.g., image characteristics such as exposure, brightness, etc., an image timestamp, and / or the like). In such an example, the image may be in a format (e.g.,RAW, JPEG, PNG, and / or the like). In some embodiments, camera 202a includes a plurality of independent cameras configured (e.g., positioned) on a vehicle to capture images for stereopsis (spatial vision). In some examples, camera 202a includes a plurality of cameras that generate image data and transmit the image data to AV computer 202f and / or a fleet management system (e.g., a fleet management system similar to fleet management system 116 of FIG. Fig. 1). In one such example, the AV computer 202f determines a depth to one or more objects within a field of view of at least two of the plurality of cameras based on the image data from the at least two cameras. In some embodiments, the cameras 202a are configured to capture images of objects within a distance from the cameras 202a (e.g., up to 100 meters, up to one kilometer, and / or the like). Accordingly, the cameras 202a include features such as sensors and lenses optimized for perceiving objects located at one or more distances from the cameras 202a.

[0042] In one embodiment, camera 202a includes at least one camera configured to capture one or more images associated with one or more traffic lights, road signs, and / or other physical objects that provide visual navigation information. In some embodiments, camera 202a generates traffic light data associated with one or more images. In some examples, camera 202a generates TLD data associated with one or more images that include a format (e.g., RAW, JPEG, PNG, and / or the like). In some embodiments, camera 202a that generates TLD data differs from other systems described herein that include cameras in that camera 202a may include one or more cameras with a wide field of view (e.g.,a wide-angle lens, a fisheye lens, a lens with a viewing angle of approximately 120 degrees or more, and / or the like) to create images over as many physical objects as possible.

[0043] The LiDAR (laser detection and ranging) sensors 202b include at least one device configured to communicate via a bus (e.g., a bus similar to bus 302 in Fig. 3) is in communication with the communication device 202e, the AV computer 202f, and / or the security control unit 202g. The LiDAR sensors 202b include a system configured to transmit light from a light emitter (e.g., a laser transmitter). Light emitted by the LiDAR sensors 202b includes light (e.g., infrared light and / or the like) that is outside the visible spectrum. In some embodiments, during operation, light emitted by the LiDAR sensors 202b strikes a physical object (e.g., a vehicle) and is reflected back to the LiDAR sensors 202b. In some embodiments, the light emitted by the LiDAR sensors 202b does not penetrate the physical objects that the light strikes. The LiDAR sensors 202b also include at least one light detector that detects the light emitted by the light emitter after the light strikes a physical object.In some embodiments, at least one computing system associated with the 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 the LiDAR sensors 202b. In some examples, the at least one computing system associated with the LiDAR sensor 202b generates an image representing 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 within the field of view of the LiDAR sensors 202b.

[0044] The radar (radio detection and ranging) sensors 202c include at least one device configured to communicate via a bus (e.g., a bus similar to bus 302 in Fig. 3) is in communication with the communication device 202e, the AV computer 202f, and / or the security control unit 202g. The radar sensors 202c include a system configured to transmit radio waves (either pulsed or continuous). The radio waves transmitted by the radar sensors 202c include radio waves that are within a predetermined spectrum. In some embodiments, during operation, radio waves transmitted by the radar sensors 202c encounter a physical object and are reflected back to the radar sensors 202c. In some embodiments, the radio waves transmitted by the radar sensors 202c are not reflected by some objects. In some embodiments, at least one data processing system associated with the radar sensors 202c generates signals representative of the objects included in a field of view of the radar sensors 202c.For example, the at least one data processing system associated with radar sensor 202c generates an image representing 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 within the field of view of radar sensors 202c.

[0045] The microphones 202d include at least one device configured to be connected to the bus 302 via a bus (e.g., a bus that is similar to the bus 302 in Fig. 3) is in communication with the communication device 202e, the AV computer 202f, and / or the security control unit 202g. The 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, the microphones 202d include transducer devices and / or similar devices. In some embodiments, one or more systems described herein may receive the data generated by the microphones 202d and determine a position of an object relative to the vehicle 200 (e.g., a distance and / or the like) based on the audio signals associated with the data.

[0046] The communication device 202e includes at least one device configured to communicate with the cameras 202a, the LiDAR sensors 202b, the radar sensors 202c, the microphones 202d, the AV computer 202f, the safety control unit 202g, and / or the DBW system 202h. For example, the communication device 202e may include a device that corresponds to the communication interface 314 of Fig. 3. In some embodiments, communication device 202e includes a vehicle-to-vehicle (V2V) communication device (e.g., a device that enables wireless communication of data between vehicles).

[0047] The AV computer 202f includes at least one device configured to communicate with the cameras 202a, the LiDAR sensors 202b, the radar sensors 202c, the microphones 202d, the communication device 202e, the security controller 202g, and / or the DBW system 202h. In some examples, the AV computer 202f includes a device such as a client device, a mobile device (e.g., a mobile phone, a tablet, and / or the like), a server (e.g., a computing device including one or more central processing units, graphics processing units, and / or the like), and / or the like. In some embodiments, the AV computer 202f is the same as or similar to the AV computer 400 described herein. Additionally or alternatively, in some embodiments, the AV computer 202f is configured to communicate with an AV system (e.g., an AV system that is similar to the remote AV system 114 of Fig. 1), a fleet management system (e.g. a fleet management system that is similar to the fleet management system 116 of Fig. 1), a V2I device (e.g., a V2I device that is similar to the V2I device 110 of Fig. 1), and / or a V2I system (e.g. a V2I system that is similar to the V2I system 118 of Fig. 1 is the same as or similar to) is in communication.

[0048] The safety control unit 202g includes at least one device configured to communicate with the cameras 202a, the LiDAR sensors 202b, the radar sensors 202c, the microphones 202d, the communication device 202e, the AV computer 202f, and / or the DBW system 202h. In some examples, the safety control unit 202g includes one or more control units (electrical control units, electromechanical control units, and / or the like) configured to generate and / or transmit control signals to operate one or more devices of the vehicle 200 (e.g., the powertrain control system 204, the steering control system 206, the braking system 208, and / or the like). In some embodiments, the security control unit 202g is configured to generate control signals that override (e.g., override) control signals generated and / or transmitted by the AV computer 202f.

[0049] The DBW system 202h includes at least one device configured to communicate with the communication device 202e and / or the AV computer 202f. In some examples, the DBW system 202h includes one or more control units (e.g., electrical control units, electromechanical control units, and / or the like) configured to generate and / or transmit control signals to operate one or more devices of the vehicle 200 (e.g., the powertrain control system 204, the steering control system 206, the braking system 208, and / or the like). Additionally or alternatively, the one or more control units of the DBW system 202h are configured to generate and / or transmit control signals to operate at least one other device (e.g., a turn signal, headlights, door locks, windshield wipers, and / or the like) of the vehicle 200.

[0050] The powertrain control system 204 includes at least one device configured to communicate with the DBW system 202h. In some examples, the powertrain control system 204 includes at least one controller, an actuator, and / or the like. In some embodiments, the powertrain control system 204 receives control signals from the DBW system 202h, and the powertrain control system 204 causes the vehicle 200 to begin moving forward, stop moving forward, begin moving backward, stop moving backward, accelerate in a direction, decelerate in a direction, turn left, turn right, and / or the like. In one example, the powertrain control system 204 causes the power supplied to an engine of the vehicle (e.g.,fuel, electricity, and / or the like) increases, remains the same, or decreases, causing at least one wheel of the vehicle 200 to rotate or not rotate.

