WO2021052185A1 - 确定智能驾驶车辆的行驶轨迹 - Google Patents

确定智能驾驶车辆的行驶轨迹 Download PDF

Info

Publication number
WO2021052185A1
WO2021052185A1 PCT/CN2020/113224 CN2020113224W WO2021052185A1 WO 2021052185 A1 WO2021052185 A1 WO 2021052185A1 CN 2020113224 W CN2020113224 W CN 2020113224W WO 2021052185 A1 WO2021052185 A1 WO 2021052185A1
Authority
WO
WIPO (PCT)
Prior art keywords
trajectory
candidate
cluster
degree
matching
Prior art date
Application number
PCT/CN2020/113224
Other languages
English (en)
French (fr)
Inventor
连世奇
丁曙光
付圣
周奕达
林伟
任冬淳
Original Assignee
北京三快在线科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京三快在线科技有限公司 filed Critical 北京三快在线科技有限公司
Publication of WO2021052185A1 publication Critical patent/WO2021052185A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle

Definitions

  • This specification relates to the field of intelligent driving technology, and in particular to a method and device for determining the driving trajectory of an intelligent driving vehicle.
  • Intelligent Drive Vehicle-Assisted Driving and Self-Driving Vehicles can be collectively referred to as Intelligent Drive.
  • Intelligent Drive As an important application of artificial intelligence technology, its role in social production and life is becoming increasingly prominent , Has become one of the main directions guiding the development of transportation technology.
  • vehicles with assisted driving functions and unmanned vehicles are mostly dependent on sensing devices for the perception of the driving environment, and are based on the data collected by the sensing devices Carry out the planning of driving trajectory.
  • an unmanned vehicle can obtain obstacles and road condition information in the driving environment through radar and/or a camera, and determine the driving trajectory according to the obtained obstacles and road condition information, so as to avoid obstacles or change the driving direction.
  • those obstacles and road conditions that cannot be detected by the sensing device are likely to bring hidden dangers to driving safety.
  • the embodiments of the present specification provide a method and device for determining the driving trajectory of an intelligent driving vehicle, so as to at least partially solve the above-mentioned problems.
  • the embodiments of this specification adopt the following technical solutions.
  • the method for determining the driving trajectory of a smart driving vehicle includes: collecting the driving trajectory of at least one reference vehicle as a reference trajectory in the collection range of the smart driving vehicle; clustering each of the reference trajectories to obtain at least A trajectory cluster, wherein each of the trajectory clusters contains at least one reference trajectory; for each candidate trajectory in at least one candidate trajectory planned in advance for the intelligent driving vehicle, the candidate trajectory and each of the trajectories are determined Cluster matching degree; based on the matching degree between each candidate trajectory and each trajectory cluster, the driving trajectory of the intelligent driving vehicle is determined from each candidate trajectory.
  • collecting the driving trajectory of the reference vehicle as the reference trajectory specifically includes: determining a pre-divided collection area in the collection range; collecting each of the reference vehicles falling into each of the collection areas The driving trajectory is used as the reference trajectory.
  • Clustering each of the reference trajectories to obtain at least one trajectory cluster specifically includes: for each collection area, clustering each of the reference trajectories falling in the collection area to obtain at least one trajectory in the collection area cluster.
  • determining the degree of matching between the candidate trajectory and the trajectory cluster specifically includes: determining the center trajectory of the trajectory cluster according to the reference trajectory in the trajectory cluster; determining the candidate trajectory and the center trajectory according to the center trajectory of the trajectory cluster The similarity of the trajectory is used as the matching degree between the candidate trajectory and the trajectory cluster.
  • determining the center trajectory of the trajectory cluster according to the reference trajectory in the trajectory cluster specifically includes: determining the average trajectory of the trajectory cluster according to the reference trajectory in the trajectory cluster as the center trajectory of the trajectory cluster; or According to the reference trajectory in the trajectory cluster, determine the average trajectory of the trajectory cluster, determine the similarity between each reference driving trajectory and the average trajectory, and determine the similarity between each reference trajectory and the average trajectory from each of the trajectory clusters. Select the center trajectory of the trajectory cluster in the reference trajectory.
  • determining the degree of matching between the candidate trajectory and the trajectory cluster specifically includes: determining the similarity between the candidate trajectory and the center trajectory of the trajectory cluster; and according to the distance between the collection area where the trajectory cluster is located and the current vehicle, And the similarity between the candidate trajectory and the center trajectory of the trajectory cluster is determined to determine the degree of matching between the candidate trajectory and the trajectory cluster. For example, set the weight of the collection area according to the distance between the collection area where the trajectory cluster is located and the intelligent driving vehicle; and multiply the product of the similarity between the candidate trajectory and the center trajectory by the weight of the collection area to determine Is the matching degree between the candidate trajectory and the trajectory cluster.
  • the value range corresponding to the collection area is set; the similarity between the candidate trajectory and the center trajectory is adjusted so that the adjusted similarity is located at Within the value range; determining the adjusted similarity as the degree of matching between the candidate trajectory and the trajectory cluster.
  • determining the driving trajectory of the intelligent driving vehicle from each candidate trajectory includes: according to preset constraint conditions and each candidate trajectory and each trajectory The matching degree of the clusters is used to determine the driving trajectory of the intelligent driving vehicle from the candidate trajectories.
  • determining the driving trajectory of the intelligent driving vehicle from each candidate trajectory specifically includes: for each candidate trajectory, according to the path of the candidate trajectory The degree of matching between each trajectory cluster in each acquisition area and the candidate trajectory is determined to determine the comprehensive matching degree of the candidate trajectory; according to the determined comprehensive matching degree of each candidate trajectory, the driving of the intelligent driving vehicle is determined from each candidate trajectory Trajectory.
  • determining the comprehensive matching degree may include: for each acquisition area that the candidate trajectory passes through, according to the weight corresponding to the acquisition area and the degree of matching between the candidate trajectory and each of the trajectory clusters in the acquisition area , Determine the total matching degree of the candidate trajectory and the acquisition area; determine the comprehensive matching degree of the candidate trajectory according to the total matching degree of the candidate trajectory and each of the acquisition areas that it passes through.
  • This manual provides a device for determining the driving trajectory of an unmanned vehicle, including: an acquisition module for acquiring each candidate trajectory pre-planned by the current vehicle at the current moment, and collecting the driving trajectories of other vehicles; a clustering module for comparing The driving trajectories of other vehicles are clustered to obtain at least one trajectory cluster, wherein each trajectory cluster contains at least one driving trajectory of other vehicles; the matching degree determination module is used for each candidate trajectory, according to each trajectory cluster The driving trajectory of other vehicles contained in the trajectory is determined to determine the degree of matching between the candidate trajectory and each trajectory cluster; the driving trajectory determination module is used to determine the degree of matching between each candidate trajectory and each trajectory cluster in each candidate trajectory The current travel trajectory of the vehicle.
  • This specification provides a computer-readable storage medium, where the storage medium stores a computer program, and when the computer program is executed by a processor, the above method for determining the driving track of an unmanned vehicle is realized.
  • An unmanned vehicle provided in this specification includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor.
  • the processor executes the program, the above-mentioned determination of the driving track of the unmanned vehicle is realized.
  • At least one trajectory cluster is determined according to the driving trajectory of other vehicles, and according to the candidate trajectories and the respective trajectories of the intelligent driving vehicle at the current moment.
  • the matching degree of the trajectory cluster determines the driving trajectory for the intelligent driving vehicle.
  • the driving trajectory of the smart driving vehicle can be determined based on the trajectory cluster formed by the regular behavior of other vehicles, so that the Intelligent driving vehicles can bypass obstacles and road conditions that cannot be detected by sensor devices, thereby effectively ensuring driving safety.
  • Fig. 1 is a flowchart of a method for determining the driving trajectory of an intelligent driving vehicle provided by an embodiment of this specification.
  • Fig. 2 is a flowchart of a method for determining the degree of matching between candidate trajectories and trajectory clusters according to an embodiment of the specification.
  • Fig. 3 is a schematic diagram of the collection area provided by the embodiment of the specification.
  • Fig. 4 is a schematic structural diagram of a device for determining the driving trajectory of an intelligent driving vehicle provided by an embodiment of this specification.
  • Fig. 5 is a schematic diagram of the intelligent driving vehicle corresponding to Fig. 1 provided by an embodiment of the specification.
  • Fig. 1 is a process of determining the driving trajectory of an intelligent driving vehicle provided by an embodiment of this specification, which may specifically include the following steps:
  • S100 Collect the driving trajectory of at least one reference vehicle as the reference trajectory in the collection range of the intelligent driving vehicle.
  • the intelligent driving vehicle may include a vehicle with a driving assistance function and/or an unmanned vehicle.
  • the reference vehicle may include other vehicles passing through the collection range of the intelligent driving vehicle.
  • the reference vehicle may be another vehicle passing through the collection range of the intelligent driving vehicle before starting to plan the driving trajectory of the intelligent driving vehicle.
  • each trajectory cluster contains at least one reference trajectory.
  • Clustering can be achieved by k-means clustering (k-means), k-center point clustering (k-medoids) and other methods.
  • clustering can be performed according to the distance between the reference trajectories, and the trajectory clusters obtained by the clustering are equivalent to the set of reference trajectories formed by the regular behavior of the reference vehicles.
  • the trajectory cluster may include one or several reference trajectories, and each reference trajectory in the trajectory cluster can be used to express at least part of the information of the trajectory cluster.
  • S104 For each candidate trajectory in at least one candidate trajectory pre-planned for the intelligent driving vehicle, determine the degree of matching between the candidate trajectory and each trajectory cluster.
  • the degree of matching between the candidate trajectory and the trajectory cluster can be determined according to the similarity between the candidate trajectory and any reference trajectory in the trajectory cluster.
