ROAD PLANNING FUSION FOR A VEHICLE

Abstract

A method for providing route planning guidance by resolving multiple behavioral predictions associated with operating a vehicle, wherein the method comprises: Installing a vehicle system in a vehicle, wherein the vehicle system provides route planning guidance based on the use of training data and fused hypotheses and / or decisions associated with the training data; Determine, by a processor, the position of the vehicle on a map containing a road network; Determine, via the processor, whether one or more agents are present within a given area of ​​the vehicle; Select, by the processor, an output trajectory for traversing the road network based on the vehicle's position on the map and the existence of one or more agents; and Control, by the processor, of vehicle operation using the output trajectory, wherein the training data includes data based on one or more predictive models used to predict a future movement associated with one or more agents, where a Canonical Correlation Analysis (CCA) algorithm is used to merge the hypotheses and / or decisions associated with the training data, where the CCA provides a mapping between one or more prediction models and the behavior of one or more drivers.

Description

INTRODUCTION

[0001] The revelation refers to route planning and, in particular, to the fusion of several trajectories to guide a vehicle through a road network.

[0002] Autonomous vehicles are able to operate and navigate without human intervention.

[0003] Autonomous vehicles, as well as some non-autonomous vehicles, use sensors such as cameras, radar, LiDAR, global positioning systems, and computer vision to perceive the vehicle's surroundings. Advanced computer control systems interpret the sensory input information to determine a vehicle's location, suitable navigation paths, and identify obstacles and relevant signage. Some autonomous vehicles update map information in real time to maintain their position even when conditions change or the vehicle enters an unfamiliar environment. Both autonomous and non-autonomous vehicles are increasingly communicating with remote computer systems and with each other via V2X communication – Vehicle-to-Everything (V2X), Vehicle-to-Vehicle (V2V), and Vehicle-to-Infrastructure (V2I).

[0004] As autonomous and semi-autonomous vehicles become increasingly common, it is crucial that each vehicle on a road network is accurately located and its position (i.e., its route) is precisely determined. Therefore, it is desirable to further improve route planning as a vehicle travels through the road network.

[0005] US 2019 / 0072966A1 discloses a predictive system and method for trajectory planning for autonomous vehicles. It comprises a training phase, in which a trajectory prediction module is trained using perception and context data of human driving behavior, and an operational phase. In this operational phase, relevant vehicle and environment features are extracted from perception data of the host vehicle, a proposed trajectory for the host vehicle is generated, and predictions are made about the trajectories of neighboring vehicles. Subsequently, the system checks for conflicts and adjusts the host vehicle's trajectory until no conflicts remain.

[0006] US 2018 / 0322642A1 discloses a system for predicting multi-agent movements. This system applies a Radon Cumulative Distribution Transform (Radon-CDT) to pairs of signature formations representing agent movements. For these pairs, components are identified using canonical correlation analysis (CCA) to learn a relationship between the signature formations. Based on this learned relationship, a counter-signature formation can be predicted for a new dataset. Subsequently, device control parameters can be adjusted based on the predicted counter-signature formation. DESCRIPTION

[0007] The object of the invention is to improve the aforementioned problems. This object is achieved by the subject matter according to claim 1. Further developments are described in the dependent claims.

[0008] In an exemplary embodiment, a method for providing route planning assistance by resolving multiple behavioral predictions associated with operating a vehicle is presented. The method includes the integration of a vehicle system into a vehicle, wherein the vehicle system provides route planning guidance based on training data using fused hypotheses and / or decisions associated with the training data. The method further includes determining the vehicle's position on a map with a road network using a processor. The method further includes determining, by the processor, whether one or more agents are present within a predefined area of ​​the vehicle.The method further includes the processor selecting an output trajectory to traverse the road network based on the vehicle's position on the map and the existence of one or more agents. The method further includes the processor controlling vehicle operation based on the output trajectory.

[0009] In addition to one or more of the features described here, one or more aspects of the described method recognize that the training data contains data based on one or more predictive models used to forecast future movement in conjunction with one or more agents. Another aspect of the method uses a Canonical Correlation Analysis (CCA) algorithm to merge the hypotheses and / or decisions associated with the training data. Another aspect of the method is that the CCA provides a mapping between one or more predictive models and the behavior of one or more drivers. Another aspect of the method is that the CCA utilizes pairwise interactions between the predictors. A further aspect of the method is that the movement information from one or more agents includes velocity, heading, and location information.Another aspect of the procedure is that one or more agents are mobile or stationary agents.