[0051] The steering control system 206 includes at least one device configured to turn one or more wheels of the vehicle 200. In some examples, the steering control system 206 includes at least one controller, an actuator, and / or the like. In some embodiments, the steering control system 206 causes the two front wheels and / or the two rear wheels of the vehicle 200 to turn left or right to cause the vehicle 200 to turn left or right.

[0052] The braking system 208 includes at least one device configured to actuate one or more brakes to cause the vehicle 200 to decelerate and / or stop. In some examples, the braking system 208 includes at least one controller and / or actuator configured to cause one or more brake calipers associated with one or more wheels of the vehicle 200 to close against a corresponding rotor of the vehicle 200. Additionally or alternatively, in some examples, the braking system 208 includes an automatic emergency braking (AEB) system, a regenerative braking system, and / or the like.

[0053] In some embodiments, the vehicle 200 includes at least one platform sensor (not explicitly shown) that measures or infers characteristics of a state or condition of the vehicle 200. In some examples, the 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.

[0054] In Fig. 3 illustrates a schematic diagram of a device 300. As shown, the device 300 includes a processor 304, a memory 306, a storage component 308, an input interface 310, an output interface 312, a communication interface 314, and a bus 302. In some embodiments, the device 300 corresponds to at least one device of the vehicles 102 (e.g., at least one device of a system of the vehicles 102), at least one device of in other embodiments Fig. 1 and / or one or more devices of the network 112 (e.g., one or more devices of a system of the network 112). In some embodiments, one or more devices of the vehicles 102 (e.g., one or more devices of a system of the vehicles 102) include at least one device of other Fig. 1 illustrated devices / objects and / or one or more devices of the network 112 (e.g., one or more devices of a system of the network 112) at least one device 300 and / or at least one component of the device 300.

[0055] Bus 302 includes a component that enables communication between the components of device 300. In some embodiments, processor 304 is implemented in hardware, software, or a combination of hardware and software. In some examples, 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 the processor 304.

[0056] The storage component 308 stores data and / or software related to the operation and use of the device 300. In some examples, the storage component 308 includes a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optical disk, a solid-state disk, and / or the like), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cassette, 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.

[0057] The input interface 310 includes a component that enables the device 300 to receive information, e.g., about 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 embodiments, the 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). The output interface 312 includes a component that provides output information from the device 300 (e.g., a display, a speaker, one or more light-emitting diodes (LEDs), and / or the like).

[0058] In some embodiments, the communication interface 314 includes a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, and / or the like) that enables the 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, the communication interface 314 enables the device 300 to receive information from another device and / or to provide information to another device. In some examples, the communication interface 314 includes an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a USB (Universal Serial Bus) interface, a Wi-Fi ®interface, a cellular network interface and / or the like.

[0059] In some embodiments, 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-volatile storage device. A non-volatile storage device includes storage space located within a single physical storage device or storage space distributed across multiple physical storage devices.

[0060] In some embodiments, 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, the 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, hard-wired circuitry is used instead of, or in combination with, software instructions to perform one or more processes described herein. Thus, embodiments described herein are not limited to any particular combination of hardware circuitry and software unless expressly stated otherwise.

[0061] Memory 306 and / or storage component 308 include a data store 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, transmitting information to, or searching information stored in the data store 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.

[0062] In some embodiments, device 300 is configured to execute software instructions stored either in memory 306 and / or in the memory of another device (e.g., another device that is the same 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 or similar to device 300), causes device 300 (e.g., at least one component of device 300) to perform one or more processes described herein. In some embodiments, a module is implemented in software, firmware, hardware, and / or the like.

[0063] The number and arrangement of the Fig. 3 is merely exemplary. In some embodiments, the device 300 may include additional components, fewer components, different components, or differently arranged components than in Fig. 3. Additionally or alternatively, a set of components (e.g., one or more components) of device 300 may perform one or more functions described as being performed by another component or set of components of device 300.

[0064] Fig. 4A illustrates an example block diagram of an AV computer 400 (sometimes referred to as an "AV stack"). As illustrated, the AV computer 400 includes a perception system 402 (sometimes referred to as a perception module), a planning system 404 (sometimes referred to as a planning module), a localization system 406 (sometimes referred to as a localization module), a control system 408 (sometimes referred to as a control module), and a database 410. In some embodiments, the perception system 402, the planning system 404, the localization system 406, the control system 408, and the database 410 are included and / or implemented in an autonomous navigation system of a vehicle (e.g., the AV computer 202f of the vehicle 200).Additionally or alternatively, in some embodiments, the perception system 402, the planning system 404, the localization system 406, the control system 408, and the database 410 are included in one or more independent systems (e.g., one or more systems that are the same or similar to the compute of an autonomous vehicle 400 and / or the like). In some examples, the perception system 402, the planning system 404, the localization system 406, the control system 408, and the database 410 are included in one or more standalone systems located in a vehicle and / or in at least one remotely located system as described herein. In some embodiments, any and / or all of the systems included in the compute of an autonomous vehicle 400 are embodied in software (e.g., software instructions stored in memory), computer hardware (e.g.,implemented 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 appreciated that in some embodiments, the autonomous vehicle compute 400 is configured to communicate with a remotely located system (e.g., an autonomous vehicle system that is the same or similar to the remote AV system 114, a fleet management system 116 that is the same or similar to the fleet management system 116, a V2I system that is the same or similar to the V2I system 118, and / or the like).

[0065] In some embodiments, perception system 402 receives data associated with at least one physical object (e.g., data used by perception system 402 to recognize 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 embodiments, the perception system 402 transmits data associated with the classification of the physical objects to the planning system 404 based on the classification of the physical objects by the perception system 402.

[0066] In some embodiments, the planning system 404 receives data associated with a destination and generates data associated with at least one route (e.g., routes 106) along which a vehicle (e.g., vehicles 102) can travel to a destination. In some embodiments, the planning system 404 periodically or continuously receives data from the perception system 402 (e.g., data associated with the classification of physical objects described above), and the planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by the perception system 402. In some embodiments, the planning system 404 receives data associated with an updated position of a vehicle (e.g.,of the vehicles 102) from the localization system 406, and the planning system 404 updates the at least one trajectory or generates at least one other trajectory based on the data generated by the localization system 406.

[0067] In some embodiments, the localization system 406 receives data associated with (e.g., representing) a location of a vehicle (e.g., vehicles 102) within an area. In some examples, the 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, the localization system 406 receives data associated with at least one point cloud from multiple LiDAR sensors, and the localization system 406 generates a combined point cloud based on each of the point clouds. In these examples, the localization system 406 compares the at least one point cloud or the combined point cloud to a two-dimensional (2D) and / or a three-dimensional (3D) map of the area stored in the database 410.The localization system 406 then determines the position of the vehicle in the area based on the localization system 406 comparing the at least one point cloud or the combined point cloud with the map. In some embodiments, the map includes a combined point cloud of the area generated prior to navigation of the vehicle. In some embodiments, maps include, but are not limited to, high-precision maps of the geometric properties of the roadway, maps describing connectivity properties of the road network, maps describing physical properties of the roadway (e.g.,Traffic speed, traffic volume, number of lanes for car and bicycle traffic, lane width, lane directions, or lane marking types and locations, or combinations thereof), as well as maps describing the spatial location of road features such as pedestrian crossings, traffic signs, or other traffic signals of various types. In some embodiments, the map is generated in real time based on the data received from the perception system.