  • the central trajectory of the trajectory cluster can be determined according to the reference trajectory in the trajectory cluster, and the similarity between the candidate trajectory and the central trajectory of the trajectory cluster can be determined as the matching degree between the candidate trajectory and the trajectory cluster.
  • the average trajectory of the trajectory cluster can be determined according to each reference trajectory in the trajectory cluster, as the center trajectory of the trajectory cluster. For example, sample each reference trajectory in the trajectory cluster in a certain order to obtain the sampling points in each reference trajectory; determine the average coordinates for the coordinates of the sampling points in the same order in each reference trajectory; finally connect sequentially in the above order With the determined average coordinates, the average trajectory of the trajectory cluster is obtained as the center trajectory of the trajectory cluster.
  • the similarity between each reference trajectory in the trajectory cluster and the average trajectory is determined, and the center trajectory of the trajectory cluster is selected according to the similarity between each reference trajectory and the average trajectory. For example, the reference trajectory with the highest similarity to the average trajectory is selected as the center trajectory of the trajectory cluster.
  • the matching degree between a candidate trajectory and each trajectory cluster can indicate which regular behavior the candidate trajectory is more similar to.
  • the matching degree between the candidate trajectory and each trajectory cluster is used as the basis, which can deviate from the dependence on the sensors installed on the intelligent driving vehicle to a certain extent.
  • S106 Determine the driving trajectory of the intelligent driving vehicle from each candidate trajectory based on the degree of matching between each candidate trajectory and each trajectory cluster.
  • the matching degree of the candidate trajectory and each trajectory cluster can be integrated to characterize the degree of adaptation of the candidate trajectory as the driving trajectory of the intelligent driving vehicle, making the characterization of the candidate trajectory more objective and comprehensive Sex.
  • the trajectory of the vehicle is more in line with the current road conditions and driving environment.
  • step S100 collecting the driving trajectory of the reference vehicle as the reference trajectory may include: determining each pre-divided collection area in the collection range of the intelligent driving vehicle; collecting reference The driving trajectory of the vehicle falling into each collection area is used as the reference trajectory.
  • the collection range of the intelligent driving vehicle may be the collection range of the sensor device.
  • the collection range of the intelligent driving vehicle may include one or several collection areas.
  • each collection area can be determined in the collection range of the intelligent driving vehicle.
  • the preset division rules may include at least one of the following: the size rules that the collection area should meet, the adaptability rules of the collection area and road conditions (for example, the collection area should not include street lights, roadblocks, etc.) that affect the driving of intelligent driving vehicles. Infrastructure), the rules of the distance between the collection area and the smart driving vehicle (for example, the number of collection areas divided within ten meters from the smart driving vehicle should be no less than eight, and the distance from the smart driving vehicle from ten to 20 meters The number of collection areas divided within should not be less than four) and so on.
  • the collection range of the intelligent driving vehicle is divided into six collection areas.
  • the reference trajectory T1 passes through the collection areas A11, A21, and A31
  • the reference trajectory T2 passes through the collection areas A11, A12, A22, and A32.
  • clustering each reference trajectory to obtain at least one trajectory cluster may include: for each acquisition area, performing clustering on each reference trajectory falling in the acquisition area. Clustering to obtain at least one trajectory cluster in the collection area.
  • a collected reference trajectory is shorter (as shown in Figure 3, the length of the reference trajectory T1 is less than the length of the reference trajectory T2), the shorter reference trajectory can be compared If there is less data, it may cause errors to directly use the collected reference trajectory for clustering.
  • clustering the collection area can reduce the number of reference trajectories. The clustering error caused by the difference in length makes the shorter reference trajectory also play a corresponding role in determining the driving trajectory of the intelligent driving vehicle.
  • Corresponding labels can be set for each reference track.
  • the tag may include the identification information (such as license plate number, etc.), type (such as van, truck, etc.), driving style (such as unmanned vehicle, human-driven vehicle, etc.), and size of the reference vehicle corresponding to the reference trajectory. For example, at least one of the width and height of the vehicle. According to the label of the reference vehicle, it can be determined whether the reference trajectory can be the object of clustering.
  • clustering each reference trajectory to obtain at least one trajectory cluster may include: setting a corresponding label for each reference trajectory, and determining the driving trajectory of the non-unmanned vehicle according to the label; Clustering the driving trajectories of non-unmanned vehicles to obtain at least one trajectory cluster. Since the driving trajectory of the unmanned vehicle largely depends on the sensor device, abandoning the driving trajectory corresponding to the unmanned vehicle in the clustering can improve the contribution of the regular behavior of the human driving the vehicle, thereby increasing the deterministic intelligence The reliability of the driving trajectory of the driving vehicle.
  • the reference trajectory may also include: a predicted possible future trajectory of the reference vehicle.
  • the possible future travel trajectory of the reference vehicle can be predicted based on the travel speed of the reference vehicle, the road conditions of the driving environment in which the reference vehicle is located, and collected travel trajectories that the reference vehicle has traveled.
  • determining the degree of matching between the candidate trajectory and the trajectory cluster, as shown in FIG. 2 may include:
  • S200 Determine the similarity between the candidate trajectory and the center trajectory of the trajectory cluster.
  • the similarity between the candidate trajectory and the central trajectory of the trajectory cluster can be used to characterize the degree of consistency between the candidate trajectory and the central trajectory of the trajectory cluster in terms of position, orientation, and curvature.
  • S202 Determine the degree of matching between the candidate trajectory and the trajectory cluster according to the distance between the collection area where the trajectory cluster is located and the intelligent driving vehicle, and the similarity between the candidate trajectory and the center trajectory of the trajectory cluster.
  • the calculation method of similarity and the calculation method of matching degree may be the same. That is, in step S104 shown in FIG. 1, the similarity between the candidate trajectory and the center trajectory of the trajectory cluster can be directly used as the degree of matching between the candidate trajectory and the trajectory cluster.
  • the calculation methods of similarity and matching can also be different.
  • the similarity between the candidate trajectory and the central trajectory of the trajectory cluster can be multiplied by the product of the weight of the collection area where the trajectory cluster is located, and the candidate trajectory and the trajectory can be determined The matching degree of the cluster.
  • a weight can be set for each collection area according to the distance of each collection area relative to the intelligent driving vehicle. The closer the distance, the greater the weight, and the farther the distance, the smaller the weight.
  • the weight of the similarity between the center trajectory in the acquisition area closer to the intelligent driving vehicle and the candidate trajectory can be increased, and the central trajectory in the acquisition area closer to the intelligent driving vehicle can be increased in determining the driving trajectory of the intelligent driving vehicle.
  • the impact of time Vice versa, the weight of the similarity between the center trajectory in the collection area farther from the smart driving vehicle and the candidate trajectory can be reduced, and the central trajectory in the collection area farther from the smart driving vehicle is used to determine the smart driving vehicle.
  • the effect of driving trajectory is used to determine the smart driving vehicle.
  • the similarity between the candidate trajectory and the center trajectory of the trajectory cluster by a certain weight as the matching degree it is also possible to set different collection areas according to the distance of each collection area relative to the intelligent driving vehicle.
  • the value range is adjusted, and the similarity between the candidate trajectory and the center trajectory of each trajectory cluster in the acquisition area is adjusted to the value range corresponding to the acquisition area.
  • the similarity between each center trajectory in the acquisition area closer to the intelligent driving vehicle and the candidate trajectory can be set to be between 50 and 80, and the center of the acquisition area farther from the intelligent driving vehicle can be set to be between 50 and 80.
  • the similarity between the trajectory and the candidate trajectory is set between ten and twenty.
  • step S106 determining the driving trajectory of the intelligent driving vehicle from the candidate trajectories based on the degree of matching between each candidate trajectory and each trajectory cluster may include: A candidate trajectory, the comprehensive matching degree of the candidate trajectory is determined according to the degree of matching between each trajectory cluster in each acquisition area passed by the candidate trajectory and the candidate trajectory; according to the determined comprehensive matching degree of each candidate trajectory, the comprehensive matching degree of each candidate trajectory is determined from each candidate trajectory. In the trajectory, the driving trajectory of the intelligent driving vehicle is determined.
  • the comprehensive matching degree of the candidate trajectory can be used to characterize: the overall matching of the candidate trajectory and all the trajectory clusters in each collection area it passes through.
  • the total matching degree of the candidate trajectory and the acquisition area can be determined according to the weight of the acquisition area and the matching degree between the candidate trajectory and each trajectory cluster in the acquisition area.
  • the matching degree of the candidate trajectory and each trajectory cluster in the acquisition area can be summed, and multiplied by the weight of the acquisition area to determine the candidate trajectory and the acquisition area.
  • the total matching degree of the region. Then, the sum of the total matching degree of the candidate trajectory and each collection area it passes through can be determined as the comprehensive matching degree of the candidate trajectory.
  • the candidate trajectory passes through the acquisition area, and then the acquisition area passed by the candidate trajectory and the acquisition area not passed by the candidate trajectory are distinguished.
  • the total matching degree between the candidate trajectory and the acquisition areas A12, A22, and A32 is adjusted to zero, that is, the candidate trajectory and the acquisition area are adjusted to zero.
  • the matching degree of each trajectory cluster in the areas A12, A22, and A32 is adjusted to zero.
  • step S106 determining the driving trajectory of the intelligent driving vehicle from the candidate trajectories based on the degree of matching between each candidate trajectory and each trajectory cluster may include: The preset constraint conditions and the matching degree of each candidate trajectory with each trajectory cluster determine the driving trajectory of the intelligent driving vehicle from each candidate trajectory.
