WO2024001393A1 - 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质 - Google Patents

用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质 Download PDF

Info

Publication number
WO2024001393A1
WO2024001393A1 PCT/CN2023/086656 CN2023086656W WO2024001393A1 WO 2024001393 A1 WO2024001393 A1 WO 2024001393A1 CN 2023086656 W CN2023086656 W CN 2023086656W WO 2024001393 A1 WO2024001393 A1 WO 2024001393A1
Authority
WO
WIPO (PCT)
Prior art keywords
decision
planning
making
information
finite state
Prior art date
Application number
PCT/CN2023/086656
Other languages
English (en)
French (fr)
Inventor
孔卫凯
邹李兵
Original Assignee
智道网联科技(北京)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 智道网联科技(北京)有限公司 filed Critical 智道网联科技(北京)有限公司
Publication of WO2024001393A1 publication Critical patent/WO2024001393A1/zh

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0015—Planning or execution of driving tasks specially adapted for safety
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2556/00—Input parameters relating to data
    • B60W2556/45—External transmission of data to or from the vehicle

Definitions

  • the present disclosure relates to the field of autonomous driving technology, and in particular, to a decision-making planning method, device, electronic equipment, and storage medium for unmanned vehicles.
  • drivers When driving, drivers not only analyze the things they see, such as road information, traffic participants, traffic rules, etc.; they also evaluate unknown risks. For example, their eyes will deliberately pay attention to the boundaries of obstacles and make appropriate inferences. Or assuming that an obstacle passes through a pedestrian or vehicle partially blocked or invisible to the eyes, it is deduced whether the vehicle can be parked safely without causing a collision.
  • the decision-making and planning method for driverless vehicles needs to cover the above scenarios, take into account traffic capacity, efficiency, and ensure safety.
  • This disclosure provides decision-making planning methods, devices, electronic equipment, and storage media for driverless vehicles to estimate possible potential risks while improving safety, traffic efficiency, and comfort.
  • the present disclosure provides a decision-making planning method for unmanned vehicles, wherein the method includes: evaluating whether there are potential risks in the unmanned vehicles in the current scenario; if there are no potential risks, using a machine Learning system processing and decision-making; when there are potential risks, use the finite state machine system to process and make decisions; when using the finite state machine system to process and make decisions, make decisions based on V2X information.
  • the present disclosure also provides a decision-making planning device for an unmanned vehicle, wherein the device includes: a risk assessment module for evaluating whether the unmanned vehicle has potential risks in the current scenario; a first decision-making module ,use
  • the second decision-making module is used to use the machine learning system to process and make decisions when there are no potential risks; the second decision-making module is used to use the finite state machine system to process and make decisions when there are potential risks, and is used to use the finite state machine to process and make decisions.
  • the system processes and makes decisions, it makes decisions based on V2X information.
  • the present disclosure also provides an electronic device, including: a processor; and a memory arranged to store computer-executable instructions, which when executed cause the processor to perform the above method.
  • the present disclosure also provides a computer-readable storage medium that stores one or more programs that, when executed by an electronic device including a plurality of application programs, cause the electronic device to The device performs the above method.
  • Figure 1 is a schematic flow chart of a decision-making planning method for unmanned vehicles in an embodiment of the present disclosure
  • Figure 2 is a schematic flow chart of the implementation of the decision planning method for driverless vehicles in an embodiment of the present disclosure
  • Figure 3 is a schematic diagram of a scenario in the decision-making planning method for unmanned vehicles in the embodiment of the present disclosure
  • Figure 4 is a schematic diagram of another scenario in the decision-making planning method for unmanned vehicles in the embodiment of the present disclosure.
  • Figure 5 is a schematic diagram of another scenario in the decision-making planning method for unmanned vehicles in the embodiment of the present disclosure.
  • Figure 6 is a schematic structural diagram of a decision-making planning device for unmanned vehicles in an embodiment of the present disclosure
  • FIG. 7 is a schematic structural diagram of an electronic device in an embodiment of the present disclosure.
  • Obstacles and occlusion areas are first obtained through sensors, and the traveling path of the unmanned vehicle is planned assuming that the blind spots pass through moving objects.
  • the sensor is any one of a camera, lidar, ultrasonic radar and millimeter wave radar.
  • the specific speed V2 overspeed cost, deceleration cost, acceleration cost, obstacle cost and acceleration change rate cost of the moving object passing through the blind spot.
  • predefined decision-making actions are performed in advance based on the performance of the sensor (confidence intervals under different detection distance intervals).
  • the driving decision set in the closer interval 1 includes braking and keeping following the car, which is the middle
  • the driving decisions set for distance interval 2 include following, changing lanes, and decelerating.
  • the driving decisions set for farther interval 3 include accelerating, following, changing lanes, and decelerating.
  • the driving decisions set for farthest interval 4 include acceleration.
  • the methods in the above related technologies can be classified as search algorithms based on finite state machines (Finite State Machine, FSM for short), or similar to rule-based systems and methods.
  • finite state machines Finite State Machine, FSM for short
  • the embodiment of the present disclosure provides a decision-making planning method for unmanned vehicles.
  • a schematic flow chart of the decision-making planning method for unmanned vehicles in the embodiment of the present disclosure is provided.
  • the method at least includes The following steps S110 to step S140:
  • Step S110 Evaluate whether the unmanned vehicle has potential risks in the current scene.
  • driverless vehicles include, but are not limited to, Robotaxi and Robobus, which can improve safety and ride comfort.
  • Step S120 Use the machine learning system to make decision planning when there are no potential risks.