[0068] In another example, the location system 406 receives GNSS (Global Navigation Satellite System) data generated by a GPS (Global Positioning System) receiver. In some examples, the location system 406 receives GNSS data associated with the location of the vehicle in the area, and the location system 406 determines a latitude and longitude of the vehicle in the area. In such an example, the location system 406 determines the position of the vehicle in the area based on the latitude and longitude of the vehicle. In some embodiments, the location system 406 generates data associated with the position of the vehicle. In some examples, the location system 406 generates data associated with the position of the vehicle based on the location system 406 determining the position of the vehicle.In such an example, the data related to the position of the vehicle includes data associated with one or more semantic properties corresponding to the position of the vehicle.

[0069] In some embodiments, 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 operate 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 braking system (e.g., braking system 208). In an example where a trajectory includes a left turn, the control system 408 transmits a control signal to cause the steering control system 206 to adjust a steering angle of the vehicle 200, thereby causing the vehicle 200 to turn left.Additionally or alternatively, the control system 408 generates and transmits control signals to cause other devices (e.g., headlights, turn signals, door locks, windshield wipers, and / or the like) of the vehicle 200 to change their state.

[0070] In some embodiments, the perception system 402, the planning system 404, the localization system 406, and / or the 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, the perception system 402, the planning system 404, the localization system 406, and / or the control system 408 implement at least one machine learning model alone or in combination with one or more of the systems noted above. In some examples, the perception system 402, the planning system 404, the localization system 406, and / or the 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). An example implementation of a machine learning model is described below with reference to . Fig. 4B-4D included.

[0071] The database 410 stores data that is transmitted to, received from, and / or updated by the perception system 402, the planning system 404, the localization system 406, and / or the control system 408. In some examples, the database 410 includes a storage component (e.g., a storage component that is associated with the storage component 308 of Fig. 3) that stores data and / or software associated with operation and utilizes at least one system of AV computer 400. In some embodiments, 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 country (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) may travel along one or more drivable regions (e.g., single-lane roads, multi-lane roads, highways, secondary roads, off-road trails, and / or the like) and may include at least one LiDAR sensor (e.g.,a LiDAR sensor that is the same as or similar to the LiDAR sensors 202b) to generate data associated with an image representing the objects contained in a field of view of the at least one LiDAR sensor.

[0072] In some embodiments, the database 410 may be implemented on a variety of devices. In some examples, the database 410 is implemented in a vehicle (e.g., a vehicle that is the same as or similar to vehicles 102 and / or vehicle 200), an AV system (e.g., an AV system that is the same as or similar to the remote AV system 114), a fleet management system (e.g., a fleet management system that is similar to the fleet management system 116 of Fig. 1), a V2I system (e.g. a V2I system that is similar to the V2I system 118 of Fig. 1 is the same as or similar to) and / or the like.

[0073] According to Fig. 4B illustrates a diagram of an implementation of a machine learning model. More specifically, a diagram of an implementation of a convolutional neural network (CNN) 420 is illustrated. For illustrative purposes, the following description of the CNN 420 is provided with respect to an implementation of the CNN 420 by the perception system 402. However, it should be understood that in some examples, the CNN 420 (e.g., one or more components of the CNN 420) is implemented by other systems different from or in addition to the perception system 402, such as the planning system 404, the localization system 406, and / or the control system 408. While the CNN 420 includes certain features as described herein, these features are provided for illustrative purposes and are not intended to limit the present disclosure.

[0074] The CNN 420 includes a plurality of convolutional layers, including a first convolutional layer 422, a second convolutional layer 424, and a convolutional layer 426. In some embodiments, the CNN 420 includes a subsampling layer 428 (sometimes referred to as a pooling layer). In some embodiments, the subsampling layer 428 and / or other subsampling layers have a dimension (i.e., a set of nodes) that is smaller than a dimension of an upstream system. Because the subsampling layer 428 has a dimension that is smaller than a dimension of an upstream layer, the CNN 420 consolidates the amount of data associated with the initial input and / or the output of an upstream layer, thereby reducing the amount of computation required by the CNN 420 to perform downstream convolutional operations.Additionally or alternatively, since the subsampling layer 428 is associated with (e.g., configured to perform) at least one subsampling function (as described below with reference to FIG. Fig. 4C and Fig. 4D), the amount of data associated with the initial input.

[0075] The perception system 402 performs convolution operations based on the perception system 402 providing respective inputs and / or outputs associated with each of the first convolutional layer 422, the second convolutional layer 424, and the convolutional layer 426 to generate respective outputs. In some examples, the perception system 402 implements the CNN 420 based on the perception system 402 providing data as input to the first convolutional layer 422, the second convolutional layer 424, and the convolutional layer 426. In such an example, the perception system 402 provides the data as input to the first convolutional layer 422, the second convolutional layer 424, and the convolutional layer 426 based on the perception system 402 receiving data from one or more different systems (e.g.,one or more systems of a vehicle that is the same as or similar to vehicle 102, a remote AV system that is the same as or similar to remote AV system 114, a fleet management system 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). A detailed description of convolution operations is provided below with reference to FIG. Fig. 4C included.

[0076] In some embodiments, the perception system 402 provides data associated with an input (referred to as an initial input) to the first convolutional layer 422, and the perception system 402 generates data associated with an output using the first convolutional layer 422. In some embodiments, the perception system 402 provides an output generated by one convolutional layer as input to another convolutional layer. For example, the perception system 402 provides the output of the first convolutional layer 422 as input to the subsampling layer 428, the second convolutional layer 424, and / or the convolutional layer 426. In such an example, the first convolutional layer 422 is referred to as an upstream layer, and the subsampling layer 428, the second convolutional layer 424, and / or the convolutional layer 426 are referred to as downstream layers.Similarly, in some embodiments, the perception system 402 provides the output of the subsampling layer 428 to the second convolutional layer 424 and / or the convolutional layer 426, and in this example, the subsampling layer 428 would be referred to as an upstream layer, and the second convolutional layer 424 and / or the convolutional layer 426 would be referred to as downstream layers.

[0077] In some embodiments, the perception system 402 processes the data associated with the input provided to the CNN 420 before the perception system 402 provides the input to the CNN 420. For example, the perception system 402 processes the data associated with the input provided to the CNN 420 based on the perception system 420 normalizing sensor data (e.g., image data, LiDAR data, radar data, and / or the like).

[0078] In some embodiments, the CNN 420 generates an output based on the perception system 420 performing convolution operations associated with each convolutional layer. In some examples, the CNN 420 generates an output based on the perception system 420 performing convolution operations associated with each convolutional layer and an initial input. In some embodiments, the perception system 402 generates the output and provides the output as a fully connected layer 430. In some examples, the perception system 402 provides the output of the convolutional layer 426 as the fully connected layer 430, where the fully connected layer 420 includes data associated with a plurality of feature values, denoted as F1, F2...FN. In this example, the output of the convolutional layer 426 includes data associated with a plurality of output feature values representing a prediction.

[0079] In some embodiments, the perception system 402 identifies a prediction from a plurality of predictions based on the perception system 402 identifying a feature value associated with the highest probability of being the correct prediction from the plurality of predictions. For example, if the fully connected layer 430 includes feature values F1, F2, ... FN and F1 is the largest feature value, the perception system 402 identifies the prediction associated with F1 as the correct prediction from the plurality of predictions. In some embodiments, the perception system 402 trains the CNN 420 to generate the prediction. In some examples, the perception system 402 trains the CNN 420 to generate the prediction based on the perception system 402 providing the CNN 420 with training data associated with the prediction.