  • preset constraint conditions are added, which can make the determined driving trajectory of the intelligent driving vehicle more consistent with the driving conditions of the intelligent driving vehicle.
  • the preset constraints may include: the curvature constraints of the candidate trajectory (for example, if the curvature of the candidate trajectory is greater than the preset curvature threshold, the candidate trajectory will not be considered because it does not meet the curvature constraints of the candidate trajectory), the intelligent The distance constraint between the driving vehicle and the reference vehicle (for example, if the distance between the intelligent driving vehicle and the reference vehicle corresponding to the reference trajectory exceeds the preset distance threshold, the reference trajectory will not be considered because it does not meet the distance constraint )Wait.
  • the basis for determining the driving trajectory of the intelligent driving vehicle may include preset constraint conditions in addition to the degree of matching between each candidate trajectory and each trajectory cluster. In this way, the method in this embodiment is able to integrate at least two factors of the matching degree and the constraint conditions, and determine the driving trajectory of the intelligent driving vehicle from each candidate trajectory.
  • each candidate trajectory and each reference trajectory can be screened according to preset constraint conditions, and the candidate trajectories and reference trajectories that do not meet the preset constraint conditions can be excluded. Then select the matching degree of each trajectory cluster obtained by referring to the trajectory to determine the driving trajectory of the intelligent driving vehicle.
  • the candidate trajectories can be screened according to the degree of matching first, and candidate trajectories that have a poor matching degree with the trajectory clusters in each acquisition area can be excluded, and then the candidate trajectories obtained after screening can be selected according to the preset constraint conditions. Determine the driving trajectory of the intelligent driving vehicle.
  • the score of the candidate trajectory can also be determined according to the constraint conditions satisfied by the candidate trajectory.
  • the score of the candidate trajectory and the matching degree between the candidate trajectory and each trajectory cluster from each candidate trajectory Determine the driving trajectory of the intelligent driving vehicle. For example, the score of the candidate trajectory and the matching degree between the candidate trajectory and each trajectory cluster are summed and calculated, and the candidate trajectory with the largest calculation result value is selected as the driving trajectory of the intelligent driving vehicle.
  • the embodiment of this specification also correspondingly provides a structural schematic diagram of a device for determining the driving trajectory of an intelligent driving vehicle, as shown in FIG. 4.
  • the device includes an acquiring module 410, a clustering module 420, a matching degree determining module 430, and a driving trajectory determining module 440.
  • the acquisition module 410 may collect the driving trajectory of at least one reference vehicle as the reference trajectory in the collection range of the intelligent driving vehicle.
  • the clustering module 420 is communicatively connected with the acquiring module 410, and can be used to cluster each reference trajectory to obtain at least one trajectory cluster, wherein each trajectory cluster includes at least one reference trajectory.
  • the matching degree determination module 430 is in communication connection with the acquisition module 410 and the clustering module 420, and can be used to determine the candidate trajectory and each trajectory cluster for each candidate trajectory in the at least one candidate trajectory planned in advance for the intelligent driving vehicle. The matching degree.
  • the driving trajectory determining module 440 is in communication connection with the matching degree determining module 430, and can be used to determine the driving trajectory of the intelligent driving vehicle from the candidate trajectories based on the degree of matching between each candidate trajectory and each trajectory cluster.
  • the acquisition module 410 may be specifically configured to determine each pre-divided collection area in the collection range of the intelligent driving vehicle, and collect the driving trajectory of each reference vehicle falling into each collection area as the reference trajectory.
  • the clustering module 420 can cluster each reference trajectory falling in the acquisition area for each acquisition area, to obtain at least one trajectory cluster in the acquisition area.
  • the matching degree determination module 430 includes a center trajectory determination sub-module 431 and a matching degree determination sub-module 432.
  • the center trajectory determination sub-module 431 and the matching degree determination sub-module 432 are communicatively connected.
  • the central trajectory determining sub-module 431 can be used to determine the central trajectory of each trajectory cluster according to the reference trajectory in the trajectory cluster.
  • the matching degree determining sub-module 432 can be used to determine the matching degree between the candidate trajectory and the trajectory cluster according to the central trajectory of the trajectory cluster.
  • the center trajectory determining submodule 431 may be used to determine the average trajectory of the trajectory cluster according to the reference trajectories in the trajectory cluster; use the determined average trajectory as the center trajectory of the trajectory cluster; or The similarity between the reference trajectory and the average trajectory, and the center trajectory of the trajectory cluster is selected from each reference trajectory of the trajectory cluster.
  • the matching degree determination submodule 432 may be used to determine the similarity between the candidate trajectory and the central trajectory of the trajectory cluster; use the determined similarity as the matching degree between the candidate trajectory and the trajectory cluster; or, according to The distance between the collection area where the trajectory cluster is located and the intelligent driving vehicle and the similarity between the candidate trajectory and the center trajectory determine the degree of matching between the candidate trajectory and the trajectory cluster.
  • the weight of the collection area can be set according to the distance between the collection area where the trajectory cluster is located and the intelligent driving vehicle; the product obtained by multiplying the similarity between the candidate trajectory and the center trajectory by the weight of the collection area, Determine the degree of matching between the candidate trajectory and the trajectory cluster.
  • the value range corresponding to the collection area can be set according to the distance between the collection area where the trajectory cluster is located and the intelligent driving vehicle; the similarity between the candidate trajectory and the center trajectory is adjusted to make the adjusted similarity Located within the value range; determining the adjusted similarity as the matching degree between the candidate trajectory and the trajectory cluster.
  • the driving trajectory determination module 440 may be used to determine the driving trajectory of the intelligent driving vehicle from the candidate trajectories according to preset constraint conditions and the degree of matching between each candidate trajectory and each trajectory cluster.
  • the embodiment of this specification also provides a computer-readable storage medium, the storage medium stores a computer program, and the computer program can be used to execute the method for determining the driving trajectory of an intelligent driving vehicle provided in FIG. 1 above.
  • the embodiment of this specification also proposes an intelligent driving vehicle as shown in FIG. 5.
  • the intelligent driving vehicle includes a processor 510, an internal bus 520, a network interface 530, a memory 540, and a non-volatile memory 550, and of course, it may also include hardware required for other services.
  • the processor 510 reads the corresponding computer program from the non-volatile memory 550 into the memory and then runs it, so as to implement the method for determining the driving trajectory of the intelligent driving vehicle described in FIG. 1.
  • the improvement of a technology can be clearly distinguished between hardware improvements (for example, improvements in circuit structures such as diodes, transistors, switches, etc.) or software improvements (improvements in method flow).
  • hardware improvements for example, improvements in circuit structures such as diodes, transistors, switches, etc.
  • software improvements improvements in method flow.
  • the improvement of many methods and processes of today can be regarded as a direct improvement of the hardware circuit structure.
  • Designers almost always get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be realized by the hardware entity module.
  • a programmable logic device for example, a Field Programmable Gate Array (Field Programmable Gate Array, FPGA)
  • PLD Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • HDL Hardware Description Language
  • ABEL Advanced Boolean Expression Language
  • AHDL Altera Hardware Description Language
  • HDCal JHDL
  • Lava Lava
  • Lola MyHDL
  • PALASM RHDL
  • VHDL Very-High-Speed Integrated Circuit Hardware Description Language
  • Verilog Verilog
  • the controller can be implemented in any suitable manner.
  • the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program codes (such as software or firmware) executable by the (micro)processor. , Logic gates, switches, application specific integrated circuits (ASICs), programmable logic controllers and embedded microcontrollers. Examples of controllers include but are not limited to the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicon Labs C8051F320, the memory controller can also be implemented as part of the memory control logic.
  • controllers in addition to implementing the controller in a purely computer-readable program code manner, it is entirely possible to program the method steps to make the controller use logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded logic.
  • the same function can be realized in the form of a microcontroller or the like. Therefore, such a controller can be regarded as a hardware component, and the devices included in it for realizing various functions can also be regarded as a structure within the hardware component. Or even, the device for realizing various functions can be regarded as both a software module for realizing the method and a structure within a hardware component.
  • a typical implementation device is a computer.
  • the computer may be, for example, a personal computer, a laptop computer, a cell phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Any combination of these devices.
  • This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules.
  • program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types.
  • This specification can also be practiced in distributed computing environments. In these distributed computing environments, tasks are performed by remote processing devices connected through a communication network.
  • program modules can be located in local and remote computer storage media including storage devices.
  • the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device.
  • the device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment.
  • the instructions provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • the computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-permanent memory in a computer readable medium, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer readable media.
  • RAM random access memory
  • ROM read-only memory
  • flash RAM flash memory
  • Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology.
  • the information can be computer-readable instructions, data structures, program modules, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Economics (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Strategic Management (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Multimedia (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Traffic Control Systems (AREA)