  • the machine learning method is used to avoid the shortcomings of the finite state machine method, such as the need to exhaustively enumerate many rules when the environment is complex, and having to sacrifice part of the performance to simplify the rules. For risk-free scenarios, it directly improves decision-making efficiency and comfort.
  • Step S130 Use finite state machine system decision-making planning when potential risks exist.
  • finite state machine system decision-making planning is used to ensure safety.
  • Step S140 When using the finite state machine system to process and make decisions, make decision planning based on the V2X information.
  • decision planning can also be carried out based on V2X information.
  • V2X information here can be obtained through the perception module of the driverless vehicle.
  • making decision planning based on V2X information also includes: when using the finite state machine system to make decision planning, if the V2X information is obtained, then based on The V2X information decision-making planning; when using the finite state machine system decision-making planning, if the V2X information is not obtained, the decision-making planning is based on the preset strategy.
  • V2X information can be obtained when using the finite state machine system for decision-making and planning, and the V2X information can be used to eliminate relevant blind spots (risks), then the decision-making and planning will be based on the V2X information.
  • V2X at the road end is used to expand the sensing range and eliminate blind spots as much as possible to improve traffic efficiency.
  • the preset strategy includes at least one of the following conservative strategies: reducing speed and shifting to a side away from the blind spot.
  • Other strategies may also be included and are not specifically limited in this disclosure. For example, emergency stop.
  • decision planning based on the V2X information includes: if it is confirmed that there is a risk in the blind area based on the V2X information, Then the preset strategy is adopted; if it is confirmed that there is no risk in the blind spot according to the V2X information, the vehicle will drive normally, and the V2X information is obtained from the roadside or other vehicles.
  • the preset strategy is adopted, that is, a conservative strategy is adopted. If it is confirmed based on the V2X information that there is no risk in the blind area, then the vehicle will drive normally, that is, if there is no risk, then the vehicle will drive normally.
  • the V2X information is obtained from the road end or other vehicles.
  • the V2X information can be received.
  • the V2X information can also be received.
  • the evaluation of whether the unmanned vehicle has potential risks in the current scene includes: if the perception system of the unmanned vehicle has no obstacle sensing and blocking blind spots, then the There is no potential risk in the current scene for the unmanned vehicle; if the perception system of the unmanned vehicle has a blind spot for obstacle sensing and occlusion, the unmanned vehicle is considered to have potential risk in the current scene.
  • the machine learning system includes a neural network-based machine learning model.
  • the specific neural network used can include machine learning models used for decision-making and planning in related technologies to improve the driving and riding experience.
  • the method further includes: generating a local planning trajectory according to the decision result.
  • a local trajectory is generated and sent to the downstream control module to control the unmanned vehicle.
  • FIG. 2 shows a schematic flowchart of the implementation process of the decision planning method for driverless vehicles in the embodiment of the present disclosure, which specifically includes the following steps:
  • decision planning is carried out based on V2X information.
  • the decision planning based on V2X information also includes:
  • V2X information When using the finite state machine system for decision-making and planning, if V2X information is obtained, the decision-making and planning will be based on the V2X information;
  • the decision-making and planning will be based on the preset strategy.
  • the decision-making plan based on the V2X information includes:
  • the preset strategy is adopted;
  • the V2X information is obtained from the roadside or other vehicles.
  • the preset strategy includes at least one of the following conservative strategies: reducing speed and shifting to a side away from the blind spot.
  • the assessment of whether the driverless vehicle has potential risks in the current scenario includes:
  • the perception system of the unmanned vehicle has no blind spots blocked by obstacles, it is considered that the unmanned vehicle has no potential risk in the current scene;
  • the unmanned vehicle If the perception system of the unmanned vehicle has a blind spot blocked by obstacles, the unmanned vehicle is considered to have potential risks in the current scene.
  • the machine learning system includes a neural network-based machine learning model.
  • the method also includes: generating a local planning trajectory according to the decision result.
  • the current scenario is shown in Figure 3: the scenario is simple, the perception system has no blind spots and is risk-free, and a machine learning system is used to process decision-making planning. These include driverless vehicle B and pedestrian A. The current scene where driverless vehicle B is driving can directly see pedestrian A and there is no blind spot.
  • the current scenario the perception system has blind spots, which means there may be potential risks.
  • the finite state machine system is used for processing.
  • conservative strategies are adopted: reducing the speed, changing lanes, etc. These include driverless vehicle B, pedestrian A, and parked vehicle C. Specifically, the obstruction of vehicle C parked on the roadside creates a blind spot in perception.
  • driverless vehicle B will adopt a conservative strategy, such as slowing down and moving away from the blind spot to avoid the risk of collision.
  • the current scenario is shown in Figure 5: if there is a blind spot in the sensing system, there may be potential risks.
  • the finite state machine system is used for processing.
  • the blind spot risk is confirmed based on the acquired V2X information. If there is a risk, a conservative strategy will be adopted. If there is no risk, the vehicle will drive normally. These include driverless vehicle B, pedestrian A, parked vehicle C and V2X information device.
  • the roadside perception system can provide effective information and eliminate vehicle-side perception blind spots, there is no need to adopt a conservative strategy. This approach balances security and efficiency.
  • the embodiment of the present disclosure also provides a decision-making and planning device 600 for unmanned vehicles.
  • a schematic structural diagram of the decision-making and planning device for unmanned vehicles in the embodiment of the present disclosure is provided.