[0010] In a further exemplary embodiment, a system for providing route planning assistance by resolving multiple behavioral predictions related to vehicle operation is presented here. The system comprises a vehicle with memory, a memory-coupled processor, a hypothesis resolver, a decision resolver, a trajectory planner, and a controller. The vehicle-associated processor is operable to operate a vehicle system within the vehicle, wherein the vehicle system provides route planning guidance based on the use of training data and fused hypotheses and / or decisions associated with the training data. The processor is also capable of determining the vehicle's location on a map of a road network.The processor is also able to determine whether one or more agents are present within a predefined area of ​​the vehicle. Furthermore, the processor is able to select an output trajectory to traverse the road network based on the vehicle's position on the map and the presence of one or more agents. The processor is also able to control the vehicle's operation based on this output trajectory.

[0011] In a further exemplary embodiment, a computer-readable storage medium for carrying out a method for providing route planning assistance by resolving multiple behavioral predictions associated with operating a vehicle is disclosed hereafter. The computer-readable storage medium comprises the installation of a vehicle system in a vehicle, wherein the vehicle system provides route planning guidance based on training data by using and fusing hypotheses and / or decisions associated with the training data. The computer-readable storage medium also includes determining the vehicle's location on a map containing a road network. The computer-readable storage medium further includes determining whether one or more agents are present within a predetermined area of ​​the vehicle.The computer-readable storage medium further includes the selection of an output trajectory to traverse the road network based on the vehicle's position on the map and the existence of one or more agents. The computer-readable storage medium also includes controlling vehicle operation based on the output trajectory.

[0012] The aforementioned features and advantages, as well as further features and advantages of the disclosure, will become apparent from the following detailed description in conjunction with the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0013] Further features, advantages and details appear only as examples in the following detailed description, which refers to the figures in which: Fig. 1 is a computing environment according to one or more embodiments; Fig. Figure 2 is a block diagram showing an example of a processing system for the practical application of the teaching given here; Fig. Figure 3 shows a schematic representation of an exemplary vehicle system according to one or more embodiments; Fig. 4 is a block diagram of vehicle components according to one or more embodiments; and Fig. Figure 5 shows a flowchart of a route planning method in which several behavioral predictions associated with operating a vehicle are resolved according to one or more embodiments. DETAILED PRESENTATION

[0014] The following description is merely exemplary and is not intended to limit the present disclosure, its application, or its use. It should be understood that throughout the figures, corresponding reference numerals indicate identical or corresponding parts and features. The term "module" as used here refers to processing circuits that may include an application-specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or grouped), memory that executes one or more software or firmware programs, a combined logic circuit, and / or other suitable components that provide the described functionality.

[0015] According to an exemplary embodiment, Fig. 1. A computing environment 50 connected to a route planning system, by resolving multiple behavioral predictions associated with the operation of a vehicle according to one or more embodiments. The computing environment 50 consists, as shown, of one or more computing devices, e.g., a server / cloud 54B and / or a vehicle on-board computer system 54N, installed in each of several autonomous or non-autonomous vehicles connected via the network 150. The one or more computing devices can communicate with each other via the network 150.

[0016] Network 150 can be, for example, a cellular network, a local area network (LAN), a wide area network (WAN) such as the internet and Wi-Fi, a dedicated short-range communication network (e.g., V2V (vehicle-to-vehicle), V2X (vehicle-to-everything), V2I (vehicle-to-infrastructure), and V2P (vehicle-to-pedestrian)), or any combination thereof, and can include wired, wireless, fiber optic, or any other connection. Network 150 can be any combination of connections and protocols that support communication between Server / Cloud 54B or the multiple vehicle on-board computer systems 54N.

[0017] When a cloud is used instead of a server, Server / Cloud 54B can serve as a remote computing resource. Server / Cloud 54B can be implemented as a service delivery model to enable convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management overhead or interaction with a service provider.