Abstract

本说明书公开了一种确定智能驾驶车辆行驶轨迹的方法及装置。根据所述方法的一个示例，通过在确定智能驾驶车辆行驶轨迹的过程中，根据至少一个参考车辆的行驶轨迹确定出至少一个轨迹簇，并根据该智能驾驶车辆的至少一个候选轨迹分别与各所述轨迹簇的匹配度，为该智能驾驶车辆确定行驶轨迹。其中，参考车辆为已经行驶通过或预测未来可能行驶通过所述智能驾驶车辆的传感装置的采集范围的车辆，不同参考车辆的行驶轨迹可反应不同的规律性行为。

Description

确定智能驾驶车辆的行驶轨迹 技术领域
本说明书涉及智能驾驶技术领域，尤其涉及一种确定智能驾驶车辆行驶轨迹的方法及装置。
背景技术
目前，车辆辅助驾驶(Vehicle-Assisted Driving)以及无人车(Self-Driving Vehicle)可统称为智能驾驶(Intelligent Drive)，其作为人工智能技术的重要应用，在社会生产、生活中的作用日益凸显，成为引导交通技术发展的主要方向之一。
在相关技术中，具有辅助驾驶功能的车辆以及无人车(以下可统称为智能驾驶车辆，Intelligent Driving Vehicle)对行驶环境的感知多依赖于传感装置，并以传感装置采集的数据为基础进行行驶轨迹的规划。例如，无人车可通过雷达和/或摄像头获取行驶环境中的障碍物以及路况信息，并根据获取的障碍物以及路况信息确定行驶轨迹，以躲避障碍物或者改变行驶方向。其中，那些传感装置无法检测的障碍物和路况，极有可能为行车安全带来隐患。
发明内容
本说明书实施例提供一种确定智能驾驶车辆行驶轨迹的方法及装置，以至少部分的解决上述问题。本说明书实施例采用下述技术方案。
本说明书提供的一种确定智能驾驶车辆行驶轨迹的方法，包括：在智能驾驶车辆的采集范围中，采集至少一个参考车辆的行驶轨迹作为参考轨迹；对各所述参考轨迹进行聚类，得到至少一个轨迹簇，其中，每个所述轨迹簇中至少包含一个参考轨迹；针对为所述智能驾驶车辆预先规划的至少一个候选轨迹中的每个候选轨迹，确定该候选轨迹与每个所述轨迹簇的匹配度；基于每个所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
可选地，采集所述参考车辆的行驶轨迹作为所述参考轨迹，具体包括：在所述采集范围中，确定预先划分的采集区域；采集各所述参考车辆落入各所述采集区域中的行驶轨迹作为所述参考轨迹。对各所述参考轨迹进行聚类，得到至少一个轨迹簇，具体包括： 针对各采集区域，对落入该采集区域中的各所述参考轨迹进行聚类，得到该采集区域中的至少一个轨迹簇。
可选地，确定该候选轨迹与轨迹簇的匹配度，具体包括：根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹；根据该轨迹簇的中心轨迹，确定该候选轨迹与该中心轨迹的相似度，作为该候选轨迹与该轨迹簇的匹配度。
可选地，根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹，具体包括：根据该轨迹簇中的参考轨迹，确定该轨迹簇的平均轨迹，作为该轨迹簇的中心轨迹；或者根据该轨迹簇中的参考轨迹，确定该轨迹簇的平均轨迹，确定每个参考行驶轨迹与该平均轨迹的相似度，并根据每个参考轨迹与该平均轨迹的相似度从该轨迹簇的各参考轨迹中选取该轨迹簇的中心轨迹。
可选地，确定该候选轨迹与该轨迹簇的匹配度，具体包括：确定该候选轨迹与该轨迹簇的中心轨迹的相似度；根据该轨迹簇所在的采集区域与所述当前车辆的距离，以及该候选轨迹与该轨迹簇的中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度。例如，根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域的权重；并将该候选轨迹与该中心轨迹的相似度乘以该采集区域的权重得到的乘积，确定为该候选轨迹与该轨迹簇的匹配度。又例如，根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域对应的取值范围；调整该候选轨迹与该中心轨迹的相似度，以使调整后的相似度位于所述取值范围内；将调整后的所述相似度确定为该候选轨迹与该轨迹簇的匹配度。
可选地，基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹，具体包括：根据预设的约束条件以及各候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
可选地，基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹，具体包括：针对每个所述候选轨迹，根据该候选轨迹途经的各采集区域中的各轨迹簇与该候选轨迹的匹配度，确定该候选轨迹的综合匹配度；根据确定出的各候选轨迹的综合匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。其中，确定所述综合匹配度，可包括：针对该候选轨迹途经的每个所述采集区域，根据该采集区域对应的权重以及该候选轨迹与该采集区域中的各所述轨迹簇的匹配度，确定该候选轨迹与该采集区域的总匹配度；根据该候选轨迹与其途经的各所述采集区域的总匹配度，确定该候选轨迹的综合匹配度。
本说明书提供一种确定无人驾驶车辆行驶轨迹的装置，包括：获取模块，用于获取当前时刻当前车辆预先规划的各候选轨迹，并采集各其他车辆的行驶轨迹；聚类模块，用于对各其他车辆的行驶轨迹进行聚类，得到至少一个轨迹簇，其中，每个轨迹簇中至少包含一个其他车辆的行驶轨迹；匹配度确定模块，用于针对每个候选轨迹，根据每个轨迹簇中包含的其他车辆的行驶轨迹，确定该候选轨迹与每个轨迹簇的匹配度；行驶轨迹确定模块，用于基于每个候选轨迹与每个轨迹簇的匹配度，在各候选轨迹中，确定所述当前车辆的行驶轨迹。
本说明书提供的一种计算机可读存储介质，所述存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述确定无人驾驶车辆行驶轨迹的方法。
本说明书提供的一种无人驾驶车辆，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现上述确定无人驾驶车辆行驶轨迹的方法。
根据本公开实施例的上述至少一个技术方案，在确定智能驾驶车辆行驶轨迹的过程中，根据其他车辆的行驶轨迹确定出至少一个轨迹簇，根据该智能驾驶车辆在当前时刻的各候选轨迹与各轨迹簇的匹配度为该智能驾驶车辆确定行驶轨迹。通过将其他车辆的规律性行为形成的轨迹簇作为确定该智能驾驶车辆行驶轨迹的依据，可有效地降低智能驾驶车辆对传感装置的检测能力的依赖程度。这样，在智能驾驶车辆所处的行驶环境中存在传感装置无法检测的障碍物和路况时，可以依据其他车辆的规律性行为形成的轨迹簇对该智能驾驶车辆的行驶轨迹进行确定，使得该智能驾驶车辆能够绕开传感装置无法检测的障碍物和路况，从而有效保障行驶安全。
附图说明
图1为本说明书实施例提供的一种确定智能驾驶车辆行驶轨迹的方法的流程图。
图2为本说明书实施例提供的一种确定候选轨迹与轨迹簇的匹配度的方法的流程图。
图3为本说明书实施例提供的采集区域的示意图。
图4为本说明书实施例提供的一种确定智能驾驶车辆行驶轨迹的装置的结构示意图。
图5为本说明书实施例提供的对应于图1的智能驾驶车辆示意图。
具体实施方式
为使本说明书的目的、技术方案和优点更加清楚，下面将结合本说明书具体实施例及相应的附图对本说明书技术方案进行清楚、完整地描述。显然，所描述的实施例仅是本说明书一部分实施例，而不是全部的实施例。基于本说明书中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本说明书保护的范围。
以下结合附图，详细说明本说明书各实施例提供的技术方案。
图1为本说明书实施例提供的一种确定智能驾驶车辆行驶轨迹的过程，具体可包括以下步骤：
S100：在智能驾驶车辆的采集范围中，采集至少一个参考车辆的行驶轨迹作为参考轨迹。
本说明书实施例的方法中，智能驾驶车辆可包括具有辅助驾驶功能的车辆和/或无人车。参考车辆可以包括途经该智能驾驶车辆的采集范围的其他车辆。例如，该参考车辆可以为在开始规划该智能驾驶车辆的行驶轨迹之前，途经该智能驾驶车辆的采集范围的其他车辆。
S102：对各参考轨迹进行聚类，得到至少一个轨迹簇。其中，每个轨迹簇中至少包含一个参考轨迹。
聚类可以通过k-平均值聚类(k-means)、k-中心点聚类(k-medoids)等方法实现。在对各参考轨迹进行聚类时，可根据各参考轨迹之间的距离进行聚类，则聚类得到的轨迹簇相当于各参考车辆的规律性行为形成的参考轨迹的集合。轨迹簇可以包括一个或者几个参考轨迹，进而轨迹簇中的每个参考轨迹均可用于表达该轨迹簇的至少部分信息。
S104：针对为该智能驾驶车辆预先规划的至少一个候选轨迹中的每个候选轨迹，确定该候选轨迹与每个轨迹簇的匹配度。
针对一个候选轨迹和一个轨迹簇，可根据该候选轨迹与该轨迹簇中任一参考轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度。
例如，可根据该轨迹簇中的参考轨迹确定该轨迹簇的中心轨迹，并确定该候选轨迹与该轨迹簇的中心轨迹的相似度，作为该候选轨迹与该轨迹簇的匹配度。
在确定该轨迹簇的中心轨迹时，可根据该轨迹簇中的每个参考轨迹确定该轨迹簇的 平均轨迹，作为该轨迹簇的中心轨迹。如，按照一定的顺序对该轨迹簇中的各参考轨迹进行采样，得到各参考轨迹中的采样点；针对各参考轨迹中排序相同的采样点的坐标确定平均坐标；最后按照上述的顺序依次连接确定出的各平均坐标，得到该轨迹簇的平均轨迹，作为该轨迹簇的中心轨迹。
