WO2021226767A1 - 一种自适应优化自动驾驶系统的方法及装置 - Google Patents

一种自适应优化自动驾驶系统的方法及装置 Download PDF

Info

Publication number
WO2021226767A1
WO2021226767A1 PCT/CN2020/089491 CN2020089491W WO2021226767A1 WO 2021226767 A1 WO2021226767 A1 WO 2021226767A1 CN 2020089491 W CN2020089491 W CN 2020089491W WO 2021226767 A1 WO2021226767 A1 WO 2021226767A1
Authority
WO
WIPO (PCT)
Prior art keywords
automatic driving
driving
driver
behavior
vehicle
Prior art date
Application number
PCT/CN2020/089491
Other languages
English (en)
French (fr)
Inventor
孙晓宇
郭洪强
王琴
张德明
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to JP2022568477A priority Critical patent/JP2023525088A/ja
Priority to EP20935690.6A priority patent/EP4137897A4/en
Priority to CN202080004187.6A priority patent/CN112654548B/zh
Priority to PCT/CN2020/089491 priority patent/WO2021226767A1/zh
Publication of WO2021226767A1 publication Critical patent/WO2021226767A1/zh
Priority to US17/982,978 priority patent/US20230063354A1/en

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W60/00Drive control systems specially adapted for autonomous road vehicles
    • B60W60/005Handover processes
    • B60W60/0053Handover processes from vehicle to occupant
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W60/00Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001Planning or execution of driving tasks
    • B60W60/0015Planning or execution of driving tasks specially adapted for safety
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W40/00Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/08Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
    • B60W40/09Driving style or behaviour
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/0098Details of control systems ensuring comfort, safety or stability not otherwise provided for
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08Interaction between the driver and the control system
    • B60W50/082Selecting or switching between different modes of propelling
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08Interaction between the driver and the control system
    • B60W50/10Interpretation of driver requests or demands
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08Interaction between the driver and the control system
    • B60W50/14Means for informing the driver, warning the driver or prompting a driver intervention
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W60/00Drive control systems specially adapted for autonomous road vehicles
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W60/00Drive control systems specially adapted for autonomous road vehicles
    • B60W60/005Handover processes
    • B60W60/0059Estimation of the risk associated with autonomous or manual driving, e.g. situation too complex, sensor failure or driver incapacity
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001Details of the control system
    • B60W2050/0019Control system elements or transfer functions
    • B60W2050/0028Mathematical models, e.g. for simulation
    • B60W2050/0029Mathematical model of the driver
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0062Adapting control system settings
    • B60W2050/0075Automatic parameter input, automatic initialising or calibrating means
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0062Adapting control system settings
    • B60W2050/0075Automatic parameter input, automatic initialising or calibrating means
    • B60W2050/0083Setting, resetting, calibration
    • B60W2050/0088Adaptive recalibration
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W50/00Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08Interaction between the driver and the control system
    • B60W50/14Means for informing the driver, warning the driver or prompting a driver intervention
    • B60W2050/146Display means
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2540/00Input parameters relating to occupants
    • B60W2540/043Identity of occupants
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2540/00Input parameters relating to occupants
    • B60W2540/21Voice
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2540/00Input parameters relating to occupants
    • B60W2540/22Psychological state; Stress level or workload
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2540/00Input parameters relating to occupants
    • B60W2540/223Posture, e.g. hand, foot, or seat position, turned or inclined
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2554/00Input parameters relating to objects
    • B60W2554/40Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404Characteristics
    • B60W2554/4045Intention, e.g. lane change or imminent movement
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2556/00Input parameters relating to data
    • B60W2556/10Historical data
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2556/00Input parameters relating to data
    • B60W2556/45External transmission of data to or from the vehicle
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B60VEHICLES IN GENERAL
    • B60WCONJOINT 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
    • B60W2754/00Output or target parameters relating to objects
    • B60W2754/10Spatial relation or speed relative to objects
    • B60W2754/30Longitudinal distance