[0080] According to Fig. 4C and Fig. 4D illustrates a diagram of exemplary operation of a CNN 440 by the perception system 402. In some embodiments, the CNN 440 (e.g., one or more components of the CNN 440) is the same as or similar to the CNN 420 (e.g., one or more components of the CNN 420) (see Fig. 4B).

[0081] In step 450, the perception system 402 provides data associated with an image as input to a CNN 440 (step 450). For example, as illustrated, the perception system 402 provides the data associated with the image to the CNN 440, where the image is a grayscale image represented as values stored in a two-dimensional (2D) array. In some embodiments, the data associated with the image may include data associated with a color image, where the color image is represented as values stored in a three-dimensional (3D) array. Additionally or alternatively, the data associated with the image may include data associated with an infrared image, a radar image, and / or the like.

[0082] In step 455, the CNN 440 performs a first convolution function. For example, the CNN 440 performs the first convolution function based on the CNN 440 providing the values representing the image as input to one or more neurons (not explicitly illustrated) included in the first convolutional layer 442. In this example, the values representing the image may correspond to values representing a region of the image (sometimes referred to as a receptive field). In some embodiments, each neuron is associated with a filter (not explicitly illustrated). A filter (sometimes referred to as a kernel) is representable as an array of values corresponding in size to the values provided as input to the neuron. In one example, a filter may be configured to identify edges (e.g., horizontal lines, vertical lines, straight lines, and / or the like).In subsequent convolutional layers, the filters associated with neurons can be designed to identify successively more complex patterns (e.g., arcs, objects, and / or the like).

[0083] In some embodiments, the CNN 440 performs the first convolution function based on the CNN 440 multiplying the values provided as input to each of the one or more neurons included in the first convolutional layer 442 by the values of the filter corresponding to each of the one or more neurons. For example, the CNN 440 may multiply the values provided as input to each of the one or more neurons included in the first convolutional layer 442 by the values of the filter corresponding to each of the one or more neurons to produce a single value or an array of values as an output. In some embodiments, the collective output of the neurons of the first convolutional layer 442 is referred to as a convolved output.In some embodiments where each neuron has the same filter, the convolved output is referred to as a feature map.

[0084] In some embodiments, the CNN 440 provides the outputs of each neuron of the first convolutional layer 442 to neurons of a downstream layer. For clarity, an upstream layer may be a layer that transmits data to another layer (referred to as a downstream layer). For example, the CNN 440 may provide the outputs of each neuron of the first convolutional layer 442 to corresponding neurons of a subsampling layer. In one example, the CNN 440 provides the outputs of each neuron of the first convolutional layer 442 to corresponding neurons of the first subsampling layer 444. In some embodiments, the CNN 440 adds a bias value to the aggregates of all values provided to each neuron of the downstream layer. For example, the CNN 440 adds a bias value to the aggregates of all values provided to each neuron of the first subsampling layer 444.In such an example, the CNN 440 determines a final value to be provided to each neuron of the first subsampling layer 444 based on the aggregates of all values provided to each neuron and an activation function associated with each neuron of the first subsampling layer 444.

[0085] In step 460, the CNN 440 performs a first subsampling function. For example, the CNN 440 may perform a first subsampling function based on the CNN 440 providing the values output by the first convolutional layer 442 to corresponding neurons of the first subsampling layer 444. In some embodiments, the CNN 440 performs the first subsampling function based on an aggregation function. In one example, the CNN 440 performs the first subsampling function based on the CNN 440 determining the maximum input among the values provided to a given neuron (referred to as a max-pooling function). In another example, the CNN 440 performs the first subsampling function based on the CNN 440 determining the average input among the values provided to a given neuron (referred to as an average-pooling function).In some embodiments, the CNN 440 generates an output based on the CNN 440 providing the values to each neuron of the first subsampling layer 444, the output sometimes referred to as a convolved subsampling output.

[0086] In step 465, the CNN 440 performs a second convolution function. In some embodiments, the CNN 440 performs the second convolution function in a similar manner to how the CNN 440 performed the first convolution function described above. In some embodiments, the CNN 440 performs the second convolution function based on the CNN 440 providing the values output by the first subsampling layer 444 as input to one or more neurons (not explicitly illustrated) included in the second convolution layer 446. In some embodiments, each neuron of the second convolution layer 446 is associated with a filter, as described above. The one or more filters associated with the second convolution layer 446 may be configured to identify more complex patterns than the filter associated with the first convolution layer 442, as described above.

[0087] In some embodiments, the CNN 440 performs the second convolution function based on the CNN 440 multiplying the values provided as input to each of the one or more neurons included in the second convolutional layer 446 by the values of the filter corresponding to each of the one or more neurons. For example, the CNN 440 may multiply the values provided as input to each of the one or more neurons included in the second convolutional layer 446 by the values of the filter corresponding to each of the one or more neurons to produce a single value or an array of values as an output.

[0088] In some embodiments, the CNN 440 provides the outputs of each neuron of the second convolutional layer 446 to neurons of a downstream layer. For example, the CNN 440 may provide the outputs of each neuron of the first convolutional layer 442 to corresponding neurons of a subsampling layer. In one example, the CNN 440 provides the outputs of each neuron of the first convolutional layer 442 to corresponding neurons of the second subsampling layer 448. In some embodiments, the CNN 440 adds a bias value to the aggregates of all values provided to each neuron of the downstream layer. For example, the CNN 440 adds a bias value to the aggregates of all values provided to each neuron of the second subsampling layer 448.In such an example, the CNN 440 determines a final value to be provided to each neuron of the second subsampling layer 448 based on the aggregates of all values provided to each neuron and an activation function associated with each neuron of the second subsampling layer 448.

[0089] In step 470, the CNN 440 performs a second subsampling function. For example, the CNN 440 may perform a second subsampling function based on the CNN 440 providing the values output by the second convolutional layer 446 to corresponding neurons of the second subsampling layer 448. In some embodiments, the CNN 440 performs the second subsampling function based on the CNN 440 using an aggregation function. In one example, the CNN 440 performs the first subsampling function based on the CNN 440 determining the maximum input or an average input among the values provided to a given neuron, as described above. In some embodiments, the CNN 440 generates an output based on the CNN 440 providing the values to each neuron of the second subsampling layer 448.

[0090] In step 475, the CNN 440 provides the output of each neuron of the second subsampling layer 448 to fully connected layers 449. For example, the CNN 440 provides the output of each neuron of the second subsampling layer 448 to fully connected layers 449 to cause the fully connected layers 449 to generate an output. In some embodiments, the fully connected layers 449 are configured to generate an output associated with a prediction (sometimes referred to as a classification). The prediction may include an indication that an object included in the image provided as input to the CNN 440 includes an object, a set of objects, and / or the like. In some embodiments, the perception system 402 performs one or more operations and / or provides the data associated with the prediction to another system described herein.

[0091] Fig. 5A illustrates a diagram of an implementation of a system for generating one or more corrected future maneuver parameters in a planner. Fig. Figure 6 is a flowchart showing an example of a process for generating corrected future maneuver parameters. Fig. 5B to 5G illustrate additional implementation details for generating a correction to future maneuver parameters. Advantageously, the present subject matter may be configured to provide an end-to-end training system that enables correction of past parameters associated with vehicle movement using information obtained from the future determinations. This means that an output of the system of the present subject matter is used as input to the processes performed by the system to correct previous and / or past input parameters and train the system to generate better and / or more precise parameters for defining vehicle maneuvers.