还可在确定上述平均轨迹后，确定该轨迹簇中每个参考轨迹与该平均轨迹的相似度，并根据每个参考轨迹与该平均轨迹的相似度选取该轨迹簇的中心轨迹。例如，选取与该平均轨迹的相似度最高的参考轨迹，作为该轨迹簇的中心轨迹。
由于通过上述步骤S102聚类得到的各轨迹簇代表了不同的规律性行为参考，因此，一个候选轨迹与各轨迹簇的匹配度可表征该候选轨迹与哪种规律性行为更为类似，后续在选择智能驾驶车辆的行驶轨迹时以候选轨迹与各轨迹簇的匹配度作为依据，可在一定程度上脱离对智能驾驶车辆上安装的传感器的依赖。
S106：基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
针对每个候选轨迹，可将该候选轨迹与各轨迹簇的匹配度进行综合，用来表征该候选轨迹作为智能驾驶车辆行驶轨迹的适应程度，使得针对该候选轨迹的表征更加客观、更具综合性。这样，能够根据一个或几个轨迹簇表达的路况信息以及参考车辆对该路况信息的应对方式，确定该候选轨迹作为智能驾驶车辆的行驶轨迹的可行性，进而使得在此基础上确定的智能驾驶车辆行驶轨迹更契合当前的道路状况和行驶环境。
在本说明书一个可选的实施例中，在步骤S100中，采集参考车辆的行驶轨迹作为参考轨迹，可以包括：在所述智能驾驶车辆的采集范围中，确定预先划分的各采集区域；采集参考车辆落入各采集区域中的行驶轨迹作为参考轨迹。
智能驾驶车辆的采集范围可以为传感装置的采集范围。智能驾驶车辆的采集范围中可以包括一个或几个采集区域。
可根据预设的划分规则，在智能驾驶车辆的采集范围中确定各采集区域。可选地，预设的划分规则可以包括以下至少一种：采集区域应满足的尺寸规则、采集区域与路况的适应性规则(如，采集区域中不应包含路灯、路障等影响智能驾驶车辆行驶的基础设施)、采集区域距智能驾驶车辆的距离规则(如，距离智能驾驶车辆的十米范围内划分的采集区域的数量应不小于八个，距离智能驾驶车辆的十米至二十米范围内划分的采集区域的数量应不小于四个)等。
如图3所示，在本说明书一个可选的实施例中，将智能驾驶车辆的采集范围划分为六个采集区域。参考轨迹T1途经采集区域A11、A21和A31，参考轨迹T2途经采集区域A11、A12、A22和A32。
在本说明书一个可选的实施例中，在步骤S102中，对各参考轨迹进行聚类，得到至少一个轨迹簇，可以包括：针对各采集区域，对落入该采集区域中的各参考轨迹进行聚类，得到该采集区域中的至少一个轨迹簇。
由于采集到的参考轨迹的长度存在差异，若采集到的一参考轨迹较短(如图3所示，参考轨迹T1的长度小于参考轨迹T2的长度)，该较短的参考轨迹可供比较的数据较少，则导致直接利用采集到的参考轨迹进行聚类较大可能会引起误差。在本说明书实施例的方法，通过将较长的参考轨迹划分在若干个不同采集区域内，而较短的参考轨迹至少能够划分在一个采集区域，针对采集区域进行聚类能够减小因参考轨迹的长度差异造成的聚类误差，使得较短的参考轨迹也能在确定智能驾驶车辆的行驶轨迹时发挥相应的作用。
可对各参考轨迹设置对应的标签。其中，标签可以包括该参考轨迹对应的参考车辆的标识信息(如，车牌号等)、类型(如，面包车、卡车等)、驾驶方式(如，无人车、人类驾驶车辆等)、尺寸(如，车辆的宽度、高度等)中的至少一种。可根据该参考车辆的标签，确定该参考轨迹是否能够作为聚类的对象。
具体地，对各参考轨迹进行聚类，得到至少一个轨迹簇，可以包括：对每个参考轨迹设置对应的标签，根据所述标签确定出非无人驾驶车辆的行驶轨迹；然后，对确定出的非无人驾驶车辆的行驶轨迹进行聚类，得到至少一个轨迹簇。由于无人车的行驶轨迹较大程度的依赖于传感装置，在聚类时舍弃无人车对应的行驶轨迹，能够提高人类驾驶车辆所表现出的规律性行为的贡献程度，进而增加确定智能驾驶车辆的行驶轨迹的可靠性。
可选地，参考轨迹除了可以是参考车辆已经行驶过的轨迹以外，还可以包括：预测出的该参考车辆的未来可能的行驶轨迹。其中，参考车辆未来可能的行驶轨迹可根据该参考车辆的行驶速度、该参考车辆所处行驶环境的路况和采集到的该参考车辆已经行驶过的行驶轨迹等信息进行预测。
在本说明书一个可选的实施例中，确定该候选轨迹与该轨迹簇的匹配度，如图2所示，可以包括：
S200：确定该候选轨迹与该轨迹簇的中心轨迹的相似度。
该候选轨迹与该轨迹簇的中心轨迹的相似度可以用于表征该候选轨迹与该轨迹簇的中心轨迹在位置、走向、曲率等多个方面的一致程度。
S202：根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，以及该候选轨迹与该轨迹簇的中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度。
在本说明书实施例中，相似度(similarity)的计算方式和匹配度(matching degree)的计算方式可以相同。即，在如图1所示的步骤S104中，可直接将候选轨迹与轨迹簇的中心轨迹的相似度，作为该候选轨迹与该轨迹簇的匹配度。
当然，相似度和匹配度的计算方式也可以不同。例如，针对一个候选轨迹和一个轨迹簇，可将该候选轨迹与该轨迹簇的中心轨迹的相似度，乘以该轨迹簇所在的采集区域的权重获得的乘积，确定为该候选轨迹与该轨迹簇的匹配度。具体的，可根据各采集区域相对于该智能驾驶车辆的距离，对各采集区域设置权重，距离越近，权重越大，距离越远，权重越小。这样，可增加距智能驾驶车辆较近的采集区域中的中心轨迹与该候选轨迹的相似度的权重，进而增加与智能驾驶车辆较近的采集区域中的各中心轨迹在确定智能驾驶车辆行驶轨迹时的影响。反之亦然，可降低距智能驾驶车辆较远的采集区域中的中心轨迹与该候选轨迹的相似度的权重，进而降低与智能驾驶车辆较远的采集区域中的各中心轨迹在确定智能驾驶车辆行驶轨迹时的影响。
当然，除了上述将候选轨迹与轨迹簇的中心轨迹的相似度乘以一定的权重作为匹配度的方法以外，还可根据各采集区域相对于该智能驾驶车辆的距离，对各采集区域设置不同的取值范围，并将候选轨迹与该采集区域中各轨迹簇的中心轨迹的相似度调整至该采集区域对应的取值范围内。如，可将距智能驾驶车辆较近的采集区域中的各中心轨迹与该候选轨迹的相似度设为五十至八十之间，而将距智能驾驶车辆较远的采集区域中的各中心轨迹与该候选轨迹的相似度设为十至二十之间。这样，也可增加与智能驾驶车辆较近的采集区域中的各中心轨迹在确定智能驾驶车辆行驶轨迹时的影响，并降低与智能驾驶车辆较远的采集区域中的各中心轨迹在确定智能驾驶车辆行驶轨迹时的影响。
在本说明书一个可选的实施例中，在步骤S106中，基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹，可以包括：针对每个候选轨迹，根据该候选轨迹途经的各采集区域中的各轨迹簇与该候选轨迹的匹配度，确定该候选轨迹的综合匹配度；根据确定出的各候选轨迹的综合匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
候选轨迹的综合匹配度可以用于表征：该候选轨迹与其途经的各采集区域中的所有轨迹簇在整体上的匹配情况。可以针对该候选轨迹途经的各采集区域，根据该采集区域的权重以及该候选轨迹与该采集区域中的各轨迹簇的匹配度，确定该候选轨迹与该采集区域的总匹配度。如，可针对该候选轨迹途经的每个采集区域，将该候选轨迹与该采集区域中的各轨迹簇的匹配度求和，并乘以该采集区域的权重，确定为该候选轨迹与该采集区域的总匹配度。然后，可将该候选轨迹与其途经的各采集区域的总匹配度之和，确定为该候选轨迹的综合匹配度。
可选地，确定候选轨迹的综合匹配度时，可针对每个采集区域判断该候选轨迹是否途经该采集区域，进而对该候选轨迹途经的采集区域和该候选轨迹未途经的采集区域进行区分。如图3所示，若一候选轨迹仅途经采集区域A11、A21和A31，则该候选轨迹与采集区域A12、A22和A32的总匹配度均被调整为零，即，将该候选轨迹与采集区域A12、A22和A32中的各轨迹簇的匹配度均被调整为零。
在本说明书一个可选的实施例中，在步骤S106中，基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹，可以包括：可根据预设的约束条件以及各候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
在确定智能驾驶车辆的行驶轨迹时，增加了预设的约束条件，能够使得确定出的智能驾驶车辆行驶轨迹更符合智能驾驶车辆的行驶条件。
预设的约束条件可以包括：候选轨迹的曲率约束(如，若该候选轨迹的曲率大于预设的曲率阈值，则该候选轨迹将因不满足候选轨迹的曲率约束而不被考虑)、该智能驾驶车辆与参考车辆的车距约束(如，若智能驾驶车辆与参考轨迹对应的参考车辆之间的距离超过预设的距离阈值，则该参考轨迹将因不满足该车距约束而不被考虑)等。
本实施例中的方法，确定智能驾驶车辆的行驶轨迹的依据，除各候选轨迹与每个轨迹簇的匹配度之外，还可以包括预设的约束条件。这样，使得本实施例中的方法能够至少综合匹配度和约束条件两个方面的因素，从各候选轨迹中确定智能驾驶车辆的行驶轨迹。
可选地，可先根据预设的约束条件对各候选轨迹和各参考轨迹进行筛选，排除不满足预设的约束条件的候选轨迹和参考轨迹，再根据筛选后获得的各候选轨迹与基于筛选后参考轨迹获得的各轨迹簇的匹配度进行选择，以确定智能驾驶车辆的行驶轨迹。或者， 可先根据匹配度对各候选轨迹进行筛选，排除与各采集区域中的轨迹簇匹配度较差的候选轨迹，再根据预设的约束条件对筛选后获得的各候选轨迹进行选择，以确定智能驾驶车辆的行驶轨迹。