Definitions

  • This application relates to the field of automatic driving technology, and in particular to a method and device for adaptively optimizing an automatic driving system.
  • Autonomous driving can refer to the configuration of one or more automatic driving modes on the vehicle.
  • the driver selects the automatic driving mode suitable for the current driving scene from one or more automatic driving modes according to his own needs, and triggers the vehicle based on the automatic driving mode Driving.
  • the number of configurations of existing automatic driving modes is limited.
  • the driving needs of all drivers in all different driving scenarios are not considered.
  • the automatic driving modes configured on the vehicle cannot meet the requirements.
  • the driving behavior of the vehicle under the control of the automatic driving system conflicts with the driving intention of the driver (such as the driving plan desired by the driver)
  • the driver usually directly takes over the vehicle and drives it manually.
  • the automatic driving plan included in the vehicle's automatic driving system is not adjusted according to the current conflicts, resulting in the driving behavior of the vehicle under the control of the automatic driving system when facing the same driving scene in the future will still be generated by the driver's driving intention.
  • Conflict can not meet the individual needs of the driver.
  • the present application provides a method and device for adaptively optimizing an automatic driving system to solve the problem of conflict between the driving behavior of a vehicle under the control of the automatic driving system and the driving intention of the driver in the prior art.
  • a method for adaptively optimizing an automatic driving system can be applied to an automatic driving device of a vehicle.
  • the automatic driving device may be an on-vehicle device, or the automatic driving device may also be applied to an on-vehicle device
  • the method may include: the automatic driving device obtains the driving intention of the driver of the vehicle that is automatically driven under the control of the automatic driving system; based on the driving intention of the driver, the automatic driving device detects the automatic driving of the target vehicle The driving behavior under the control of the driving system conflicts with the driving intention of the driver.
  • the automatic driving system is updated so that the driving behavior of the target vehicle under the control of the updated automatic driving system matches the driving intention of the driver.
  • the driving intention is characterized by characteristic parameters corresponding to the driver's behavior; the driver's behavior includes one or more of the driver's operating behavior, visual behavior, emotional behavior, and physical posture behavior.
  • the driver's driving intention can be determined from multiple angles according to the driver's multiple characteristic behaviors, which is comprehensive and accurate.
  • the driving behavior of the target vehicle under the control of the automatic driving system conflicts with the driving intention of the driver, including: the characteristic parameter corresponding to the driver's behavior exceeds the preset range, and/or the driver
  • the time that the characteristic parameter corresponding to the behavior of the driver exceeds the preset range is greater than or equal to the first preset value, and/or the number of times the characteristic parameter corresponding to the driver's behavior exceeds the preset range is greater than or equal to the second preset value.
  • the driver’s behavior can accurately reflect the driver’s driving intention, based on this, when the driver’s behavior corresponds to a feature parameter that exceeds the preset range, and/or the driver’s behavior corresponds to When the time and/or the number of times that the characteristic parameter exceeds the preset range is greater than or equal to the preset value, it can be accurately determined that the driving intention of the driver conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving device obtains the driving intention of the driver and the driving behavior of the target vehicle under the control of the automatic driving system according to the driving intention of the driver and the driving behavior of the target vehicle under the control of the automatic driving system.
  • the first characteristic parameter of the driving data corresponding to the driving behavior; the automatic driving device inputs the first characteristic parameter into the preset neural network model used to determine the driving behavior matching the driving intention of the driver to obtain the first driving behavior;
  • the first driving behavior is to update the automatic driving system.
  • the automatic driving device can characterize the driving intention of the driver and the driving of the vehicle under the control of the automatic driving system.
  • the characteristic parameters of the driving data corresponding to the behavior are input into the preset neural network model to obtain the first behavior characteristic, and the automatic driving system is updated according to the first behavior characteristic, which is simple and convenient.
  • the automatic driving device presents one or more questions to the driver; the automatic driving device receives answers from the driver to the one or more questions, and the answers to the one or more questions are used to indicate whether Update the automatic driving system; when the answers to the one or more questions are used to instruct to update the automatic driving system, the automatic driving device updates the automatic driving system according to the first driving behavior.
  • the driver's driving intention can be further confirmed according to one or more questions. Because the driver’s answer to the one or more questions can reflect the driver’s true driving intention, the driver’s driving intention can be determined more accurately, and furthermore, whether to update the automatic driving system can be determined according to the driver’s answer .
  • the automatic driving system is updated according to the driving behavior that matches the driving intention. So that the driving behavior of the vehicle under the control of the updated automatic driving system meets the driving intention of the driver.
  • the automatic driving device uses the first driving behavior as the driving behavior of the target vehicle under the control of the updated automatic driving system; If the behavior does not conform to the safe driving behavior of the target vehicle, the automatic driving device will perform the driving behavior of the target vehicle under the control of the updated automatic driving system for the safe driving behavior.
  • the driving behavior of the vehicle under the control of the updated automatic driving system matches the driving intention of the driver, and the driving behavior conforms to the driving behavior of the vehicle
  • the driving behavior can be regarded as the behavior under the control of the updated automatic driving system.
  • the safe driving behavior can be regarded as the behavior under the control of the updated automatic driving system. In this way, the vehicle can drive safely under the control of the automatic driving system.
  • a device for adaptively optimizing an automatic driving system is provided.
  • the device is applied to an automatic driving device or a chip or a system on a chip in an automatic driving device. It can also be used in an automatic driving device to implement the first aspect or the first Any possible design of the functional modules of the described method.
  • the device can realize the functions performed by the automatic driving device in the above-mentioned aspects or various possible designs, and the functions can be realized by hardware executing corresponding software.
  • the hardware or software includes one or more modules corresponding to the above-mentioned functions.
  • the device of the self-adaptive optimization automatic driving system includes a communication unit and a processing unit.
  • the communication unit is used to obtain the driving intention of the driver of the target vehicle, where the target vehicle is a vehicle that is automatically driven under the control of an automatic driving system.
  • the processing unit is used to detect that there is a conflict between the driving behavior of the target vehicle under the automatic driving control and the driving intention of the driver based on the driving intention of the driver.
  • the processing unit is also used to update the automatic driving system, so that the driving behavior of the target vehicle under the control of the updated automatic driving system matches the driving intention of the driver.
  • the specific implementation of the device for self-adaptive optimization of the automatic driving system can refer to the behavioral function of the automatic driving device in the method for self-adaptive optimization of the automatic driving system provided in the first aspect or any possible design of the first aspect, here Do not repeat it again. Therefore, the provided device for adaptively optimizing an automatic driving system can achieve the same beneficial effects as the first aspect or any possible design of the first aspect.
  • an automatic driving device may be a vehicle-mounted device or a chip or a system on a chip in the vehicle-mounted device.
  • the automatic driving device can realize the functions performed by the above-mentioned aspects or various possible designs.
  • the functions can be realized by hardware.
  • the automatic driving device may include: a processor and a communication Interface, the processor is used to run a computer program or instructions to implement the method for adaptively optimizing the automatic driving system as described in the first aspect and any one of the possible implementation manners of the first aspect.
  • the automatic driving device may further include a memory, and the memory is used to store computer execution instructions and data necessary for the automatic driving device.
  • the processor executes the computer-executable instructions stored in the memory, so that the automatic driving device executes the self-adaptive optimization automatic control described in the first aspect or any one of the possible designs of the first aspect. The method of driving the system.
  • a computer-readable storage medium may be a readable non-volatile storage medium.
  • the computer-readable storage medium stores computer instructions or programs. During operation, the computer can execute the method for adaptively optimizing the automatic driving system described in the first aspect or any one of the possible designs of the foregoing aspects.
  • a computer program product containing instructions which when running on a computer, enables the computer to execute the adaptively optimized automatic driving system described in the first aspect or any one of the possible designs of the above aspects Methods.
  • an automatic driving device may be an automatic driving device or a chip or a system on a chip in the automatic driving device.
  • the automatic driving device includes one or more processors and one or more memories. .
  • the one or more memories are coupled with the one or more processors, and the one or more memories are used to store computer program codes, and the computer program codes include computer instructions.
  • the automatic driving device is caused to execute the method for adaptively optimizing the automatic driving system as described in the first aspect or any possible design of the first aspect.
  • a chip system in a seventh aspect, includes a processor and a communication interface.
  • the chip system can be used to implement the functions performed by the automatic driving device in the first aspect or any possible design of the first aspect.
  • the processor is used to obtain the operating characteristics of the driver through the communication interface.
  • the technical effects brought about by any one of the design methods of the second aspect to the seventh aspect may refer to the technical effects brought about by the above-mentioned first aspect or any possible design of the first aspect, and will not be repeated.
  • a communication system in an eighth aspect, includes an automatic driving device and a vehicle to everything (V2X) server.
  • the automatic driving device is in communication connection with the V2X server, and the automatic driving device can be used to implement the foregoing
  • the V2X server may be used to provide a plurality of information for the automatic driving device, for example, the plurality of information may include characteristic information of the driver , The driving behavior of the vehicle under the control of the automatic driving system, etc.
  • FIG. 1 is a schematic structural diagram of a communication system provided by an embodiment of this application.
  • FIG. 2 is a schematic structural diagram of a vehicle provided by an embodiment of the application.
  • FIG. 3 is a schematic structural diagram of an automatic driving device provided by an embodiment of the application.
  • FIG. 4 is a schematic structural diagram of a V2X server provided by an embodiment of the application.
  • FIG. 5 is an example flowchart of a method for adaptively optimizing an automatic driving system according to an embodiment of the application
  • FIG. 6a is an example diagram of a driver's operation of the accelerator pedal of a vehicle provided by an embodiment of the application;
  • FIG. 6b is an example diagram of a driver's operation of a deceleration pedal of a vehicle provided by an embodiment of the application;
  • FIG. 6c is an example diagram of a driver's operation of a steering wheel of a vehicle provided by an embodiment of the application.
  • FIG. 7 is a schematic structural diagram of a vehicle provided by an embodiment of the application.
  • FIG. 8 is a schematic diagram of a display interface of a vehicle provided by an embodiment of the application.
  • FIG. 9 is an example flowchart of another method for adaptively optimizing an automatic driving system provided by an embodiment of the application.
  • FIG. 10 is a schematic structural diagram of an automatic driving device 1000 provided by an embodiment of this application.
  • FIG. 11 is a schematic structural diagram of a communication system provided by an embodiment of this application.
  • Traffic information It can also be called key elements, traffic characteristics, traffic characteristics, etc., without restriction.
  • the traffic information can be used to indicate the traffic environment on the road during the driving of the vehicle.
  • the traffic information may include one or more of vehicle information, obstacle information, pedestrian information, and information about the traffic environment on the road on which the vehicle is traveling.
  • vehicle information, obstacle information, and pedestrian information can indicate the position and relative speed of other vehicles, pedestrians, and obstacles relative to the vehicle.
  • vehicle information may include the number of vehicles, the speed of the vehicle, the distance from the vehicle to the vehicle, and the type of the vehicle.
  • Obstacle information can include road height limit bars, guardrails, and so on.
  • Pedestrian information can include the number of pedestrians, pedestrian speed, and the distance between the pedestrian and the vehicle.
  • the traffic environment information can include road information, lighting conditions, and weather conditions.
  • the road information may be highways, national highways, provincial highways, urban roads, or rural roads.
  • Road information can also include closed road signs, traffic lights, lane lines, closed roads and non-closed roads, etc., which are not restricted.
  • the traffic sign can indicate the upper limit of the road speed limit, the lower limit of the speed limit, and stop driving ahead. Traffic lights can indicate whether to stop, turn left or turn right.
  • the lane line can indicate the driving direction of the vehicle, the turning radius, the direction of the lanes on both sides, whether it is allowed to change lanes, etc.
  • the driver when the driver is driving the vehicle on the road, facing different traffic information, the driver can control the vehicle to perform driving behavior corresponding to the traffic information.
  • Driving behavior can mean that when a vehicle is driving on a road, it needs to complete different driving actions in different traffic environments before it can reach the destination. The different actions can be following the car, keeping the lane, changing lanes and cutting in, being cut in by other vehicles while driving, and so on. For example, when the vehicle is cut in by another vehicle while driving, the driving behavior of the vehicle is to reduce the speed when the vehicle is cut in.
  • the vehicle can perform one or more driving behaviors under one traffic information.
  • the one or more driving behaviors may be used to reflect the driving intention of the driver.
  • the driving behavior of the vehicle may include slowing down and adjusting the driving direction according to the radius of the curve.
  • the traffic information and one or more driving behaviors of the vehicle corresponding to the traffic information may constitute a driving scene.
  • Driving scene It can also be called the driving function domain.
  • the driving scene may refer to clustering multiple traffic information with the same or similar characteristics and the driving behavior of the vehicles under the traffic information according to a preset clustering algorithm to obtain multiple categories, wherein each of the multiple categories
  • Each category may include a type of traffic information and at least one driving behavior corresponding to the traffic information.
  • One category corresponds to one driving scene.
  • the vehicle has different driving behaviors under the control of the automatic driving system.
  • the preset clustering algorithm may be a k-means clustering algorithm (k-means).
  • k-means k-means clustering algorithm
  • the k-means clustering algorithm can refer to the prior art, which will not be repeated here.
  • the driving scene may include: direct traffic on urban peak sections, closed roads forward traffic, unclosed roads slow traffic, urban road intersections, non-urban road intersections, and the like.
  • the traffic information and driving behavior included in the above driving scene can be as shown in Table 1.
  • the traffic information in each driving scene may also include other information, such as road vehicle information, pedestrian information, etc.
  • other information such as road vehicle information, pedestrian information, etc.
  • Table 1 can also include other driving scenarios, which are not limited.
  • the driver's driving intention refers to the operation that the driver performs on the vehicle or the driving behavior that the driver wants the vehicle to perform when the vehicle is running.
  • the driving behavior can include following a car, changing lanes to the left, changing lanes to the right, or overtaking.
  • the driving behavior that the driver wants the vehicle to perform is overtaking.
  • the driving operation of the driver is to control the accelerator pedal so that the vehicle performs the driving behavior of accelerating.
  • the driving operation of the driver controlling the accelerator pedal and the driving behavior of the vehicle performing acceleration can be called natural driving data.
  • Natural driving data refers to the general designation of the driving behavior of the vehicle and the characteristics of the driver's operation of the vehicle when the driver is driving the vehicle.
  • the operating features may include controlling the steering wheel, controlling the accelerator pedal, and controlling the deceleration pedal, etc., which are not limited.
  • Auto-driving mode It means that the driver enters the destination through the auto-driving system, and the auto-driving system can determine a reasonable driving route based on preset parameters.
  • the driving route may include a plurality of traffic information, for example, the plurality of traffic information may include a plurality of curves, a plurality of road junctions, and so on.
  • the automatic driving system can control the vehicle equipped with the automatic driving system to perform corresponding driving behaviors according to different driving scenarios on the driving route. For example, at the curve of the driving route, the automatic driving system can adjust the driving behavior of the vehicle according to the curvature of the road. For example, the automatic driving system can adjust the rotation direction, rotation angle, and speed of the vehicle to make the vehicle Smooth and safe driving.
  • the vehicle is configured on certain driving scenarios.
  • the automatic driving mode cannot meet the driving needs of the driver, or, in the automatic driving technology, for scenes that the automatic driving system cannot handle, the automatic driving system will remind the driver to take over the driving right of the vehicle and drive manually.
  • the driver needs to switch the automatic driving mode to manual driving.
  • the driving behavior of the vehicle under the control of the automatic driving system conflicts with the driver's driving intention (such as the driving plan desired by the driver)
  • the driver usually directly takes over the vehicle and drives manually.
  • the driving behavior of the vehicle under the control of the automatic driving mode is not adjusted according to the current conflict, which causes the driving behavior of the vehicle under the control of the automatic driving system to conflict with the driver's driving intention when facing the same driving scene in the future. , Can not meet the individual needs of the driver.
  • an embodiment of the present application provides a method for adaptively optimizing an automatic driving system, which is applied to an automatic driving device of a vehicle, including: the automatic driving device obtains the driving intention of the driver; and based on the driving intention of the driver, The automatic driving device detects that the driving behavior of the vehicle under the control of the automatic driving system conflicts with the driving intention of the driver, and the automatic driving device updates the automatic driving system so that the driving behavior of the vehicle under the control of the updated automatic driving system matches driving Driver’s driving intention.
  • the automatic driving system when it is detected that the driving intention of the driver of a vehicle that is automatically driven under the control of the automatic driving system conflicts with the driving behavior of the vehicle under the control of the automatic driving system of the vehicle, the automatic driving system is updated so that the vehicle is updated
  • the driving behavior under the control of the subsequent automatic driving system matches the driver's driving intention.
  • the subsequent driving behavior under the control of the updated automatic driving system can meet the driving needs of the driver.
  • Fig. 1 shows an example diagram of a communication system provided by an embodiment of the present application.
  • the communication system may include a vehicle 10 and a V2X server 20.
  • the vehicle 10 may communicate with the V2X server 20 using V2X communication technology.
  • the vehicle 20 can also communicate with the V2X server 20 via a wireless link, for example, via a fifth generation (5G) network, or via other methods, without limitation.
  • 5G fifth generation
  • the vehicle 10 may be an intelligent network driving (intelligent network driving) vehicle, which is a typical vehicle networking terminal.
  • vehicle 10 may specifically execute the method for adaptively optimizing the automatic driving system in the embodiment of the present application through its internal functional units or devices.
  • the vehicle 10 may include an automatic driving device for executing the method for adaptively optimizing the automatic driving system provided by the embodiment of the present application.
  • the automatic driving device may communicate with other components of the vehicle 10 through a controller area network (CAN) bus.
  • CAN controller area network
  • the V2X server 20 can provide the vehicle 10 with multiple driving behaviors of the vehicle under the control of the automatic driving system and/or the natural driving data of multiple drivers, etc., and can also be used to receive image information from the driver of the vehicle 10, and The image information of the driver is recognized to obtain the body posture behavior of the driver.
  • the body posture behavior may include whether the driver is using a mobile phone, whether the driver is driving fatigued, and so on.
  • the V2X server 20 may also send the driver's body posture behavior to the vehicle 10.
  • the V2X server 20 may be a physical server or a virtual server, such as a cloud server, etc., which is not limited. The specific structure of the V2X server 20 will be described in detail in the embodiment shown in FIG. 4.
  • FIG. 2 is a schematic structural diagram of a vehicle 10 provided by an embodiment of this application.
  • the vehicle 10 may include an automatic driving device 101, a vehicle body gateway 102, a vehicle body antenna 103, and the like.
  • the automatic driving device 101 may be communicatively connected with the body antenna 103 through a radio frequency (RF) cable.
  • RF radio frequency
  • the vehicle 10 may include more or fewer components than those shown in FIG. 2, or the vehicle 10 may include a combination of certain components shown in FIG. 2, or the vehicle 10 may include Split parts of the parts shown, etc.
  • the vehicle 10 may also include a domain controller (DC) and a multi-domain controller (multi-domain controller,
  • the components shown in FIG. 2 may be implemented in hardware, software, or a combination of software and hardware.
  • the automatic driving device 101 in the vehicle 10 may be a car networking chip or the like, and the specific structure of the automatic driving device 101 will be described in detail in the embodiment shown in FIG. 3.
  • the automatic driving device 101 may be called an on-board unit (OBU), an on-board terminal, etc., for example, the automatic driving device 101 may be a telematics BOX (T-Box).
  • the automatic driving device 101 is mainly used to execute the method for adaptively optimizing the automatic driving system provided in the embodiments of the present application.
  • the body gateway 102 is mainly used for receiving and sending vehicle information, and the body gateway 102 may be connected to the automatic driving device 101 through a CAN bus.
  • the vehicle body gateway 102 may obtain from the automatic driving device 101 the updated automatic driving system obtained after the automatic driving device 101 executes the method for adaptively optimizing the automatic driving system provided in this embodiment of the application, and the vehicle is under the control of the automatic driving system
  • the obtained driving behavior of the updated automatic driving system and the driving behavior of the vehicle under the control of the automatic driving system are sent to other components of the vehicle 10.
  • the body antenna 103 may have a built-in communication antenna, and the communication antenna is responsible for signal reception and transmission.
  • the communication antenna can send vehicle information of a vehicle to an automatic driving device of another vehicle, and can also receive vehicle information sent from another automatic driving device.
  • FIG. 3 is a schematic structural diagram of an automatic driving device 101 provided by an embodiment of this application.
  • the automatic driving device 101 may include an automatic driving perception module 1011, a smart cockpit interaction module 1012, a driving intention conflict detection module 1013, and an automatic driving control module 1014.
  • the automatic driving perception module 1011 can be used to recognize and divide the traffic information and driving behavior of the vehicle 10; Perform voice recognition; the driving intention conflict detection module 1013 can be used to determine the driving intention of the driver, and detect whether the driving intention of the driver conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving control module 1014 may be used to control the driving behavior of the vehicle during automatic driving.
  • the smart cockpit interaction module 1012 may use a long-short term memory-deep neural networks (LSTM-DNN) model to recognize the voice information of the driver.
  • LSTM-DNN long-short term memory-deep neural networks
  • the smart cockpit interaction module 1012 is equipped with an LSTM-DNN model, and the autopilot device can obtain the voice information of the people inside the vehicle through the sound collection device, and input the voice information of the people inside the vehicle into the LSTM of the smart cockpit interaction module 1012 -In the DNN model, so that the intelligent cockpit interaction module 1012 recognizes the voice information of persons located inside the vehicle.
  • the following descriptions related to the recognition of voice information can be referred to here, and will not be repeated in the following.
  • the automatic driving device 101 may further include a driver online adaptive module 1015.
  • the driver online adaptive module 1015 can be used to adjust the driving behavior of the vehicle under the control of the automatic driving system in the driving scene according to the driving intention of the driver of the vehicle in the driving scene, and to adjust the driving behavior of the vehicle under the control of the automatic driving system after the adjustment.
  • the driving behavior under control and the driving scene are sent to the automatic driving control module 1014.
  • the automatic driving control module 1014 After the automatic driving control module 1014 receives the adjusted driving behavior of the vehicle under the control of the automatic driving system and the information of the driving scene from the driver online adaptive module 1015, it can update the automatic driving system, for example, The driving behavior of the vehicle under the control of the automatic driving system is replaced with the driving behavior of the adjusted vehicle under the control of the automatic driving system.
  • the structure illustrated in FIG. 3 does not constitute a specific limitation on the automatic driving device 101.
  • the automatic driving device 101 may include more or fewer components than those shown in FIG. 3, or the automatic driving device 101 may include a combination of certain components shown in FIG.
  • the driving device 101 may include split parts of the parts shown in FIG. 3 and the like.
  • the components shown in FIG. 3 can be implemented in hardware, software, or a combination of software and hardware.
  • the device shown in FIG. 3 may also be a chip or a chip system in an automatic driving device.
  • the chip system can be composed of chips, and can also include chips and other discrete devices.
  • FIG. 4 is a schematic structural diagram of a V2X server 20 according to an embodiment of this application.
  • the V2X server 20 may include a driver preference and historical operation database 2011, and a parameter module 2012 with the highest driver acceptance.
  • the driver preference and historical operation database 2011 is used to store natural driving data of multiple drivers within a preset time period and at least one driving behavior of the vehicle under the control of the automatic driving system.
  • the parameter module 2012 with the highest driver acceptance is used to store the most frequently used automatic driving mode among the at least one automatic driving mode.
  • the V2X server 20 can also update the data in the driver preference and historical operation database 2011 and the parameter module 2012 with the highest driver acceptance according to a preset time period.
  • the preset time period may be preset by the V2X server 20.
  • the V2X server 20 may also update the data in the driver preference and historical operation database 2011 and the parameter module 2012 with the highest driver acceptance through interaction with the automatic driving device 101. For example, after the automatic driving device updates the driving behavior of the vehicle under the control of the automatic driving system, it sends the driving behavior of the vehicle under the control of the updated automatic driving system to the V2X server.
  • the V2X server After the V2X server receives the driving behavior of the vehicle under the control of the updated automatic driving system from the automatic driving device, it can replace the driving behavior of the vehicle under the control of the automatic driving system in the V2X server with the updated automatic driving behavior of the vehicle. Driving behavior under the control of the driving system.
  • the V2X server 20 may also include the driver online adaptive module 1015 in FIG. 2, where the function of the driver online adaptive module 1015 can be referred to the above description, which will not be repeated here.
  • the structure illustrated in FIG. 4 does not constitute a specific limitation on the V2X server 20.
  • the V2X server 20 may include more or less components than those shown in FIG. 4, or the V2X server 20 may include a combination of certain components shown in FIG. 4, or the V2X server 20 It may include split parts of the parts shown in FIG. 4 and the like.
  • the components shown in FIG. 4 can be implemented in hardware, software, or a combination of software and hardware.
  • the device shown in FIG. 4 may also be a chip or a chip system in a V2X server.
  • the chip system can be composed of chips, and can also include chips and other discrete devices.
  • words such as “exemplary” or “for example” are used as examples, illustrations, or illustrations. Any embodiment or design solution described as “exemplary” or “for example” in the embodiments of the present application should not be construed as being more preferable or advantageous than other embodiments or design solutions. To be precise, words such as “exemplary” or “for example” are used to present related concepts in a specific manner.
  • first and second are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined with “first” and “second” may explicitly or implicitly include one or more of these features. In the description of the embodiments of the present application, unless otherwise specified, “plurality” means two or more.
  • the execution subject of the method for adaptively optimizing the automatic driving system may be an automatic driving device, for example, the automatic driving device may be the automatic driving device in FIG. 2 or FIG. 3.
  • the execution subject of the method may also be the chip or the system on chip in the automatic driving device, which is not limited.
  • the following describes an example in which the execution subject of the method for adaptively optimizing an automatic driving system is an automatic driving device.
  • FIG. 5 is a schematic flowchart of a method for adaptively optimizing an automatic driving system according to an embodiment of the application. As shown in Figure 5, the method may include:
  • Step 501 The automatic driving device acquires the driving intention of the driver of the target vehicle.
  • the target vehicle is a vehicle that is automatically driven under the control of an automatic control system.
  • the target vehicle may be the vehicle in FIG. 1, and the target vehicle may have the components described in FIG. 2.
  • the automatic driving device may include one or more modules in FIG. 3.
  • the driver can be a driver who has driven the target vehicle, or a driver who has not driven the target vehicle, or can be described as a new driver.
  • the automatic driving device can detect whether the driver has driven the target vehicle or is a new driver based on the driver's characteristic information. One feature information is used to identify a driver.
  • the driver's driving intention may be characterized by characteristic parameters corresponding to the driver's behavior. That is, the automatic driving device can detect the driver's driving intention based on the driver's behavior.
  • the driver's behavior may include one or more of the driver's operating behavior, visual behavior, emotional behavior, and body posture behavior.
  • the operation behavior of the driver may refer to the manual operation behavior of the driver on the vehicle during the automatic driving of the vehicle.
  • the operation behavior of the driver may include the operation behavior of the driver on the steering wheel of the vehicle, the operation behavior of the driver on the accelerator pedal of the vehicle, and the operation behavior of the driver on the deceleration pedal of the vehicle.
  • the automatic driving device can obtain the driver's driving intention in a variety of ways.
  • the multiple methods may include: the automatic driving device can input characteristic parameters corresponding to the driver's behavior into the driving intention model to obtain the driver's driving intention; the automatic driving device can determine the driver's driving intention according to the driver's operating behavior ; The automatic driving device can determine the driver's driving intention based on the driver's emotional behavior.
  • the automatic driving device can also determine the driver's driving intention in other ways, without restriction.
  • the automatic driving device inputs the characteristic parameters corresponding to the driver's behavior into the driving intention model to obtain the driving intention of the driver.
  • the input parameter of the driving intention model is the characteristic parameter of the driver's behavior
  • the output parameter of the driving intention is the driving intention of the driver.
  • the driving intention model can be preset by the automatic driving device.