  • the decision-making planning device 600 for an unmanned vehicle at least includes: a risk assessment module 610, a first decision-making module 620 and a second decision-making module 630, wherein:
  • the risk assessment module 610 is specifically configured to assess whether the unmanned vehicle has potential risks in the current scenario.
  • driverless vehicles include, but are not limited to, Robotaxi and Robobus, which can improve safety and ride comfort.
  • the first decision-making module 620 is specifically configured to use a machine learning system to make decision planning when there are no potential risks.
  • the machine learning method is used to avoid the shortcomings of the finite state machine method, such as the need to exhaustively enumerate many rules when the environment is complex, and having to sacrifice part of the performance to simplify the rules. For risk-free scenarios, it directly improves decision-making efficiency and comfort.
  • the second decision-making module 630 is specifically configured to use finite state machine system decision planning when there are potential risks.
  • finite state machine system decision-making planning is used to ensure safety.
  • the message can be improved without using conservative strategies.
  • decision planning is carried out based on V2X information.
  • decision planning can also be carried out based on V2X information.
  • V2X information here can be obtained through the perception module of the driverless vehicle.
  • the above-mentioned decision-making and planning device for unmanned vehicles can implement each step of the decision-making and planning method for unmanned vehicles provided in the aforementioned embodiments.
  • the explanations are all applicable to the decision-making planning device for autonomous vehicles, and will not be described again here.
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
  • the electronic device includes a processor and optionally an internal bus, a network interface, and a memory.
  • the memory may include memory, such as high-speed random access memory (Random-Access Memory, RAM), or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
  • RAM random access memory
  • non-volatile memory such as at least one disk memory.
  • the electronic equipment may also include other hardware required by the business.
  • the processor, network interface and memory can be connected to each other through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect, a peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture, extended industrial standard architecture) bus, etc.
  • the bus can be divided into address bus, data bus, control bus, etc. For ease of presentation, only one bidirectional arrow is used in Figure 7, but it does not mean that there is only one bus or one type of bus.
  • Memory used to store programs.
  • a program may include program code including computer operating instructions.
  • Memory may include internal memory and non-volatile memory and provides instructions and data to the processor.
  • the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, at the logical level Forming a decision planning device for driverless vehicles.
  • the processor executes the program stored in the memory and is specifically used to perform the following operations:
  • decision planning is carried out based on V2X information.
  • the method executed by the decision-making planning device for an unmanned vehicle disclosed in the embodiment shown in FIG. 1 of the present disclosure can be applied in a processor or implemented by the processor.
  • the processor may be an integrated circuit chip that has signal processing capabilities.
  • each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor.
  • the above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital Signal Processor, DSP), special integrated Circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • CPU central processing unit
  • NP Network Processor
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • a general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
  • the steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly implemented by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other mature storage media in this field.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
  • the electronic device can also execute the method performed by the decision-making and planning device for unmanned vehicles in Figure 1, and realize the functions of the decision-making and planning device for unmanned vehicles in the embodiment shown in Figure 1.
  • the embodiment of the present disclosure is in This will not be described again.
  • Embodiments of the present disclosure also provide a computer-readable storage medium that stores one or more programs, the one or more programs include instructions, and the instructions when executed by an electronic device including multiple application programs
  • the electronic device can be caused to execute the method executed by the decision-making planning device for an unmanned vehicle in the embodiment shown in Figure 1, and is specifically used to execute:
  • decision planning is carried out based on V2X information.
  • embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the disclosure may take the form of an entirely hardware implementation, an entirely software implementation, or an implementation combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions
  • the device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.
  • These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device.
  • Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.
  • a 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
  • Memory may include non-permanent storage in computer-readable media, random access memory (RAM) and/or non-volatile memory in the form of 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 random access memory
  • Computer-readable media includes both persistent and non-volatile, removable and non-removable media that can be implemented by any method or technology for storage of information.
  • Information may be computer-readable instructions, data structures, modules of programs, 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, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic tape cassette, tape magnetic disk storage or other magnetic storage device or any other non-transmission medium that may be used to store information that can be accessed by a computing device.
  • computer-readable media does not include transitory media, such as modulated data signals and carrier waves.
  • embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the disclosure may take the form of an entirely hardware implementation, an entirely software implementation, or an implementation combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种用于无人驾驶车辆的决策规划方法，包括评估所述无人驾驶车辆在当前场景中是否存在潜在风险（S110）；在不存在潜在风险的情况下，使用机器学习系统决策规划（S120）；在存在潜在风险的情况下，使用有限状态机系统决策规划（S130）；在使用有限状态机系统决策规划时，根据V2X信息进行决策规划（S140），从而在保证安全性的同时提高驾驶乘坐体验，可用于Robotaxi、Robobus。还涉及一种用于无人驾驶车辆的决策规划装置及电子设备、存储介质。