当然，也可针对各候选轨迹，根据该候选轨迹满足的约束条件确定该候选轨迹的分值，根据该候选轨迹的分值以及该候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定智能驾驶车辆的行驶轨迹。例如，对该候选轨迹的分值以及该候选轨迹与每个轨迹簇的匹配度进行求和计算，选取计算结果数值最大的候选轨迹作为智能驾驶车辆的行驶轨迹。
基于图1所示的业务执行方法，本说明书实施例还对应提供一种确定智能驾驶车辆行驶轨迹的装置的结构示意图，如图4所示。
图4为本说明书实施例提供的一种确定智能驾驶车辆行驶轨迹的装置的结构示意图，所述装置包括获取模块410、聚类模块420、匹配度确定模块430和行驶轨迹确定模块440。
其中，获取模块410，可以在智能驾驶车辆的采集范围中，采集至少一个参考车辆的行驶轨迹作为参考轨迹。
聚类模块420与获取模块410通信连接，，可以用于对各参考轨迹进行聚类，得到至少一个轨迹簇，其中，每个轨迹簇中至少包含一个参考轨迹。
匹配度确定模块430与获取模块410以及聚类模块420通信连接，可以用于针对为所述智能驾驶车辆预先规划的至少一个候选轨迹中的每个候选轨迹，确定该候选轨迹与每个轨迹簇的匹配度。
行驶轨迹确定模块440与匹配度确定模块430通信连接，可以用于基于每个候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
可选地，获取模块410可以具体用于在所述智能驾驶车辆的采集范围中，确定预先划分的各采集区域，并采集各参考车辆落入各采集区域中的行驶轨迹作为所述参考轨迹。这样使得，聚类模块420能够针对各采集区域，对落入该采集区域中的各参考轨迹进行聚类，得到该采集区域中的至少一个轨迹簇。
可选地，匹配度确定模块430包括中心轨迹确定子模块431和匹配度确定子模块432。中心轨迹确定子模块431和匹配度确定子模块432通信连接。中心轨迹确定子模块431，可以用于针对每个轨迹簇，根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹。匹 配度确定子模块432，可以用于根据该轨迹簇的中心轨迹，确定该候选轨迹与该轨迹簇的匹配度。
可选地，中心轨迹确定子模块431可以用于根据该轨迹簇中的参考轨迹，确定该轨迹簇的平均轨迹；将确定出的平均轨迹，作为该轨迹簇的中心轨迹；或者，根据每个参考轨迹与该平均轨迹的相似度，从该轨迹簇的各参考轨迹中选取该轨迹簇的中心轨迹。
可选地，匹配度确定子模块432可以用于确定该候选轨迹与该轨迹簇的中心轨迹的相似度；将确定出的相似度，作为该候选轨迹与该轨迹簇的匹配度；或者，根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离以及该候选轨迹与该中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度。例如，可根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域的权重；并将该候选轨迹与该中心轨迹的相似度乘以该采集区域的权重得到的乘积，确定为该候选轨迹与该轨迹簇的匹配度。又例如，可根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域对应的取值范围；调整该候选轨迹与该中心轨迹的相似度，以使调整后的相似度位于所述取值范围内；将调整后的所述相似度确定为该候选轨迹与该轨迹簇的匹配度。
可选地，行驶轨迹确定模块440可以用于根据预设的约束条件以及各候选轨迹与每个轨迹簇的匹配度，从各候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
本说明书实施例还提供了一种计算机可读存储介质，该存储介质存储有计算机程序，计算机程序可用于执行上述图1提供的确定智能驾驶车辆的行驶轨迹的方法。
基于图1所示的业务执行的方法，本说明书实施例还提出了如图5所示的一种智能驾驶车辆。如图5，在硬件层面，该智能驾驶车辆包括处理器510、内部总线520、网络接口530、内存540以及非易失性存储器550，当然还可能包括其他业务所需要的硬件。处理器510从非易失性存储器550中读取对应的计算机程序到内存中然后运行，以实现上述图1所述的确定智能驾驶车辆的行驶轨迹的方法。
当然，除了软件实现方式之外，本说明书并不排除其他实现方式，比如逻辑器件抑或软硬件结合的方式等等，也就是说以下处理流程的执行主体并不限定于各个逻辑单元，也可以是硬件或逻辑器件。
在20世纪90年代，对于一个技术的改进可以很明显地区分是硬件上的改进(例如，对二极管、晶体管、开关等电路结构的改进)还是软件上的改进(对于方法流程的改进)。然而，随着技术的发展，当今的很多方法流程的改进已经可以视为硬件电路结构的直接 改进。设计人员几乎都通过将改进的方法流程编程到硬件电路中来得到相应的硬件电路结构。因此，不能说一个方法流程的改进就不能用硬件实体模块来实现。例如，可编程逻辑器件(Programmable Logic Device,PLD)(例如现场可编程门阵列(Field Programmable Gate Array，FPGA))就是这样一种集成电路，其逻辑功能由用户对器件编程来确定。由设计人员自行编程来把一个数字系统“集成”在一片PLD上，而不需要请芯片制造厂商来设计和制作专用的集成电路芯片。而且，如今，取代手工地制作集成电路芯片，这种编程也多半改用“逻辑编译器(logic compiler)”软件来实现，它与程序开发撰写时所用的软件编译器相类似，而要编译之前的原始代码也得用特定的编程语言来撰写，此称之为硬件描述语言(Hardware Description Language，HDL)，而HDL也并非仅有一种，而是有许多种，如ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language)等，目前最普遍使用的是VHDL(Very-High-Speed Integrated Circuit Hardware Description Language)与Verilog。本领域技术人员也应该清楚，只需要将方法流程用上述几种硬件描述语言稍作逻辑编程并编程到集成电路中，就可以很容易得到实现该逻辑方法流程的硬件电路。
控制器可以按任何适当的方式实现，例如，控制器可以采取例如微处理器或处理器以及存储可由该(微)处理器执行的计算机可读程序代码(例如软件或固件)的计算机可读介质、逻辑门、开关、专用集成电路(Application Specific Integrated Circuit，ASIC)、可编程逻辑控制器和嵌入微控制器的形式，控制器的例子包括但不限于以下微控制器：ARC 625D、Atmel AT91SAM、Microchip PIC18F26K20以及Silicone Labs C8051F320，存储器控制器还可以被实现为存储器的控制逻辑的一部分。本领域技术人员也知道，除了以纯计算机可读程序代码方式实现控制器以外，完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件，而对其内包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至，可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
上述实施例阐明的系统、装置、模块或单元，具体可以由计算机芯片或实体实现，或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的，计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这 些设备中的任何设备的组合。本说明书可以在由计算机执行的计算机可执行指令的一般上下文中描述，例如程序模块。一般地，程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、组件、数据结构等等。也可以在分布式计算环境中实践本说明书，在这些分布式计算环境中，由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中，程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
为了描述的方便，描述以上装置时以功能分为各种单元分别描述。当然，在实施本说明书时可以把各单元的功能在同一个或多个软件和/或硬件中实现。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中，计算设备包括一个或多个处理器(CPU)、输入/输出接口、网 络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
还需要说明的是，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
以上所述仅为本说明书的实施例而已，并不用于限制本说明书。对于本领域技术人员来说，本说明书可以有各种更改和变化。凡在本说明书的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本说明书的权利要求范围之内。