  • a method for determining the driving intention model may be obtained by training the behavior characteristics of multiple drivers in a preset time period and the corresponding natural driving data through a hidden Markov model.
  • the training method of the hidden Markov model can refer to the prior art, which will not be repeated here.
  • the automatic driving device may input the characteristic parameters of the driver's behavior into the above-mentioned driving intention model to obtain the driving intention of the driver.
  • the automatic driving device can compare the driving intention of the driver with the driving behavior of the vehicle under the control of the automatic driving system. If the driving intention of the driver is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system, the automatic driving device can determine the driver's driving behavior. The driving intention conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving device can determine the driver's driving intention according to the driver's operating behavior.
  • the driver's operating characteristics may include the driver's operation of the accelerator pedal, the driver's operation of the deceleration pedal, and the driver's operation of the steering wheel.
  • the method for the automatic driving device to determine the driving intention of the driver according to the operating characteristics of the driver will be described in detail below.
  • the automatic driving device may obtain the parameter characteristics corresponding to the driver's behavior through one or more sensors.
  • the sensor may include one or more of an accelerator pedal position sensor, a deceleration pedal position sensor, and a steering wheel torque sensor.
  • the accelerator pedal position sensor is used to detect the position change information of the accelerator pedal of the target vehicle
  • the deceleration pedal position sensor is used to detect the position change information of the deceleration pedal of the target vehicle
  • the steering wheel torque sensor is used to detect the angle information of the steering wheel rotation of the target vehicle.
  • the automatic driving device can determine the driver's operation behavior on the accelerator pedal based on the change information of the accelerator pedal.
  • the automatic driving device can determine the driver's operation behavior on the deceleration pedal through the change information of the deceleration pedal.
  • the automatic driving device can determine the driver's operation behavior on the steering wheel through the torque change information of the steering wheel. The following is based on the driver's operation behavior of the automatic driving device. Detect the driver’s driving intention to explain:
  • the automatic driving device obtains the driver's driving intention according to the driver's operation behavior on the steering wheel of the vehicle.
  • the automatic driving device detects that the position of the accelerator pedal of the target vehicle has moved from the first position to the second position, when the accelerator pedal of the target vehicle is in the first position, the speed of the target vehicle is the first position. Vehicle speed; when the accelerator pedal of the target vehicle is in the second position, the vehicle speed of the target vehicle is the second vehicle speed, and the second vehicle speed is greater than the first vehicle speed. Then the automatic driving device can determine that the driver's driving intention is to increase the driving speed of the target vehicle in the driving scene.
  • the automatic driving device obtains the driver's driving intention according to the driver's operation behavior on the deceleration pedal.
  • the automatic driving device detects that the position of the deceleration pedal of the target vehicle has moved from the third position to the fourth position, when the deceleration pedal of the target vehicle is in the third position, the speed of the target vehicle is the third position.
  • Vehicle speed When the deceleration pedal of the target vehicle is in the fourth position, the vehicle speed of the target vehicle is the fourth vehicle speed, and the fourth vehicle speed is less than the third vehicle speed. Then the automatic driving device can determine that the driver's driving intention is to reduce the driving speed of the target vehicle in the driving scene.
  • the automatic driving device obtains the driver's driving intention according to the driver's operation behavior on the steering wheel.
  • the torque change information of the steering wheel may include the change information of the first direction and the change information of the second direction of the steering wheel.
  • the first direction and the second direction are opposite directions. For example, as shown in FIG. 6c, when the steering wheel rotates in the first direction, the target vehicle rotates to the left of the traveling direction; when the steering wheel rotates in the second direction, The target vehicle turns to the right in the direction of travel.
  • the automatic driving device may determine that the driver's driving intention is that the target vehicle is expected to turn in the first direction.
  • the automatic driving device can determine the driver's driving intention according to the driver's emotional behavior.
  • the automatic driving device may determine that the driver's driving intention is not satisfied with the driving behavior of the target vehicle under the control of the automatic driving system in the driving scene.
  • the automatic driving device can also determine whether the driver is focused on the driving behavior of the target vehicle and whether it is performing the second task through the driver's body posture behavior.
  • the automatic driving device determines that the driving intention of the driver is not satisfied with the driving behavior of the vehicle under the control of the automatic driving system. That is, in this driving scenario, the driving intention of the driver conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the driver's visual behavior and body posture behavior can be used to reflect the driver's concentration on the driving behavior of the target vehicle.
  • the driver's visual behavior may include the driver's head behavior and gaze behavior.
  • the driver's head behavior may include the direction of the driver's head rotation, the angular speed of rotation, the angle of rotation, and so on.
  • the driver's gaze behavior includes the driver's blinking frequency, the duration of eye closure, and so on.
  • the driver's visual behavior may also include the driver's other eye behavior, which is not limited.
  • the driver's body posture behavior can also be used to determine the driver's second behavior.
  • the second behavior may be behaviors other than the driving behavior of focusing on the target vehicle.
  • the second behavior may include behaviors such as the driver using a mobile phone, and the driver talking with people in other locations.
  • the emotional behavior of the driver is used to reflect whether the driver is satisfied with the driving behavior of the target vehicle under the control of the automatic driving system.
  • the driver's emotional behavior can include the driver's facial expressions and language features.
  • the automatic driving device may obtain various images of the driver within a preset time period through the camera device.
  • the automatic driving device can recognize the multiple images to obtain the driver's visual behavior, body posture behavior, and facial expressions.
  • the preset time period can be preset according to needs.
  • the automatic driving device may obtain the driver's voice information through the sound collection device.
  • the sound collection device may be a microphone or the like.
  • the automatic driving device can recognize the driver's voice information to obtain the driver's language characteristics.
  • the facial expression of the driver may include a first facial expression and a second facial expression.
  • the language features of the driver may include a first language feature and a second language feature.
  • the automatic driving device may be preset with a first facial expression library and a second facial expression library. After acquiring the facial expression of the driver, the automatic driving device matches the facial expression of the driver with each facial expression in the first facial expression library and the second facial expression library, and determines that the driver’s facial expression belongs to the first The facial expression is still the second facial expression.
  • the automatic driving device may determine that the driver’s facial expression belongs to the first facial expression .
  • the automatic driving device may determine that the driver's facial expression belongs to the second facial expression.
  • the first preset threshold and the second preset threshold may be preset according to needs and are not limited.
  • Multiple sound collection devices may be provided in the target vehicle.
  • the multiple sound collection devices are in communication connection with the automatic driving device.
  • the automatic driving device can acquire the voice of a person located inside the target vehicle through the multiple sound collection devices, and recognize the voice of the person inside the target vehicle to determine the source of the sound. For example, the automatic driving device can identify the source of the sound based on the audio of the sound, and determine the source of the sound. When it is determined that the source of the sound includes multiple locations and the multiple locations include the driver's location, the automatic driving device determines that the driver is talking with people in other locations.
  • the target vehicle may be provided with a sound collection device 1 and a sound collection device 2.
  • the sound collection device 1 is used to collect the voices of the driver and the person located at the front passenger position
  • the sound collection device 2 is used to collect the voices of the person located in the back row of the target vehicle.
  • the sound collection device 1 and the sound collection device 2 can send the collected sounds to the automatic driving device for the automatic driving device to recognize the sound, so that the automatic driving device can determine whether the driver is talking with people in other locations.
  • the sound collection device 1 can collect the voice of the driver, and the sound collection device 2 can collect the voice of the person in the back row of the target vehicle, and the automatic driving device can determine that the driver is talking with the person in the back row.
  • the target vehicle can also be provided with a sound collection device at each seat, so that the automatic driving device can more accurately determine the source of the sound.
  • the method for determining the first language feature and the second language feature of the driver can refer to the above-mentioned method for determining the first facial expression and the second facial expression, which will not be repeated here.
  • the first facial feature and the first language feature can be used to characterize the driver's emotional behavior as positive emotional behavior.
  • the driver's emotional behavior is a positive emotional behavior
  • the driver's driving intention is: to be satisfied with the driving behavior of the target vehicle under the automatic control system.
  • the first facial feature may include happiness, satisfaction, etc.
  • the first language feature may include good, very good, etc.
  • the second facial feature and the second language feature can be used to characterize the driver's emotional behavior as negative emotional behavior.
  • the driver's emotional behavior is negative emotional behavior
  • the driver's driving intention is: not satisfied with the driving behavior of the target vehicle under the automatic control system.
  • the second facial expression may include worry, anger, etc.
  • Second language features can include really bad, not easy to use, and so on.
  • the automatic driving device may input the characteristic parameters corresponding to the driver's behavior into the driving intention model to obtain the driving intention of the driver.
  • Step 502 Based on the driving intention of the driver, the automatic driving device detects that there is a conflict between the driving behavior of the target vehicle under the automatic control system and the driving intention of the driver.
  • the automatic driving device can detect whether there is a conflict between the driver's driving intention and the driving behavior of the vehicle under the automatic control system according to the characteristic parameters corresponding to the driver's behavior. For example, when the driver’s relief corresponding characteristic parameter exceeds the preset range, and/or the driver’s corresponding characteristic parameter exceeds the preset range for a time greater than or equal to the first preset value, and/or the driver’s corresponding characteristic parameter When the number of times exceeding the preset range is greater than or equal to the second preset value, the automatic driving device may detect that there is a conflict between the driver's driving intention and the driving behavior of the vehicle under the automatic control system.
  • the number of times the driver's corresponding characteristic parameter exceeds the preset range is greater than or equal to the second preset value can also be described as the number of times the driver's corresponding characteristic parameter exceeds the preset range within the preset time is greater than or equal to the second default value.
  • the preset time can be set in advance according to needs.
  • the change range of the accelerator pedal corresponding to the automatic driving system of the target vehicle is the first range.
  • the first range can be preset according to needs.
  • the automatic driving device can detect the driving behavior of the target vehicle under the control of the automatic driving system and the driver's driving There is a conflict of intentions.
  • the change range of the deceleration pedal corresponding to the automatic driving system of the target vehicle is the second range.
  • the second range can be preset according to needs.
  • the automatic driving device can detect the driving behavior of the target vehicle under the control of the automatic driving system and the driver's driving There is a conflict of intentions.
  • the driving behavior of the target vehicle under the control of the automatic driving system is: the threshold of the angle at which the target vehicle rotates in the first direction is the first angle.
  • the first angle can be preset according to needs.
  • the automatic driving device can detect that the driving behavior of the target vehicle under the control of the automatic driving system conflicts with the driving intention of the driver.
  • the first threshold to the sixth threshold can be preset as needed.
  • the automatic driving device can determine the conflict time between the driver's driving intention and the driving behavior of the vehicle under the control of the automatic driving system /The number of conflicts, if the conflict time is greater than or equal to the preset time/the number of conflicts is greater than or equal to the preset number, the automatic driving device can determine that the driver's driving intention conflicts with the driving behavior of the target vehicle under the control of the automatic driving system.
  • the preset time and the preset number of times can be preset and are not limited.
  • the following takes the automatic driving device to determine the driver's driving intention according to the driver's visual behavior as an example, and the method for the automatic driving device to determine the conflict time and the number of conflicts between the driver's driving intention and the driving behavior of the vehicle under the control of the automatic driving system. Specific instructions.
  • the automatic driving device determines the conflict time between the driver's driving intention and the vehicle's driving behavior under the control of the automatic driving system.
  • the automatic driving device can count the conflict time between the driver's driving intention and the driving behavior of the vehicle under the control of the automatic driving system. If the conflict time is greater than or equal to the preset time, the automatic driving device can determine the driver's driving intention and the vehicle in automatic Conflict of driving behavior under the control of the driving system. If the conflict time is less than the preset time, the automatic driving device can determine that the driver's driving intention does not conflict with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving device is provided with a timer. Once the automatic driving device detects that the driver’s driving intention is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system, it can trigger the timer to start timing. The steps are:
  • the automatic driving device obtains the driver's visual behavior, and inputs the visual behavior into the above-mentioned driver's driving intention model to obtain the driver's driving intention.
  • the automatic driving device can repeat the above steps until the automatic driving device determines the driver's behavior according to the driver's visual behavior.
  • the driving intention is consistent with the driving behavior of the vehicle under the control of the automatic driving system.
  • the timer is triggered to stop timing.
  • the automatic driving device determines the conflict time between the driving intention of the driver and the driving behavior of the vehicle under the control of the automatic driving system according to the time when the timer starts and the time when the timer stops.
  • the automatic driving device can periodically or randomly obtain the driver's visual behavior, and determine the driver's driving intention, without limitation.
  • the automatic driving device determines the driver The driving intention of the vehicle conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving device can detect the target vehicle's The driving behavior under the control of the automatic driving system conflicts with the driver's driving intention.
  • the automatic driving device is provided with a counter. Once the automatic driving device detects that the driver’s driving intention is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system, the automatic driving device acquires multiple visual behaviors of the driver within a preset period of time, and combines the multiple visual behaviors Input into the aforementioned driving intention model to obtain the driving intention of the driver corresponding to each of the multiple visual behaviors. The automatic driving device can respectively compare whether the driving intention of the driver corresponding to each visual behavior is consistent with the driving behavior of the vehicle under the control of the automatic driving system.
  • the counter will record the first value; if the driving intention of the driver corresponding to a visual behavior is the same as that of the vehicle under the control of the automatic driving system If the driving behavior is consistent, the counter records the second value. If the number of the first values counted by the counter is greater than or equal to the preset value, the automatic driving device may determine that the driving intention of the driver conflicts with the driving behavior of the vehicle under the control of the automatic driving system.
  • the preset time period and preset value are set in advance and are not limited.
  • the automatic driving device determines that the driver’s driving intention at the first moment is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system.
  • the visual behavior of the driver corresponding to the time such as: visual behavior 1, visual behavior 2,..., visual behavior N, where N is a positive integer greater than or equal to 1, and the time gap between the N moments can be the same , Can also be different, and there is no restriction.
  • the automatic driving device can determine the driving intention of the driver corresponding to each of the N visual behaviors according to the above-mentioned driving intention model of the driver.
  • the driving intention of the driver determined by the automatic driving device includes driving intention 1. 2,..., driving intention N, where visual behavior corresponds to driving intention one-to-one, for example, visual behavior 1 corresponds to driving intention 1, visual behavior 2 corresponds to driving intention 2, and visual behavior N corresponds to visual behavior N.
  • the automatic driving device respectively compares whether the above N driving intentions are consistent with the driving behavior of the vehicle under the control of the automatic driving system. Among them, if they are inconsistent, the counter can record the first value, for example, the first value is 1. If they are consistent, the counter can record the second value, for example, the second value can be 0. For example, the automatic driving device compares whether the driving intention 1 is consistent with the driving behavior of the vehicle under the control of the automatic driving system. Whether the driving behavior under the control of the driving system is consistent, if not, the counter can be increased by 1 on the basis of 1, that is, the counter can be recorded as 2. Until the automatic driving device has compared the N driving intentions with the driving behavior of the vehicle under the control of the automatic driving system.
  • the automatic driving device detects that the driver's driving intention is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system, the automatic driving device acquires the driver's visual behavior and inputs the visual behavior In the above driving intention model, the driving intention of the driver corresponding to the visual behavior is obtained. The automatic driving device compares whether the driving intention of the driver corresponding to the visual behavior is consistent with the driving behavior of the vehicle under the control of the automatic driving system.
  • the counter can record the first value; if the driver's driving intention is consistent with the driving behavior of the vehicle under the control of the automatic driving system, the counter You can record the second value.
  • the automatic driving device can cyclically execute the above steps 3 to 4 within a preset time period, and the automatic driving device can determine that the driver’s driving intention and the vehicle’s driving intention within the preset time period are under the control of the automatic driving system The number of inconsistent driving behaviors.
  • the automatic driving device determines that the driver’s driving intention at the first moment is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system. Whether the driver’s driving intention corresponding to the driver’s visual behavior 2 at the moment is consistent with the driving behavior of the vehicle under the control of the automatic driving system, if inconsistent, the counter is recorded as 1; if it is consistent, the counter does not count or is recorded as 0. In this way, the automatic driving device can repeat the above steps 4 and 4 to determine the number of times the driver's driving intention is inconsistent with the driving behavior of the vehicle under the control of the automatic driving system within the preset time period.
  • the automatic driving device can determine that the driver’s driving intention is: dissatisfied with the target vehicle’s driving behavior under the control of the automatic driving system, that is, , The automatic driving device can detect that the driving behavior of the target vehicle under the control of the automatic driving system conflicts with the driving intention of the driver.
  • Step 503 The automatic driving device updates the automatic driving system.
  • the driving behavior of the target vehicle under the control of the updated automatic driving system matches the driving intention of the driver.
  • the automatic driving device can obtain the first characteristic parameter according to the driving intention of the driver and the driving data corresponding to the driving behavior of the target vehicle under the control of the automatic driving system; the automatic driving device inputs the first characteristic parameter
  • the neural network model is preset to obtain the first driving behavior; the automatic driving device updates the automatic driving system according to the first driving behavior.
  • the first characteristic parameter represents the driving intention of the driver and the driving data corresponding to the driving behavior of the target vehicle under the control of the automatic driving system.
  • the first feature parameter may be a feature vector.
  • the feature vector may include multiple parameters, and each parameter of the multiple parameters may be used to represent a piece of driving data.
  • the first characteristic parameter may include the distance between the target vehicle and other vehicles, the speed of the target vehicle, the driving direction, and the steering angle.
  • the preset neural network model is used to determine the driving behavior that matches the driver's driving intention.
  • the preset neural network model can be preset.
  • the preset neural network model may be a deep feedforward neural network regression model.
  • the preset neural network model is obtained by training based on multiple historical driving data of the driver.
  • the input of the preset neural network model is the characteristic parameter, and the output is the first driving behavior that satisfies the driver's driving intention. That is, the first driving behavior is a driving behavior that satisfies the driver's driving intention.
  • the first driving behavior may include one or more driving data.
  • the one or more driving data may be the speed of the vehicle, the distance to the preceding vehicle, and the like.
  • the training method of the preset neural network model can refer to the prior art, which will not be repeated here.
  • the automatic driving device may input the above-mentioned characteristic parameters into the preset neural network model in the form of a characteristic vector to obtain an output vector corresponding to the characteristic vector.
  • the output vector may include one or more parameters, and each of the one or more parameters corresponds to a driving behavior.
  • the automatic driving device updates the automatic driving system according to the first driving behavior means that the automatic driving device can compare the first driving behavior with the driving behavior of the vehicle under the control of the automatic driving system, and update the automatic driving system according to the comparison result.
  • the automatic driving device can replace the driving behavior of the vehicle under the control of the automatic driving system with the first driving behavior.
  • the driving behavior of the target vehicle under the control of the automatic driving system is to keep the speed of the target vehicle at the second speed.
  • the first vehicle speed is greater than the second vehicle speed.
  • the automatic driving device may replace the vehicle speed of the target vehicle under the control of the automatic driving system with the first vehicle speed, so that the vehicle speed of the target vehicle under the control of the automatic driving system is the first vehicle speed.
  • the automatic driving device may cyclically execute the above steps 501 to 503 until the target vehicle stops running, where stopping the target vehicle includes: the driver controls the target vehicle to stop running, Or the driver turns off the automatic driving system of the target vehicle, for example, the driver manually drives.
  • the automatic driving device updates the automatic driving system when it detects that the driving intention of the driver of the automatic driving vehicle under the control of the automatic driving system conflicts with the driving behavior of the vehicle under the control of the automatic driving system of the vehicle, So that the driving behavior of the vehicle under the control of the updated automatic driving system matches the driving intention of the driver. Based on this, the subsequent driving behavior under the control of the updated automatic driving system can meet the driving needs of the driver. Thus, the problem of conflict between the driving behavior of the vehicle under the control of the automatic driving system and the driving intention of the driver is solved.
  • the method provided in the embodiment of the present application may further include: the automatic driving device detects whether the driver of the target vehicle is a new driver.
  • the new driver refers to a person who has not driven the target vehicle.
  • the automatic driving device when the automatic driving device detects that the target vehicle is started, the automatic driving device can obtain the characteristic information of the driver, and match the characteristic information of the driver with the characteristic information of multiple drivers possessed by the automatic driving device. If the characteristic information of the driver does not match the characteristic information of each driver in the characteristic information of the multiple drivers, the automatic driving device determines that the driver of the target vehicle is a new driver; if the characteristic of the driver If the information matches the characteristic information of one driver among the characteristic information of the plurality of drivers, the automatic driving device determines that the driver of the target vehicle is the driver who has driven the target vehicle.
  • the automatic driving device may also determine the driving of the target vehicle under the control of the automatic driving system according to the characteristic information of the driver of the target vehicle. model.
  • the automatic driving device determines the automatic driving mode of the target vehicle according to the characteristic information of the driver. That is, if the driver of the target vehicle is a driver who has driven the target vehicle, the automatic driving device may determine the automatic driving mode of the target vehicle according to the characteristic information of the driver.
  • the automatic driving device may obtain the automatic driving mode used by the driver from the V2X server shown in FIG. 4, and use the used automatic driving mode as the automatic driving mode of the target vehicle.
  • the automatic driving device may also store the historical automatic driving mode used by the driver, and the automatic driving device may use the stored automatic driving mode used by the driver as the automatic driving mode of the target vehicle without limitation.
  • the automatic driving device may use the preset automatic driving mode as the automatic driving mode of the target vehicle.
  • the preset automatic driving mode can be pre-configured in the automatic driving device.
  • the automatic driving device can also obtain the preset automatic driving mode from the historical automatic driving mode used by multiple drivers in the V2X server, without limitation. For example, the automatic driving device may select the most frequently used automatic driving mode from among the historical automatic driving modes used by multiple drivers as the preset automatic driving mode.
  • the method provided in the embodiment of the present application may further include: the automatic driving device obtains the driving behavior of the target vehicle under the control of the automatic driving system.
  • the target vehicle has different driving behaviors under the control of the automatic driving system.
  • the driving scene and the driving behavior of the target vehicle corresponding to the driving scene under the control of the automatic driving system may be as shown in Table 1, which will not be repeated.
  • the automatic driving device may obtain the driving intention of the driver in the first driving scene and the driving behavior of the target vehicle corresponding to the first driving scene under the control of the automatic control system. If the automatic driving device detects that there is a conflict between the driver's driving intention in the first driving scene and the driving behavior of the target vehicle under the control of the automatic control system, the automatic driving device may update the automatic driving system.
  • the automatic driving device may update the automatic driving during the driving of the target vehicle.
  • the automatic driving device can also update the automatic driving behavior corresponding to the first driving scene in the automatic driving system after the target vehicle stops driving.
  • the automatic driving device can update the first driving scene in the automatic driving system after the target vehicle reaches the destination.
  • the autonomous driving behavior corresponding to a driving scene is not restricted.
  • the automatic driving device may record multiple driving scenarios that conflict with the driving intention of the driver and the driving intention of the driver.
  • the driving scene recorded by the automatic driving device includes driving scene 1 and driving scene 2.
  • driving scene 1 and driving scene 2 are driving scenes in which the driving behavior of the vehicle under the control of the automatic driving system conflicts with the driving intention of the driver.
  • the automatic driving device may update the driving behavior of the vehicle corresponding to driving scene 1 under the control of the automatic driving system and the driving behavior of the vehicle corresponding to driving scene 2 under the control of the automatic driving system after the automatic driving of the vehicle ends.
  • the automatic driving device may acquire the first driving scene of the target vehicle in multiple ways.
  • the multiple methods may include: the automatic driving device acquires the first driving scene of the target vehicle through the equipment of the target vehicle, such as: automatic The driving device can obtain the first driving scene of the target vehicle through a variety of sensors.
  • the automatic driving device obtains the first driving scene of the target vehicle from the interaction of other devices.
  • the automatic driving device can obtain the first driving scene of the target vehicle through the interaction with the V2X server. The interaction to obtain the first driving scene of the target vehicle.
  • the automatic driving device may also obtain the first driving scene of the target vehicle according to one or more of the above-mentioned multiple methods, which is not limited. The above multiple methods are described below:
  • the automatic driving device can obtain the first driving scene of the target vehicle through a variety of sensors.
  • the automatic driving device can collect the traffic information of the target vehicle driving road through various sensors, and integrate the collected traffic information of the target vehicle driving road to obtain the first driving scene of the target vehicle.
  • various sensors may include laser radar, millimeter wave radar, ultrasonic radar, monocular or binocular cameras, etc., which are not limited.
  • the automatic driving device can obtain the speed of vehicles around the target vehicle and the speed of pedestrians through lidar.
  • the automatic driving device can obtain road condition information, the number of vehicles, the number of pedestrians, etc. of the target vehicle's driving road through monocular or binocular cameras, and there is no restriction.
  • the automatic driving device can obtain the first driving scene of the target vehicle through interaction with the V2X server.
  • the automatic driving device may send request information for requesting the first driving scene to the V2X server.
  • the V2X server receives the request information for requesting the first driving scene from the automatic driving device.
  • the V2X server determines the first driving scene according to the request message, and sends the first driving scene to the automatic driving device.
  • the automatic driving device can send the coordinate information of the target vehicle (such as global positioning system (GPS) coordinates) to the V2X server through the body antenna in Figure 2, and the V2X server receives the coordinate information of the target vehicle, According to the coordinate information, the driving scene corresponding to the coordinate information is determined.
  • the V2X server sends the first driving scene corresponding to the coordinate information to the automatic driving device.
  • the automatic driving device may receive the first driving scene corresponding to the coordinate information through the vehicle body antenna.
  • the automatic driving device can obtain the first driving scene of the target vehicle by interacting with other vehicles.
  • this manner can refer to the manner in which the first driving scene of the target vehicle is acquired by the interaction between the automatic driving device and the V2X server, which will not be repeated here.
  • the automatic driving device may acquire the behavior characteristics of the driver in the first driving scene, and determine the driving intention of the driver according to the behavior characteristics of the driver.
  • the driver's behavior characteristics can be used to characterize whether the driver is satisfied with the driving behavior of the target vehicle under the control of the automatic driving system.
  • the driver's behavior characteristics can include the driver's operating characteristics, visual behavior, emotional behavior, and physical
  • the automatic driving device can determine the driver's driving intention based on one or more of these behavior characteristics.