Description

用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质
相关申请的交叉引用
本公开要求于2022年06月30日提交中国专利局的申请号为CN202210778938.0、名称为“用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本公开中。
技术领域
本公开涉及自动驾驶技术领域，尤其涉及一种用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质。
背景技术
驾驶者在驾驶时，不光对于看到的事物进行分析，如道路信息、交通参与者、交通规则等；同时也在评估未知的风险，比如眼睛会刻意的留意障碍物边界，并作出适当推理，或者假定障碍物遮挡部分或者眼睛看不到的地方穿出行人或者车辆，推理能否安全停车而不导致出现碰撞。
自动驾驶汽车也面临这种情况，车身周围不管是部署超声波雷达、激光雷达、毫米波雷达、相机等，总会出现被障碍物本身遮挡的区域，可以称之为障碍物感知遮挡盲区或障碍物遮挡区域；或其他可能会出现的需要紧急停车的场景。
无人驾驶车辆的决策规划方法，需要覆盖上述场景，并兼顾通行能力、效率，保证安全性。
发明内容
本公开提供了用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质，以实现对可能存在的潜在风险进行预估，同时提升安全性、通行效率以及舒适性。
本公开提供一种用于无人驾驶车辆的决策规划方法，其中，所述方法包括：评估所述无人驾驶车辆在当前场景中是否存在潜在风险；在不存在潜在风险的情况下，使用机器学习系统处理并决策；在存在潜在风险的情况下，使用有限状态机系统处理并决策；在使用有限状态机系统处理并决策时，根据V2X信息进行决策。
本公开还提供一种用于无人驾驶车辆的决策规划装置，其中，所述装置包括：风险评估模块，用于评估所述无人驾驶车辆在当前场景中是否存在潜在风险；第一决策模块，用 于在不存在潜在风险的情况下，使用机器学习系统处理并决策；第二决策模块，用于在存在潜在风险的情况下，使用有限状态机系统处理并决策，以及用于在使用有限状态机系统处理并决策时，根据V2X信息进行决策。
本公开还提供一种电子设备，包括：处理器；以及被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器执行上述方法。
本公开还提供一种计算机可读存储介质，所述计算机可读存储介质存储一个或多个程序，所述一个或多个程序当被包括多个应用程序的电子设备执行时，使得所述电子设备执行上述方法。
本公开采用的上述至少一个技术方案能够达到以下有益效果：
在顶层首先评估所述无人驾驶车辆在当前场景中是否存在潜在风险，在不存在潜在风险的情况下，使用机器学习系统决策规划；在存在潜在风险的情况下，使用有限状态机系统决策规划。此外，在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划。通过本公开为无人驾驶车辆提供一种可行的决策规划方法，使用有限状态机保证安全性，使用机器学习提高驾驶乘坐体验。
附图说明
此处所说明的附图用来提供对本公开的进一步理解，构成本公开的一部分，本公开的示意性实施方式及其说明用于解释本公开，并不构成对本公开的不当限定。在附图中：
图1为本公开实施方式中用于无人驾驶车辆的决策规划方法流程示意图；
图2为本公开实施方式中用于无人驾驶车辆的决策规划方法实现流程示意图；
图3为本公开实施方式中用于无人驾驶车辆的决策规划方法中的一种场景情况示意图；
图4为本公开实施方式中用于无人驾驶车辆的决策规划方法中的另一种场景情况示意图；
图5为本公开实施方式中用于无人驾驶车辆的决策规划方法中的又一种场景情况示意图；
图6为本公开实施方式中用于无人驾驶车辆的决策规划装置结构示意图；
图7为本公开实施方式中一种电子设备的结构示意图。
具体实施方式
为使本公开的目的、技术方案和优点更加清楚，下面将结合本公开具体实施方式及相应的附图对本公开技术方案进行清楚、完整地描述。显然，所描述的实施方式仅是本公开 一部分实施方式，而不是全部的实施方式。基于本公开中的实施方式，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方式，都属于本公开保护的范围。
发明人研究时发现，在相关技术中的一些方案中，a.首先通过传感器获取障碍物及遮挡区域(即盲区)，假设盲区穿出移动物体的情况下，规划无人驾驶车辆的行进路径和速度V1；b.所述传感器为摄像头、激光雷达、超声波雷达和毫米波雷达中的任意一者。c.移动物体从盲区中穿出的特定速度V2、超速代价、减速代价、加速代价、障碍物代价和加速度变化率代价。
在相关技术中的另一些方案中，提前根据传感器的性能(不同探测距离区间下的置信区间)进行预定义决策动作，如：较近区间1设置的行驶决策包括刹车和保持跟车，为中距离区间2设置的行驶决策包括跟车、变道和减速，为较远区间3设置的行驶决策包括加速、跟车、变道和减速，为最远区间4设置的行驶决策包括加速。
上述相关技术中的方法可以归类为基于有限状态机(finite state machine，简称FSM)，或类似于基于规则的系统和方法的搜索算法。
而相关技术中的再一些方案中：由于驾驶环境或移动机器人所处的环境的动态和复杂性质，很难规划出所有可能的情况并针对每种可能的情况来制定规则或有限状态机FSM。这还需要大量的有限状态机FSM，而这些有限状态机FSM可能难以开发、测试和验证。在此基础上提出用机器学习的方法，设置一个或多个神经网络，用于驱动自动驾驶的规划轨迹决策行为。
发明人认为，在相关技术中的解决方案中在通行效率、舒适性和安全性上不能兼顾。可以通过采取在有限状态机中增加保守的决策规则，来提高安全性；通过采用机器学习的方法来提高自动驾驶的通行效率和舒适性。本公开中提供了有限状态机和机器学习的混合决策方法。
以下结合附图，详细说明本公开各实施方式提供的技术方案。
本公开实施方式提供了一种用于无人驾驶车辆的决策规划方法，如图1所示，提供了本公开实施方式中用于无人驾驶车辆的决策规划方法流程示意图，所述方法至少包括如下的步骤S110至步骤S140：
步骤S110，评估所述无人驾驶车辆在当前场景中是否存在潜在风险。
在无人驾驶车辆的决策模块的最顶层先要评估所述无人驾驶车辆在当前场景中是否存在潜在风险。也就是说，既不假设存在风险，也不用根据传感器的结果进行风险应对。
需要注意的是，无人驾驶车辆包括但不限于，Robotaxi、Robobus，能够提高安全性和乘坐舒适性。