Claims (16)

  1. 一种确定智能驾驶车辆行驶轨迹的方法，包括：
    在智能驾驶车辆的采集范围中，采集至少一个参考车辆的行驶轨迹作为参考轨迹；
    对各所述参考轨迹进行聚类，得到至少一个轨迹簇，其中，每个所述轨迹簇中至少包含一个参考轨迹；
    针对为所述智能驾驶车辆预先规划的至少一个候选轨迹中的每个候选轨迹，确定该候选轨迹与每个所述轨迹簇的匹配度；
    基于每个所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
  2. 如权利要求1所述的方法，其中，采集所述参考车辆的行驶轨迹作为所述参考轨迹，包括：
    在所述采集范围中，确定预先划分的采集区域；
    采集所述参考车辆落入各所述采集区域中的行驶轨迹作为所述参考轨迹。
  3. 如权利要求2所述的方法，其中，对各所述参考轨迹进行聚类，得到至少一个轨迹簇，包括：
    针对各所述采集区域，对落入该采集区域中的各所述参考轨迹进行聚类，得到该采集区域中的至少一个轨迹簇。
  4. 如权利要求3所述的方法，其中，确定该候选轨迹与所述轨迹簇的匹配度，包括：
    根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹；
    根据该轨迹簇的中心轨迹，确定该候选轨迹与该轨迹簇的匹配度。
  5. 如权利要求4所述的方法，其中，根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹，包括：
    根据该轨迹簇中的参考轨迹，确定该轨迹簇的平均轨迹；
    将所述平均轨迹确定为该轨迹簇的中心轨迹。
  6. 如权利要求4所述的方法，其中，根据该轨迹簇中的参考轨迹，确定该轨迹簇的中心轨迹，包括：
    根据该轨迹簇中的参考轨迹，确定该轨迹簇的平均轨迹；
    确定每个所述参考轨迹与该平均轨迹的相似度；
    根据每个所述参考轨迹与该平均轨迹的相似度，从该轨迹簇的各所述参考轨迹中选取该轨迹簇的中心轨迹。
  7. 如权利要求4所述的方法，其中，根据该轨迹簇的中心轨迹，确定该候选轨迹与该轨迹簇的匹配度，具体包括：
    确定该候选轨迹与该中心轨迹的相似度；
    将该候选轨迹与该中心轨迹的相似度，确定为该候选轨迹与该轨迹簇的匹配度。
  8. 如权利要求4所述的方法，其中，根据该轨迹簇的中心轨迹，确定该候选轨迹与该轨迹簇的匹配度，具体包括：
    确定该候选轨迹与该中心轨迹的相似度；
    根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，以及该候选轨迹与该中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度。
  9. 如权利要求8所述的方法，其中，根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，以及该候选轨迹与该中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度，包括：
    根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域的权重；
    将该候选轨迹与该中心轨迹的相似度乘以该采集区域的权重得到的乘积，确定为该候选轨迹与该轨迹簇的匹配度。
  10. 如权利要求8所述的方法，其中，根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，以及该候选轨迹与该中心轨迹的相似度，确定该候选轨迹与该轨迹簇的匹配度，包括：
    根据该轨迹簇所在的采集区域与所述智能驾驶车辆的距离，设置该采集区域对应的取值范围；
    调整该候选轨迹与该中心轨迹的相似度，以使调整后的相似度位于所述取值范围内；
    将调整后的所述相似度确定为该候选轨迹与该轨迹簇的匹配度。
  11. 如权利要求1所述的方法，其中，基于每个所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹，包括：
    根据预设的约束条件以及各所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
  12. 如权利要求2所述的方法，其中，基于每个所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹，包括：
    针对每个所述候选轨迹，根据该候选轨迹途经的各所述采集区域中的各所述轨迹簇与该候选轨迹的匹配度，确定该候选轨迹的综合匹配度；
    根据确定出的各所述候选轨迹的综合匹配度，从各所述候选轨迹中确定所述智能驾 驶车辆的行驶轨迹。
  13. 如权利要求12所述的方法，其中，根据该候选轨迹途经的各所述采集区域中的各所述轨迹簇与该候选轨迹的匹配度，确定该候选轨迹的综合匹配度，包括：
    针对该候选轨迹途经的每个所述采集区域，根据该采集区域对应的权重以及该候选轨迹与该采集区域中的各所述轨迹簇的匹配度，确定该候选轨迹与该采集区域的总匹配度；
    根据该候选轨迹与其途经的各所述采集区域的总匹配度，确定该候选轨迹的综合匹配度。
  14. 一种确定智能驾驶车辆行驶轨迹的装置，包括：
    获取模块，用于在智能驾驶车辆的采集范围中，采集至少一个参考车辆的行驶轨迹作为参考轨迹；
    聚类模块，用于对各所述参考轨迹进行聚类，得到至少一个轨迹簇，其中，每个所述轨迹簇中至少包含一个参考轨迹；
    匹配度确定模块，用于针对为所述智能驾驶车辆预先规划的至少一个候选轨迹中的每个候选轨迹，确定该候选轨迹与每个所述轨迹簇的匹配度；
    行驶轨迹确定模块，用于基于每个所述候选轨迹与每个所述轨迹簇的匹配度，从各所述候选轨迹中确定所述智能驾驶车辆的行驶轨迹。
  15. 一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时实现上述权利要求1-13任一所述的方法。
  16. 一种智能驾驶车辆，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述程序时实现上述权利要求1-13所述的方法。
PCT/CN2020/113224 2019-09-19 2020-09-03 确定智能驾驶车辆的行驶轨迹 WO2021052185A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910885705.9A CN110674723B (zh) 2019-09-19 2019-09-19 一种确定无人驾驶车辆行驶轨迹的方法及装置
CN201910885705.9 2019-09-19