  • the relevant description of the driver's behavior characteristics and the implementation of the on-board device to determine the driver's driving intention according to the driver's behavior characteristics can be referred to the technical solution shown in FIG. 5.
  • the automatic driving device when the automatic driving device detects that there is a conflict between the driving intention of the driver of the vehicle in the first driving scene and the driving behavior of the vehicle under the control of the automatic driving system, the automatic driving device can use the automatic driving system
  • the corresponding driving behavior is adjusted to a driving behavior that meets the driver's driving intention in the first driving scene, so that the subsequent automatic driving device faces the type of driving scene and the driving behavior of the vehicle under the control of the automatic driving system can satisfy the driver Demand.
  • the method provided in the embodiment of the present application may further include: the automatic driving device determines the automatic driving mode of the target vehicle in the first driving scene.
  • the relevant description of the automatic driving mode can refer to the above, and it will not be repeated.
  • the automatic driving device determining the automatic driving mode of the target vehicle in the first driving scene may include: the automatic driving device automatically selects, for example: the automatic driving device automatically selects from a plurality of pre-configured automatic driving modes according to some relevant information of the first driving scene
  • the automatic driving mode of the target vehicle is selected in the, or the driver manually selects, for example, when the driver wants automatic driving, the driver manually selects the appropriate automatic driving mode, and triggers the automatic driving device according to the automatic driving selected by the driver Mode for automatic driving.
  • this method can refer to the following method 1 or method 2.
  • Method 1 Automatic selection by the autopilot device.
  • the automatic driving device can select an automatic driving mode from one or more automatic driving modes according to the characteristic information of the driver. Among them, the characteristic information of a driver is used to uniquely identify a driver.
  • the one or more automatic driving modes may be an automatic driving mode stored by the automatic driving device, or may be an automatic driving mode obtained by the automatic driving device from the V2X server in FIG. 2.
  • the automatic driving device may obtain the image information of the driver through the camera device, and the image information may be video information or picture information.
  • the image information may include facial features or pupil features of the driver.
  • the automatic driving device can identify the driver's characteristic information by recognizing the driver's image information, or the automatic driving device can send the driver's image information to the V2X server so that the V2X server can recognize the driver's image information , Get the driver's characteristic information.
  • the automatic driving device may match the characteristic information of the driver with the characteristic information of multiple drivers possessed by the automatic driving device, and determine the automatic driving mode that matches the characteristic information of the driver.
  • the characteristic information of each driver in the characteristic information of the plurality of drivers has a corresponding automatic driving mode.
  • the automatic driving device may use the automatic driving mode corresponding to the characteristic information of the driver of the target vehicle with the characteristic information of the driver of the target vehicle whose similarity is greater than the preset value among the characteristic information of the multiple drivers as the driver of the target vehicle Autopilot mode.
  • the preset value can be preset by the automatic driving device and is not limited.
  • the automatic driving device may also send the acquired characteristic information of the driver to the V2X server.
  • the V2X server may compare the characteristic information of the driver with one or more automatic driving information. The characteristic information of the driver corresponding to the mode is matched, and the automatic driving mode that matches the characteristic information of the driver is selected from one or more automatic driving modes.
  • the V2X server may send the automatic driving mode that matches the characteristic information of the driver to the automatic driving device.
  • the automatic driving device receives the automatic driving mode matching the driver's characteristic information from the V2X server, it can control the target vehicle to automatically drive according to the automatic driving mode matching the driver's characteristic information.
  • the automatic driving mode may be an automatic driving mode manually selected by the driver after starting the vehicle.
  • the target vehicle may be provided with a display device.
  • the display device may be triggered to display the display interface shown in FIG. 8, and the driver may select the automatic driving mode through the display interface according to requirements.
  • the automatic driving device detects the operation instruction of the driver, it can determine the automatic driving mode corresponding to the operation instruction according to the operation instruction of the driver.
  • the display interface in FIG. 8 may include multiple touch buttons from “Mode 1" to "Mode 6", and each touch button corresponds to an automatic driving mode.
  • the automatic driving device can be triggered to obtain the automatic driving mode corresponding to the touch button.
  • the touch button "Mode 5" the automatic driving device can be triggered to obtain the automatic driving mode corresponding to the touch button "Mode 5".
  • the driver can manually select the automatic driving mode through the terminal corresponding to the target vehicle, and trigger the terminal corresponding to the target vehicle to notify the automatic driving device of the selected automatic driving mode.
  • the driving device determines the automatic driving mode of the target vehicle in the first driving scene according to the notification from the terminal.
  • the terminal corresponding to the automatic driving device may include any device that controls the target vehicle, such as a smart phone, a mobile hard disk, a personal notebook, a tablet computer, etc., without limitation.
  • the driver uses the mobile phone to send the automatic driving mode of the terminal such as the mobile phone to the automatic driving device in a wired or wireless manner.
  • the wired method may be a universal serial bus (universal serial bus).
  • Serial bus, USB), Type-C, etc. are connected to automatic driving devices.
  • the wireless method can be connected to the autopilot device through Bluetooth, wireless fidelity (wireless fidelity, WiFi), etc., and there is no restriction.
  • the automatic driving device can receive the automatic driving mode from the smart phone.
  • the method may further include:
  • the automatic driving device determines the first characteristic parameter according to the driving intention of the driver and the driving behavior of the vehicle under the control of the automatic driving system.
  • the first characteristic parameter may be a characteristic vector.
  • driving scenes are as follows: urban rush forward follow-up, urban peak forward follow-up, and urban peak vehicle actively cuts in, the automatic driving device is based on the driver’s driving intention and the vehicle’s The driving behavior under the control of the automatic driving system is explained by the determined feature vector.
  • the automatic driving device may indicate the driving intention and vehicle information of the driver with numbers or characters, so as to determine the characteristic parameters corresponding to the driving scene.
  • v+ means that the driver’s driving intention is to increase the speed of the vehicle
  • v- means that the driver’s driving intention is to reduce the speed of the vehicle
  • s means that the driver’s driving intention is to pull over
  • s means that the driver’s driving intention is to left.
  • Lanes move closer r means that the driver’s driving intention is to move closer to the right lane
  • l+v means that the driver’s driving intention is to increase the speed of the vehicle when moving closer to the left lane
  • lv means that the driver’s driving intention is to lower when moving closer to the left lane
  • the driver's driving intention can also be expressed by other driving, which is not restricted.
  • the vehicle type can be represented by numbers or characters. For example, taking the number to indicate the vehicle type as an example, "1" can indicate that the vehicle type is a car, "2" can indicate that the vehicle type is a truck, and "3" can indicate that the vehicle type is a bus, and there is no restriction.
  • Driving scene 1 The driving scene is going forward and following the car at the peak of the city.
  • the automatic driving device can construct a feature vector based on the vehicle information of the target vehicle and the vehicle information of the preceding vehicle.
  • the feature vector determined by the automatic driving device may be:
  • d f represents a distance in front of the vehicle and the target vehicle
  • the vehicle speed V F represents the front of the vehicle
  • v e represents the target vehicle speed
  • i f represents the type of the vehicle front of the vehicle.
  • Each variable in the feature vector may include multiple dimensions.
  • the size of the vehicle may include three dimensions: vehicle length, vehicle width, and vehicle height.
  • the vehicle type of the vehicle may include cars, trucks, buses, and so on.
  • Driving scene 2 The driving scene is the traffic lane keeping in the city peak.
  • the driving parameters of the target vehicle are related to the vehicle information of the preceding vehicle, the left vehicle, and the right vehicle.
  • the vehicle information and the vehicle information of the vehicle on the right construct a feature vector.
  • the feature vector determined by the vehicle device may be:
  • f ⁇ v-, d f , v f , v e , s f , i f , i l , s l , v lv , v ll , d l , i r , s r , v rv , v rl , dr ⁇ .
  • d f, v f, v e, s f, i f can be described with reference to a scene, and is not repeated here.
  • i l represents the type of the left vehicle
  • s l represents the size of the left vehicle
  • v lv represents the lateral speed of the left vehicle
  • v ll represents the longitudinal speed of the left vehicle
  • d l represents the lateral relative between the left vehicle and the target vehicle distance.
  • the lateral relative distance refers to the horizontal distance between two vehicles.
  • i r denotes the type of the vehicle to the right
  • s r indicates the size of the right of the vehicle
  • v rv right direction represents the lateral speed of the vehicle
  • v rl represents the longitudinal speed of the vehicle to the right
  • d r represents the lateral rightward of the vehicle relative to the target vehicle distance.
  • the automatic driving device can complement the feature vector according to the following criteria:
  • the type of the vehicle in front is a bicycle; the size of the vehicle in front is small enough, such as ⁇ 0,0,0 ⁇ ; the distance between the vehicle in front and the target vehicle is a larger value, such as 3000m; the speed of the vehicle in front is a larger value, such as 800km/h.
  • the automatic driving device's method of complementing the parameters in the feature vector can refer to the above-mentioned automatic driving device for when there is no vehicle in front of the target vehicle, the automatic driving device compares the feature vector The processing method of complementing the parameters in, will not be repeated here.
  • Driving scene 3 The driving scene is the active cut-in of the city peak vehicle.
  • the automatic driving device can construct a feature vector based on the vehicle information of the target vehicle and the cut-in vehicle.
  • the feature vector determined by the automatic driving device may be:
  • d l represents the longitudinal distance between the target vehicle and the vehicle being cut into
  • d v represents the lateral distance between the target vehicle and the vehicle being cut into
  • v ev represents the cut speed of the target vehicle
  • v el represents the longitudinal distance when the target vehicle cuts in.
  • i represents the type of vehicle being cut into
  • s represents the size of the vehicle being cut into.
  • the automatic driving device inputs the characteristic parameters into the preset neural network model to obtain the first driving behavior.
  • the preset neural network model is obtained by training based on the driver's historical driving data in multiple driving scenarios.
  • the input of the preset neural network model is the characteristic parameter, and the output is the first driving behavior that satisfies the driver's driving intention. That is, the first driving behavior is a driving behavior that satisfies the driver's driving intention.
  • the first driving behavior may include one or more driving data.
  • the one or more driving data may be the speed of the vehicle, the distance to the preceding vehicle, and the like.
  • the preset neural network model may be preset.
  • the preset neural network model may be a deep feedforward neural network regression model.
  • the preset neural network model can be obtained by training based on multiple sample feature parameters, and one sample feature parameter corresponds to the driving behavior of a driver.
  • the training method of the preset neural network model can refer to the prior art, which will not be repeated here.
  • the automatic driving device may input the above-mentioned characteristic parameters into the preset neural network model in the form of a characteristic vector to obtain an output vector corresponding to the characteristic vector.
  • the output vector may include one or more parameters, and each of the one or more parameters corresponds to a driving behavior.
  • the driver’s driving intention is to reduce the speed of the vehicle.
  • the distance between the target vehicle and the vehicle in front is 5m
  • the speed of the vehicle in front of the target vehicle is 40km/h
  • the speed of the target vehicle is 50km/h
  • the length of the vehicle in front is 40km/h. It is 4.6m, 1.7m wide and 1.4m high.
  • the vehicle type of the vehicle in front is a sedan.
  • the automatic driving device inputs the feature vector into the preset neural network model, and the output vector obtained is ⁇ 40, 0, 0, 0 ⁇ .
  • 40 indicates that the speed of the target vehicle is 40km/h
  • the last three 0s respectively indicate that the target vehicle does not slow down, does not pull over, and does not exit the following mode. That is, the first driving behavior is to control the speed of the target vehicle to 40 km/h.
  • the automatic driving device presents one or more questions to the driver. And receive the driver's answer to the one or more questions. The answer to the one or more questions can be used to indicate whether to update the automatic driving system.
  • the automatic driving device may select one or more questions from a preset question library according to the first driving behavior and the first driving scene.
  • the automatic driving device may play the one or more questions through the voice playback device, or may display the one or more questions through the display device.
  • the preset question library is determined according to the driving behaviors of multiple drivers within a preset time.
  • the realization process of this realization method can be:
  • the automatic driving device compares the driving behavior of the vehicle under the control of the automatic driving system with the first driving behavior, and determines whether the driving behavior of the vehicle under the control of the automatic driving system is consistent/same as the first driving behavior. For example, if the difference between the driving behavior of the vehicle under the control of the automatic driving system and the first driving behavior satisfies the preset condition/does not exceed the preset range, the automatic driving device determines that the driving behavior of the vehicle under the control of the automatic driving system is different from that of the first driving behavior.
  • the first driving behavior is consistent/same; if the difference between the driving behavior of the vehicle under the control of the automatic driving system and the first driving behavior does not meet the preset condition or exceeds the preset range, the automatic driving device determines that the vehicle is in the automatic driving system
  • the driving behavior under control is inconsistent/different from the first driving behavior.
  • the vehicle is under the control of the automatic driving system to increase the speed of the target vehicle to 80km/h
  • the first driving behavior is to increase the speed of the target vehicle to 81km/h.
  • the driving behavior of the vehicle under the control of the automatic driving system is to control the distance between the target vehicle and the preceding vehicle to be 10m
  • the first driving behavior is to control the target vehicle to the front
  • the distance of the vehicle is 15m
  • the preset condition is that the distance between the control target vehicle and the preceding vehicle does not exceed 12m. Since 15m is greater than 12m, that is, the first driving behavior does not meet the preset condition, the automatic driving device determines that the driving behavior of the vehicle under the control of the automatic driving system is inconsistent/different from the first driving behavior.
  • the automatic driving device can determine that the driving behavior of the vehicle under the control of the automatic driving system does not conflict with the driving intention of the driver;
  • the driving behavior under the control of the automatic driving system is inconsistent/different from the first driving behavior.
  • the automatic driving device can compare the driving behavior of the vehicle under the control of the automatic driving system with the first driving behavior, and determine the driving behavior of the vehicle under the control of the automatic driving system.
  • driving data that is inconsistent with the driving data of the first driving behavior and change information of the inconsistent driving data.
  • the automatic driving device matches the inconsistent/different driving data and the change information of the inconsistent/different driving data with a preset problem library, and determines the problem corresponding to the inconsistent/different driving data.
  • the automatic driving device broadcasts in the form of voice, or displays the problem corresponding to the inconsistent driving data through the display device.
  • the automatic driving device receives the driver's voice through a microphone, recognizes the driver's voice, and determines the driver's answer to the one or more questions. Wherein, the driver's voice is the driver's answer to the question corresponding to the inconsistent driving data.
  • the automatic driving device can play the question corresponding to the inconsistent driving data again, and the automatic driving device can also use the preset answer as the driver's answer .
  • the preset time can be preset by the automatic driving device and is not limited. For example, taking the preset time of 10s as an example, if the autopilot device does not receive the driver's voice within 10s, the preset answer of the autopilot device is an affirmative answer, for example, the affirmative answer may be "yes".
  • the automatic driving device may update the automatic driving system.
  • the answer to the one or more questions is "yes"
  • the answer to the one or more questions can be used to instruct to update the automatic driving system.
  • the questions in the question library may be set as simple and direct questions, for example, the set questions may be judged questions.
  • the answer given by the driver can be "yes" or "no".
  • the corresponding question library may be as shown in Table 2.
  • the autopilot device determines the driver’s driving The intention is to increase the speed of the vehicle.
  • the automatic driving device compares the driving behavior of the vehicle under the control of the automatic driving system (the speed of the target vehicle is 50km/h) with the first driving behavior (the speed of the target vehicle is 40km/h) to determine that the vehicle is controlled by the automatic driving system
  • the driving data in which the next driving behavior is inconsistent with the first driving behavior is the speed of the target vehicle, and the change information of the speed of the target vehicle is that the speed is reduced from 50km/h to 40km/h.
  • the automatic driving device matches the vehicle speed of the target vehicle and the change information of the vehicle speed with the vehicle speed and the problem in Table 2, and determines that the problem corresponding to the vehicle speed of the target vehicle is Problem 2 in Table 2.
  • the automatic driving device plays question 2 in Table 2 in the form of voice.
  • the automatic driving device receives the driver's voice through a microphone, and recognizes the driver's voice to determine the driver's answer.
  • the automatic driving device determines that the driver's answer is "yes" based on the recognized voice, the automatic driving device can determine that the driver's driving intention is to reduce the speed of the target vehicle.
  • the automatic driving device reduces the speed of the target vehicle to 40km/h, and updates the speed of the vehicle in the driving behavior under the control of the automatic driving system to 40km/h .
  • the automatic driving device determines that the driver's answer is "No" based on the recognized voice, the automatic driving device can determine that the driver's driving intention is not to reduce the speed of the target vehicle.
  • the automatic driving device can also continue to play question 1 in Table 2. If the automatic driving device determines that the answer of the driver corresponding to question 1 in Table 2 is "No", the automatic driving device does not update the driving behavior of the vehicle under the control of the automatic driving system.
  • the corresponding problem library may be as shown in Table 3.
  • Question 2 Does it slow down?
  • Question 3 Do you move closer to the left lane?
  • Question 4 Do you move closer to the right lane?
  • Question 5 Do you pull over and park?
  • Question 6 Do you want to exit lane keeping mode?
  • the corresponding problem database may be as shown in Table 4.
  • Question 1 Whether to increase the cutting speed?
  • Question 2 Whether to reduce the cutting speed?
  • Question 3 Whether to increase the cutting distance?
  • Question 4 Whether to reduce the cutting distance?
  • Question 5 Whether to increase the cutting distance and cutting speed at the same time?
  • Question 6 Whether to reduce the cutting distance and increase the cutting speed?
  • Question 7 Whether to increase the cutting distance and reduce the cutting speed?
  • Question 8 Whether to reduce the cutting distance and cutting speed at the same time?
  • Question 9 Do you pull over and park?
  • Question 10 Do you want to exit the active cut-in mode?
  • the automatic driving device may also detect whether the first driving behavior is included in the preset driving behavior.
  • the preset driving behavior may be a safe driving behavior that conforms to the target vehicle.
  • the automatic driving device may regard the first driving behavior as the driving behavior of the target vehicle under the control of the updated automatic driving system; if the first driving behavior does not conform to the target vehicle For safe driving behavior, the automatic driving device can regard the safe driving behavior as the driving behavior of the target vehicle under the control of the updated automatic driving system.
  • the automatic driving device determines that the first driving behavior is to increase the speed of the target vehicle to 80 km/h. However, if the maximum speed of the restricted vehicle on a certain road on which the target vehicle is traveling is 60km/h, the automatic driving device detects that the first driving behavior does not conform to the safe driving behavior of the target vehicle, and the automatic driving device can take the speed of 60km/h as The speed of the vehicle under the control of the updated autopilot system. That is, in the subsequent automatic driving process of the target vehicle, the speed of the target vehicle under the control of the automatic driving system is 60 km/h.
  • the automatic driving device may send prompt information, which is used to prompt the driver that the target vehicle is going to speed.
  • the method provided in the embodiment of the present application may further include: the automatic driving device determines the characteristic information of the driver of the target vehicle in the second driving scene. When the first driving scene matches the second driving scene, and the characteristic information of the driver in the first driving scene is consistent with the characteristic information of the driver in the second driving scene, the automatic driving device determines the driving of the target vehicle in the second driving scene The behavior is the driving behavior under the updated autopilot system.
  • the automatic driving mode used by the target vehicle in the second driving scene is the updated automatic driving mode.
  • the matching of the first driving scene and the second driving scene means that the similarity of the first driving scene is greater than or equal to a preset value.
  • the characteristic information of the driver in the first driving scene is consistent with the characteristic information of the driver in the second driving scene means that the driver of the target vehicle in the first driving scene and the driver of the target vehicle in the second driving scene are the same driver .
  • the automatic driving device when facing the same or similar driving scene in the subsequent, can control the target vehicle to drive according to the driving behavior of the updated automatic driving system, which solves the problem of the same or similar driving scene in the first driving scene.
  • the driving behavior of the vehicle under the control of the automatic driving system still conflicts with the driver's driving intention.
  • FIG. 5 The method shown in FIG. 5 will be described in detail below in conjunction with the communication system shown in FIG. 1.
  • another method for adaptively optimizing an automatic driving system may include:
  • FIG. 5 The method shown in FIG. 5 will be described in detail below in conjunction with the communication system shown in FIG. 1.
  • the method for adaptively optimizing an automatic driving system may include:
  • Step 901 The automatic driving device detects whether the driver of the target vehicle is a new driver.
  • Step 902 The automatic driving device determines the automatic driving mode of the target vehicle according to the characteristic information of the driver.
  • the automatic driving device may determine the automatic driving mode of the target vehicle according to the characteristic information of the driver.
  • the automatic driving device may obtain the automatic driving mode used by the driver from the V2X server shown in FIG. 4, and use the used automatic driving mode as the automatic driving mode of the target vehicle.
  • the automatic driving device may also store the historical automatic driving mode used by the driver, and the automatic driving device may use the stored automatic driving mode used by the driver as the automatic driving mode of the target vehicle without limitation.
  • Step 903 The automatic driving device obtains characteristic parameters corresponding to the behavior of the driver in the first driving scene.
  • Step 904 The automatic driving device determines the driver's driving intention according to the characteristic parameter corresponding to the driver's behavior.
  • Step 905 The automatic driving device determines the first characteristic parameter according to the driving intention of the driver and the driving behavior of the vehicle under the control of the automatic driving system.
  • Step 906 The automatic driving device inputs the first characteristic parameter into the preset neural network model to obtain the first driving behavior.
  • Step 907 The automatic driving device updates the automatic driving system according to the first driving behavior.
  • the method provided in the embodiments of the present application is introduced from the perspective of an automatic driving device.
  • the automatic driving device includes a hardware structure and/or software module corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiment of the present application may divide the function modules of the automatic driving device according to the foregoing method examples.
  • each function module may be divided corresponding to each function, or two or more functions may be integrated into one processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • FIG. 10 shows a schematic diagram of a possible structure of the device (denoted as the automatic driving device 100) involved in the above embodiment.
  • the automatic driving device 100 includes a communication unit 1002 and a processing unit. 1001, may further include a storage unit 1003.
  • the schematic structural diagram shown in FIG. 10 may be used to illustrate the structure of the automatic driving device involved in the foregoing embodiment.
  • the schematic structural diagram shown in FIG. 10 may be used to illustrate the structure of the V2X server involved in the foregoing embodiment.
  • the processing unit 1001 is used to control and manage the actions of the automatic driving device, for example, the processing unit 1001 is used to execute FIG. 5 Step 501, step 502, step 503 in FIG. 9, step 901, step 902, step 904, step 905, step 906, step 906, step 907 in FIG. 9, step 903 in FIG. 9 is executed through the communication unit 1002, and/ Or actions performed by the automatic driving device in other processes described in the embodiments of this application.
  • the processing unit 1001 may communicate with other network entities through the communication unit 1002, for example, communicate with the V2X server 20 shown in FIG. 1.
  • the storage unit 1003 is used to store program codes and data of the automatic driving device.
  • the automatic driving device 100 may be an automatic driving device or a chip in the automatic driving device.
  • the processing unit 1001 may be a processor or a controller, and the communication unit 1002 may be a communication interface, a transceiver, a transceiver, a transceiver circuit, a transceiver, and the like.
  • the communication interface is a general term and may include one or more interfaces.
  • the storage unit 1003 may be a memory.
  • the processing unit 1001 may be a processor or a controller, and the communication unit 1002 may be an input interface and/or an output interface, a pin or a circuit, or the like.
  • the storage unit 1003 may be a storage unit (for example, a register, a cache, etc.) in the chip, or a storage unit (for example, read-only memory (ROM), Random access memory (random access memory, RAM for short, etc.).
  • the processing unit 1001 is used to control and manage the actions of the V2X server.
  • the processing unit 1001 is used to execute the embodiments of the present application.
  • the processing unit 1001 may communicate with other network entities through the communication unit 1002, for example, communicate with the vehicle 10 shown in FIG. 1.
  • the storage unit 1003 is used to store the program code and data of the V2X server.
  • the automatic driving device 100 may be a V2X server or a chip in the V2X server.
  • the processing unit 1001 may be a processor or a controller, and the communication unit 1002 may be a communication interface, a transceiver, a transceiver, a transceiver circuit, a transceiver, and the like.
  • the communication interface is a general term and may include one or more interfaces.
  • the storage unit 1003 may be a memory.
  • the processing unit 1001 may be a processor or a controller, and the communication unit 1002 may be an input interface and/or an output interface, a pin or a circuit, or the like.
  • the storage unit 1003 may be a storage unit (for example, a register, a cache, etc.) in the chip, or a storage unit (for example, read-only memory (ROM), Random access memory (random access memory, RAM for short, etc.).
  • the integrated unit in FIG. 10 is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solutions of the embodiments of the present application are essentially or the part that contributes to the prior art, or all or part of the technical solutions can be embodied in the form of software products, and the computer software products are stored in a storage
  • the medium includes several instructions to make a computer device (which can be a personal computer, a server, or a first access network device, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application .
  • Storage media for storing computer software products include: U disk, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk and other media that can store program codes.
  • the unit in FIG. 10 may also be referred to as a module, for example, the processing unit may be referred to as a processing module.
  • FIG. 11 shows an example diagram of a communication system provided by an embodiment of the present application, which includes a V2X server 1101 and an automatic driving device 1102.
  • the V2X server 1101 is configured to execute the actions performed by the V2X server in the foregoing embodiment. For example, the V2X server 1101 sends a preset automatic driving mode to the automatic driving device.
  • the automatic driving device 1102 is used to perform the actions performed on the automatic driving device in the above-mentioned embodiment.
  • the automatic driving device 1102 is used to perform the steps in FIG. 5 and FIG. 9.
  • each step in the method provided in this embodiment can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software.
  • the steps of the method disclosed in combination with the embodiments of the present application may be directly embodied as being executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor.
  • the processor in this application may include but is not limited to at least one of the following: central processing unit (CPU), microprocessor, digital signal processor (DSP), microcontroller (microcontroller unit, MCU), or Various computing devices such as artificial intelligence processors that run software.
  • Each computing device may include one or more cores for executing software instructions for calculation or processing.
  • the processor can be a single semiconductor chip, or it can be integrated with other circuits to form a semiconductor chip. For example, it can form an SoC (on-chip) with other circuits (such as codec circuits, hardware acceleration circuits, or various bus and interface circuits). System), or it can be integrated into the ASIC as a built-in processor of an ASIC, and the ASIC integrated with the processor can be packaged separately or together with other circuits.
  • the processor may further include necessary hardware accelerators, such as field programmable gate array (FPGA) and PLD (programmable logic device) , Or a logic circuit that implements dedicated logic operations.
  • FPGA field programmable gate array
  • the memory in the embodiments of the present application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory , RAM) or other types of dynamic storage devices that can store information and instructions, and may also be electrically erasable programmable read-only memory (EEPROM).
  • ROM read-only memory
  • RAM random access memory
  • EEPROM electrically erasable programmable read-only memory
  • the memory can also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital universal discs, Blu-ray discs, etc.) , Disk storage media or other magnetic storage devices, or any other media that can be used to carry or store desired program codes in the form of instructions or data structures and that can be accessed by a computer, but are not limited thereto.
  • CD-ROM compact disc read-only memory
  • optical disc storage including compact discs, laser discs, optical discs, digital universal discs, Blu-ray discs, etc.
  • Disk storage media or other magnetic storage devices or any other media that can be used to carry or store desired program codes in the form of instructions or data structures and that can be accessed by a computer, but are not limited thereto.
  • the above embodiments it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • a software program it can be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions.
  • the computer program instructions When the computer program instructions are loaded and executed on the computer, the processes or functions described in the embodiments of the present application are generated in whole or in part.
  • the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • Computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • computer instructions may be transmitted from a website, computer, server, or data center through a cable (such as Coaxial cable, optical fiber, digital subscriber line (digital subscriber line, referred to as DSL) or wireless (such as infrared, wireless, microwave, etc.) transmission to another website site, computer, server or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or may include one or more data storage devices such as a server or a data center that can be integrated with the medium.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, and a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Navigation (AREA)