[0092] As mentioned above, a vehicle (e.g., an autonomous vehicle) contains sensors that monitor various parameters associated with the vehicle. For example, some sensors monitor / detect changes in the vehicle's environment, while others monitor / detect various aspects related to the vehicle's operation. All information / data communicated by the sensors to the vehicle's control unit (or other processing component) is used by the planner component ("Planner") of the control unit to determine path, direction, speed, and / or other movement and / or maneuvering parameters. Planners are typically subject to errors, and to mitigate their effects (or amplify the effects of errors), the present subject matter trains an error-correcting data model to perform self-correction of the planner.During training, received vehicle maneuver values are used as input and compared with a predefined set of reference vehicle maneuver values (e.g., predefined by an "expert driver") to determine an error correction value. The error correction value and a specific future maneuver value are backpropagated into the data model as input for continuous, recursive training.

[0093] Fig. 5A illustrates an example of a system 500 implementing a self-correcting scheduler, according to some embodiments of the present subject matter. The system 500 may be integrated into a vehicle (e.g., the Fig. 1 shown vehicle 102, which in Fig. 2, etc.). The system 500 includes one or more sensors 502, a vehicle control unit 504, and a drive-by-wire (DBW) component 506. The system 500 may also include other components associated with the operation of an autonomous vehicle (as described above). The vehicle control unit 504 may include a feature extractor and rasterizer component 508, a planner component 510, a neural network component 512 (similar to the components described above in connection with Fig. 4B-4D) and a reference maneuver component 514.

[0094] The vehicle's sensors 502 monitor various parameters and / or sets of parameters associated with the vehicle. The parameters may include, but are not limited to, parameters related to the condition of the vehicle, e.g., heading, driving speed, etc. The parameters may also include, but are not limited to, parameters related to the status of the vehicle, e.g., tire pressure, oil level, transmission fluid temperature, etc. The vehicle's sensors (e.g., camera, LIDAR, SONAR, etc.) may further monitor various parameters associated with an environment surrounding the vehicle. These parameters may include, but are not limited to, parameters associated with other vehicles (e.g., speed, direction, etc.) and / or other objects (e.g., a pedestrian entering a roadway in front of the vehicle).At 501, the sensors 502 provide data for one or more measured / monitored parameters to the vehicle control unit 504.

[0095] The sensor data 501 can be provided to the feature extractor and rasterization component 508 of the vehicle control unit 504. The feature extractor and rasterization component 508 extracts one or more features from the received data, which also includes data related to one or more previous maneuvers of the vehicle. The features refer to data received from the vehicle control unit (e.g., direction, speed, etc.). The features are rasterized and then provided to the planner component 510. The reference maneuver parameters 514 are also provided to the planner component 510.

[0096] Using the planner component's neural network 512, the planner component 510 recursively trains a maneuver data model and determines one or more future maneuver parameters (which may correspond, for example, to "increase speed by 2 miles per hour," "turn 2 degrees left," etc.) that the vehicle may execute. In addition to determining future maneuver parameters, the planner component 510 compares the received input data, such as previously determined maneuver parameters, with the reference maneuver parameters 514 provided to the planner component 510. Comparing the previously determined maneuver parameters with the reference maneuver parameters 514 allows the planner component 510 to determine one or more correction values to correct or adjust the future maneuver parameters determined by the planner component 510 (e.g., corresponding to "turn 1.5 degrees left instead of 2 degrees left," etc.).In particular, during the comparison, one or more features related to the vehicle and / or the environment (e.g., route centerline proximity, course agreement, ego shifts, etc.) are extracted from the future output signal (i.e., output signal 503), and a vector containing values of these features is generated and rasterized. The planner component 510 then compares the generated / rasterized vector with a vector containing reference values 514 to determine an error correction vector, which is fed back in a loop along with a future maneuver output signal to repeat the process described above. The generated output signal 503 corresponds to the determined future maneuver parameter(s) and one or more correction values. The above process is then repeated.

[0097] The output signal 503 (i.e., the future maneuver parameter(s) and the error correction values) is fed back as input to the planner component 510 during a training phase to reduce the accumulating effect of errors. The process can be repeated continuously or at specific times.

[0098] In some embodiments, the generated output signal may be transmitted from the vehicle control unit 504 to the drive-by-wire component 506 to enable the vehicle to execute the particular maneuver taking the particular correction(s) into account. The above training process may be performed by the vehicle control unit 504 while the vehicle is operating (i.e., during driving time). Alternatively or additionally, the training process may also be performed while the vehicle is not operating.

[0099] Fig. Figure 5B illustrates additional details of the self-correcting scheduler system 500 as implemented by the control unit 504 (as shown in Fig. 5A), according to some embodiments of the present subject matter. As shown in Fig. As shown in Figure 5B, the feature extractor and rasterization component 508 may include one or more feature extractor and rasterization components 1, 2, ...N 508 (a, b, ...n). The feature extractor and rasterization components 508 (a, b, ...n) receive one or more previous and / or current maneuver parameters (also referred to as attitudes / poses) 509.

[0100] Additionally, one or more parameters related to the scene context 513 are provided to the feature extractor and rasterization components 508 (a, b, ...n). The scene context parameters 513 may correspond to parameters associated with the vehicle's surroundings. As a non-limiting example, such parameters may relate to, for example, a distance of the vehicle from the curb, the presence of a pedestrian on the sidewalk, and / or any other parameters. The scene context parameters 513 may also be extracted and rasterized.

[0101] After extraction and rasterization, parameters 509 and 513 serve as input for the planner component 510 (and its neural network 512, as in Fig. 5A). Using their neural network 512 (as shown in Fig. 5A), the planner component 510 generates one or more future attitude parameters 511 as output. The future attitude parameters 511 may indicate one or more future maneuvers that the vehicle may perform. At 515, the future attitude parameters 511 may be sampled and concatenated, and at 503, transferred or backpropagated to the feature extractor and rasterizer components 508 (a, b, ...n) to be input as past attitude parameters 509. Thus, the process may continue recursively, as mentioned above.

[0102] The planner component 510 may be configured to include one or more modules to process various inputs (e.g., past pose parameters 509, scene context parameters 513 extracted from features and rasterized) and combine them to produce an output corresponding to future poses of the ego vehicle. Note that at each processing loop or step (i.e., from inputting past poses to processing past poses, generating future poses, and inputting the generated future poses as past poses), the vehicle updates its pose with a future pose, and this pose becomes the vehicle's past pose. This feedback mechanism creates recursion, where the output is influenced by previous outputs. The feedback mechanism is advantageous because it mimics real-world decision conditions.In addition, the configuration of the system 500 (as shown in . Fig. 5A) a realistic autonomous vehicle design that can bridge a behavioral gap between training and real-world inference.

[0103] Fig. Figure 5C illustrates a training network 520 used by the autonomous vehicle system (e.g., as shown in Fig. 5A and B) during training, according to some embodiments of the present subject matter. The training network 520 is a recursive network and includes one or more feature extractor and rasterizer components 508 (a, b, ..., n), the planner component 510, and the sampling and concatenation component 515. The elements 508, 510, and 515 are engaged in processing inputs and generating outputs during the time intervals t = 0, 1, .... Due to the recursive nature of the Fig. 5C, an output of the sampling and concatenation component 515 generated at time t = 0 serves as input to the feature extractor and rasterization components 508 at time t = 1.