[0018] Fig. Figure 2 shows an example of a processing system 200 for implementing the teaching. The processing system 200 can form at least part of one or more computing devices, such as a server / cloud 54B, and / or the vehicle on-board computer system 54N. The processing system 200 can comprise one or more central processing units (processors) 201a, 201b, 201c, etc. (collectively or generally referred to as processor(s) 201). The processors 201 are connected to the system memory 214 and various other components via a system bus 213. The read-only memory (ROM) 202 is connected to the system bus 213 and can contain a basic input / output system (BIOS) that controls certain basic functions of the processing system 200.

[0019] Fig. Figure 2 further shows an input / output (I / O) adapter 207 and a network adapter 206 connected to the system bus 213. The I / O adapter 207 can be a small SCSI (Computer System Interface) adapter that communicates with a hard disk 203 and / or other storage drives 205 or other similar components. The I / O adapter 207, hard disk 203, and other storage drives 205 are collectively referred to here as mass storage 204. The operating system 220 for execution on the processing system 200 can be stored in the mass storage 204. The network adapter 206 connects the system bus 213 to an external network 216, which can be network 150, enabling the processing system 200 to communicate with other such systems. A screen (e.g.,A display monitor (215) can be connected to the system bus 213 via the display adapter 212. The system bus may contain a graphics adapter to improve the performance of graphics-intensive applications and a video controller. In one configuration, network adapter 206, I / O adapter 207, and display adapter 212 can be connected to one or more I / O buses that are connected to the system bus 213 via an intermediate bus bridge (not shown). Suitable I / O buses for connecting peripheral devices such as disk controllers, network adapters, and graphics cards typically include common protocols such as Peripheral Component Interconnect (PCI). Additional input / output devices are shown connected to the system bus 213 via the operator panel adapter 208 and the display adapter 212. A microphone 209, the steering wheel / dashboard controls 210 and the speaker 211 can all be connected to the system bus 213 via the user interface adapter 208, which...B. may contain a Super I / O chip that integrates multiple device adapters into a single integrated circuit.

[0020] The Processing System 200 can additionally include a Graphics Processing Unit 230. The Graphics Processing Unit 230 is a specialized electronic circuit for manipulating and modifying memory to accelerate the creation of images in a frame buffer intended for display output. In general, the Graphics Processing Unit 230 is highly efficient at manipulating computer graphics and image processing and has a highly parallel architecture, making it more effective than general-purpose CPUs for algorithms that process large blocks of data in parallel.

[0021] The processing system contains 200, as in Fig. 2 configured, the processing capability in the form of processors 201, the storage capability including system memory 214 and mass storage 204, input devices such as microphone 209 and steering wheel / dashboard controls 210, and the output capability including speakers 211 and display monitor 215. In one embodiment, a portion of the system memory 214 and the mass storage 204 together store an operating system to manage the functions of the various in Fig. to coordinate the 2 components shown.

[0022] Fig. Figure 3 shows the components of a system 300 belonging to autonomous or non-autonomous vehicles with the vehicle on-board computer system 54N, in one or more embodiments. The vehicle 310 typically consists of a chassis 312, a body 314, the front wheels 316, and the rear wheels 318. The body 314 can be mounted on the chassis 312 and can largely enclose components of the vehicle 310. The body 314 and the chassis 312 can together form a frame. The wheels 316 and 318 are each rotatably coupled to the chassis 312 near a corner of the body 314.

[0023] The route planning system, which resolves multiple behavioral predictions associated with operating a vehicle, can be integrated into vehicle 310. Vehicle 310 is represented as a passenger car, but it should be noted that vehicle 310 could be a different type of vehicle, such as a motorcycle, truck, sport utility vehicle (SUV), recreational vehicle (RV), ship, aircraft, etc.

[0024] The 310 vehicle can operate at various levels of vehicle automation, such as Level 4 or Level 5. Operation in a Level 4 system indicates a "high level of automation," meaning that all aspects of the dynamic driving task are performed by an automated driving system, depending on the driving mode, even if a human driver does not respond appropriately to a request for intervention. Operation in a Level 5 system signifies "full automation" and refers to the full-time performance of an automated driving system for all aspects of the dynamic driving task under all road and environmental conditions that a human driver could handle.

[0025] The vehicle 310 can also include a drive system 320, a transmission system 322, a steering system 324, a braking system 326, a sensor system 328, an actuator system 330, at least one data storage device 332, at least one control unit 334, and a communication system 336. The drive system 320 can be an internal combustion engine, an electric machine such as a traction motor, and / or a fuel cell drive system. The transmission system 322 can be configured to transmit power from the drive system 320 to the vehicle wheels 316 and 318 according to selectable speed ratios. The transmission system 322 can include a stepped automatic transmission, a continuously variable transmission, or other suitable transmissions.