Abstract

一种自适应优化自动驾驶系统的方法及装置,涉及自动驾驶技术领域,解决了车辆在自动驾驶系统的控制下的驾驶行为与驾驶员的驾驶意图冲突的问题,可以应用在智能汽车、网联汽车上。具体方案为:自动驾驶装置获取目标车辆的驾驶员的驾驶意图(步骤501);基于驾驶员的驾驶意图,自动驾驶装置检测到车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突(步骤502);自动驾驶装置更新自动驾驶系统,以使得目标车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图(步骤503)。本方法及装置用于车辆的自动驾驶的过程中。

Description

一种自适应优化自动驾驶系统的方法及装置 技术领域
本申请涉及自动驾驶技术领域,尤其涉及一种自适应优化自动驾驶系统的方法及装置。
背景技术
随着自动驾驶技术的发展,自动驾驶技术给人们驾驶车辆提供的许多便利。自动驾驶可以指将一种或者多种自动驾驶模式配置在车辆上,驾驶员根据自身需求从一种或者多种自动驾驶模式中选择适合当前行驶场景的自动驾驶模式,基于该自动驾驶模式触发车辆的行驶。
但是,现有自动驾驶模式的配置数量是有限的,在配置自动驾驶模式时未考虑到所有不同行驶场景下所有驾驶员的驾驶需求,在某些行驶场景下车辆上配置的自动驾驶模式不能满足驾驶员的驾驶需求。此时,驾驶员需要将自动驾驶模式切换到手动驾驶。
例如,当某行驶场景下,车辆在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图(如驾驶员希望的驾驶规划)产生冲突时,驾驶员通常会直接接管车辆,手动驾驶。但车辆的自动驾驶系统所包括的自动驾驶规划并没有根据当前的冲突进行调整,导致在以后面对相同的行驶场景时车辆在自动驾驶系统控制下的驾驶行为依旧会和驾驶员的驾驶意图产生冲突,不能满足驾驶员的个性化需求。
发明内容
本申请提供一种自适应优化自动驾驶系统的方法及装置,以解决现有技术中车辆在自动驾驶系统的控制下的驾驶行为和驾驶员的驾驶意图冲突的问题。
为达到上述目的,本申请采用如下技术方案:
第一方面,提供一种自适应优化自动驾驶系统的方法,该方法可以应用于车辆的自动驾驶装置,例如,该自动驾驶装置可以为车载装置,或者,该自动驾驶装置也可以应用于车载装置的芯片或片上系统,该方法可以包括:自动驾驶装置获取在自动驾驶系统的控制下自动驾驶的车辆的驾驶员的驾驶意图;基于驾驶员的驾驶意图,自动驾驶装置检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突,更新自动驾驶系统,以使得目标车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。
基于第一方面提供的自适应优化自动驾驶系统的方法,在检测到在自动驾驶系统控制下自动驾驶的车辆的驾驶员的驾驶意图与车辆的自动驾驶系统控制下的车辆的驾驶行为冲突时,更新自动驾驶系统,以使得车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。基于此,后续在更新后的自动驾驶系统的控制下的驾驶行为可以满足驾驶员的驾驶需求。从而,解决了在自动驾驶系统控制下的车辆的驾驶行为与驾驶员的驾驶意图冲突的问题。
一种可能的实现方式中,驾驶意图由驾驶员的行为对应的特征参数表征;驾驶员的行为包括:驾驶员的操作行为、视觉行为、情绪行为、身体姿态行为中的一种或者多种。
基于该可能的实现方式,可以根据驾驶员的多个特征行为,从多角度确定驾驶员的驾驶意图,全面准确。
一种可能的实现方式中,目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突,包括:驾驶员的行为对应的特征参数超过预设范围,和/或,驾驶员的行为对应的特征参数超过预设范围的时间大于或等于第一预设值,和/或,驾驶员的行为对应的特征参数超过预设范围的次数大于或等于第二预设值。
基于该可能的实现方式中,由于驾驶员的行为可以准确的反映出驾驶员的驾驶意图,基于此,当驾驶员的行为对应的特征参数超过预设范围,和/或,驾驶员的行为对应的特征参数超过预设范围的时间和/或次数大于或等于预设值时,可以准确的确定驾驶员的驾驶意图与车辆的在自动驾驶系统控制下的驾驶行为冲突。
一种可能的实现方式中,自动驾驶装置根据驾驶员的驾驶意图以及目标车辆在自动驾驶系统控制下的驾驶行为,得到用于表征驾驶员的驾驶意图以及目标车辆在该自动驾驶系统控制下的驾驶行为对应的驾驶数据的第一特征参数;自动驾驶装置将第一特征参数输入用于确定匹配驾驶员的驾驶意图的驾驶行为的预设神经网络模型,得到第一驾驶行为;自动驾驶装置根据第一驾驶行为,更新自动驾驶系统。
基于该可能的实现方式中,由于预设神经网络模型可以为预先配置在车辆的自动驾驶装置中,因此,自动驾驶装置可以将表征驾驶员的驾驶意图以及车辆在该自动驾驶系统控制下的驾驶行为对应的驾驶数据的特征参数输入预设神经网络模型,得到第一行为特征,并根据第一行为特征更新自动驾驶系统,简单方便。
一种可能的实现方式中,自动驾驶装置向驾驶员呈现一个或多个问题;自动驾驶装置接收来自驾驶员针对该一个或多个问题的回答,该一个或多个问题的回答用于指示是否更新自动驾驶系统;当该一个或多个问题的回答用于指示更新自动驾驶系统时,自动驾驶装置根据第一驾驶行为更新自动驾驶系统。
基于该可能的实现方式中,在确定驾驶员的驾驶意图之后,可以根据一个或多个问题进一步对驾驶员的驾驶意图进行确认。由于,驾驶员对该一个或多个问题的回答可以反映驾驶员的真实的驾驶意图,如此,可以更加准确的确定驾驶员的驾驶意图,进而,可以根据驾驶员的答案确定是否更新自动驾驶系统。在该一个或多个问题的回答用于指示更新自动驾驶系统时,根据与驾驶意图匹配的驾驶行为更新自动驾驶系统。以使得车辆的在更新后的自动驾驶系统控制下的驾驶行为满足驾驶员的驾驶意图。
一种可能的实现方式中,若第一驾驶行为符合目标车辆的安全驾驶行为,则自动驾驶装置将第一驾驶行为作为目标车辆在更新后的自动驾驶系统控制下的驾驶行为;若第一驾驶行为不符合目标车辆的安全驾驶行为,则自动驾驶装置将该安全驾驶行为目标车辆在更新后的自动驾驶系统控制下的驾驶行为。
基于该可能的实现方式中,为了保证车辆在自动驾驶系统的控制下安全行驶,当车辆在更新后的自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图匹配,且该驾驶行为符合车辆的安全驾驶行为时,可以将该驾驶行为作为更新后的自动驾驶系统控制下的行为。或者,当该驾驶行为不符合车辆的安全驾驶行为时,可以将安全驾驶行为作为更新后的自动驾驶系统控制下的行为。如此,车辆在自动驾驶系统的控制下可以安全行驶。
第二方面,提供一种自适应优化自动驾驶系统的装置,该装置应用于自动驾驶装置或者自动驾驶装置中的芯片或者片上系统,还可以为自动驾驶装置中用于实现第一方面或第一方面的任一可能的设计所述的方法的功能模块。该装置可以实现上述各方面或者各可能 的设计中自动驾驶装置所执行的功能,所述功能可以通过硬件执行相应的软件实现。所述硬件或软件包括一个或多个上述功能相应的模块。如:该自适应优化自动驾驶系统的装置包括通信单元以及处理单元。
通信单元,用于获取目标车辆的驾驶员的驾驶意图,其中,目标车辆为在自动驾驶系统的控制下自动驾驶的车辆。
处理单元,用于基于驾驶员的驾驶意图,检测目标车辆的在自动驾驶控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
该处理单元,还用于更新自动驾驶系统,以使得目标车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。
其中,该自适应优化自动驾驶系统的装置的具体实现方式可以参考第一方面或第一方面的任一可能的设计提供的自适应优化自动驾驶系统的方法中自动驾驶装置的行为功能,在此不再重复赘述。因此,该提供的自适应优化自动驾驶系统的装置可以达到与第一方面或者第一方面的任一可能的设计相同的有益效果。
第三方面,提供了一种自动驾驶装置,该自动驾驶装置可以为车载装置或者车载装置中的芯片或者片上系统。该自动驾驶装置可以实现上述各方面或者各可能的设计自动驾驶装置所执行的功能,所述功能可以通过硬件实现,如:一种可能的设计中,该自动驾驶装置可以包括:处理器和通信接口,处理器用于运行计算机程序或指令,以实现如第一方面和第一方面的任一种可能的实现方式中所描述的自适应优化自动驾驶系统的方法。
在又一种可能的设计中,自动驾驶装置还可以包括存储器,存储器用于保存自动驾驶装置必要的计算机执行指令和数据。当该自动驾驶装置运行时,该处理器执行该存储器存储的该计算机执行指令,以使该自动驾驶装置执行上述第一方面或者第一方面的任一种可能的设计所述的自适应优化自动驾驶系统的方法。
第四方面,提供了一种计算机可读存储介质,该计算机可读存储介质可以为可读的非易失性存储介质,该计算机可读存储介质存储有计算机指令或者程序,当其在计算机上运行时,使得计算机可以执行上述第一方面或者上述方面的任一种可能的设计所述的自适应优化自动驾驶系统的方法。
第五方面,提供了一种包含指令的计算机程序产品,当其在计算机上运行时,使得计算机可以执行上述第一方面或者上述方面的任一种可能的设计所述的自适应优化自动驾驶系统的方法。
第六方面,提供了一种自动驾驶装置,该自动驾驶装置可以为自动驾驶装置或者自动驾驶装置中的芯片或者片上系统,该自动驾驶装置包括一个或者多个处理器以及和一个或多个存储器。所述一个或多个存储器与所述一个或多个处理器耦合,所述一个或多个存储器用于存储计算机程序代码,所述计算机程序代码包括计算机指令,当所述一个或多个处理器执行所述计算机指令时,使得所述自动驾驶装置执行如上述第一方面或者第一方面的任一可能的设计所述的自适应优化自动驾驶系统的方法。
第七方面,提供了一种芯片系统,该芯片系统包括处理器以及通信接口,该芯片系统可以用于实现上述第一方面或第一方面的任一可能的设计中自动驾驶装置所执行的功能,例如处理器用于通过通信接口获取驾驶员的操作特征。
其中,第二方面至第七方面中任一种设计方式所带来的技术效果可参见上述第一方面 或者第一方面的任一种可能的设计所带来的技术效果,不再赘述。
第八方面,提供了一种通信系统,该通信系统包括自动驾驶装置和任何事物(vehicle to everything,V2X)服务器,该自动驾驶装置与该V2X服务器通信连接,该自动驾驶装置可以用于实现上述第一方面或第一方面的任一可能的设计中自动驾驶装置所执行的功能,该V2X服务器可以用于为自动驾驶装置提供多个信息,例如,该多个信息可以包括驾驶员的特征信息、车辆在自动驾驶系统控制下的驾驶行为等。
附图说明
图1为本申请实施例提供的一种通信系统的结构示意图;
图2为本申请实施例提供的一种车辆的结构示意图;
图3为本申请实施例提供的一种自动驾驶装置的结构示意图;
图4为本申请实施例提供的一种V2X服务器的结构示意图;
图5为本申请实施例提供的一种自适应优化自动驾驶系统的方法的流程示例图;
图6a为本申请实施例提供的一种驾驶员对车辆的加速踏板的操作示例图;
图6b为本申请实施例提供的一种驾驶员对车辆的减速踏板的操作示例图;
图6c为本申请实施例提供的一种驾驶员对车辆的方向盘的操作示例图;
图7为本申请实施例提供的一种车辆的结构示意图;
图8为本申请实施例提供的一种车辆的显示界面的示意图;
图9为本申请实施例提供的另一种自适应优化自动驾驶系统的方法的流程示例图;
图10为本申请实施例提供的一种自动驾驶装置1000的结构示意图;
图11为本申请实施例提供的一种通信系统的结构示意图。
具体实施方式
在描述本申请实施例之前,对本申请实施例涉及的名词术语进行解释说明:
交通信息:也可以称为关键要素、交通特征、通行特征等,不予限制。交通信息可以用于表示车辆行驶过程中的行驶道路的交通环境。具体的,交通信息可以包括车辆的行驶道路上的车辆信息、障碍物信息、行人信息以及交通环境的信息中的一个或多个。
其中,车辆信息,障碍物信息及行人信息可以指示其他车辆、行人、障碍物相对自车的位置,相对速度等。具体的,车辆信息可以包括车辆数量,车辆速度,车辆到本车的距离以及车辆类型。障碍物信息可以包括道路的限高杆、护栏等。行人信息可以包括行人数量、行人速度以及行人到本车的距离。交通环境的信息可以包括道路信息、灯光条件以及天气情况。例如,道路信息可以为高速公路、国道、省道、城市道路或者乡村道路等。道路信息还可以包括封道路指示牌、红绿灯、车道线、闭道路和非封闭道路等,不予限制。
其中,交通指示牌可以指示道路限速上限,限速下限,前方停止行车等。红绿灯可以指示是否停车,左转或右转。车道线可以指示车辆行驶方向,转弯半径,两边车道方向,是否允许换道等。
这样,驾驶员在驾驶车辆在道路上行驶的过程中,面对不同的交通信息,驾驶员可以控制车辆执行与交通信息对应的驾驶行为。
驾驶行为:也可以称为驾驶主体任务。驾驶行为可以是指车辆在道路中行驶的过程中,需要在不同的交通环境下完成不同的驾驶动作,才能够到达目的地。该不同的动作可以为跟车、车道保持、变道切入、行驶时被其他车辆切入等。例如,在车辆行驶时被其他车辆 切入时,本车的驾驶行为是在被切入时降低车速。
车辆在一个交通信息下可以执行一个或多个驾驶行为。该一个或多个驾驶行为可以用于反映驾驶员的驾驶意图。
例如,在交通信息包括行驶道路为弯道时,车辆的驾驶行为可以包括减速行驶以及根据弯道半径调整行驶方向等。该交通信息以及该交通信息对应的车辆的一个或多个驾驶行为可以构成一个行驶场景。
行驶场景:也可以称为驾驶功能域。行驶场景可以是指按照预设聚类算法将具有多个具有相同或相似特征的交通信息以及在该交通信息下车辆的驾驶行为进行聚类,得到多个类别,其中,该多个类别中每个类别可以包括一种交通信息以及该交通信息对应的至少一个驾驶行为。一个类别对应一个行驶场景。在不同的行驶场景下,车载在自动驾驶系统控制下具有不同的驾驶行为。
其中,预设聚类算法可以为k均值聚类算法(k-means clustering algorithm,k-means)。k-means聚类算法可以参照现有技术,此处不再赘述。
示例性的,如表1所示,行驶场景可以包括:城区高峰路段顺行、封闭道路顺行、非封闭道路缓行、城市道路交汇处通行、非城市道路交汇处通行等。上述行驶场景包括的交通信息以及驾驶行为可以如表1所示。
表1
Figure PCTCN2020089491-appb-000001
其中,表1中,每个行驶场景中的交通信息还可以包括其他信息,如道路的车辆信息、行人信息等,具体描述可以参照上述交通信息的描述,不予赘述。表1还可以包括其他行驶场景,不予限制。
驾驶员的驾驶意图:驾驶员的驾驶意图是指车辆在行驶时,驾驶员对车辆进行的操作或者驾驶员想要车辆执行的驾驶行为。其中,驾驶行为可以包括跟车、左换道、右换道或超车等。例如,在某一行驶场境下,驾驶员想要车辆执行的驾驶行为是超车。当车辆执行超车的驾驶行为时,驾驶员的驾驶操作为控制加速踏板,以使得车辆执行加速的驾驶行为。驾驶员控制加速踏板的驾驶操作以及车辆执行加速的驾驶行为,可以称为自然驾驶数据。
自然驾驶数据:自然驾驶数据是指当驾驶员驾驶车辆行驶时,车辆的驾驶行为和驾驶员对车辆的操作特征的统称。
其中,操作特征可以包括控制方向盘、控制加速踏板以及控制减速踏板等,不予限制。
自动驾驶模式:是指驾驶员通过自动驾驶系统输入目的地,自动驾驶系统可以根据预 设参数,确定合理的行驶路线。该行驶路线可以包括多个交通信息,例如,该多个交通信息可以包括多个弯道、多个道路交汇处等。
自动驾驶系统可以根据行驶路线上不同的行驶场景控制配置有该自动驾驶系统的车辆执行相应的驾驶行为。例如,在该行驶路线的弯道处,自动驾驶系统可以根据道路的曲率对该车辆的驾驶行为进行调整,例如,自动驾驶系统可以对车辆的转动方向、转动角度以及车速进行调整,以使得车辆平稳安全的行驶。
现有技术中,由于车辆配置的自动驾驶模式的数量是有限的,且在为车辆配置自动驾驶模式时未考虑到所有场景下所有驾驶员的驾驶需求,在某些行驶场景下车辆上配置的自动驾驶模式不能满足驾驶员的驾驶需求,或者,在自动驾驶技术中,对于自动驾驶系统无法处理的场景,自动驾驶系统会提醒驾驶员接管车辆的驾驶权,手动驾驶。此时,驾驶员需要将自动驾驶模式切换到手动驾驶。例如,当某行驶场景下,车辆在自动驾驶系统控制下执行的驾驶行为与驾驶员的驾驶意图(如驾驶员希望的驾驶规划)产生冲突时,驾驶员通常会直接接管车辆,手动驾驶。但车辆在自动驾驶模式控制下的驾驶行为并没有根据当前的冲突进行调整,导致在以后面对相同的行驶场景时车辆在自动驾驶系统控制下的驾驶行为依旧会和驾驶员的驾驶意图产生冲突,不能满足驾驶员的个性化需求。
为了解决上述技术问题,本申请实施例提供了一种自适应优化自动驾驶系统的方法,应用于车辆的自动驾驶装置,包括:自动驾驶装置获取驾驶员的驾驶意图;基于驾驶员的驾驶意图,自动驾驶装置检测到车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突,自动驾驶装置更新自动驾驶系统,以使得车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。
基于该技术方案,在检测到在自动驾驶系统控制下自动驾驶的车辆的驾驶员的驾驶意图与车辆的自动驾驶系统控制下的车辆的驾驶行为冲突时,更新自动驾驶系统,以使得车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。基于此,后续在更新后的自动驾驶系统的控制下的驾驶行为可以满足驾驶员的驾驶需求。从而,解决了在自动驾驶系统控制下的车辆的驾驶行为与驾驶员的驾驶意图冲突的问题。
下面将结合附图对本申请实施例的实施方式进行详细描述。
图1示出的是本申请实施例提供的一种通信系统的示例图。如图1所示,该通信系统可以包括车辆10和V2X服务器20。车辆10可以采用V2X通信技术与V2X服务器20进行通信。车辆20还可以通过无线链路与V2X服务器20相互通信,比如,通过第五代(5th generation,5G)网络相互通信,也可以通过其他方式通信,不予限制。
其中,车辆10可以为智能网联驾驶(intelligent network driving)车辆,是一种典型的车联网终端。本领域技术人员应该理解的是,车辆10具体可以通过其内部的功能单元或装置执行本申请实施例的自适应优化自动驾驶系统的方法。例如车辆10中可包括用于执行本申请实施例提供的自适应优化自动驾驶系统的方法的自动驾驶装置。自动驾驶装置可通过控制器局域网络(controller area network,CAN)总线与车辆10的其他部件通信连接。车辆10的具体结构将在图2所示的实施例中详细描述。
V2X服务器20可以为车辆10提供车辆在自动驾驶系统控制下的多个驾驶行为和/或多个驾驶员的自然驾驶数据等,还可以用于接收来自车辆10的驾驶员的图像信息,并对驾驶员的图像信息进行识别,得到驾驶员的身体姿态行为,例如,身体姿态行为可以包括驾 驶员是否在使用手机,驾驶员是否疲劳驾驶等。V2X服务器20还可以将驾驶员的身体姿态行为发送给车辆10。其中,该V2X服务器20可以为实体服务器,也可以为虚拟服务器,例如,云服务器等,不予限制。V2X服务器20的具体结构将在图4所示的实施例中详细描述。
需要说明的是,本申请实施例描述的通信系统是为了更加清楚的说明本申请实施例的技术方案,并不构成对于本申请实施例提供的技术方案的限定,本领域普通技术人员可知,随着通信系统的演变和其他通信系统的出现,本申请实施例提供的技术方案对于类似的技术问题,同样适用。
请参考图2,为本申请实施例提供的一种车辆10的结构示意图。如图2所示,车辆10可以包括自动驾驶装置101、车身网关102、车身天线103等。其中,自动驾驶装置101可以通过射频(radio frequency,RF)电缆与车身天线103通信连接。
可以理解的是,图2示意的结构并不构成对车辆10的具体限定。在一些实施例中,车辆10可以包括比图2所示部件更多或更少的部件,或者,车辆10可以包括图2示出的某些部件的组合部件,或者,车辆10可以包括图2所示部件的拆分部件等。如:车辆10还可以包括域控制器(domain controller,DC)、多域控制器(multi-domain controller,
MDC)等。图2示出的部件可以以硬件、软件或软件和硬件的组合实现。如:车辆10中的自动驾驶装置101可以为车联网芯片等,自动驾驶装置101的具体结构将在图3所示的实施例中详细描述。
自动驾驶装置101可以称为车载单元(on board unit,OBU)、车载终端等,例如,自动驾驶装置101可以为车载盒子(telematics BOX,T-Box)。自动驾驶装置101主要用于执行本申请实施例提供的自适应优化自动驾驶系统的方法。
车身网关102主要用于车辆信息的接收和发送,车身网关102可以通过CAN总线与自动驾驶装置101连接。示例性的,车身网关102可以从自动驾驶装置101获取自动驾驶装置101执行本申请实施例提供的自适应优化自动驾驶系统的方法后得到的更新后的自动驾驶系统以及车辆在自动驾驶系统控制下的驾驶行为,将获取到的更新后的自动驾驶系统以及车辆在自动驾驶系统控制下的驾驶行为发送给车辆10的其他部件。
车身天线103可以内置通信天线,通信天线负责信号的接收和发送。例如,通信天线可以将车辆的车辆信息发送给其他车辆的自动驾驶装置,也可以接收来其他自动驾驶装置发送的车辆信息等。
请参考图3,为本申请实施例提供的一种自动驾驶装置101的结构示意图。如图3所示,自动驾驶装置101可以包括自动驾驶感知模块1011、智能座舱交互模块1012、驾驶意图冲突检测模块1013、自动驾驶控制模块1014。
其中,自动驾驶感知模块1011可以用于对车辆10的交通信息以及驾驶行为的识别和划分;智能座舱交互模块1012主要用于获取驾驶员的图像信息以及语音信息,还可以对驾驶员的语音信息进行语音识别;驾驶意图冲突检测模块1013可以用于确定驾驶员的驾驶意图,以及检测驾驶员的驾驶意图与车辆的在自动驾驶系统控制下的驾驶行为是否冲突。自动驾驶控制模块1014可以用于控制车辆的自动驾驶时的驾驶行为。
例如,智能座舱交互模块1012可以利用长短时记忆单元-深度学习(long-short term memory-deep neural networks,LSTM-DNN)模型对驾驶员的语音信息进行识别。比如,智 能座舱交互模块1012配置有LSTM-DNN模型,自动驾驶装置可以通过声音采集装置获取位于车辆内部的人员的语音信息,并将位于车辆内部的人员的语音信息输入智能座舱交互模块1012的LSTM-DNN模型中,以使得智能座舱交互模块1012对位于车辆内部的人员的语音信息进行识别。下文中涉及与语音信息的识别相关的描述,均可以参照此处,后续不再赘述。
一种可能的设计中,如图3所示,自动驾驶装置101还可以包括驾驶员在线自适应模块1015。驾驶员在线自适应模块1015可以用于根据行驶场景下的车辆的驾驶员的驾驶意图调整该行驶场景下车辆的在自动驾驶系统控制下的驾驶行为,并将调整后的车辆的在自动驾驶系统控制下的驾驶行为以及该行驶场景发送给自动驾驶控制模块1014。
自动驾驶控制模块1014接收到来自驾驶员在线自适应模块1015的调整后的车辆的在自动驾驶系统控制下的驾驶行为以及该行驶场景的信息后,可以对该自动驾驶系统进行更新,例如,将该车辆在自动驾驶系统控制下的驾驶行为替换为调整后的车辆在自动驾驶系统控制下的驾驶行为。
可以理解的是,图3示意的结构并不构成对自动驾驶装置101的具体限定。在另一些实施例中,自动驾驶装置101可以包括比图3所示部件更多或更少的部件,或者,自动驾驶装置101可以包括图3示出的某些部件的组合部件,或者,自动驾驶装置101可以包括图3所示部件的拆分部件等。图3示的部件可以以硬件、软件或软件和硬件的组合实现。
本申请实施例中,图3所示的装置也可以为自动驾驶装置中的芯片或芯片系统。芯片系统可以由芯片构成,也可以包括芯片和其他分立器件。
请参考图4,为本申请实施例提供的一种V2X服务器20的结构示意图。如图4所示,V2X服务器20可以包括驾驶员偏好及历史操作数据库2011、驾驶员接受度最高的参数模块2012。
其中,驾驶员偏好及历史操作数据库2011用于存储预设时间段内多个驾驶员的自然驾驶数据,以及车辆在自动驾驶系统控制下的至少一个驾驶行为。驾驶员接受度最高的参数模块2012用于存储该至少一个自动驾驶模式中使用次数最多的自动驾驶模式。
进一步的,V2X服务器20还可以按照预设时间周期更新驾驶员偏好及历史操作数据库2011以及驾驶员接受度最高的参数模块2012中的数据。其中,预设时间周期可以为V2X服务器20预先设置的。或者,V2X服务器20也可以通过与自动驾驶装置101的交互更新驾驶员偏好及历史操作数据库2011以及驾驶员接受度最高的参数模块2012中的数据。例如,当自动驾驶装置更新车辆在自动驾驶系统控制下的驾驶行为之后,向V2X服务器发送车辆在更新后的自动驾驶系统控制下的驾驶行为。V2X服务器在接收到来自自动驾驶装置的车辆在更新后的自动驾驶系统控制下的驾驶行为之后,可以将V2X服务器中具有的车辆在自动驾驶系统控制下的驾驶行为替换为车辆在更新后的自动驾驶系统控制下的驾驶行为。
一种可能的设计中,V2X服务器20还可以包括图2中的驾驶员在线自适应模块1015,其中,驾驶员在线自适应模块1015的功能可以参照上述描述,此处不再赘述。
可以理解的是,图4示意的结构并不构成对V2X服务器20的具体限定。在另一些实施例中,V2X服务器20可以包括比图4所示部件更多或更少的部件,或者,V2X服务器20可以包括图4示出的某些部件的组合部件,或者,V2X服务器20可以包括图4所示部 件的拆分部件等。图4示的部件可以以硬件、软件或软件和硬件的组合实现。
本申请实施例中,图4所示的装置也可以为V2X服务器中的芯片或芯片系统。芯片系统可以由芯片构成,也可以包括芯片和其他分立器件。
在本申请实施例中,“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言,使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
以下,术语“第一”、“第二”仅用于描述目的,而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此,限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个该特征。在本申请实施例的描述中,除非另有说明,“多个”的含义是两个或两个以上。
以下实施例中的方法均可以在具有上述硬件结构的自动驾驶装置中实现。以下结合图1所示的通信系统,对本申请实施例提供的自适应优化自动驾驶系统的方法进行详细介绍。
需要说明的是,本申请实施例中,自适应优化自动驾驶系统的方法的执行主体可以为自动驾驶装置,例如,该自动驾驶装置可以为图2或图3中的自动驾驶装置。该方法的执行主体也可以是该自动驾驶装置中的芯片或片上系统,不予限制。下面以自适应优化自动驾驶系统的方法的执行主体为自动驾驶装置为例进行描述。
图5为本申请实施例提供的一种自适应优化自动驾驶系统的方法的流程示意图。如图5所示,该方法可以包括:
步骤501、自动驾驶装置获取目标车辆的驾驶员的驾驶意图。
其中,目标车辆为在自动控制系统控制下自动驾驶的车辆。目标车辆可以为图1中的车辆,目标车辆可以具有图2所述的部件。自动驾驶装置可以包括图3中的一个或多个模块。
其中,驾驶员可以为驾驶过目标车辆的驾驶员,也可以为没有驾驶过该目标车辆的驾驶员或者可以描述为新的驾驶员。自动驾驶装置可以根据驾驶员的特征信息检测驾驶员为驾驶过目标车辆的驾驶员,还是新的驾驶员。一个特征信息用于标识一个驾驶员。
一种可能的实现方式中,驾驶员的驾驶意图可以由驾驶员的行为对应的特征参数表征。也即,自动驾驶装置可以根据驾驶员的行为,检测驾驶员的驾驶意图。
其中,驾驶员的行为可以包括驾驶员的操作行为、视觉行为、情绪行为、身体姿态行为中的一种或多种。
其中,驾驶员的操作行为可以是指车辆在自动行驶过程中,驾驶员对车辆的手动操作行为。例如,驾驶员的操作行为可以包括驾驶员对车辆的方向盘的操作行为、驾驶员对车辆的加速踏板的操作行为、驾驶员对车辆的减速踏板的操作行为。
自动驾驶装置可以通过多种方式获取驾驶员的驾驶意图。例如,该多种方式可以包括:自动驾驶装置可以将驾驶员的行为对应的特征参数输入驾驶意图模型,得到驾驶员的驾驶意图;自动驾驶装置可以根据驾驶员的操作行为确定驾驶员的驾驶意图;自动驾驶装置可以根据驾驶员的情绪行为确定驾驶员的驾驶意图。自动驾驶装置还可以通过其他方式确定驾驶员的驾驶意图,不予限制。下面对这些方式进行说明:
方式1、自动驾驶装置将驾驶员的行为对应的特征参数输入驾驶意图模型,得到驾驶员 的驾驶意图。
其中,驾驶意图模型的输入参数为驾驶员的行为的特征参数,驾驶意图的输出参数为驾驶员的驾驶意图。驾驶意图模型可以自动驾驶装置预先设置的。
例如,驾驶意图模型的一种确定方法可以为:通过隐马尔科夫模型对预设时间段内多个驾驶员的行为特征以及对应的自然驾驶数据训练得到。隐马尔科夫模型的训练方法可以参照现有技术,此处不再赘述。
具体的,自动驾驶装置可以将驾驶员的行为的特征参数输入上述驾驶意图模型,得到驾驶员的驾驶意图。自动驾驶装置可以比较驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为,若驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致,则自动驾驶装置可以确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突。