[0104] In particular, the processing can start with an initial parameter set that contains past layers x0 - at time t = 0. The feature extractor and rasterization components 508 and the planner component 510 generate an output x0 + and forward them to the sampling and concatenation component 515. The output of the sampling and concatenation component 515, x1 - , becomes the input to the feature extractor and rasterization components 508 at time t = 1, as in Fig. 5C. Thus, each subsequent time, the previous location parameters become the last location parameters fed to the network. During training, the recursive process can be performed a predetermined number of times, t. Alternatively or additionally, the process can be performed indefinitely and / or for as long as desired.

[0105] In addition to providing future location parameters as input, system 500 also determines an error, which is backpropagated along with the output future location parameters. In some cases, the error can only be backpropagated if there is a subset of differentiable components at each stage. The error can be determined as the change in output relative to the change in input. For example, if a function (e.g., related to feature extraction and rasterization) is defined as y = f(x), the change or derivative is determined as dy = f'(x)dx, and thus dx = dy / f'(x). Thus, if there is an error in the output dx, the error in the input can also be determined and ultimately used to update the input dx. This output is useful in distinguishing between different data components during the process performed by system 500.

[0106] In some embodiments, some features (e.g., route centerline proximity, heading agreement, ego shifts) can be easily distinguished because they are based on differentiable functions. For other features (e.g., binary features, e.g., on drivable area), an approximation can be made using differentiable, piecewise differentiable counterparts. Differentiating features allows establishing a relationship between the change in output and the change in input, so that a corresponding correction signal can be backpropagated. Feature discrimination relationships are discussed below.

[0107] The following equation relates a change in input to the output: Δy=f(x+Δx)−f(x) where x is the input, y is the output, and Δ represents changes in these respective variables. If the function f is differentiable, Δx can be easily obtained from Δy. However, if the function is not differentiable, Δx can be obtained as follows (assuming that the function f is an inverse function f -1 has): f−1(Δy)=f−1(f(x+Δx)−f(x)) f−1(Δy)=f−1(f(x+Δx))−f−1(f(x)) f−1(Δy)=x+Δx−x f−1(Δy)=Δx

[0108] Thus, the inverse function can be applied to the error signal Δy to determine a correction for Δx. In some non-limiting embodiments, the function pairs f and f -1 be selected as follows: f: pd f(µ = x) corresponding to a probability distribution with a mean equal to x and f -1: E[.] corresponding to the expected value. Thus, a raster image can be generated using f(x). The system 500 can use the generated raster image together with other inputs and, during backpropagation, generate a correction of the image from which, by applying Δx, f -1 (Δy) can be obtained.

[0109] In some embodiments, differentiation of the features may be used to ensure that inputs to the self-correcting planner are differentiable (e.g., different from a state of the ego vehicle). Otherwise, the backpropagation processes may be inaccurate and / or fail if the inputs are the same. In particular, the differentiability of the parameters may be determined with respect to previous inputs to the process of the subject matter and / or to corrections generated as a result of the process of the subject matter. Furthermore, in some exemplary embodiments, an anchor and / or a reference may be defined for the purpose of differentiability, wherein such an anchor / reference may be linked to one or more reference (e.g.,“Expert”) features associated with one or more expert maneuvers of the vehicle and / or other parameters associated with vehicle maneuvers.

[0110] Fig. 5D illustrates a process 530 for generating correction(s) to input signals according to some embodiments of the present subject matter. As in Fig. 5D, the feature extractor and rasterization component 508 and the planner component 510 of the Fig. 5A may be configured to perform process 530. Process 530 may begin with an input 531, e.g., x = (0, 0). A function f 521 may be applied to input 531 and passed to feature extractor and rasterizer component 508 to generate an image 525. Image 525 is passed to planner component 510. A reference parameter (e.g., "loss" determined based on "expert driver" parameters) may be applied to the output of planner component 510. The output of the planner component may be updated to generate an updated image 527 by feature extractor and rasterizer component 508. At this point, an inverse of function f 521, i.e., function f -1523, applied to generate a correction signal 533, i.e., Δx = (3, 1). The process 530 may be performed using an additional raster level, which may be added to the raster set to backpropagate the attitude correction signal. Therefore, the information that may be fed to the planner may be redundant and already present in other raster levels. However, the specific raster level used for propagating the correction signal may allow the attitude correction information to be transmitted during backpropagation. Alternatively or additionally, backpropagation may also provide a correction signal via other levels.

[0111] Fig. 5E illustrates an example of an experimental implementation 540 of the self-correcting scheduler system 500 (as in Fig. 5A) according to some embodiments of the present subject matter. As shown in Fig. 5E, the feature extractor and rasterization components 1, 2, ...N 508 (a, b, ...n) receive one or more previous and / or current maneuver parameters (also referred to as locations) 509. In this case, the feature extractor and rasterization component 508a is configured to process past maneuvers of the vehicle (“PastManeuvers”), the feature extractor and rasterization component 508b is configured to process a relative distance to the vehicle's destination (“RelativeDistanceToDestination”), which may include an intended destination of the vehicle (e.g., lane change), speed change (e.g., an increase in speed to 45 mph), etc., and the feature extractor and rasterization component 508n is configured to process relative lane center (“RelativeLaneCenter”), which may be associated, for example, with a position of the vehicle relative to the center lane of a road. can be.

[0112] Parameters related to the scene context 513 are passed to the feature extractor and rasterization components 508 (a, b, ...n). After extraction and rasterization, the parameters 509 and 513 are passed to the planner component 510 (and its neural network 512) to generate one or more future location parameters 511 as output. The future location parameters 511 are sampled and concatenated at 515 and passed at 503 or backpropagated to the feature extractor and rasterization components 508 (a, b, ...n) to become past location parameters 509 and begin the above process again.

[0113] In some embodiments, during feature extraction and rasterization by components 508, one or more feature vectors may be generated by components 508. Fig. 5F illustrates examples of feature vectors that may be generated by the feature extractor and rasterization components 508, according to some embodiments of the present subject matter. As in Fig. 5F, component 508a may receive input data (e.g., previous locations 509, scene context 513, etc.) 551a, and in particular, the data may represent past maneuvers. Data 551a may be structured, for example, as a table (or in any other desired form). Table 551a may contain time-based values of Δx, Δy, and Δheading determined at times t (e.g., t=1, 3, 4, 5, etc.). After processing by component 508a, table 551a may be converted into a feature vector 552a containing a time-sorted arrangement of values in table 551a.

[0114] Similarly, component 508b may receive input data (e.g., past locations 509, scene context 513, etc.) 551b, and in particular, the data may represent a relative distance to a target. Data 551b may also be structured, e.g., in the form of a table (or any other desired form). Table 551b may include time-based values of x, y, and heading as determined at time t (e.g., t=1). After processing by component 508b, table 551b may be converted into a feature vector 552b containing an array of values in table 551b.

[0115] Component 508n may also receive input data (e.g., past locations 509, scene context 513, etc.) 551n, particularly related to relative lane center data. Data 551n may also be in the form of a table (or other form). Table 551n may include time-based values of Δx, Δy, and ΔHeading as determined at time t (e.g., t=1). Following processing, table 551n may be converted into a feature vector 552n containing an array of values in table 551n.

[0116] The feature vectors 552a, 552b, 552n are then combined or concatenated into a feature vector 554 that combines values from the tables 551a, 551b, 551n. The vector 554 is then input to the planner component 510, and more specifically, to the neural network 512. The neural network includes a first input layer 541, a first hidden layer 542, a second output layer 543, and an output layer 544. It is understood that the neural network 512 can be arranged in any desired manner. The neural network 512 processes the feature vector 554 to generate an output, which is then backpropagated and becomes the input to the components 508. As part of the processing by the neural network 512, the position values of the feature vector at each time interval t are compared with reference parameter values 514 to determine error correction values or losses 560, as in Fig. 5G is shown.