评估潜在风险，对于感知系统的性能限制比如感知距离、甚至是未知的潜在风险，进行决策规划，提高安全性、舒适性。
步骤S120，在不存在潜在风险的情况下，使用机器学习系统决策规划。
在不存在潜在风险的情况下，采用机器学习方法，避免了有限状态机方法中环境复杂时需要穷举的规则很多、不得不牺牲一部分性能以简化规则等缺点。对于无风险的场景，直接提升了决策的效率、提高了舒适性。
步骤S130，在存在潜在风险的情况下，使用有限状态机系统决策规划。
当所述自动驾驶车辆有潜在风险时使用有限状态机系统决策规划，保证安全性。
如果潜在风险且不可消除的情况，则需要使用保守策略保证安全性。
此外，如果潜在风险可消除的情况下，则可以不使用保守策略，提高效率。
步骤S140，在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划。
在使用机器学习系统决策规划或者使用有限状态机系统决策规划之后，如果在使用有限状态机系统处理并决策时，还可以根据V2X信息进行决策规划。
需要注意的是，这里的V2X信息可以通过无人驾驶车辆的感知模块获取。
在本公开的一个实施方式中，所述在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划，还包括：在使用有限状态机系统决策规划时，如果获取得到V2X信息，则根据所述V2X信息决策规划；在使用有限状态机系统决策规划时，如果未获取得到V2X信息，则根据预设策略决策规划。
具体实施时，如果使用有限状态机系统决策规划时，可以获得V2X信息，且V2X信息可用于消除相关盲区(风险)，则根据所述V2X信息决策规划。
在有危险的场景时，采用传统的有限状态机以保证安全性；同时借助路端的V2X扩大感知范围，尽量消除盲区，以提高通行效率。
可选地，如果使用有限状态机系统决策规划时，如果未获取得到V2X信息，则需要按照保守策略进行决策规划。
可选地，所述预设策略至少包括以下之一的保守策略：降低速度、向远离盲区一侧偏移。还可以包括其他策略，在本公开中并不进行具体限定。比如，紧急停车。
在本公开的一个实施方式中，所述在使用有限状态机系统处理并决策时，如果获取得到V2X信息，则根据所述V2X信息决策规划，包括：如果根据所述V2X信息确认盲区存在风险，则采用所述预设策略；如果根据所述V2X信息确认盲区不存在风险，则正常行驶，所述V2X信息通过路端或他车获得。
如果通过所述V2X信息可以确认盲区存在风险，则采用所述预设策略，即采用保守策 略，如果根据所述V2X信息确认盲区不存在风险，则正常行驶，即无风险则正常行驶。
需要注意的是，所述V2X信息通过路端或他车获得。当无人驾驶车辆进入路侧RSU设备的覆盖范围内，可以接收得到所述V2X信息。或者，当无人驾驶车辆进入他车的覆盖范围内，也可以接收得到所述V2X信息。
在本公开的一个实施方式中，所述评估所述无人驾驶车辆在当前场景中是否存在潜在风险，包括：如果所述无人驾驶车辆的感知系统无障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中不存在潜在风险；如果所述无人驾驶车辆的感知系统有障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中存在潜在风险。
在本公开的一个实施方式中，所述机器学习系统包括基于神经网络的机器学习模型。
具体使用的神经网络可以包括相关技术中的用于决策规划的机器学习模型，提高驾驶乘坐体验。
在本公开的一个实施方式中，所述方法还包括：根据决策结果，生成局部规划轨迹。
根据决策规划结果，生成局部轨迹，发给下游的控制模块，以控制无人驾驶车辆。
如图2所示为本公开实施方式中用于无人驾驶车辆的决策规划方法实现流程示意图，其具体包括如下步骤：
评估所述无人驾驶车辆在当前场景中是否存在潜在风险；
在不存在潜在风险的情况下，使用机器学习系统决策规划；
在存在潜在风险的情况下，使用有限状态机系统决策规划；
在使用有限状态机系统决策规划时，根据V2X信息进行决策规划。
所述在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划，还包括：
在使用有限状态机系统决策规划时，如果获取得到V2X信息，则根据所述V2X信息决策规划；
在使用有限状态机系统决策规划时，如果未获取得到V2X信息，则根据预设策略决策规划。
所述在使用有限状态机系统处理并决策时，如果获取得到V2X信息，则根据所述V2X信息决策规划，包括：
如果根据所述V2X信息确认盲区存在风险，则采用所述预设策略；
如果根据所述V2X信息确认盲区不存在风险，则正常行驶，所述V2X信息通过路端或他车获得。
所述预设策略至少包括以下之一的保守策略：降低速度、向远离盲区一侧偏移。
所述评估所述无人驾驶车辆在当前场景中是否存在潜在风险，包括：
如果所述无人驾驶车辆的感知系统无障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中不存在潜在风险；
如果所述无人驾驶车辆的感知系统有障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中存在潜在风险。
所述机器学习系统包括基于神经网络的机器学习模型。
所述方法还包括：根据决策结果，生成局部规划轨迹。
如图3所示当前场景：场景简单，感知系统无盲区即无风险，采用机器学习系统处理决策规划。其中包括无人驾驶车辆B、行人A。无人驾驶车辆B行驶的当前场景可以直接看到行人A并无盲区。
如图4所示当前场景：感知系统有盲区即可能有潜在风险，在未获取到V2X信息的情况下，采用有限状态机系统处理，具体采用保守策略：降低速度、换道等。其中包括无人驾驶车辆B、行人A、停靠车辆C。具体而言，路旁停靠车辆C遮挡产生感知盲区，为防止行人A从盲区穿出，无人驾驶车辆B会采取保守策略，比如，减速、向远离盲区一侧偏移，从而避免碰撞风险。
此外，还包括其他需要紧急停车的场景，也需要避免碰撞风险。
如图5所示当前场景：感知系统有盲区即可能有潜在风险，采用有限状态机系处理，基于获取的V2X信息确认盲区风险，有风险则采用保守策略，无风险则正常行驶。其中包括无人驾驶车辆B、行人A、停靠车辆C以及V2X信息device。除此以外，如果路侧的感知系统能够提供有效信息，消除车端感知盲区，则无需采用保守策略。这种方法兼顾了安全性和效率。
本公开实施方式还提供了用于无人驾驶车辆的决策规划装置600，如图6所示，提供了本公开实施方式中用于无人驾驶车辆的决策规划装置的结构示意图，所述用于无人驾驶车辆的决策规划装置600至少包括：风险评估模块610、第一决策模块620以及第二决策模块630，其中：