方式2、自动驾驶装置可以根据驾驶员的操作行为确定驾驶员的驾驶意图。
其中,驾驶员的操作特征可以包括驾驶员对加速踏板的操作、驾驶员对减速踏板的操作以及驾驶员对方向盘的操作。下面对自动驾驶装置根据驾驶员的操作特征确定驾驶员的驾驶意图的方法具体说明。
例如,自动驾驶装置可以通过一个或多个传感器获取驾驶员的行为对应的参数特征。比如,传感器可以包括加速踏板位置传感器、减速踏板位置传感器、方向盘扭矩传感器中的一个或多个。
其中,加速踏板位置传感器用于检测目标车辆的加速踏板的位置变化信息,减速踏板位置传感器用于检测目标车辆的减速踏板的位置变化信息,方向盘扭矩传感器用于检测目标车辆的方向盘转动的角度信息。自动驾驶装置可以通过加速踏板的变化信息确定驾驶员对加速踏板的操作行为。自动驾驶装置可以通过减速踏板的变化信息确定驾驶员对减速踏板的操作行为。自动驾驶装置可以通过方向盘的扭矩变化信息确定驾驶员对方向盘的操作行为。下面对自动驾驶装置根据驾驶员的操作行为。检测驾驶员的驾驶意图进行说明:
1、自动驾驶装置根据驾驶员对车辆的转向盘的操作行为,获取驾驶员的驾驶意图。
如图6a所示,若自动驾驶装置检测到目标车辆的加速踏板的位置从第一位置移动至第二位置,其中,当目标车辆的加速踏板位于第一位置时,目标车辆的车速为第一车速;当目标车辆的加速踏板位于第二位置时,目标车辆的车速为第二车速,且第二车速大于第一车速。则自动驾驶装置可以确定驾驶员的驾驶意图为提高该行驶场景下目标车辆的行驶速度。
2、自动驾驶装置根据驾驶员对减速踏板的操作行为,获取驾驶员的驾驶意图。
如图6b所示,若自动驾驶装置检测到目标车辆的减速踏板的位置从第三位置移动至第四位置,其中,当目标车辆的减速踏板位于第三位置时,目标车辆的车速为第三车速;当目标车辆的减速踏板位于第四位置时,目标车辆的车速为第四车速,且第四车速小于第三车速。则自动驾驶装置可以确定驾驶员的驾驶意图为降低该行驶场景下目标车辆的行驶速度。
3、自动驾驶装置根据驾驶员对方向盘的操作行为,获取驾驶员的驾驶意图。
其中,方向盘的扭矩变化信息可以包括方向盘的第一方向的变化信息以及第二方向的变化信息。其中,第一方向与第二方向为相反的方向,例如,如图6c所示,当方向盘向第一方向转动时,目标车辆朝行驶方向的左方转动;当方向盘向第二方向转动时,目标车辆 朝行驶方向的右方转动。
例如,当自动驾驶装置通过方向盘扭矩传感器检测到方向盘向第一方向转动的角度为α,则自动驾驶装置可以确定驾驶员的驾驶意图为:期望目标车辆朝第一方向转向。
方式3、自动驾驶装置可以根据驾驶员的情绪行为确定驾驶员的驾驶意图。
当驾驶员的情绪行为为负面情绪行为时,自动驾驶装置可以确定驾驶员的驾驶意图为不满意该行驶场景下目标车辆在自动驾驶系统控制下的驾驶行为。
进一步的,自动驾驶装置还可以通过驾驶员的身体姿态行为,确定驾驶员是否专注于目标车辆的驾驶行为,以及是否在执行第二任务。
若驾驶员没有在执行第二任务,则自动驾驶装置确定驾驶员的驾驶意图为不满意车辆在自动驾驶系统控制下的驾驶行为。也即,在该行驶场景下,驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突。
其中,驾驶员的视觉行为以及身体姿态行为可以用于反映驾驶员对目标车辆的驾驶行为的专注度。
驾驶员的视觉行为可以包括驾驶员的头部行为以及注视行为。例如,驾驶员的头部行为可以包括驾驶员的头部转动方向、转动的角速度、转动角度等。驾驶员的注视行为包括驾驶员的眨眼频率、眼睛的闭合持续时间等。当然,驾驶员的视觉行为还可以包括驾驶员的其他眼部行为,不予限制。
驾驶员的身体姿态行为还可以用于确定驾驶员的第二行为。例如,第二行为可以为除专注于目标车辆的驾驶行为之外的其他行为。例如,第二行为可以包括驾驶员使用手机,驾驶员与其他位置上的人员交谈等行为。
其中,驾驶员的情绪行为用于反映驾驶员是否满意目标车辆的在自动驾驶系统控制下的驾驶行为。其中,驾驶员的情绪行为可以包括驾驶员的面部表情以及语言特征。
一种示例,自动驾驶装置可以通过摄像装置获取预设时间段内驾驶员的多种图像。自动驾驶装置可以对该多张图像进行识别,得到驾驶员的视觉行为、身体姿态行为以及面部表情。其中,该预设时间段可为根据需要预先设置的。
又一种示例,自动驾驶装置可以通过声音采集装置获取驾驶员的语音信息。例如,声音采集装置可以为麦克风等。自动驾驶装置可以对该驾驶员的语音信息进行识别,得到驾驶员的语言特征。
其中,驾驶员的面部表情可以包括第一面部表情和第二面部表情。驾驶员的语言特征可以包括第一语言特征和第二语言特征。
例如,自动驾驶装置可以预先设置有第一面部表情库和第二面部表情库。自动驾驶装置在获取驾驶员的面部表情后,将驾驶员的面部表情分别与第一面部表情库以及第二面部表情库中的每个面部表情进行匹配,确定驾驶员的面部表情属于第一面部表情还是第二面部表情。
例如,当驾驶员的面部表情与第一面部表情库中的某一个面部表情的相似度大于或等于第一预设阈值时,自动驾驶装置可以确定驾驶员的面部表情属于第一面部表情。或者,当驾驶员的面部表情与第二面部表情库中的某一个面部表情的相似度大于或等于第二预设阈值时,自动驾驶装置可以确定驾驶员的面部表情属于第二面部表情。其中,第一预设阈值和第二预设阈值可以为根据需要预先设置的,不予限制。
目标车辆内可以设置有多个声音采集装置。该多个声音采集装置与自动驾驶装置通信连接。自动驾驶装置可以通过该多个声音采集装置获取位于目标车辆内部的人的声音,并对目标车辆内部的人的声音进行识别,确定声音的来源。例如,自动驾驶装置可以根据声音的音频对声音的来源进行识别,确定声音的来源。当确定声音的来源包括多个位置且该多个位置包括驾驶员的位置时,自动驾驶装置确定驾驶员在与其他位置上的人员交谈。
例如,如图7所示,以目标车辆包括前排车座和后排车座为例,该目标车辆可以设置有声音采集装置1和声音采集装置2。声音采集装置1用于采集驾驶员以及位于副驾驶位置的人的声音,声音采集装置2用于采集位于目标车辆后排的人员的声音。声音采集装置1和声音采集装置2可以将采集的声音发送到自动驾驶装置,用于自动驾驶装置对声音进行识别,从而,自动驾驶装置可以确定驾驶员是否在与其他位置上的人员交谈。比如,声音采集装置1可以采集到驾驶员的声音,且声音采集装置2可以采集到位于目标车辆后排的人员的声音,自动驾驶装置可以确定驾驶员在与后排上的人员交谈。其中,目标车辆还可以在每个车座处各设置一个声音采集装置,以使得自动驾驶装置可以更加精确的确定声音的来源。
驾驶员的第一语言特征和第二语言特征的确定方法可以参照上述第一面部表情和第二面部表情的确定方法,此处不再赘述。
需要说明的是,第一面部特征以及第一语言特征可以用于表征驾驶员的情绪行为为正面情绪行为。当驾驶员的情绪行为为正面情绪行为时,驾驶员的驾驶意图为:满意目标车辆的在自动控制系统下的驾驶行为。
例如,第一面部特征可以包括高兴、满意等,第一语言特征可以包括不错、挺好等。
需要说明的是,第二面部特征以及第二语言特征可以用于表征驾驶员的情绪行为为负面情绪行为。当驾驶员的情绪行为为负面情绪行为时,驾驶员的驾驶意图为:不满意目标车辆的在自动控制系统下的驾驶行为。
例如,第二面部表情可以包括发愁、生气等。第二语言特征可以包括真差、不好用等。
另一种可能的实现方式中,自动驾驶装置可以驾驶员的行为对应的特征参数输入到驾驶意图模型,得到驾驶员的驾驶意图。
步骤502、基于驾驶员的驾驶意图,自动驾驶装置检测到目标车辆的在自动控制系统下的驾驶行为与驾驶员的驾驶意图存在冲突。
一种可能的实现方式中,自动驾驶装置可以根据驾驶员的行为对应的特征参数,检测驾驶员的驾驶意图与车辆在自动控制系统下的驾驶行为是否存在冲突。比如,当驾驶员的欣慰对应特征参数超过预设范围,和/或驾驶员的对应的特征参数超过预设范围的时间大于或等于第一预设值,和/或驾驶员的对应的特征参数超过预设范围的次数大于或等于第二预设值时,自动驾驶装置可以检测到驾驶员的驾驶意图与车辆在自动控制系统下的驾驶行为存在冲突。
其中,驾驶员的对应的特征参数超过预设范围的次数大于或等于第二预设值也可以描述为在预设时间内驾驶员的对应的特征参数超过预设范围的次数大于或等于第二预设值。预设时间可以根据需要设置预先设置。
例如,结合图6a所示,目标车辆的在自动驾驶系统对应的加速踏板的变化范围为第一范围。其中,第一范围可以为根据需要预先设置的。
若加速踏板从第一位置移动至第二位置的变化范围超过第一范围,和/或加速踏板从第一位置移动至第二位置的变化范围超过第一范围的时间超过第一阈值,和/或加速踏板从第一位置移动至第二位置的变化范围超过第一范围的次数超过第二阈值,则自动驾驶装置可以检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
又例如,结合图6b所示,目标车辆的在自动驾驶系统对应的减速踏板的变化范围为第二范围。其中,第二范围可以为根据需要预先设置的。
若减速踏板从第三位置移动至第四位置的变化范围超过第二范围,和/或减速踏板从第三位置移动至第四位置的变化范围超过第二范围的时间超过第三阈值,和/或减速踏板从第三位置移动至第四位置的变化范围超过第二范围的次数超过第四阈值,则自动驾驶装置可以检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
又例如,结合图6c所示,目标车辆在自动驾驶系统控制下的驾驶行为是:目标车辆向第一方向转动的角度的阈值为第一角度。其中,第一角度可以为根据需要预先设置的。
若方向盘向第一方向转动的角度为α大于第一角度,和/或方向盘向第一方向转动的角度为α大于第一角度的时间超过第五阈值,和/或方向盘向第一方向转动的角度为α大于第一角度的次数超过第六阈值,则自动驾驶装置可以检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
需要说明的是,上述三个例子中,第一阈值~第六阈值可以根据需要预先设置的。
进一步的,为了更准确地确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突,自动驾驶装置可以确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为的冲突时间/冲突次数,若冲突时间大于或等于预设时间/冲突次数大于或等于预设次数,自动驾驶装置可以确定驾驶员的驾驶意图与目标车辆的在自动驾驶系统控制下的驾驶行为冲突。其中,预设时间以及预设次数可以为预先设置的,不予限制。下面以自动驾驶装置根据驾驶员的视觉行为确定驾驶员的驾驶意图为例,分别对自动驾驶装置确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为的冲突时间以及冲突次数的方法具体说明。
1、自动驾驶装置确定驾驶员的驾驶意图与车辆的在自动驾驶系统控制下的驾驶行为的冲突时间。
自动驾驶装置可以统计驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为的冲突时间,若冲突时间大于或等于预设时间,则自动驾驶装置可以确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突。若冲突时间小于预设时间,则自动驾驶装置可以确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不冲突。
一种示例中,自动驾驶装置设置有计时器,一旦自动驾驶装置检测到驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致时,可以触发计时器开始计时,自动驾驶装置执行的步骤为:
自动驾驶装置获取驾驶员的视觉行为,并将视觉行为输入上述驾驶员的驾驶意图模型中,得到驾驶员的驾驶意图。
其中,若自动驾驶装置确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为还是不一致,自动驾驶装置可以重复执行上述步骤,直至自动驾驶装置根据驾驶员的视觉行为确定的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为一致。当自动驾 驶装置确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为一致时,触发计时器停止计时。自动驾驶装置根据该计时器开始计时的时间与停止计时的时间,确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为的冲突时间。
例如,若计时器开始计时,自动驾驶装置可以周期性或随机性的获取驾驶员的视觉行为,并确定驾驶员的驾驶意图,不予限制。
基于上述的示例,若自动驾驶装置根据驾驶员的视觉行为确定的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为的冲突时间大于或等于预设时间,则自动驾驶装置确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突。
另一种可能的实现方式中,当驾驶员的视觉行为对应的特征参数超过预设范围,和/或以及身体姿态行为对应的特征参数超过预设范围时,自动驾驶装置可以检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
例如,自动驾驶装置设置有计数器。一旦自动驾驶装置检测到驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致时,自动驾驶装置获取预设时间段内驾驶员的多个的视觉行为,并将该多个视觉行为输入到上述驾驶意图模型中,得到该多个视觉行为中每个视觉行为对应的驾驶员的驾驶意图。自动驾驶装置可以分别比较每个视觉行为对应的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为是否一致。
若一个视觉行为对应的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致,则计数器记第一数值;若一个视觉行为对应的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为一致,则计数器记第二数值。若计数器统计的第一数值的数量大于或等于预设值,则自动驾驶装置可以确定驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为冲突。其中,预设时间段和预设值为预先设置的,不予限制。
例如,在第一时刻,自动驾驶装置确定第一时刻驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致,自动驾驶装置可以获取第一时刻之后的预设时间段内的N个时刻对应的驾驶员的视觉行为,如:分别为视觉行为1,视觉行为2,…,视觉行为N,其中,N为大于或等于1的正整数,该N个时刻之间的时间间隙可以相同,也可以不相同,不予限制。
自动驾驶装置可以根据上述驾驶员的驾驶意图模型确定该N个视觉行为中每个视觉行为对应的驾驶员的驾驶意图,例如,自动驾驶装置确定的驾驶员的驾驶意图包括驾驶意图1,驾驶意图2,…,驾驶意图N,其中,视觉行为与驾驶意图一一对应,如,视觉行为1对应驾驶意图1,视觉行为2对应驾驶意图2,视觉行为N对应视觉行为N。
自动驾驶装置分别比较上述N个驾驶意图与车辆在自动驾驶系统控制下的驾驶行为是否一致。其中,若不一致,计数器可以记第一数值,如第一数值为1;若一致,计数器可以记第二数值,如第二数值可以为0。如,自动驾驶装置比较驾驶意图1和车辆在自动驾驶系统控制下的驾驶行为是否一致,若不一致,计数器可以记1,若一致,计数器可以记0;自动驾驶装置比较驾驶意图2和车辆在自动驾驶系统控制下的驾驶行为是否一致,若不一致,计数器可以在1的基础上增加1,也即,计数器可以记2。直至自动驾驶装置比较完该N个驾驶意图与车辆在自动驾驶系统控制下的驾驶行为。
又一种可能的实现方式中,一旦自动驾驶装置检测到驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致时,自动驾驶装置获取驾驶员的视觉行为,并将该视觉 行为输入到上述驾驶意图模型中,得到该视觉行为对应的驾驶员的驾驶意图。自动驾驶装置比较该视觉行为对应的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为是否一致。
其中,若驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致,则计数器可以记第一数值;若驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为一致,则计数器可以记第二数值。
需要说明的是,自动驾驶装置可以在预设时间段内循环的执行上述步骤3~步骤4,进而自动驾驶装置可以确定预设时间段内驾驶员的驾驶意图与车辆的在自动驾驶系统控制下的驾驶行为不一致的次数。
例如,在第一时刻,自动驾驶装置确定第一时刻驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致,自动驾驶装置可以根据步骤3和步骤4确定第一时刻之后的第二时刻的驾驶员的视觉行为2对应的驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为是否一致,若不一致,计数器记1;若一致,计数器不计数或记0。这样,自动驾驶装置可以重复执行上述步骤4和步骤4,确定预设时间段内驾驶员的驾驶意图与车辆在自动驾驶系统控制下的驾驶行为不一致的次数。
又一种可能的实现方式中,当驾驶员的情绪行为为负面情绪行为时,自动驾驶装置可以确定驾驶员的驾驶意图为:不满意目标车辆的在自动驾驶系统控制下的驾驶行为,也即,自动驾驶装置可以检测到目标车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图存在冲突。
步骤503、自动驾驶装置更新自动驾驶系统。
其中,目标车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。
一种可能的实现方式中,自动驾驶装置可以根据驾驶员的驾驶意图以及目标车辆在自动驾驶系统控制下的驾驶行为对应的驾驶数据,得到第一特征参数;自动驾驶装置将第一特征参数输入预设神经网络模型,得到第一驾驶行为;自动驾驶装置根据第一驾驶行为,更新自动驾驶系统。
其中,第一特征参数表征驾驶员的驾驶意图以及目标车辆在自动驾驶系统控制下的驾驶行为对应的驾驶数据。
例如,第一特征参数可以为特征向量。该特征向量可以包括多个参数,该多个参数中每个参数可以用于表示一个驾驶数据。比如,第一特征参数可以包括目标车辆与其他车辆的间距、目标车辆的车速、行驶方向以及转向角度等。
其中,预设神经网络模型用于确定匹配驾驶员的驾驶意图的驾驶行为。该预设神经网络模型可以为预先设置的。比如,该预设神经网络模型可以为深度前馈神经网络回归模型。
例如,该预设神经网络模型为根据驾驶员的多个历史驾驶数据训练得到。该预设神经网络模型的输入为特征参数,输出为满足驾驶员的驾驶意图的第一驾驶行为。也即,第一驾驶行为为满足驾驶员的驾驶意图的驾驶行为。第一驾驶行为可以包括一个或多个驾驶数据,例如,该一个或多个驾驶数据可以为车辆的车速,与前车的距离等。预设神经网络模型的训练方法可以参照现有技术,此处不再赘述。
可以理解的是,本申请实施例中,自动驾驶装置可以将上述特征参数以特征向量的形式输入预设神经网络模型,得到与该特征向量对应的输出向量。其中,该输出向量可以包 括一个或多个参数,该一个或多个参数中每个参数对应一个驾驶行为。
其中,自动驾驶装置根据第一驾驶行为,更新自动驾驶系统是指自动驾驶装置可以将第一驾驶行为与车辆的在自动驾驶系统控制下的驾驶行为进行对比,根据对比结果,更新自动驾驶系统,如,自动驾驶装置可以将车辆的在自动驾驶系统控制下的驾驶行为替换为第一驾驶行为。
例如,以自动驾驶装置确定的第一驾驶行为为将目标车辆的车速提高到第一车速为例。目标车辆的在自动驾驶系统控制下的驾驶行为是保持目标车辆的车速为第二车速。其中,第一车速大于第二车速。自动驾驶装置可以将目标车辆的在自动驾驶系统控制下的车速替换为第一车速,以使得目标车辆的在自动驾驶系统控制下的车速为第一车速。
本申请实施例中,在目标车辆的自动驾驶过程中,自动驾驶装置可以循环执行上述步骤501~步骤503,直至目标车辆停止行驶,其中,目标车辆停止行驶包括:驾驶员控制目标车辆停止行驶,或者驾驶员关闭目标车辆的自动驾驶系统,比如,驾驶员手动驾驶。
基于图5的技术方案,自动驾驶装置在检测到在自动驾驶系统控制下自动驾驶的车辆的驾驶员的驾驶意图与车辆的自动驾驶系统控制下的车辆的驾驶行为冲突时,更新自动驾驶系统,以使得车辆在更新后的自动驾驶系统控制下的驾驶行为匹配驾驶员的驾驶意图。基于此,后续在更新后的自动驾驶系统的控制下的驾驶行为可以满足驾驶员的驾驶需求。从而,解决了在自动驾驶系统控制下的车辆的驾驶行为与驾驶员的驾驶意图冲突的问题。
在图5所示方法的一种可能的实现方式中,本申请实施例提供的方法,还可以包括:自动驾驶装置检测目标车辆的驾驶员是否为新的驾驶员。
其中,新的驾驶员是指没有驾驶过该目标车辆的人员。
一种示例中,当自动驾驶装置检测到目标车辆启动时,自动驾驶装置可以获取驾驶员的特征信息,并将驾驶员的特征信息与自动驾驶装置具有的多个驾驶员的特征信息进行匹配。若该驾驶员的特征信息与该多个驾驶员的特征信息中每个驾驶员的特征信息均不匹配,则自动驾驶装置确定目标车辆的驾驶员为新的驾驶员;若该驾驶员的特征信息与该多个驾驶员的特征信息中一个驾驶员的特征信息匹配,则自动驾驶装置确定目标车辆的驾驶员为驾驶过该目标车辆的驾驶员。
在图5所示方法的另一种可能的实现方式中,本申请实施例提供的方法中,自动驾驶装置还可以根据目标车辆的驾驶员的特征信确定目标车辆在自动驾驶系统控制下的驾驶模式。
其中,若目标车辆的驾驶员不是新的驾驶员,自动驾驶装置根据驾驶员的特征信息确定目标车辆的自动驾驶模式。也就是说,若目标车辆的驾驶员为驾驶过该目标车辆的驾驶员,自动驾驶装置可以根据该驾驶员的特征信息确定目标车辆的自动驾驶模式。
例如,自动驾驶装置可以从图4所示的V2X服务器中获取该驾驶员使用过的自动驾驶模式,并将该使用过的自动驾驶模式作为目标车辆的自动驾驶模式。
自动驾驶装置也可以存储有该驾驶员使用的历史的自动驾驶模式,自动驾驶装置可以将存储的该驾驶员使用的自动驾驶模式作为目标车辆的自动驾驶模式,不予限制。
其中,若目标车辆的驾驶员为新的驾驶员,自动驾驶装置可以将预设的自动驾驶模式作为目标车辆的自动驾驶模式。其中,预设的自动驾驶模式可以预先配置在自动驾驶装置中。自动驾驶装置也可以从V2X服务器中的多个驾驶员使用的历史的自动驾驶模式中获取 预设的自动驾驶模式,不予限制。例如,自动驾驶装置可以从多个驾驶员使用的历史的自动驾驶模式中选择使用次数最多的自动驾驶模式作为预设的自动驾驶模式。
在图5所示方法的又一种可能的实现方式中,本申请实施例提供的方法,还可以包括:自动驾驶装置获取目标车辆的在自动驾驶系统控制下的驾驶行为。
其中,在不同的行驶场景下,目标车辆的在自动驾驶系统控制下的具有不同的驾驶行为。具体的,行驶场景以及行驶场景对应的目标车辆在自动驾驶系统控制下的驾驶行为可以如表1所示,不予赘述。
例如,以第一行驶场景为例,自动驾驶装置可以获取第一行驶场景下驾驶员的驾驶意图,以及第一行驶场景对应的目标车辆在自动控制系统控制下的驾驶行为。若自动驾驶装置检测到第一行驶场景下驾驶员的驾驶意图与目标车辆的在自动控制系统控制下的驾驶行为存在冲突,则自动驾驶装置可以更新自动驾驶系统。
需要说明的是,本申请实施例中,在自动驾驶装置确定第一行驶场景下目标车辆的驾驶行为与驾驶员的驾驶意图冲突后,自动驾驶装置可以在目标车辆的行驶过程中,更新自动驾驶系统中第一行驶场景对应的驾驶行为。自动驾驶装置也可以在目标车辆停止行驶后,更新自动驾驶系统中第一行驶场景对应的自动驾驶行为,如,自动驾驶装置可以在目标车辆到达目的地后,自动驾驶装置更新自动驾驶系统中第一行驶场景对应的自动驾驶行为,不予限制。
例如,在目标车辆在自动行驶的过程中,自动驾驶装置可以记录多个与驾驶员的驾驶意图冲突的行驶场景以及驾驶员的驾驶意图。比如,自动驾驶装置记录的行驶场景包括行驶场景1、行驶场景2。其中,行驶场景1和行驶场景2为车辆的在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图冲突的行驶场景。自动驾驶装置可以在车辆结束自动驾驶之后,更新行驶场景1对应的车辆的在自动驾驶系统控制下的驾驶行为,以及更新行驶场景2对应的车辆的在自动驾驶系统控制下的驾驶行为。
示例性的,自动驾驶装置可以通过多种方式获取目标车辆的第一行驶场景,例如,该多种方式可以包括:自动驾驶装置通过目标车辆的设备获取目标车辆的第一行驶场景,如:自动驾驶装置可以通过多种传感器获取目标车辆的第一行驶场景。自动驾驶装置从其他装置的交互获取目标车辆的第一行驶场景,如:自动驾驶装置可以通过与V2X服务器的交互获取目标车辆的第一行驶场景,自动驾驶装置可以通过与其他车辆的自动驾驶装置的交互获取目标车辆的第一行驶场景。自动驾驶装置也可以根据上述多种方式的一种或多种获取目标车辆的第一行驶场景,不予限制。下面对上述多种方式进行说明:
1、自动驾驶装置可以通过多种传感器获取目标车辆的第一行驶场景。
自动驾驶装置可以通过多种传感器采集目标车辆行驶道路的交通信息,并对采集到目标车辆行驶道路的交通信息进行整合,得到目标车辆的第一行驶场景。
本申请实施例中,多种传感器可以包括激光雷达,毫米波雷达,超声波雷达,单目或双目摄像头等,不予限制。
例如,自动驾驶装置可以通过激光雷达获取目标车辆周围的车辆的车速以及行人的车速等。自动驾驶装置可以通过单目或双目摄像头获取目标车辆行驶道路的路况信息、车辆的数量,行人的数量等,不予限制。
2、自动驾驶装置可以通过与V2X服务器的交互获取目标车辆的第一行驶场景。
自动驾驶装置可以向V2X服务器发送用于请求第一行驶场景的请求信息。相应的,V2X服务器接收来自自动驾驶装置的用于请求第一行驶场景的请求信息。V2X服务器根据该请求消息确定第一行驶场景,并将该第一行驶场景发送给自动驾驶装置。
例如,自动驾驶装置可以通过图2中的车身天线将目标车辆的坐标信息(如全球定位系统(global positioning system,GPS)坐标)发送给V2X服务器,V2X服务器在接收到目标车辆的坐标信息后,根据该坐标信息确定与该坐标信息对应的行驶场景。V2X服务器向自动驾驶装置发送与该坐标信息对应的第一行驶场景。相应的,自动驾驶装置可以通过车身天线接收与该坐标信息对应的第一行驶场景。
3、自动驾驶装置可以通过与其他车辆进行交互获取目标车辆的第一行驶场景。
例如,该方式可以参照自动驾驶装置与V2X服务器的交互获取目标车辆的第一行驶场景的方式,此处不再赘述。
示例性的,自动驾驶装置获取到第一行驶场景后,可以获取第一行驶场景下驾驶员的行为特征,根据驾驶员的行为特征确定驾驶员的驾驶意图。
其中,驾驶员的行为特征可以用于表征驾驶员是否满意目标车辆的在自动驾驶系统控制下的驾驶行为,如:驾驶员的行为特征可以包括驾驶员的操作特征、视觉行为、情绪行为、身体姿态行为等,此时,自动驾驶装置可以根据这些行为特征中的一种或多种确定驾驶员的驾驶意图。具体的,驾驶员的行为特征的相关描述、以及车载在装置根据驾驶员的行为特征确定驾驶员的驾驶意图的实现方式可参照图5的技术方案所示。
基于该可能的实现方式,在自动驾驶装置检测到第一行驶场景下车辆的驾驶员的驾驶意图与车辆的在自动驾驶系统控制下的驾驶行为存在冲突时,自动驾驶装置可以将该自动驾驶系统对应的驾驶行为调整为满足第一行驶场景下驾驶员的驾驶意图的驾驶行为,以便于后续自动驾驶装置在面对类型的行驶场景时车辆的在自动驾驶系统控制下的驾驶行为可以满足驾驶员的需求。从而,解决了车辆在自动驾驶系统控制下的驾驶行为与驾驶员的驾驶意图冲突的问题。
在图5所示方法的又一种可能的实现方式中,本申请实施例提供的方法中,还可以包括:自动驾驶装置确定第一行驶场景下目标车辆的自动驾驶模式。
其中,自动驾驶模式的相关描述可参照上述,不予赘述。
其中,自动驾驶装置确定第一行驶场景下目标车辆的自动驾驶模式可以包括:自动驾驶装置自动选择,如:自动驾驶装置根据第一行驶场景的一些相关信息自动从预配置的多个自动驾驶模式中选择出目标车辆的自动驾驶模式;或者,驾驶员手动选择,如:在驾驶员希望自动驾驶时,由驾驶员手动选择适合的自动驾驶模式,并触发自动驾驶装置根据驾驶员选择的自动驾驶模式进行自动驾驶。具体的,该方式可参照下述方式一或方式二中所述。
方式一、自动驾驶装置自动选择。
自动驾驶装置可以根据驾驶员的特征信息,从一个或多个自动驾驶模式中选择的自动驾驶模式。其中,一个驾驶员的特征信息用于唯一标识一个驾驶员。该一个或多个自动驾驶模式可以为自动驾驶装置存储的自动驾驶模式,也可以为自动驾驶装置从图2中的V2X服务器处获取的自动驾驶模式。
示例性的,自动驾驶装置可以通过摄像装置获取驾驶员的图像信息,图像信息可以为 视频信息或图片信息。该图像信息可以包括驾驶员的面部特征或者瞳孔特征。自动驾驶装置可以通过对驾驶员的图像信息进行识别,确定驾驶员的特征信息,或者,自动驾驶装置可以将驾驶员的图像信息发送给V2X服务器,以使得V2X服务器对驾驶员的图像信息进行识别,得到驾驶员的特征信息。