[0117] Fig. 6 illustrates an example process 600 for generating a self-correction in a planner component of a vehicle according to some embodiments of the present subject matter. The process 600 may be different from that shown in Fig. 5A-G. At 602, at least a first parameter (e.g., current, past maneuver(s) of the ego vehicle) is received from the vehicle control unit 504. The first parameters are associated with one or more previous maneuvers of the vehicle. At least a second parameter associated with a maneuver target of the vehicle is also received. The parameters are provided to the feature extractor and rasterization components 508.

[0118] At 604, the planner component 510 generates (e.g., after receiving one or more feature vectors from the components 508, as in Fig. 5F) a future maneuver parameter corresponding to a future maneuver of the vehicle (e.g., entry of the ego vehicle into an adjacent lane, etc.). The planner component 510 uses its neural network 512 to generate future maneuver parameters.

[0119] At 606, at least one data model (as described above with respect to Fig.5B-G) by comparing the generated future maneuver parameter with one or more reference maneuver parameters (e.g., reference maneuver 514, as determined by an "expert driver"). As a result of the training, a corrected future maneuver parameter (e.g., future maneuver and correction 503) is generated at 608. The corrected future maneuver parameter includes a future maneuver of the vehicle and a correction to the future maneuver of the vehicle. The corrected future maneuver parameter(s) are then carried forward or backpropagated to be included in the training process as past maneuver parameters.

[0120] Further non-limiting aspects or embodiments are listed in the numbered clauses below:

[0121] Clause 1: A method comprising, using at least one processor, receiving at least a first set of parameters associated with one or more previous maneuvers of a vehicle and at least one second set of parameters associated with a maneuver target of the vehicle; generating, using the at least one processor and the first and second sets of parameters, at least one future maneuver parameter corresponding to at least one future maneuver of the vehicle; training, using the at least one processor, training at least one data model by comparing the at least one generated future maneuver parameter with one or more reference maneuver parameters;Using the at least one processor, generating at least one corrected future maneuver parameter based on the training, wherein the corrected future maneuver parameter includes at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle; wherein the at least one first parameter set includes the generated at least one corrected future maneuver parameter, the generated at least one corrected future maneuver parameter is used to correct at least one first parameter in the at least one first parameter set, and the training is performed using the corrected at least one first parameter in the at least one first parameter set.

[0122] Clause 2: The method of clause 1, further comprising executing the generated corrected at least one future maneuver of the vehicle using the at least one processor.

[0123] Clause 3: The method of any preceding clause, wherein receiving further comprises continuously receiving, using the at least one processor, the generated at least one corrected future maneuver parameter following each execution to correct one or more previously received first parameters in the at least one first parameter set.

[0124] Clause 4: The method of clause 3, further comprising receiving at least one third parameter associated with a state of the vehicle, wherein the training further comprises training, using the at least one processor, the at least one data model using the future maneuver parameter, the at least one third parameter, and the one or more reference maneuver parameters.

[0125] Clause 5: The method of Clause 3, further comprising repeating, using the at least one processor, the receiving, comparing, training and generating the at least one corrected future maneuver parameter.

[0126] Clause 6: The method of Clause 5, wherein the first and / or second set of parameters correspond to at least one feature associated with at least one maneuver of the vehicle.

[0127] Clause 7: The method of Clause 6, further comprising screening the at least one feature.

[0128] Clause 8: The method of Clause 7, wherein training further comprises training, using the at least one processor, the at least one data model using the at least one rasterized feature.

[0129] Clause 9: The method of Clause 2, wherein said executing further comprises executing, using said at least one processor, said at least one corrected future maneuver of said vehicle while said vehicle is in operation.

[0130] Clause 10: A method according to any one of the preceding clauses, wherein the at least one first set of parameters and the at least one second set of parameters include at least one of the following: a speed, a position, an acceleration, a direction of movement, and any combination thereof, of the vehicle.

[0131] Clause 11: The method of any preceding clause, wherein at least one received first parameter in the at least one first parameter set is differentiable from at least one of the following: at least one previously received first parameter in the at least one first parameter set, and at least one corrected first parameter.

[0132] Clause 12: The method of any preceding clause, wherein at least one received first parameter in the at least one first parameter set is differentiable from at least one first anchor parameter in the at least one first parameter set, wherein the at least one first anchor parameter is associated with at least one of the following: at least one reference feature associated with at least one maneuver of the vehicle, and at least one reference maneuver parameter in the one or more reference maneuver parameters.

[0133] Clause 13: A system comprising at least one processor and at least one non-transitory storage medium storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations according to any of the preceding clauses 1 to 12.

[0134] Clause 14: At least one non-volatile storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations according to any of the preceding clauses 1 to 12.

[0135] In the foregoing description, aspects and embodiments of the present disclosure have been described with reference to numerous specific details that may vary from implementation to implementation. Accordingly, the description and drawings are to be considered illustrative rather than restrictive. The sole and exclusive indicator of the scope of the invention, and of what is intended by applicants to be the scope of the invention, is the literal and equivalent scope of the set of claims emerging from this application, in the specific form in which such claims arise, including subsequent amendments. All definitions of terms contained in such claims expressly set forth herein apply to the meaning of the terms used in the claims.In addition, when the term "further comprising" is used in the foregoing description or in the following claims, what follows that phrase may be an additional step or entity, or a sub-step / sub-entity of a previously stated step or entity.

[0136] The following aspects are also part of the invention: 1. A method comprising: receiving, using at least one processor, at least a first set of parameters associated with one or more previous maneuvers of a vehicle and at least a second set of parameters associated with a maneuver target of the vehicle; using the at least one processor and using the first and second parameter sets, generating at least one future maneuver parameter corresponding to at least one future maneuver of the vehicle; using the at least one processor, training at least one data model by comparing the at least one generated future maneuver parameter with one or more reference maneuver parameters; using the at least one processor and based on the training, generating at least one corrected future maneuver parameter, wherein the corrected future maneuver parameter includes at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle; wherein the at least one first parameter set contains the generated at least one corrected future maneuver parameter, the generated at least one corrected future maneuver parameter is used to correct at least one first parameter in the at least one first parameter set, and the training is performed using the corrected at least one first parameter in the at least one first parameter set. 2. The method of aspect 1, further comprising executing the generated corrected at least one future maneuver of the vehicle using the at least one processor. 3. The method of any preceding aspect, wherein receiving further comprises continuously receiving, using the at least one processor, the generated at least one corrected future maneuver parameter following each execution to correct one or more previously received first parameters in the at least one first parameter set. 4. The method of aspect 3, further comprising receiving at least one third parameter associated with a state of the vehicle; wherein the training further comprises training, using the at least one processor, the at least one data model using the future maneuver parameter, the at least one third parameter, and the one or more reference maneuver parameters. 5. The method according to aspect 3, further comprising: repeating the receiving, comparing, training and generating of the at least one corrected future maneuver parameter using the at least one processor. 6. The method of aspect 5, wherein the first and / or second parameter set corresponds to at least one feature associated with at least one maneuver of the vehicle. 7. The method of aspect 6, further comprising rasterizing the at least one feature. 8. The method of aspect 7, wherein training further comprises training the at least one data model using the at least one rasterized feature using the at least one processor. 9. The method of aspect 2, wherein said performing further comprises: executing the at least one corrected future maneuver of the vehicle using the at least one processor while the vehicle is in operation. 10. The method according to any one of the preceding aspects, wherein the at least one first parameter set and the at least one second parameter set include at least one of the following: a speed, a position, an acceleration, a direction of movement, and any combination thereof, of the vehicle. 11. The method according to any one of the preceding aspects, wherein at least one received first parameter in the at least one first parameter set is differentiable from at least one of the following: at least one previously received first parameter in the at least one first parameter set, and at least one corrected first parameter. 12. The method of any one of the preceding aspects, wherein at least one received first parameter in the at least one first parameter set is differentiable from at least one first anchor parameter in the at least one first parameter set, wherein the at least one first anchor parameter is associated with at least one of the following: at least one reference feature associated with at least one maneuver of the vehicle, and at least one reference maneuver parameter in the one or more reference maneuver parameters. 13. System comprising: at least one processor and at least one non-volatile storage medium storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations according to any one of the preceding aspects 1 to 12. 14. At least one non-volatile storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations according to any one of the preceding aspects 1 to 12.