Publications (1)

Publication Number Publication Date
WO2021052185A1 true WO2021052185A1 (zh) 2021-03-25

Family

ID=69076895

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/113224 WO2021052185A1 (zh) 2019-09-19 2020-09-03 确定智能驾驶车辆的行驶轨迹

Country Status (2)

Country Link
CN (1) CN110674723B (zh)
WO (1) WO2021052185A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113470079A (zh) * 2021-07-15 2021-10-01 浙江大华技术股份有限公司 一种落脚区域的输出方法、装置及电子设备
CN113703453A (zh) * 2021-08-24 2021-11-26 京东鲲鹏(江苏)科技有限公司 确定行驶轨迹的方法、装置、电子设备及存储介质
CN114119301A (zh) * 2021-11-03 2022-03-01 支付宝(杭州)信息技术有限公司 基于共享车辆的自学车处理方法及装置
CN117349688A (zh) * 2023-12-01 2024-01-05 中南大学 一种基于峰值轨迹的轨迹聚类方法、装置、设备及介质

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110674723B (zh) * 2019-09-19 2022-07-15 北京三快在线科技有限公司 一种确定无人驾驶车辆行驶轨迹的方法及装置
CN111079721B (zh) * 2020-03-23 2020-07-03 北京三快在线科技有限公司 一种障碍物的轨迹预测方法及装置
CN111114543B (zh) * 2020-03-26 2020-07-03 北京三快在线科技有限公司 一种轨迹预测方法及装置
CN111476616B (zh) * 2020-06-24 2020-10-30 腾讯科技（深圳）有限公司 轨迹确定方法、装置、电子设备及计算机存储介质
US11814075B2 (en) * 2020-08-26 2023-11-14 Motional Ad Llc Conditional motion predictions
CN113593219B (zh) * 2021-06-30 2023-02-28 北京百度网讯科技有限公司 车流的统计方法、装置、电子设备及存储介质
CN113635912B (zh) * 2021-09-10 2023-02-24 阿波罗智能技术(北京)有限公司 车辆控制方法、装置、设备、存储介质及自动驾驶车辆
WO2023070258A1 (zh) * 2021-10-25 2023-05-04 华为技术有限公司 一种车辆的轨迹规划方法、装置及车辆
CN114295119B (zh) * 2021-12-31 2024-02-23 北京三快在线科技有限公司 一种构建地图的方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103246706A (zh) * 2013-04-09 2013-08-14 哈尔滨工程大学 一种道路网络空间中车辆对象移动轨迹聚类的方法
CN105015547A (zh) * 2014-04-28 2015-11-04 丰田自动车株式会社 驾驶辅助装置
WO2016063687A1 (ja) * 2014-10-24 2016-04-28 株式会社デンソー 隣接車線監視装置
CN109583151A (zh) * 2019-02-20 2019-04-05 百度在线网络技术（北京）有限公司 车辆的行驶轨迹预测方法及装置
CN110674723A (zh) * 2019-09-19 2020-01-10 北京三快在线科技有限公司 一种确定无人驾驶车辆行驶轨迹的方法及装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5838901B2 (ja) * 2012-04-25 2016-01-06 トヨタ自動車株式会社 物体識別装置及び物体識別方法
EP3885217A1 (en) * 2015-02-10 2021-09-29 Mobileye Vision Technologies Ltd. Sparse map for autonomous vehicle navigation
CN105117737A (zh) * 2015-10-08 2015-12-02 北京邮电大学 一种基于车辆轨迹向量确定车辆实际方向的方法及装置
CN107084735A (zh) * 2017-04-26 2017-08-22 电子科技大学 适用于减少冗余导航的导航路径框架
US11113547B2 (en) * 2017-05-31 2021-09-07 Baidu Usa Llc Planning control in response to a driving obstruction during operation of an autonomous driving vehicle (ADV)
CN108303108B (zh) * 2017-12-05 2020-05-22 华南理工大学 一种基于车辆历史轨迹的个性化路线推荐方法
CN109034448B (zh) * 2018-06-14 2022-02-11 重庆邮电大学 基于车辆轨迹语义分析和深度信念网络的轨迹预测方法
CN109703568B (zh) * 2019-02-19 2020-08-18 百度在线网络技术（北京）有限公司 自动驾驶车辆行驶策略实时学习的方法、装置和服务器
CN109598816B (zh) * 2019-02-20 2022-06-07 百度在线网络技术（北京）有限公司 自动驾驶车辆行为特征收集方法、装置、服务器及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103246706A (zh) * 2013-04-09 2013-08-14 哈尔滨工程大学 一种道路网络空间中车辆对象移动轨迹聚类的方法
CN105015547A (zh) * 2014-04-28 2015-11-04 丰田自动车株式会社 驾驶辅助装置
WO2016063687A1 (ja) * 2014-10-24 2016-04-28 株式会社デンソー 隣接車線監視装置
CN109583151A (zh) * 2019-02-20 2019-04-05 百度在线网络技术（北京）有限公司 车辆的行驶轨迹预测方法及装置
CN110674723A (zh) * 2019-09-19 2020-01-10 北京三快在线科技有限公司 一种确定无人驾驶车辆行驶轨迹的方法及装置

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113470079A (zh) * 2021-07-15 2021-10-01 浙江大华技术股份有限公司 一种落脚区域的输出方法、装置及电子设备
CN113703453A (zh) * 2021-08-24 2021-11-26 京东鲲鹏(江苏)科技有限公司 确定行驶轨迹的方法、装置、电子设备及存储介质
CN114119301A (zh) * 2021-11-03 2022-03-01 支付宝(杭州)信息技术有限公司 基于共享车辆的自学车处理方法及装置
CN117349688A (zh) * 2023-12-01 2024-01-05 中南大学 一种基于峰值轨迹的轨迹聚类方法、装置、设备及介质
CN117349688B (zh) * 2023-12-01 2024-03-19 中南大学 一种基于峰值轨迹的轨迹聚类方法、装置、设备及介质

Also Published As

Publication number Publication date
CN110674723A (zh) 2020-01-10
CN110674723B (zh) 2022-07-15

Similar Documents

Publication Publication Date Title
WO2021052185A1 (zh) 确定智能驾驶车辆的行驶轨迹
WO2021190484A1 (zh) 一种障碍物的轨迹预测方法及装置
JP6738932B2 (ja) シミュレーションプラットフォームに配置された、機械学習モデルを訓練するためのシステム及び方法
JP7338052B2 (ja) 軌跡予測方法、装置、機器及び記憶媒体リソース
JP6602352B2 (ja) 自律走行車用の計画フィードバックに基づく決定改善システム
JP6722280B2 (ja) 自律走行車の交通予測における予測軌跡の評価フレームワーク
US9291474B2 (en) System and method for providing global positioning system (GPS) feedback to a user
US20210403034A1 (en) Systems and Methods for Optimizing Trajectory Planner Based on Human Driving Behaviors
WO2018140327A1 (en) Computer-assisted or autonomous driving with region-of-interest determination for traffic light analysis
WO2022213867A1 (zh) 目标物识别
CN111126362B (zh) 一种预测障碍物轨迹的方法及装置
CN113968243B (zh) 一种障碍物轨迹预测方法、装置、设备及存储介质
WO2022252574A1 (zh) 一种故障检测的方法、装置、存储介质及电子设备
CN111062372B (zh) 一种预测障碍物轨迹的方法及装置
EP4164932A1 (en) Systems and methods for disengagement prediction and triage assistant
CN112309233B (zh) 一种道路边界的确定、道路切分方法及装置
CN112629550B (zh) 一种预测障碍物轨迹以及模型训练的方法及装置
WO2019207639A1 (ja) 行動選択装置、行動選択プログラム及び行動選択方法
CN114332808A (zh) 一种预测转向意图的方法及装置
Westny et al. Vehicle behavior prediction and generalization using imbalanced learning techniques
CN114485706A (zh) 一种路线规划方法、装置、存储介质及电子设备
US20220314980A1 (en) Obstacle tracking method, storage medium and unmanned driving device
JP2021511999A (ja) 自動運転車両の３ポイントターンの最適プランナー切り替え方法
CN112987754B (zh) 一种无人设备的控制方法、装置、存储介质及电子设备
WO2021175319A1 (zh) 出入口标注和路线规划

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20865612

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20865612

Country of ref document: EP

Kind code of ref document: A1