例如,自动驾驶装置可以根据驾驶员的特征信息与自动驾驶装置具有的多个驾驶员的特征信息进行匹配,确定与驾驶员的特征信息匹配的自动驾驶模式。其中,该多个驾驶员的特征信息中每个驾驶员的特征信息具有对应的自动驾驶模式。如,自动驾驶装置可以将该多个驾驶员的特征信息中与目标车辆的驾驶员的特征信息相似度大于预设值的驾驶员的特征信息对应的自动驾驶模式,作为目标车辆的驾驶员的自动驾驶模式。其中,预设值可以为自动驾驶装置预先设置的,不予限制。
又例如,自动驾驶装置也可以将获取的驾驶员的特征信息发送给V2X服务器,V2X服务器在接收到驾驶员的特征信息后,可以根据该驾驶员的特征信息与具有的一个或多个自动驾驶模式对应的驾驶员的特征信息进行匹配,从一个或多个自动驾驶模式选择确定与驾驶员的特征信息匹配的自动驾驶模式。V2X服务器在确定与驾驶员的特征信息匹配的自动驾驶模式之后,可以将该与驾驶员的特征信息匹配的自动驾驶模式发送给自动驾驶装置。相应的,自动驾驶装置接收来自V2X服务器的与驾驶员的特征信息匹配的自动驾驶模式后,可以根据该与驾驶员的特性信息匹配的自动驾驶模式控制目标车辆进行自动行驶。具体的,可以参照上述自动驾驶装置确定目标车载的驾驶员的自动驾驶模式的描述,此处不再赘述。
方式二、驾驶员手动选择。
方式二中,一种可能的设计中,自动驾驶模式可以为驾驶员在启动车辆后手动选择的自动驾驶模式。例如,目标车辆可以设置有显示装置,当自动驾驶装置检测到目标车辆启动时,可以触发该显示装置显示图8所示的显示界面,驾驶员可以根据需求通过该显示界面选择自动驾驶模式。当自动驾驶装置检测到驾驶员的操作指令后,可以根据驾驶员的操作指令确定与该操作指令对应的自动驾驶模式。
例如,图8中的显示界面可以包括“模式1”~“模式6”多个触摸按键,每个触摸按键对应一种自动驾驶模式。当驾驶员点击某一个触摸按键时,可以触发自动驾驶装置获取该触摸按键对应的自动驾驶模式。如,当驾驶员点击触摸按键“模式5”,可以触发自动驾驶装置可以获取触摸按键“模式5”对应的自动驾驶模式。
方式二中,又一种可能的设计中,驾驶员可以通过与目标车辆对应的终端手动选择自动驾驶模式,并触发与目标车辆对应的终端将选择出的自动驾驶模式通知给自动驾驶装置,自动驾驶装置根据该终端的通知确定第一行驶场景下目标车辆的自动驾驶模式。
其中,与自动驾驶装置对应的终端可以包括控制目标车辆的任一装置,如:智能手机、移动硬盘、个人笔记本、平板电脑等,不予限制。
例如,以终端为智能手机为例,驾驶员通过手机通过有线的方式或无线的方式将手机等终端具有的自动驾驶模式发送给自动驾驶装置,例如,有线的方式可以为通用串行总线(universal serial bus,USB)、Type-C等连接自动驾驶装置。无线的方式可以通过蓝牙、无线保真(wireless fidelity,WiFi)等连接自动驾驶装置,不予限制。相应的,自动驾驶装置可以接收来自智能手机的自动驾驶模式。
在图5所示方法的又一种可能的实现方式中,本申请实施例提供的方法中,在步骤 503之前,该方法还可以包括:
S1、自动驾驶装置根据驾驶员的驾驶意图以及车辆在自动驾驶系统控制下的驾驶行为,确定第一特征参数。
其中,第一特征参数可以为特征向量。下面以第一特征参数为特征向量为例,分别对行驶场景为城市高峰顺行跟车、城市高峰顺行跟车、城市高峰车辆主动切入时,自动驾驶装置根据驾驶员的驾驶意图以及车辆在自动驾驶系统控制下的驾驶行为,确定的特征向量进行说明。
需要说明的是,自动驾驶装置在确定驾驶员的驾驶意图以后,自动驾驶装置可以将驾驶员的驾驶意图以及车辆信息用数字或字符表示,以便于确定行驶场景对应的特征参数。
例如,v+表示驾驶员的驾驶意图为提高车辆的车速,v-表示驾驶员的驾驶意图为降低车辆的车速,s表示驾驶员的驾驶意图为靠边停车,s表示驾驶员的驾驶意图为向左车道靠拢,r表示驾驶员的驾驶意图为向右车道靠拢,l+v表示驾驶员的驾驶意图为向左车道靠拢时提高车辆的车速,l-v表示驾驶员的驾驶意图为向左车道靠拢时降低车辆的车速等。驾驶员的驾驶意图还可以用其他行驶表示,不予限制。
又例如,车辆类型可以用数字或字符表示。例如,以数字表示车辆类型为例,“1”可以表示车辆类型为轿车、“2”可以表示车辆类型为卡车、“3”可以表示车辆类型为公交车,不予限制。
行驶场景1、行驶场景为城市高峰顺行跟车。
由于车辆在顺行跟车时,目标车辆的行驶参数与前车的车辆信息相关,自动驾驶装置可以根据目标车辆的车辆信息和前方车辆的车辆信息构建特征向量。例如,自动驾驶装置确定的特征向量可以为:
f={v+,d f,v f,v e,s f,1}。
其中,d f表示目标车辆与前方车辆的距离,v f表示前方车辆的车速,v e表示目标车辆的车速,s f前方车辆的大小,i f表示前方车辆的车辆类型。该特征向量中每个变量可以为包括多个维度,例如,车辆的大小可以包括:车长、车宽、车高等三个维度。车辆的车辆类型可以包括轿车、卡车、公交车等。
行驶场景2、行驶场景为城市高峰顺行车道保持。
由于目标车辆在车道保持时,目标车辆的行驶参数与前方车辆、左方车辆以及右方车辆的车辆信息相关,自动驾驶装置可以根据目标车辆的车辆信息、前方车辆的车辆信息、左方车辆的车辆信息以及右方车辆的车辆信息构建特征向量。例如,车辆装置确定的特征向量可以为:
f={v-,d f,v f,v e,s f,i f,i l,s l,v lv,v ll,d l,i r,s r,v rv,v rl,d r}。
其中,d f,v f,v e,s f,i f,可以参照行驶场景1的描述,此处不再赘述。
i l表示左方车辆的类型,s l表示左方车辆的大小,v lv表示左方车辆的横向车速,v ll表示左方车辆的纵向车速,d l表示左方车辆与目标车辆的横向相对距离。其中,横向相对距离是指两个车辆之间的水平距离。
i r表示右方车辆的类型,s r表示右方车辆的大小,v rv表示右方车辆的横向车速,v rl表示右方车辆的纵向车速,d r表示右方车辆与目标车辆的横向相对距离。
进一步的,对于目标车辆的前方或左右方没有车辆时,为了保证每一个主体任务下的 每一个行驶场景的特征向量的维度相等,自动驾驶装置可以对特征向量按照以下准则做补齐处理:
对特征向量中的参数赋予一个预设值,以保证赋值后的特征向量对目标车辆的行驶无任何影响。以目标车辆的前方没有车辆为例,可以做如下设置:
前方车辆的类型为自行车;前方车辆的大小足够小,如{0,0,0};前方车辆与目标车辆的距离为较大的数值,如3000m;前方车辆的速度为较大的数值,如800km/h。
其中,对于目标车辆的左方或右方没有车辆时,自动驾驶装置对特征向量中的参数的补齐处理方法可以参照上述自动驾驶装置对目标车辆的前方没有车辆时,自动驾驶装置对特征向量中的参数的补齐处理方法,此处不再赘述。
行驶场景3:行驶场景为城市高峰车辆主动切入。
由于车辆主动切入时,目标车辆的行驶参数与被切入车辆的车辆信息相关,自动驾驶装置可以根据目标车辆的车辆信息和被切入车辆的车辆信息构建特征向量。例如,自动驾驶装置确定的特征向量可以为:
f={l-v,d l,d v,v ev,v el,v l,i,s}
其中,d l表示目标车辆与被切入车辆之间的纵向距离,d v表示目标车辆与被切入车辆之间的横向距离、v ev表示目标车辆的切入速度,v el表示目标车辆切入时的纵向速度,v l表示被切入车辆的纵向速度,i表示被切入车辆的类型,s表示被切入车辆的大小。
S2、自动驾驶装置将特征参数输入预设神经网络模型,得到第一驾驶行为。
其中,该预设神经网络模型为根据多个行驶场景下驾驶员的历史驾驶数据训练得到。该预设神经网络模型的输入为特征参数,输出为满足驾驶员的驾驶意图的第一驾驶行为。也即,第一驾驶行为为满足驾驶员的驾驶意图的驾驶行为。第一驾驶行为可以包括一个或多个驾驶数据,例如,该一个或多个驾驶数据可以为车辆的车速,与前车的距离等。
例如,该预设神经网络模型可以为预先设置的。例如,该预设神经网络模型可以为深度前馈神经网络回归模型。
其中,该预设神经网络模型可以根据多个样本特征参数训练得到,一个样本特征参数对应一个驾驶员的驾驶行为。预设神经网络模型的训练方法可以参照现有技术,此处不再赘述。
可以理解的是,本申请实施例中,自动驾驶装置可以将上述特征参数以特征向量的形式输入预设神经网络模型,得到与该特征向量对应的输出向量。其中,该输出向量可以包括一个或多个参数,该一个或多个参数中每个参数对应一个驾驶行为。
例如,当行驶场景为城市高峰顺行跟车时,特征参数对应的特征向量为f={w,d f,v f,v e,s f,i f}={v-,5,40,50,(4.6,1.7,1.4),1}。也就是说,驾驶员的驾驶意图为降低车辆的车速,目标车辆与前方车辆的距离为5m,目标车辆的前方车辆的车速为40km/h,目标车辆的车速为50km/h,前方车辆的长为4.6m,宽为1.7m,高为1.4m,前方车辆的车辆类型为轿车。自动驾驶装置将该特征向量输入预设神经网络模型,得到的输出向量为{40,0,0,0}。其中,40表示目标车辆的车速为40km/h,后三个0分别表示目标车辆不减速、不靠边停车、不退出跟车模式。也即,第一驾驶行为是:控制目标车辆的车速为40km/h。
S3、自动驾驶装置向驾驶员呈现一个或多个问题。并接收驾驶员针对该一个或多个问题的回答。其中,该一个或多个问题的回答可以用于指示是否更新自动驾驶系统。
例如,自动驾驶装置可以根据第一驾驶行为以及第一行驶场景,从预设问题库中选择一个或多个问题。自动驾驶装置可以通过语音播放装置播放该一个或多个问题,或者可以通过显示装置显示该一个或多个问题。
其中,该预设问题库为根据预设时间内多个驾驶员的驾驶行为确定。该实现方式的实现过程可以为:
1、自动驾驶装置将车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为对比,确定车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为是否一致/相同。例如,若车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为的差值满足预设条件/不超过预设范围,则自动驾驶装置确定车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为一致/相同;若车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为的差值不满足预设条件或超过预设范围,则自动驾驶装置确定车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为不一致/不相同。
比如,第一行驶场景下,如目标车辆执行超车时,车辆在自动驾驶系统的控制下的是将目标车辆的车速提高到80km/h,第一驾驶行为是将目标车辆的车速提高到81km/h,预设范围为{-2km/h,2km/h}。由于80km/h-81km/h=-1km/h,也即车辆在自动驾驶系统的控制下的车速与第一驾驶行为对应的车速的差值在该预设范围内,则自动驾驶装置确定车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为一致/相同。
又比如,第一行驶场景下,如目标车辆执行跟车时,车辆在自动驾驶系统的控制下的驾驶行为是控制目标车辆与前方车辆的距离为10m,第一驾驶行为是控制目标车辆与前方车辆的距离为15m,预设条件是控制目标车辆与前方车辆的距离不超过12m。由于15m大于12m,也即第一驾驶行为不满足预设条件,则自动驾驶装置确定车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为不一致/不相同。
若车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为一致/相同,则自动驾驶装置可以确定车辆在自动驾驶系统的控制下的驾驶行为与驾驶员的驾驶意图不冲突;若车辆在自动驾驶系统的控制下的驾驶行为与第一驾驶行为不一致/不相同,自动驾驶装置可以比较车辆在自动驾驶系统控制下的驾驶行为与第一驾驶行为,确定车辆在自动驾驶系统的控制下的驾驶行为的多个驾驶数据中与第一驾驶行为的驾驶数据不一致的驾驶数据,以及该不一致的驾驶数据的变化信息。
2、自动驾驶装置将该不一致/不相同的驾驶数据以及该不一致/不相同的驾驶数据的变化信息与预设问题库进行匹配,确定该不一致/不相同的驾驶数据对应的问题。
3、自动驾驶装置通过语音的形式播放,或者通过显示装置显示该不一致的驾驶数据对应的问题。自动驾驶装置通过麦克风接收驾驶员的语音,并对该驾驶员的语音进行识别,确定驾驶员针对该一个或多个问题的回答。其中,该驾驶员的语音为驾驶员对该不一致的驾驶数据对应的问题的回答。
需要说明的,若自动驾驶装置在预设时间内未接收到驾驶员的语音,自动驾驶装置可以再次播放该不一致的驾驶数据对应的问题,自动驾驶装置也可以将预设答案作为驾驶员的回答。其中,预设时间可以为自动驾驶装置预先设置的,不予限制。例如,以预设时间为10s为例,若自动驾驶装置在10s内未接收到驾驶员的语音,自动驾驶装置的预设答案为肯定答案,比如,肯定答案可以为“是”。
4、当该一个或多个问题的回答用于指示更新自动驾驶系统时,自动驾驶装置可以更新自动驾驶系统。
例如,若该一个或多个问题的回答为“是”,则该一个或多个问题的回答可以用于指示更新自动驾驶系统。
本申请实施例中,为了降低确定驾驶员的驾驶意图的复杂程度,问题库中的问题可以设置为简单并且直接的问题,如,设置的问题可以为判断式的问题。相应的,驾驶员给出的答案可以为“是”或“不是”。
一种示例中,当行驶场景为城市高峰顺行跟车时,对应的问题库可以如表2所示。
表2
问题1 是否加速?
问题2 是否减速?
问题3 是否靠边停车?
问题4 是否退出跟车模式?
需要说明的是,当自动驾驶装置根据表2中的问题播放的问题为问题1时,若自动驾驶装置接收到驾驶员的回答为“是”,也就是说,自动驾驶装置确定驾驶员的驾驶意图为提高车辆的速度。
下面结合表2中问题以及上述例子的特征向量和输出向量,对上述实现方式的过程进行说明:
1、自动驾驶装置将车辆在自动驾驶系统控制下的驾驶行为(目标车辆的车速为50km/h)与第一驾驶行为(目标车辆的车速为40km/h)对比,确定车辆在自动驾驶系统控制下的驾驶行为与第一驾驶行为不一致的驾驶数据为目标车辆的车速,且目标车辆的车速的变化信息为车速从50km/h减少至40km/h。
2、自动驾驶装置将目标车辆的车速以及车速的变化信息与车速与表2中问题进行匹配,确定目标车辆的车速对应的问题为表2中的问题2。
3、自动驾驶装置以语音的形式播放表2中的问题2。自动驾驶装置通过麦克风接收驾驶员的语音,并对该驾驶员的语音进行识别,确定驾驶员的答案。
若自动驾驶装置根据识别后的语音,确定驾驶员的答案为“是”,则自动驾驶装置可以确定驾驶员的驾驶意图是降低目标车辆的车速。
4、在确定驾驶员的答案为“是”的情况下,自动驾驶装置将目标车辆的车速降低为40km/h,并将车辆在自动驾驶系统控制下的驾驶行为中的车速更新为40km/h。
若自动驾驶装置根据识别后的语音,确定驾驶员的答案为“否”,则自动驾驶装置可以确定驾驶员的驾驶意图是不降低目标车辆的车速。
在驾驶员的答案为“否”的情况下,自动驾驶装置还可以继续播放表2中的问题1。若自动驾驶装置确定表2中的问题1对应的驾驶员的答案为“否”,自动驾驶装置不更新车辆在自动驾驶系统控制下的驾驶行为。
另一种示例中,当行驶场景为城市高峰顺行车道保持时,对应的问题库可以如表3所示。
表3
问题1 是否加速?
问题2 是否减速?
问题3 是否向左车道靠拢?
问题4 是否向右车道靠拢?
问题5 是否靠边停车?
问题6 是否退出车道保持模式?
又一种示例中,当行驶场景为城市高峰顺行车辆主动切入时,对应的问题库可以如表4所示。
表4
问题1 是否提高切入速度?
问题2 是否减小切入速度?
问题3 是否提高切入距离?
问题4 是否减小切入距离?
问题5 是否同时提高切入距离和切入速度?
问题6 是否减少切入距离,提高切入速度?
问题7 是否提高切入距离,减少切入速度?
问题8 是否同时减少切入距离和切入速度?
问题9 是否靠边停车?
问题10 是否退出主动切入模式?
需要说明的是,自动驾驶装置根据表3以及表4中的问题更新车辆在自动驾驶系统控制下的驾驶行为的过程可以参照上述表2,此处不再赘述。上述表2、表3以及表4中的问题仅为示例性的,每个表还可以包括其他问题,不予限制。
进一步的,自动驾驶装置在确定第一驾驶行为之后,自动驾驶装置还可以检测第一驾驶行为是否包括在预设的驾驶行为内。其中,该预设的驾驶行为可以为符合目标车辆的安全驾驶行为。
其中,若第一驾驶行为符合目标车辆的安全驾驶行为,则自动驾驶装置可以将第一驾驶行为作为目标车辆在更新后的自动驾驶系统控制下的驾驶行为;若第一驾驶行为不符合目标车辆的安全驾驶行为,则自动驾驶装置可以将安全驾驶行为作为目标车辆在更新后的自动驾驶系统控制下的驾驶行为。
例如,自动驾驶装置确定第一驾驶行为为将目标车辆的车速提高到80km/h。但是,目标车辆行驶的某一道路限制车辆的最高车速为60km/h,则自动驾驶装置检测到第一驾驶行为不符合目标车辆的安全驾驶行为,则自动驾驶装置可以将车速为60km/h作为车辆的在更新后的自动驾驶系统控制下的车速。也即,在目标车辆的后续的自动行驶过程中,目标车辆的在自动驾驶系统控制下的车速为60km/h。
进一步的,自动驾驶装置可以发送提示信息,该提示信息用于提示驾驶员目标车辆将要超速行驶。
进一步的,本申请实施例提供的方法还可以包括:自动驾驶装置确定第二行驶场景下目标车辆的驾驶员的特征信息。当第一行驶场景和第二行驶场景匹配,且第一行驶场景下驾驶员的特征信息与第二行驶场景下驾驶员的特征信息一致时,自动驾驶装置确定第二行驶场景下目标车辆的驾驶行为是在更新后的自动驾驶系统下的驾驶行为。
其中,目标车辆在第二行驶场景下使用的自动驾驶模式为更新后的自动驾驶模式。
其中,第一行驶场景和第二行驶场景匹配是指第一行驶场景的相似度大于或等于预设值。第一行驶场景下驾驶员的特征信息与第二行驶场景下驾驶员的特征信息一致是指第一行驶场景下目标车辆的驾驶员和第二行驶场景下目标车辆的驾驶员为同一个驾驶员。
基于该可能的实现方式,在后续面对相同或相似的行驶场景时,自动驾驶装置可以根据更新后的自动驾驶系统的驾驶行为控制目标车辆行驶,解决了在与第一行驶场景相同或相似的行驶场景下,车辆在自动驾驶系统控制下的驾驶行为仍然与驾驶员的驾驶意图冲突的问题。
下面结合图1所示通信系统,对图5所示方法进行详细描述。
如图9所示,为本申请实施例提供的又一种自适应优化自动驾驶系统的方法,该方法可以包括:
下面结合图1所示通信系统,对图5所示方法进行详细描述。
如图9所示,为本申请实施例提供的自适应优化自动驾驶系统的方法,该方法可以包括:
步骤901、自动驾驶装置检测目标车辆的驾驶员是否为新的驾驶员。
步骤902、自动驾驶装置根据驾驶员的特征信息确定目标车辆的自动驾驶模式。
也就是说,若目标车辆的驾驶员为驾驶过该目标车辆的驾驶员,自动驾驶装置可以根据该驾驶员的特征信息确定目标车辆的自动驾驶模式。
例如,自动驾驶装置可以从图4所示的V2X服务器中获取该驾驶员使用过的自动驾驶模式,并将该使用过的自动驾驶模式作为目标车辆的自动驾驶模式。自动驾驶装置也可以存储有该驾驶员使用的历史的自动驾驶模式,自动驾驶装置可以将存储的该驾驶员使用的自动驾驶模式作为目标车辆的自动驾驶模式,不予限制。
步骤903、自动驾驶装置获取第一行驶场景下驾驶员的行为对应的特征参数。
步骤904、自动驾驶装置根据驾驶员的行为对应的特征参数确定驾驶员的驾驶意图。
步骤905、自动驾驶装置根据驾驶员的驾驶意图以及车辆在自动驾驶系统控制下的驾驶行为,确定第一特征参数。
步骤906、自动驾驶装置将第一特征参数输入预设神经网络模型,得到第一驾驶行为。
步骤907、自动驾驶装置根据第一驾驶行为,更新自动驾驶系统。
上述步骤901~步骤907的描述请参照图5的技术方法,不予赘述。
本申请上述实施例中的各个方案在不矛盾的前提下,均可以进行结合。
上述本申请提供的实施例中,从自动驾驶装置的角度对本申请实施例提供的方法进行了介绍。可以理解的是,自动驾驶装置为了实现上述本申请实施例提供的方法中的各功能,自动驾驶装置包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到,结合本文中所公开的实施例描述的各示例的算法步骤,本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对自动驾驶装置进行功能模块的划分,例如,可以对应各个功能划分各个功能模块,也可以将两个或两个以上的功能集成在一个处理模块 中。上述集成的模块既可以采用硬件的形式实现,也可以采用软件功能模块的形式实现。需要说明的是,本申请实施例中对模块的划分是示意性的,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。
在采用集成的单元的情况下,图10示出了上述实施例中所涉及的装置(记为自动驾驶装置100)的一种可能的结构示意图,该自动驾驶装置100包括通信单元1002和处理单元1001,还可以包括存储单元1003。图10所示的结构示意图可以用于示意上述实施例中所涉及自动驾驶装置的结构。图10所示的结构示意图可以用于示意上述实施例中所涉及V2X服务器的结构。
当图10所示的结构示意图用于示意上述实施例中所涉及的自动驾驶装置的结构时,处理单元1001用于对自动驾驶装置的动作进行控制管理,例如,处理单元1001用于执行图5中的步骤501、步骤502、步骤503,图9中的步骤901、步骤902、步骤904、步骤905、步骤906、步骤906、步骤907,通过通信单元1002执行图9中的步骤903,和/或本申请实施例中所描述的其他过程中的自动驾驶装置执行的动作。处理单元1001可以通过通信单元1002与其他网络实体通信,例如,与图1中示出的V2X服务器20通信。存储单元1003用于存储自动驾驶装置的程序代码和数据。
当图10所示的结构示意图用于示意上述实施例中所涉及的自动驾驶装置的结构时,自动驾驶装置100可以是自动驾驶装置,也可以是自动驾驶装置内的芯片。
其中,当自动驾驶装置100为自动驾驶装置时,处理单元1001可以是处理器或控制器,通信单元1002可以是通信接口、收发器、收发机、收发电路、收发装置等。其中,通信接口是统称,可以包括一个或多个接口。存储单元1003可以是存储器。当自动驾驶装置100为自动驾驶装置内的芯片时,处理单元1001可以是处理器或控制器,通信单元1002可以是输入接口和/或输出接口、管脚或电路等。存储单元1003可以是该芯片内的存储单元(例如,寄存器、缓存等),也可以是自动驾驶装置内的位于该芯片外部的存储单元(例如,只读存储器(read-onlymemory,简称ROM)、随机存取存储器(random access memory,简称RAM)等)。
当图10所示的结构示意图用于示意上述实施例中所涉及的V2X服务器的结构时,处理单元1001用于对V2X服务器的动作进行控制管理,例如,处理单元1001用于执行本申请实施例中所描述的其他过程中的V2X服务器执行的动作。处理单元1001可以通过通信单元1002与其他网络实体通信,例如,与图1中示出的车辆10通信。存储单元1003用于存储V2X服务器的程序代码和数据。
当图10所示的结构示意图用于示意上述实施例中所涉及的V2X服务器的结构时,自动驾驶装置100可以是V2X服务器,也可以是V2X服务器内的芯片。
其中,当自动驾驶装置100为V2X服务器时,处理单元1001可以是处理器或控制器,通信单元1002可以是通信接口、收发器、收发机、收发电路、收发装置等。其中,通信接口是统称,可以包括一个或多个接口。存储单元1003可以是存储器。当自动驾驶装置100为自动驾驶装置内的芯片时,处理单元1001可以是处理器或控制器,通信单元1002可以是输入接口和/或输出接口、管脚或电路等。存储单元1003可以是该芯片内的存储单元(例如,寄存器、缓存等),也可以是自动驾驶装置内的位于该芯片外部的存储单元(例如,只读存储器(read-onlymemory,简称ROM)、随机存取存储器(random access memory, 简称RAM)等)。
图10中的集成的单元如果以软件功能模块的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请实施例的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者第一接入网设备等)或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。存储计算机软件产品的存储介质包括:U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
图10中的单元也可以称为模块,例如,处理单元可以称为处理模块。
如图11所示,图11示出了本申请实施例提供的一种通信系统示例图,包括V2X服务器1101和自动驾驶装置1102。
V2X服务器1101用于执行上述实施例中V2X服务器执行的动作,例如,V2X服务器1101向自动驾驶装置发送预设的自动驾驶模式。
自动驾驶装置1102用于执行上述实施例在自动驾驶装置执行的动作,例如,自动驾驶装置1102用于执行图5和图9中的步骤。
在实现过程中,本实施例提供的方法中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。结合本申请实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成,或者用处理器中的硬件及软件模块组合执行完成。
本申请中的处理器可以包括但不限于以下至少一种:中央处理单元(central processing unit,CPU)、微处理器、数字信号处理器(DSP)、微控制器(microcontroller unit,MCU)、或人工智能处理器等各类运行软件的计算设备,每种计算设备可包括一个或多个用于执行软件指令以进行运算或处理的核。该处理器可以是个单独的半导体芯片,也可以跟其他电路一起集成为一个半导体芯片,例如,可以跟其他电路(如编解码电路、硬件加速电路或各种总线和接口电路)构成一个SoC(片上系统),或者也可以作为一个ASIC的内置处理器集成在所述ASIC当中,该集成了处理器的ASIC可以单独封装或者也可以跟其他电路封装在一起。该处理器除了包括用于执行软件指令以进行运算或处理的核外,还可进一步包括必要的硬件加速器,如现场可编程门阵列(field programmable gate array,FPGA)、PLD(可编程逻辑器件)、或者实现专用逻辑运算的逻辑电路。
本申请实施例中的存储器,可以包括如下至少一种类型:只读存储器(read-only memory,ROM)或可存储静态信息和指令的其他类型的静态存储设备,随机存取存储器(random access memory,RAM)或者可存储信息和指令的其他类型的动态存储设备,也可以是电可擦可编程只读存储器(Electrically erasable programmabler-only memory,EEPROM)。在某些场景下,存储器还可以是只读光盘(compact disc read-only memory,CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质,但不限于此。
在上述实施例中,可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件程序实现时,可以全部或部分地以计算机程序产品的形式来实现。该计算机程 序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时,全部或部分地产生按照本申请实施例所述的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。计算机指令可以存储在计算机可读存储介质中,或者从一个计算机可读存储介质向另一个计算机可读存储介质传输,例如,计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line,简称DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可以用介质集成的服务器、数据中心等数据存储设备。可用介质可以是磁性介质(例如,软盘、硬盘、磁带),光介质(例如,DVD)、或者半导体介质(例如固态硬盘(solid state disk,简称SSD))等。
尽管在此结合各实施例对本申请进行了描述,然而,在实施所要求保护的本申请过程中,本领域技术人员通过查看附图、公开内容、以及所附权利要求书,可理解并实现公开实施例的其他变化。在权利要求中,“包括”(comprising)一词不排除其他组成部分或步骤,“一”或“一个”不排除多个的情况。单个处理器或其他单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施,但这并不表示这些措施不能组合起来产生良好的效果。
尽管结合具体特征及其实施例对本申请进行了描述,显而易见的,在不脱离本申请的精神和范围的情况下,可对其进行各种修改和组合。相应地,本说明书和附图仅仅是所附权利要求所界定的本申请的示例性说明,且视为已覆盖本申请范围内的任意和所有修改、变化、组合或等同物。显然,本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样,倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内,则本申请也意图包含这些改动和变型在内。