在本公开的一个实施方式中，所述风险评估模块610具体用于：评估所述无人驾驶车辆在当前场景中是否存在潜在风险。
在无人驾驶车辆的决策模块的最顶层先要评估所述无人驾驶车辆在当前场景中是否存在潜在风险。也就是说，既不假设存在风险，也不用根据传感器的结果进行风险应对。
需要注意的是，无人驾驶车辆包括但不限于，Robotaxi、Robobus，能够提高安全性和乘坐舒适性。
评估潜在风险，对于感知系统的性能限制比如感知距离、甚至是未知的潜在风险，进 行决策规划，提高安全性、舒适性。
在本公开的一个实施方式中，所述第一决策模块620具体用于：在不存在潜在风险的情况下，使用机器学习系统决策规划。
在不存在潜在风险的情况下，采用机器学习方法，避免了有限状态机方法中环境复杂时需要穷举的规则很多、不得不牺牲一部分性能以简化规则等缺点。对于无风险的场景，直接提升了决策的效率、提高了舒适性。
在本公开的一个实施方式中，所述第二决策模块630具体用于：在存在潜在风险的情况下，使用有限状态机系统决策规划。
当所述自动驾驶车辆有潜在风险时使用有限状态机系统决策规划，保证安全性。
如果潜在风险且不可消除的情况，则需要使用保守策略保证安全性。
此外，如果潜在风险可消除的情况下，则可以不使用保守策略，提高消息。
在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划。
在使用机器学习系统决策规划或者使用有限状态机系统决策规划之后，如果在使用有限状态机系统处理并决策时，还可以根据V2X信息进行决策规划。
需要注意的是，这里的V2X信息可以通过无人驾驶车辆的感知模块获取。
能够理解，上述用于无人驾驶车辆的决策规划装置，能够实现前述实施方式中提供的用于无人驾驶车辆的决策规划方法的各个步骤，关于用于无人驾驶车辆的决策规划方法的相关阐释均适用于用于无人驾驶车辆的决策规划装置，此处不再赘述。
图7是本公开的一个实施方式电子设备的结构示意图。请参考图7，在硬件层面，该电子设备包括处理器，可选地还包括内部总线、网络接口、存储器。其中，存储器可能包含内存，例如高速随机存取存储器(Random-Access Memory，RAM)，也可能还包括非易失性存储器(non-volatile memory)，例如至少1个磁盘存储器等。当然，该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接，该内部总线可以是ISA(Industry Standard Architecture，工业标准体系结构)总线、PCI(Peripheral Component Interconnect，外设部件互连标准)总线或EISA(Extended Industry Standard Architecture，扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示，图7中仅用一个双向箭头表示，但并不表示仅有一根总线或一种类型的总线。
存储器，用于存放程序。具体地，程序可以包括程序代码，所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器，并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行，在逻辑层面上 形成用于无人驾驶车辆的决策规划装置。处理器，执行存储器所存放的程序，并具体用于执行以下操作：
评估所述无人驾驶车辆在当前场景中是否存在潜在风险；
在不存在潜在风险的情况下，使用机器学习系统决策规划；
在存在潜在风险的情况下，使用有限状态机系统决策规划；
在使用有限状态机系统决策规划时，根据V2X信息进行决策规划。
上述如本公开图1所示实施方式揭示的用于无人驾驶车辆的决策规划装置执行的方法可以应用于处理器中，或者由处理器实现。处理器可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器，包括中央处理器(Central Processing Unit，CPU)、网络处理器(Network Processor，NP)等；还可以是数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field－Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本公开实施方式中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本公开实施方式所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。
该电子设备还可执行图1中用于无人驾驶车辆的决策规划装置执行的方法，并实现用于无人驾驶车辆的决策规划装置在图1所示实施方式的功能，本公开实施方式在此不再赘述。
本公开实施方式还提出了一种计算机可读存储介质，该计算机可读存储介质存储一个或多个程序，该一个或多个程序包括指令，该指令当被包括多个应用程序的电子设备执行时，能够使该电子设备执行图1所示实施方式中用于无人驾驶车辆的决策规划装置执行的方法，并具体用于执行：
评估所述无人驾驶车辆在当前场景中是否存在潜在风险；
在不存在潜在风险的情况下，使用机器学习系统决策规划；
在存在潜在风险的情况下，使用有限状态机系统决策规划；
在使用有限状态机系统决策规划时，根据V2X信息进行决策规划。
本领域内的技术人员应明白，本公开的实施方式可提供为方法、系统、或计算机程序产品。因此，本公开可采用完全硬件实施方式、完全软件实施方式、或结合软件和硬件方面的实施方式的形式。而且，本公开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。
本公开是参照根据本公开实施方式的方法、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。
本领域技术人员应明白，本公开的实施方式可提供为方法、系统或计算机程序产品。因此，本公开可采用完全硬件实施方式、完全软件实施方式或结合软件和硬件方面的实施方式的形式。而且，本公开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。
以上所述仅为本公开的实施方式而已，并不用于限制本公开。对于本领域技术人员来说，本公开可以有各种更改和变化。凡在本公开的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本公开的权利要求范围之内。