Claims

[1] Method (600) comprising: Receiving (602), using at least one processor, a first set of parameters associated with one or more previous maneuvers of a vehicle and a second set of parameters associated with a maneuver goal of the vehicle; Generating (604), using the at least one processor and using the first parameter set and the second parameter set, at least one future maneuver parameter corresponding to one or more future maneuvers of the vehicle; Training (606), using the at least one processor, at least one data model by comparing the at least one future maneuver parameter with one or more reference maneuver parameters identified as expert driver maneuver parameters; Generating (608), using the at least one processor and based on the training, at least one corrected future maneuver parameter, wherein the corrected future maneuver parameter comprises at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle based on error correction values indicating a difference between the at least one future maneuver parameter and the one or more reference maneuver parameters; Generating, using the at least one processor, a signal for controlling the vehicle to execute the at least one corrected future maneuver using the at least one corrected future maneuver parameter; and Using the signal, causing the vehicle to operate at least in part based on the corrected future maneuver parameter. [2] The method (600) of claim 1, further comprising executing, using the at least one processor, the corrected at least one future maneuver of the vehicle. [3] The method (600) of claim 1, wherein receiving (602) further comprises: continuously receiving, using the at least one processor, the at least one corrected future maneuver parameter following each execution to correct one or more previously received first parameters in the first parameter set. [4] The method (600) of claim 3, further comprising receiving at least one third parameter associated with a state of the vehicle; wherein training further comprises: training, using the at least one processor, the at least one data model using the future maneuver parameter, the at least one third parameter, and the one or more reference maneuver parameters. [5] The method (600) of claim 3, further comprising: Repeating, using the at least one processor, training the at least one data model before generating the at least one corrected future maneuver parameter. [6] The method (600) of claim 5, wherein the first parameter set and the second parameter set correspond to at least one feature associated with at least one maneuver of the vehicle. [7] The method (600) of claim 6, further comprising rasterizing the at least one feature as at least one rasterized feature. [8] The method (600) of claim 7, wherein the training further comprises: Training, using the at least one processor, the at least one data model using the at least one rasterized feature. [9] The method (600) of claim 2, wherein said performing further comprises: Executing, using the at least one processor, the at least one corrected future maneuver of the vehicle while the vehicle is in operation. [10] The method (600) of claim 1, wherein the first parameter set and the second parameter set comprise a speed and / or a position and / or an acceleration and / or a direction of movement, and / or any combination thereof, of the vehicle. [11] The method (600) of claim 1, wherein at least one received first parameter in the first parameter set is differentiable from at least one of the following: at least one previously received first parameter in the first parameter set and at least one corrected first parameter. [12] The method (600) of claim 1, wherein expert driver maneuver parameters are associated with at least one of the following: at least one reference feature associated with at least one maneuver of the vehicle, and at least one reference maneuver parameter in the one or more reference maneuver parameters. [13] System comprising: at least one processor and at least one non-volatile storage medium storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations that include: Receiving (602) a first set of parameters associated with one or more previous maneuvers of a vehicle and a second set of parameters associated with a maneuver target of the vehicle; Generating (604), using the first parameter set and the second parameter set, at least one future maneuver parameter corresponding to one or more future maneuvers of the vehicle; Training (606) at least one data model by comparing the at least one future maneuver parameter with one or more reference maneuver parameters identified as expert driver maneuver parameters; Generating (608), based on the training, at least one corrected future maneuver parameter, wherein the corrected future maneuver parameter comprises at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle based on error correction values indicating a difference between the at least one future maneuver parameter and the one or more reference maneuver parameters; Generating a signal for controlling the vehicle to execute the at least one corrected future maneuver using the at least one corrected future maneuver parameter; and Using the signal, causing the vehicle to operate at least in part based on the corrected future maneuver parameter. [14] The system of claim 13, wherein the steps further comprise executing the corrected at least one future maneuver of the vehicle. [15] The system of claim 13, wherein receiving (602) further comprises continuously receiving the at least one corrected future maneuver parameter following each execution to correct one or more previously received first parameters in the first parameter set. [16] The system of claim 15, wherein the steps further comprise receiving at least one third parameter associated with a condition of the vehicle; wherein the training further comprises training the at least one data model using the future maneuver parameter, the at least one third parameter, and the one or more reference maneuver parameters. [17] The system of claim 13, wherein the first set of parameters and the second set of parameters correspond to at least one feature associated with at least one maneuver of the vehicle. [18] The system of claim 17, wherein the steps further comprise rasterizing the at least one feature as at least one rasterized feature. [19] The system of claim 18, wherein training further comprises training the at least one data model using the at least one rasterized feature. [20] The system of claim 14, wherein said executing further comprises: Executing at least one corrected future maneuver of the vehicle during operation of the vehicle. [21] The system of claim 13, wherein the first set of parameters and the second set of parameters comprise a speed and / or a position and / or an acceleration and / or a direction of movement and / or any combination thereof of the vehicle. [22] The system of claim 13, wherein at least one received first parameter in the first parameter set is differentiable from at least one of the following: at least one previously received first parameter in the first parameter set and at least one corrected first parameter. [23] The system of claim 13, wherein expert driver maneuver parameters are associated with at least one of the following: at least one reference feature associated with at least one maneuver of the vehicle, and at least one reference maneuver parameter in the one or more reference maneuver parameters. [24] Non-volatile storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations that include: Receiving (602) a first set of parameters associated with one or more previous maneuvers of a vehicle and a second set of parameters associated with a maneuver target of the vehicle; Generating (604), using the first parameter set and the second parameter set, at least one future maneuver parameter corresponding to one or more future maneuvers of the vehicle; Training (606) at least one data model by comparing the at least one future maneuver parameter with one or more reference maneuver parameters identified as expert driver maneuver parameters; Generating (608), based on the training, at least one corrected future maneuver parameter, wherein the corrected future maneuver parameter comprises at least one future maneuver of the vehicle and at least one correction of the at least one future maneuver of the vehicle based on error correction values indicating a difference between the at least one future maneuver parameter and the one or more reference maneuver parameters; Generating a signal for controlling the vehicle to execute the at least one corrected future maneuver using the at least one corrected future maneuver parameter; and Using the signal, causing the vehicle to operate at least in part based on the corrected future maneuver parameter. [25] The non-volatile storage medium of claim 24, wherein the steps further comprise executing the corrected at least one future maneuver of the vehicle. [26] The non-volatile storage medium of claim 25, wherein receiving further comprises continuously receiving the at least one corrected future maneuver parameter following each execution to correct one or more previously received first parameters in the first parameter set.

Images