Claims (15)

  1. 一种自适应优化自动驾驶系统的方法,其特征在于,包括:
    获取目标车辆的驾驶员的驾驶意图,所述目标车辆为在所述自动驾驶系统的控制下自动驾驶的车辆;
    基于所述驾驶员的驾驶意图,检测到所述目标车辆的在所述自动驾驶系统控制下的驾驶行为与所述驾驶员的驾驶意图存在冲突;
    更新所述自动驾驶系统,以使得所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为匹配所述驾驶员的驾驶意图。
  2. 根据权利要求1所述的自适应优化自动驾驶系统的方法,其特征在于,所述驾驶意图由所述驾驶员的行为对应的特征参数表征;所述驾驶员的行为包括:所述驾驶员的操作行为、视觉行为、情绪行为、身体姿态行为中的一种或者多种。
  3. 根据权利要求2所述的自适应优化自动驾驶系统的方法,其特征在于,所述目标车辆的在所述自动驾驶系统控制下的驾驶行为与所述驾驶员的驾驶意图存在冲突,包括:
    所述驾驶员的行为对应的特征参数超过预设范围,和/或,
    所述驾驶员的行为对应的特征参数超过预设范围的时间大于或等于第一预设值,和/或,
    所述驾驶员的行为对应的特征参数超过预设范围的次数大于或等于第二预设值。
  4. 根据权利要求1-3任一项所述的自适应优化自动驾驶系统的方法,其特征在于,所述更新所述自动驾驶系统,包括:
    根据所述驾驶员的驾驶意图以及所述目标车辆在所述自动驾驶系统控制下的驾驶行为,得到第一特征参数,所述第一特征参数表征所述驾驶员的驾驶意图以及所述目标车辆在所述自动驾驶系统控制下的驾驶行为对应的驾驶数据;
    将所述第一特征参数输入预设神经网络模型,得到第一驾驶行为,所述预设神经网络模型用于确定匹配所述驾驶员的驾驶意图的驾驶行为;
    根据所述第一驾驶行为,更新所述自动驾驶系统。
  5. 根据权利要求4所述的自适应优化自动驾驶系统的方法,其特征在于,在根据所述第一驾驶行为,更新所述自动驾驶系统之前,所述方法还包括:
    向所述驾驶员呈现一个或多个问题;
    接收来自所述驾驶员针对所述一个或多个问题的回答,所述一个或多个问题的回答用于指示是否更新所述自动驾驶系统;
    所述根据所述第一驾驶行为,更新所述自动驾驶系统包括:当所述一个或多个问题的回答用于指示更新所述自动驾驶系统时,根据所述第一驾驶行为更新所述自动驾驶系统。
  6. 根据权利要求4或5所述的自适应优化自动驾驶系统的方法,其特征在于,所述根据所述第一驾驶行为更新所述自动驾驶系统,包括:
    若所述第一驾驶行为符合所述目标车辆的安全驾驶行为,则将所述第一驾驶行为作为所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为;
    若所述第一驾驶行为不符合所述目标车辆的安全驾驶行为,则将所述安全驾驶行为作为所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为。
  7. 一种自适应优化自动驾驶系统的装置,其特征在于,包括:
    通信单元,用于获取目标车辆的驾驶员的驾驶意图,所述目标车辆为在所述自动驾驶系统的控制下自动驾驶的车辆;
    处理单元,用于基于所述驾驶员的驾驶意图,检测到所述目标车辆的在所述自动驾驶系统控制下的驾驶行为与所述驾驶员的驾驶意图存在冲突;
    所述处理单元,还用于更新所述自动驾驶系统,以使得所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为匹配所述驾驶员的驾驶意图。
  8. 根据权利要求7所述的自适应优化自动驾驶系统的装置,其特征在于,所述驾驶意图由所述驾驶员的行为对应的特征参数表征;所述驾驶员的行为包括:所述驾驶员的操作行为、视觉行为、情绪行为、身体姿态行为中的一种或者多种。
  9. 根据权利要求7所述的自适应优化自动驾驶系统的装置,其特征在于,所述目标车辆的在所述自动驾驶系统控制下的驾驶行为与所述驾驶员的驾驶意图存在冲突,包括:
    所述驾驶员的行为对应的特征参数超过预设范围,和/或,
    所述驾驶员的行为对应的特征参数超过预设范围的时间大于或等于第一预设值,和/或,
    所述驾驶员的行为对应的特征参数超过预设范围的次数大于或等于第二预设值。
  10. 根据权利要求7-9任一项所述的自适应优化自动驾驶系统的装置,其特征在于,所述处理单元,具体用于:
    根据所述驾驶员的驾驶意图以及所述目标车辆在所述自动驾驶系统控制下的驾驶行为,得到第一特征参数,所述第一特征参数表征所述驾驶员的驾驶意图以及所述目标车辆在所述自动驾驶系统控制下的驾驶行为对应的驾驶数据;
    将所述第一特征参数输入预设神经网络模型,得到第一驾驶行为,所述预设神经网络模型用于确定匹配所述驾驶员的驾驶意图的驾驶行为;
    根据所述第一驾驶行为,更新所述自动驾驶系统。
  11. 根据权利要求10所述的自适应优化自动驾驶系统的装置,其特征在于,
    所述通信单元,还用于:
    向所述驾驶员呈现一个或多个问题;
    接收来自所述驾驶员针对所述一个或多个问题的回答,所述一个或多个问题的回答用于指示是否更新所述自动驾驶系统;
    所述处理单元,具体用于:
    当所述一个或多个问题的回答用于指示更新所述自动驾驶系统时,根据所述第一驾驶行为更新所述自动驾驶系统。
  12. 根据权利要求10或11所述的自适应优化自动驾驶系统的装置,其特征在于,所述处理单元,还用于:
    若所述第一驾驶行为符合所述目标车辆的安全驾驶行为,则将所述第一驾驶行为作为所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为;
    若所述第一驾驶行为不符合所述目标车辆的安全驾驶行为,则将所述安全驾驶行为作为所述目标车辆在更新后的所述自动驾驶系统控制下的驾驶行为。
  13. 一种自动驾驶装置,其特征在于,所述自动驾驶装置包括一个或多个处理器和一个或多个存储器;所述一个或多个存储器与所述一个或多个处理器耦合,所述一个或多个 存储器用于存储计算机程序代码或计算机指令;
    当所述一个或多个处理器执行计算机指令时,使得所述自动驾驶装置执行如权利要求1-6任一项的自适应优化自动驾驶系统的方法。
  14. 一种计算机可读存储介质,其特征在于,计算机可读存储介质存储有计算机指令或程序,当计算机指令或程序在计算机上运行时,使得计算机执行如权利要求1-6任一项的自适应优化自动驾驶系统的方法。
  15. 一种芯片,其特征在于,包括:处理器和通信接口,所述处理器通过所述通信接口与存储器耦合,当所述处理器执行所述存储器中的计算机程序或指令时,使得如权利要求1-6任一项所述的自适应优化自动驾驶系统的方法被执行。
PCT/CN2020/089491 2020-05-09 2020-05-09 一种自适应优化自动驾驶系统的方法及装置 WO2021226767A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2022568477A JP2023525088A (ja) 2020-05-09 2020-05-09 自動運転システムを適応的に最適化するための方法および装置
EP20935690.6A EP4137897A4 (en) 2020-05-09 2020-05-09 METHOD AND DEVICE FOR THE SELF-ADAPTIVE OPTIMIZATION OF AN AUTOMATIC DRIVING SYSTEM
CN202080004187.6A CN112654548B (zh) 2020-05-09 2020-05-09 一种自适应优化自动驾驶系统的方法及装置
PCT/CN2020/089491 WO2021226767A1 (zh) 2020-05-09 2020-05-09 一种自适应优化自动驾驶系统的方法及装置
US17/982,978 US20230063354A1 (en) 2020-05-09 2022-11-08 Method and Apparatus for Adaptively Optimizing Autonomous Driving System

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/089491 WO2021226767A1 (zh) 2020-05-09 2020-05-09 一种自适应优化自动驾驶系统的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/982,978 Continuation US20230063354A1 (en) 2020-05-09 2022-11-08 Method and Apparatus for Adaptively Optimizing Autonomous Driving System

Publications (1)

Publication Number Publication Date
WO2021226767A1 true WO2021226767A1 (zh) 2021-11-18

Family

ID=75368451

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/089491 WO2021226767A1 (zh) 2020-05-09 2020-05-09 一种自适应优化自动驾驶系统的方法及装置

Country Status (5)

Country Link
US (1) US20230063354A1 (zh)
EP (1) EP4137897A4 (zh)
JP (1) JP2023525088A (zh)
CN (1) CN112654548B (zh)
WO (1) WO2021226767A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114771652A (zh) * 2022-04-15 2022-07-22 江铃汽车股份有限公司 一种自动驾驶转向控制方法及系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20230138610A1 (en) * 2021-11-02 2023-05-04 Robert Bosch Gmbh Customizing Operational Design Domain of an Autonomous Driving System for a Vehicle Based on Driver's Behavior

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140297115A1 (en) * 2013-04-01 2014-10-02 Kia Motors Corporation System and method for controlling vehicle driving mode
CN104859652A (zh) * 2014-02-26 2015-08-26 通用汽车环球科技运作有限责任公司 用于自动驾驶的方法和系统
CN105365851A (zh) * 2015-12-15 2016-03-02 中车大连机车研究所有限公司 智能机车司机控制装置
CN107315359A (zh) * 2016-04-27 2017-11-03 华为技术有限公司 车辆控制方法及系统
CN107662611A (zh) * 2017-11-06 2018-02-06 吉林大学 一种基于驾驶员情绪识别的自动驾驶模式切换系统
CN108803623A (zh) * 2017-10-22 2018-11-13 怀效宁 一种自动驾驶车辆个性化行车的方法和行车合法化的系统
CN111038521A (zh) * 2018-10-11 2020-04-21 顾泽苍 一种自动驾驶“意识决定”模型的构成方法

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102011100277A1 (de) * 2011-05-03 2012-11-08 Conti Temic Microelectronic Gmbh Verfahren zur Konfiguration und Personalisierung von Steuergerätverhalten
DE112014006584B4 (de) * 2014-04-14 2022-09-22 Mitsubishi Electric Corporation Fahrassistenzvorrichtung und Fahrassistenzverfahren
CN104590274A (zh) * 2014-11-26 2015-05-06 浙江吉利汽车研究院有限公司 一种驾驶行为自适应系统及驾驶行为自适应方法
US9688271B2 (en) * 2015-03-11 2017-06-27 Elwha Llc Occupant based vehicle control
WO2016170785A1 (ja) * 2015-04-21 2016-10-27 パナソニックIpマネジメント株式会社 情報処理システム、情報処理方法、およびプログラム
EP3357780B1 (en) * 2015-09-30 2022-10-26 Sony Group Corporation Driving control device, driving control method, and program
CN107223101A (zh) * 2016-06-23 2017-09-29 驭势科技(北京)有限公司 车辆自动驾驶方法和车辆自动驾驶系统
DE102016224291A1 (de) * 2016-12-06 2018-06-07 Trw Automotive Gmbh Verfahren zur rechnergestützten Adaption eines vorgegebenen teilautomatisierten Fahrsystems eines Kraftfahrzeugs
US10346888B2 (en) * 2017-06-16 2019-07-09 Uber Technologies, Inc. Systems and methods to obtain passenger feedback in response to autonomous vehicle driving events
US10532749B2 (en) * 2017-08-11 2020-01-14 Uatc, Llc Systems and methods to adjust autonomous vehicle parameters in response to passenger feedback
US10668925B2 (en) * 2017-09-05 2020-06-02 Baidu Usa Llc Driver intention-based lane assistant system for autonomous driving vehicles
JP6748059B2 (ja) * 2017-12-28 2020-08-26 本田技研工業株式会社 車両の制御装置
CN109242251B (zh) * 2018-08-03 2020-03-06 百度在线网络技术(北京)有限公司 行车行为安全性检测方法、装置、设备及存储介质
JP7172321B2 (ja) * 2018-09-12 2022-11-16 トヨタ自動車株式会社 運転評価装置、運転評価システム、運転評価方法、及び運転評価用コンピュータプログラム
KR20200135630A (ko) * 2019-05-23 2020-12-03 현대자동차주식회사 자율 주행 차량의 제어장치 및 그 방법
CN110471411A (zh) * 2019-07-26 2019-11-19 华为技术有限公司 自动驾驶方法和自动驾驶装置
CN111016924B (zh) * 2019-12-12 2022-07-12 长城汽车股份有限公司 自动驾驶车辆的远程驾驶控制方法、装置及远程驾驶系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140297115A1 (en) * 2013-04-01 2014-10-02 Kia Motors Corporation System and method for controlling vehicle driving mode
CN104097634A (zh) * 2013-04-01 2014-10-15 现代自动车株式会社 用于控制车辆驾驶模式的系统和方法
CN104859652A (zh) * 2014-02-26 2015-08-26 通用汽车环球科技运作有限责任公司 用于自动驾驶的方法和系统
CN105365851A (zh) * 2015-12-15 2016-03-02 中车大连机车研究所有限公司 智能机车司机控制装置
CN107315359A (zh) * 2016-04-27 2017-11-03 华为技术有限公司 车辆控制方法及系统
CN108803623A (zh) * 2017-10-22 2018-11-13 怀效宁 一种自动驾驶车辆个性化行车的方法和行车合法化的系统
CN107662611A (zh) * 2017-11-06 2018-02-06 吉林大学 一种基于驾驶员情绪识别的自动驾驶模式切换系统
CN111038521A (zh) * 2018-10-11 2020-04-21 顾泽苍 一种自动驾驶“意识决定”模型的构成方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4137897A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114771652A (zh) * 2022-04-15 2022-07-22 江铃汽车股份有限公司 一种自动驾驶转向控制方法及系统

Also Published As

Publication number Publication date
EP4137897A1 (en) 2023-02-22
CN112654548B (zh) 2022-04-05
US20230063354A1 (en) 2023-03-02
JP2023525088A (ja) 2023-06-14
CN112654548A (zh) 2021-04-13
EP4137897A4 (en) 2023-05-31

Similar Documents

Publication Publication Date Title
US11388482B2 (en) Display control system and method to generate a virtual environment in a vehicle
US11079758B2 (en) Systems and methods for incentivizing user-aided improvement of autonomous vehicle control systems and methods of operating a vehicle using the same
US11685386B2 (en) System and method for determining a change of a customary vehicle driver
CN112249032B (zh) 一种自动驾驶决策方法、系统、设备及计算机存储介质
WO2017057059A1 (ja) 運転制御装置、および運転制御方法、並びにプログラム
JP7329755B2 (ja) 支援方法およびそれを利用した支援システム、支援装置
US20230063354A1 (en) Method and Apparatus for Adaptively Optimizing Autonomous Driving System
WO2018058266A1 (zh) 一种基于移动终端的提示方法及系统
JPWO2017057060A1 (ja) 運転制御装置、および運転制御方法、並びにプログラム
JP2019043495A (ja) 自動運転調整装置、自動運転調整システム、及び自動運転調整方法
CN109690657A (zh) 学习和适应的智能教程
KR102209421B1 (ko) 자율 주행 차량과 이를 이용한 주행 제어 시스템 및 방법
US20210092086A1 (en) System and method for facilitating driver communication via an audio centric network
WO2021028533A1 (en) Method, device, medium, and vehicle for providing individual driving experience
WO2022205159A1 (zh) 一种车辆控制方法及装置
CN118046915A (zh) 基于乘员行为和情绪提供反馈的交通载具系统和方法
WO2021229671A1 (ja) 走行支援装置および走行支援方法
TWI831488B (zh) 應用在車輛駕駛輔助的先進緊急剎車制動系統及其控制方法
US11935407B1 (en) Dynamic loose vehicle formation method based on real-time online navigation map
WO2022194246A9 (zh) 控制方法、装置和系统
US20240294180A1 (en) Systems and methods for determining trust across mobility platforms
CN114572219B (zh) 自动超车方法、装置、车辆、存储介质及芯片
US20240174087A1 (en) Server and System for Controlling Driving of a Vehicle, and Method Thereof
WO2023001068A1 (zh) 一种车辆驾驶方法及装置
CN117311515A (zh) 一种车外交互方法、装置、电子设备及车辆

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: 20935690

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022568477

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 202237065108

Country of ref document: IN

ENP Entry into the national phase

Ref document number: 2020935690

Country of ref document: EP

Effective date: 20221116

NENP Non-entry into the national phase

Ref country code: DE