Claims (10)

  1. 一种用于无人驾驶车辆的决策规划方法，其中，所述方法包括：
    评估所述无人驾驶车辆在当前场景中是否存在潜在风险；
    在不存在潜在风险的情况下，使用机器学习系统决策规划；
    在存在潜在风险的情况下，使用有限状态机系统决策规划；
    在使用有限状态机系统决策规划时，根据V2X信息进行决策规划。
  2. 如权利要求1所述方法，其中，所述在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划，还包括：
    在使用有限状态机系统决策规划时，如果获取得到V2X信息，则根据所述V2X信息决策规划；
    在使用有限状态机系统决策规划时，如果未获取得到V2X信息，则根据预设策略决策规划。
  3. 如权利要求2所述方法，其中，所述在使用有限状态机系统处理并决策时，如果获取得到V2X信息，则根据所述V2X信息决策规划，包括：
    如果根据所述V2X信息确认盲区存在风险，则采用所述预设策略；
    如果根据所述V2X信息确认盲区不存在风险，则正常行驶，所述V2X信息通过路端或他车获得。
  4. 如权利要求2所述方法，其中，所述预设策略至少包括以下之一的保守策略：降低速度、向远离盲区一侧偏移。
  5. 如权利要求1所述方法，其中，所述评估所述无人驾驶车辆在当前场景中是否存在潜在风险，包括：
    如果所述无人驾驶车辆的感知系统无障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中不存在潜在风险；
    如果所述无人驾驶车辆的感知系统有障碍物感知遮挡盲区，则认为所述无人驾驶车辆在当前场景中存在潜在风险。
  6. 如权利要求1至5任一项所述方法，其中，所述机器学习系统包括基于神经网络的机器学习模型。
  7. 如权利要求1所述方法，其中，所述方法还包括：
    根据决策结果，生成局部规划轨迹。
  8. 一种用于无人驾驶车辆的决策规划装置，其中，所述装置包括：
    风险评估模块，用于评估所述无人驾驶车辆在当前场景中是否存在潜在风险；
    第一决策模块，用于在不存在潜在风险的情况下，使用机器学习系统决策规划；
    第二决策模块，用于在存在潜在风险的情况下，使用有限状态机系统决策规划，以及用于在使用有限状态机系统处理并决策时，根据V2X信息进行决策规划。
  9. 一种电子设备，包括：
    处理器；以及
    被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器执行所述权利要求1～7之任一所述方法。
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储一个或多个程序，所述一个或多个程序当被包括多个应用程序的电子设备执行时，使得所述电子设备执行所述权利要求1～7之任一所述方法。
PCT/CN2023/086656 2022-06-30 2023-04-06 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质 WO2024001393A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210778938.0A CN115027500B (zh) 2022-06-30 2022-06-30 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质
CN202210778938.0 2022-06-30

Publications (1)

Publication Number Publication Date
WO2024001393A1 true WO2024001393A1 (zh) 2024-01-04

Family

ID=83128679

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/086656 WO2024001393A1 (zh) 2022-06-30 2023-04-06 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质

Country Status (2)