[0026] The braking system 326 can be configured to supply braking torque to the vehicle wheels 316 and 318. The braking system 326 can use friction brakes, wire brakes, a regenerative braking system such as an electric motor, and / or other suitable braking systems. The steering system 324 influences the position of the vehicle wheels 316 and 318.

[0027] The sensor system 328 may include one or more sensor devices 340a-340n that detect the observable conditions of the external and / or internal environment of the vehicle 310. The sensors 340a-340n may include, among others, speed, radar, LiDAR, global positioning systems, optical cameras, thermal imaging cameras, ultrasonic sensors, inertial measurement devices, and / or other sensors. The actuator system 330 includes one or more actuator devices 342a-342n that control, but are not limited to, one or more vehicle features, such as the drive system 320, the transmission system 322, the steering system 324, and the braking system 326. In various embodiments, the vehicle features may also include, but are not limited to, internal and / or external features of the vehicle, such as doors, trunk, and cabin features like air conditioning, music, lighting, etc. (not numbered).

[0028] The 328 sensor system can acquire a variety of vehicle measurements and / or other information. The 340a-340n sensors can generate measurements representing the position, speed, and / or acceleration of the vehicle 310. The 340a-340n sensors can also generate measurements representing lateral acceleration, yaw rate, etc. The 340a-340n sensors can utilize a variety of different sensors and sensor technologies, including those that use wheel speed, vehicle speed, accelerometer position, gear position, shift lever position, accelerometers, engine speed, engine power, throttle position, and inertial measurement unit (IMU) output, etc.The 340a-340n sensors can be used to determine the vehicle speed relative to the ground by directing radar, laser and / or other signals at known stationary objects and analyzing the reflected signals, or by using feedback from a navigation unit that has GPS and / or telematics capabilities via a telematics module, which can be used to monitor the position, movement, status and behavior of the vehicle.

[0029] The Communication System 336 can be configured to wirelessly transmit information to and from other Units 348, such as, but not limited to, other vehicles (“V2V” communication), infrastructure (“V2I” communication), remote systems, and / or personal devices. The Communication System 336 can be a wireless communication system configured for communication over a wireless local area network (WLAN) using the IEEE 802.11 standard or through cellular data communication. However, additional or alternative communication methods, such as a dedicated channel for short-range communication (DSRC), are also considered within the scope of this instruction.DSRC channels refer to one-way or two-way short- to medium-range wireless communication channels specifically designed for automotive applications and a corresponding set of protocols and standards.

[0030] The data storage device 332 can store data for the automatic control of the autonomous vehicle 310. The data storage device 332 can also store defined maps of the navigable environment. The defined maps can be obtained from a remote system. For example, the defined maps can be compiled by the remote system and transmitted to the autonomous vehicle 310 (wirelessly and / or via cable) and stored in the data storage device 332. Route information can also be stored in the data storage device 332; that is, a series of road segments (geographically linked to one or more of the defined maps) that together define a route a user can take to travel from a starting point (e.g., the user's current location) to a destination. The data storage device 332 can be part of the controller 334, separate from the controller 334, or part of the controller 334 and part of a separate system.

[0031] The controller 334 can include at least one processor 344 and a computer-readable storage device or medium 346. The processor 344 can be any custom or commercially available processor, central processing unit (CPU), graphics processing unit (GPU), auxiliary processor among several processors connected to the controller 334, semiconductor-based microprocessor (in the form of a microchip or chipset), a microprocessor, any combination thereof, or generally any instruction-executing device.

[0032] The instructions can contain one or more separate programs, each containing an ordered list of executable instructions for implementing logical functions. When executed by the processor 344, the instructions receive and process signals from the sensor system 328, perform logic, calculations, methods, and / or algorithms for the automatic control of the components of the autonomous vehicle 310, and generate control signals to the actuator system 330 to automatically control the components of the autonomous vehicle 310 based on the logic, calculations, methods, and / or algorithms.

[0033] The 310 vehicle may also include a safety control module (not shown), an infotainment / entertainment control module (not shown), a telematics module (not shown), a GPS module (not shown) (GLONASS may also be used), etc. The safety control module may provide various functions for crash or collision detection, avoidance, and / or mitigation. For example, the safety control module may provide and / or implement collision warnings, lane departure warnings, autonomous or semi-autonomous braking, autonomous or semi-autonomous steering, airbag deployment, active crumple zones, seatbelt pretensioners or load limiters, and automatic notification of emergency responders in the event of an accident, etc.

[0034] The infotainment / entertainment control module can provide the occupants of the 310 vehicle with a combination of information and entertainment. This information and entertainment can include, for example, music, websites, movies, television programs, video games, and / or other content.

[0035] The telematics module can utilize wireless voice and / or data communication via a wireless carrier system (not shown) and a wireless network (not shown) to enable the vehicle 310 to offer a range of different services, including those related to navigation, telephony, emergency assistance, diagnostics, infotainment, and more. The telematics module can also utilize cellular communication according to the GSM, W-CDMA, or CDMA standards and wireless communication according to one or more protocols implemented according to the 3G or 4G standards, or other wireless protocols such as IEEE 802.11, WiMAX, or Bluetooth. When used for packet-switched data communication such as TCP / IP, the telematics module can be configured with a static IP address or set up to automatically obtain a dynamically assigned IP address from another device on the network, e.g., a network device.from a router or from a network address server (e.g. a DHCP server).

[0036] The GPS module can receive radio signals from multiple GPS satellites (not shown). From these received radio signals, the GPS module can determine a vehicle position, which can be used to provide navigation and other location-based services. Navigation information can be displayed on a screen in the vehicle 310 (e.g., Display 215) or verbally, as is the case with turn-by-turn navigation. Navigation services can be provided via a dedicated onboard navigation module (which may be part of the GPS module), or some or all navigation services can be provided via the telematics module. This allows the vehicle 310's position information to be sent to a remote location to provide the vehicle 310 with navigation maps, map annotations (points of interest, restaurants, etc.), route calculations, and similar information.

[0037] Fig. 4, with further reference to Fig. Figures 1-3 show a behavior path planning resolution system 400, which is connected to any of a variety of autonomous or non-autonomous vehicles that incorporate the vehicle onboard computer system 54N. The behavior path planning resolution system 400 can comprise several components (e.g., a controller 410, which may be controller 334, a hypothesis resolver 430, a decision resolver 415, and a trajectory planner 405). The behavior-based path planning resolution system 400 can provide path planning assistance for a vehicle.

[0038] The behavior-based path planning resolution system 400 can initially be trained to make path planning decisions based on data reflecting the decisions and actions of drivers operating a vehicle on a road network with respect to a specific driving situation (e.g., operating at a road fork, three-way stop, intersection, highway on-ramp, highway off-ramp, turnaround, etc.) and / or a specific location or location type (e.g., highway, two-lane road, left-turn lane, urban area, etc.). Driver actions can occur in response to interaction with other mobile or stationary objects, traffic signs, traffic lights, road geometry, work zones, traffic, etc. (i.e., behavior).Once the Behavioral Path Planning Resolution System 400 has received a quantity of training data above a predefined threshold, it can be integrated into the vehicle 310. The Behavioral Path Planning Resolution System 400 integrated into the vehicle 310 can then be used to make decisions about vehicle operation (i.e., steering, braking, accelerating, etc.) based on the resolution of multiple hypotheses and / or decisions while the vehicle 310 is operating autonomously or semi-autonomously.

[0039] The Behavioral Path Planning Resolution System 400 can use a variety of models for movement behavior (i.e., predictive models (e.g., kinematics, tree regression, GMM-HMM, etc.)) using training or live data to develop multiple hypotheses (e.g., 435 and 440), where each hypothesis can be a path prediction for an agent within a given distance from the vehicle 310. Each path prediction includes a velocity, heading, and position prediction, as well as a calculated trajectory for each agent.

[0040] Any hypothesis can be entered into the Hypothesis Resolver 430. Each hypothesis can be a spatial trajectory of the agent moving from one location to another on a map. The Hypothesis Resolver 430 can select and output the best hypothesis (the selection is based on the accuracy of each hypothesis prediction for a certain period in the past) from the multitude of hypotheses entered into the Hypothesis Resolver 430 (e.g., Hypothesis 435 and Hypothesis 440). The best hypothesis can be a fusion of several hypotheses. Accordingly, the Hypothesis Resolver 430 can output a predicted path based on the fusion of the predicted paths associated with the multiple hypotheses.