Country Link
CN (1) CN115027500B (zh)
WO (1) WO2024001393A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115027500B (zh) * 2022-06-30 2024-05-14 智道网联科技(北京)有限公司 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10268191B1 (en) * 2017-07-07 2019-04-23 Zoox, Inc. Predictive teleoperator situational awareness
CN110969848A (zh) * 2019-11-26 2020-04-07 武汉理工大学 一种对向双车道下基于强化学习的自动驾驶超车决策方法
US10649453B1 (en) * 2018-11-15 2020-05-12 Nissan North America, Inc. Introspective autonomous vehicle operational management
WO2020113187A1 (en) * 2018-11-30 2020-06-04 Sanjay Rao Motion and object predictability system for autonomous vehicles
CN111679660A (zh) * 2020-06-16 2020-09-18 中国科学院深圳先进技术研究院 一种融合类人驾驶行为的无人驾驶深度强化学习方法
CN113511215A (zh) * 2021-05-31 2021-10-19 西安电子科技大学 一种混合自动驾驶决策方法、设备及计算机存储介质
CN113568416A (zh) * 2021-09-26 2021-10-29 智道网联科技(北京)有限公司 无人车轨迹规划方法、装置和计算机可读存储介质
CN115027500A (zh) * 2022-06-30 2022-09-09 智道网联科技(北京)有限公司 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6839326B1 (en) * 2000-09-26 2005-01-04 Nokia Corporation Antenna phase estimation algorithm for WCDMA closed loop transmitter antenna diversity system
US8515609B2 (en) * 2009-07-06 2013-08-20 Honeywell International Inc. Flight technical control management for an unmanned aerial vehicle
US9020873B1 (en) * 2012-05-24 2015-04-28 The Travelers Indemnity Company Decision engine using a finite state machine for conducting randomized experiments
US10446273B1 (en) * 2013-08-12 2019-10-15 Cerner Innovation, Inc. Decision support with clinical nomenclatures
WO2018102425A1 (en) * 2016-12-02 2018-06-07 Starsky Robotics, Inc. Vehicle control system and method of use
US11402839B2 (en) * 2017-03-07 2022-08-02 Robert Bosch Gmbh Action planning system and method for autonomous vehicles
US10564638B1 (en) * 2017-07-07 2020-02-18 Zoox, Inc. Teleoperator situational awareness
US10809722B2 (en) * 2018-01-29 2020-10-20 Telenav, Inc. Navigation system with route prediction mechanism and method of operation thereof
US11169536B2 (en) * 2018-04-09 2021-11-09 SafeAI, Inc. Analysis of scenarios for controlling vehicle operations
EP3575172A1 (en) * 2018-05-31 2019-12-04 Visteon Global Technologies, Inc. Adaptive longitudinal control using reinforcement learning
EP3726497A1 (en) * 2019-04-15 2020-10-21 Zenuity AB Autonomous decisions in traffic situations with planning control
US20210252698A1 (en) * 2020-02-14 2021-08-19 Nvidia Corporation Robotic control using deep learning

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10268191B1 (en) * 2017-07-07 2019-04-23 Zoox, Inc. Predictive teleoperator situational awareness
US10649453B1 (en) * 2018-11-15 2020-05-12 Nissan North America, Inc. Introspective autonomous vehicle operational management
WO2020113187A1 (en) * 2018-11-30 2020-06-04 Sanjay Rao Motion and object predictability system for autonomous vehicles
CN110969848A (zh) * 2019-11-26 2020-04-07 武汉理工大学 一种对向双车道下基于强化学习的自动驾驶超车决策方法
CN111679660A (zh) * 2020-06-16 2020-09-18 中国科学院深圳先进技术研究院 一种融合类人驾驶行为的无人驾驶深度强化学习方法
CN113511215A (zh) * 2021-05-31 2021-10-19 西安电子科技大学 一种混合自动驾驶决策方法、设备及计算机存储介质
CN113568416A (zh) * 2021-09-26 2021-10-29 智道网联科技(北京)有限公司 无人车轨迹规划方法、装置和计算机可读存储介质
CN115027500A (zh) * 2022-06-30 2022-09-09 智道网联科技(北京)有限公司 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Automotive Product Development", 31 May 2020, ISBN: 978-7-111-65274-8, article VIVEK, D. B.: "Vehicle Networking or V2X Technology", pages: 88 - 91, XP009551943 *

Also Published As

Publication number Publication date
CN115027500B (zh) 2024-05-14
CN115027500A (zh) 2022-09-09

Similar Documents

Publication Publication Date Title
JP7047089B2 (ja) セルラーネットワークベース運転支援方法及び交通制御ユニット
US11212654B2 (en) Coordinated driving through driver-to-driver V2X communication
Hamid et al. Internet of vehicle (IoV) applications in expediting the implementation of smart highway of autonomous vehicle: A survey
Sontges et al. Worst-case analysis of the time-to-react using reachable sets
Yoo et al. A stackelberg game theoretic driver model for merging
WO2024001393A1 (zh) 用于无人驾驶车辆的决策规划方法、装置及电子设备、存储介质
US11150656B2 (en) Autonomous vehicle decision making
WO2021051959A1 (zh) 车辆控制的方法、装置、控制器和智能汽车
US20220032906A1 (en) Device and method for reducing collision risk
Abdel-Aty et al. Evaluation of automated emergency braking system's avoidance of pedestrian crashes at intersections under occluded conditions within a virtual simulator
Yoo et al. A stackelberg game theoretic model of lane-merging
CN114368394A (zh) 攻击基于车联网的v2x设备的方法、装置及存储介质
Duperret et al. Formal design of a provably safe robotic roundabout system
CN112590778A (zh) 车辆控制的方法、装置、控制器和智能汽车
US20230159025A1 (en) Managing Vehicle Behavior Based On Predicted Behavior Of Other Vehicles
Ma et al. Legal Decision-making for Highway Automated Driving
CN116279450B (zh) 车辆控制方法、装置及电子设备、存储介质
Ali Decision making and control for automotive safety
Gassmann et al. An online safety guard for intelligent transportation systems
Skruch Control systems in semi and fully automated cars
JP6809551B2 (ja) 車両アプリケーションによるリソース競合を判定する方法
Wang et al. Mutually Beneficial Cooperative Driving to Solve Traffic Congestion through V2V Communication
TWI812344B (zh) 基於先進駕駛輔助系統中行車狀態的駕駛威脅分析控制系統及其方法
US20230162597A1 (en) Managing Vehicle Behavior Based On Predicted Behavior Of Other Vehicles
US20230128456A1 (en) Adaptive trust calibration

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: 23829583

Country of ref document: EP

Kind code of ref document: A1