[0041] Hypothesis fusion can be performed considering N, where N is the total number of data points, with each data point representing a single path k. As mentioned earlier, each hypothesis of the multiple hypotheses represents a different model and a different true observation. The multiple hypotheses can be represented as a set of vectors Si: (Xji,Yji)∈Si where S i ∈ R N Accordingly, every hypothesis S i is a 2D trajectory, described by X, Y on the map. Accordingly, the fusion technique described here can find a highly probable path for a given set of individual paths, S, based on the many hypotheses.

[0042] The Hypothesis Resolver 430 can use a Canonical Correlation Analysis (CCA) algorithm to select the best-predicted path. The equation for the CCA might be the following, based, for example, on two predictors (Hypothesis X predicts the path and Hypothesis Y predicts the path) and an observed path, i.e., a mapped trajectory (Z): CCA=argmaxs{[[UTXTYV]√UTXXTU√VTYYTV] +[UTXTZZW√UTXXTU√WTZZTW+VTYTZW√VTYYTV√WTZZTW] where parameters U, V, and W are the upper k components of the CCA, which can be thought of as weights learned within the CCA model. Parameters X and Y are predicted paths for two different hypotheses. Parameter Z is the confirmed path.

[0043] A first summand represents a pairwise correlation of the two predictors, and a second summand represents predictor-observation correlations. The result of the CCA provides a complete joint embedding that can be used to integrate all path predictions and observations into a single, fused path based on human decisions and actions derived from the training data. The joint embedding is a space in which the trajectories are mapped into a fused (shared) embedding, which can be used to determine the mapping between the fused trajectories and the observed trajectories.

[0044] The output of hypothesis resolver 430 (i.e., the best-predicted future path for a given agent) is used to generate multiple decisions (e.g., Decision 1(420) and Decision M(425)). Each generated decision can consider the best-predicted future path for every agent within the specified range of vehicle 310. Each decision can compute an output trajectory that can be used to plan a path for vehicle 310. Each decision can provide a valid / realistic (human) output trajectory, even if an average output trajectory would be invalid (e.g., selecting a left fork or a right fork when an average indicates selecting a path between the left and right forks).

[0045] Each decision can be inputted into a Decision Resolver 415. The Decision Resolver 415 can select the best decision, that is, a decision that most closely mimics human behavior with respect to a given set of events. The Decision Resolver 415's selection of the best decision can also utilize the CCA to merge multiple decisions in a similar manner to the fusion performed by the Hypothesis Resolver 430. The Decision Resolver 415 can input the best decision / fused decision into the Trajectory Planner 405. The Trajectory Planner 405 can generate a path / trajectory for the vehicle 310 to traverse a road network using the output trajectory associated with the provided decision. The Trajectory Planner 405 can input the path / trajectory into the Controller 410.The control unit 410 can use the received path / trajectory to make decisions about vehicle operation, causing the vehicle 310 to travel through the road network.

[0046] Fig. Figure 5 shows a flowchart of a Method 500 for implementing a procedure for providing route planning assistance by resolving multiple behavioral predictions associated with operating a vehicle according to one or more embodiments. In Block 505, a system (e.g., the Behavioral Route Planning Resolution System 400) can receive data from each of a plurality of agents during a training phase. The received data can include speed, heading, and location information. In Block 510, the system can generate training data from the received data. In Block 515, the system can combine hypotheses and / or decisions generated using the training data.

[0047] In Block 520, a vehicle system or part thereof can be trained using the generated training data and the fused hypotheses or decisions. The training can be based on simulations of agents (mobile and stationary) interacting on or along a road network. The simulations can be based on random permutations of means, vehicles, and road types. In Block 525, the trained vehicle system can be installed in a vehicle, for example, the vehicle onboard computer system 54N. In Block 530, the vehicle onboard computer system 54N can determine the vehicle's location on a map of the road network during operation (i.e., while driving through the road network). In Block 535, the vehicle onboard computer system 54N can determine whether one or more agents (mobile or stationary) are present within a predefined area of ​​the vehicle.

[0048] In block 540, the vehicle onboard computer system 54N can use the trained vehicle system to select an output trajectory to traverse the road network with respect to the vehicle's specific position on the map and the presence of one or more agents. In block 545, the vehicle onboard computer system 54N can use the selected output to control the vehicle's operation and execute the output trajectory.

[0049] Accordingly, the embodiments disclosed here describe a system capable of resolving multiple behavioral predictions used for path planning. The embodiments presented here can address the challenge of combining multiple trajectories and delivering a single output that integrates key information in the form of a fused trajectory. The fused trajectory implicitly addresses the interdependencies between input and output trajectories, in addition to the intra-internal dependencies between the input trajectories.

[0050] The system described here can utilize individual models, each providing a valid path prediction with unique, desirable properties representing all or a subset of predictors. Once all path predictions have been generated, the input feature vectors are constructed for modeling canonical correlation analysis (CCA). While CCA can be performed on any set of informative features, the implementations described here can use any set of (x, y), where x and y are the trajectories of distinct predictors. Accordingly, not every feature array is concatenated into a single feature vector; instead, each is explicitly represented, thus enabling pairwise correlations of the input space.

[0051] The technical effects and advantages of the disclosed embodiments include, but are not limited to, the use of behavioral patterns of human vehicle operation derived from training data to control vehicle operation, such as steering, braking, and acceleration. Accordingly, autonomous and non-autonomous vehicles using the disclosed embodiments operate with increased safety, as the driving actions reflect highly human decision-making when faced with similar situations and / or locations. Consequently, after training the system, real-world applications such as autonomous driving can be influenced to navigate a road network safely.

[0052] The present disclosure may relate to a system, a method, and / or a computer-readable storage medium. The computer-readable storage medium may contain computer-readable program instructions that cause a processor to perform aspects of the present disclosure.

[0053] The computer-readable storage medium can be a tangible device capable of storing instructions for use by a command-executing device. The computer-readable storage medium can be, for example, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or a suitable combination of the foregoing, without limitation.A non-exhaustive list of more specific examples of a computer-readable storage medium includes the following: a portable computer floppy disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read storage (CD-ROM), digital versatile disk (DVD), memory stick, mechanically coded device, and any suitable combination of the foregoing.

[0054] A computer-readable storage medium, such as the one used here, is not to be interpreted as a temporary signal in itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves that propagate through a waveguide or other transmission media (e.g., light pulses traveling through a fiber optic cable), or electrical signals that are transmitted over a line.

[0055] The computer-readable program instructions can also be loaded onto a computer, other programmable data processing device, or other device to execute a series of operational steps on the computer, other programmable device, or other device to create a computer-implemented process, such that the instructions executed on the computer, other programmable device, or other device implement the functions / actions specified in the flowchart and / or block or blocks.

[0056] While the foregoing disclosure has been described with reference to exemplary embodiments, it is understood by those skilled in the art that various modifications can be made and elements thereof can be replaced by equivalents without departing from the scope of application. Furthermore, many modifications can be made to adapt a particular situation or material to the teachings of the disclosure without deviating from its essential scope. It is therefore intended that the present disclosure is not limited to the individual disclosed embodiments, but encompasses all embodiments that fall within the scope of this embodiment.

Claims

[1] A method for providing route planning guidance by resolving multiple behavioral predictions associated with the operation of a vehicle, the method comprising: Installing a vehicle system in a vehicle, wherein the vehicle system provides route planning guidance based on the use of training data and fused hypotheses and / or decisions associated with the training data; Determine, by a processor, the position of the vehicle on a map containing a road network; Determine, via the processor, whether one or more agents are present within a given area of ​​the vehicle; Select, by the processor, an output trajectory for traversing the road network based on the vehicle's position on the map and the existence of one or more agents; and Control, by the processor, of vehicle operation using the output trajectory, wherein the training data includes data based on one or more predictive models used to predict a future movement associated with one or more agents, where a Canonical Correlation Analysis (CCA) algorithm is used to merge the hypotheses and / or decisions associated with the training data, where the CCA provides a mapping between one or more predictive models and the behavior of one or more drivers. [2] The method according to claim 1, wherein the CCA uses pairwise interactions between predictors. [3] The method according to claim 1, wherein the motion information from one or more agents includes speed, course and location information. [4] A computer-readable storage medium containing program instructions, wherein the program instructions are readable by a processor to cause the processor to execute a method according to any one of claims 1 to 3.

Images