WO2021109164A1 - 自动驾驶中车辆变道至目标车道前车之后的预测方法 - Google Patents

自动驾驶中车辆变道至目标车道前车之后的预测方法 Download PDF

Info

Publication number
WO2021109164A1
WO2021109164A1 PCT/CN2019/123927 CN2019123927W WO2021109164A1 WO 2021109164 A1 WO2021109164 A1 WO 2021109164A1 CN 2019123927 W CN2019123927 W CN 2019123927W WO 2021109164 A1 WO2021109164 A1 WO 2021109164A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
lane
willingness
changing
change
Prior art date
Application number
PCT/CN2019/123927
Other languages
English (en)
French (fr)
Inventor
杜光辉
经建峰
袁雁城
张尧文
Original Assignee
格物汽车科技(苏州)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 格物汽车科技(苏州)有限公司 filed Critical 格物汽车科技(苏州)有限公司
Publication of WO2021109164A1 publication Critical patent/WO2021109164A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation 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
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation 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
    • B60W40/02—Estimation 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 ambient conditions
    • B60W40/04—Traffic conditions

Definitions

  • the present invention relates to the technical field of automatic driving, and in particular to a method for predicting after a vehicle changes lanes to a target lane before the vehicle in automatic driving.
  • Autonomous driving includes four modules: prediction, decision-making, planning, and control.
  • the prediction module predicts whether the vehicle will change lanes in the future and the lane change trajectory in the case of lane changes based on the vehicle driving state data and lane environment data in each lane.
  • Lane change includes left lane change and right lane change.
  • Left and right lane changes include lane change before the vehicle in front of the target lane and lane change after the vehicle in front of the target lane.
  • the decision-making module calculates the next expected state of the vehicle based on the predicted output results of the prediction module, environmental information, navigation information, the driver's behavior module, the vehicle dynamics model, etc.; the planning module plans the current state and the next expected state Vehicle driving trajectory; the control module calculates the corresponding accelerator, braking and steering according to the planned driving trajectory.
  • the present invention provides a prediction method for a vehicle to change lanes to a target lane before the vehicle in automatic driving, which combines the pre-lane change vehicle, the first car in front of the lane where the pre-change vehicle is located, and the target lane in front of the pre-change vehicle
  • the driving state data of the first car and the first car behind the pre-change vehicle in the target lane calculates the willingness of the pre-lane change vehicle to the vehicle in front of the target lane, and predicts based on the willingness Whether the pre-changing vehicle will change lanes to the target lane in the future, effectively eliminate the factors that interfere with the decision of the vehicle by surrounding vehicles, improve the confidence of the decision of the vehicle; at the same time, effectively reduce the data analysis of the prediction module in the automatic driving system
  • the amount of calculation reduces the operating difficulty and operating cost of the automatic driving system, and effectively improves the lag time of decision-making.
  • the present invention provides a method for predicting a vehicle's lane change to the target lane after the vehicle in front of the target lane in automatic driving, which includes the following steps:
  • the pre-changing vehicle the first car in front of the lane where the pre-changing vehicle is located, the first car in the target lane in front of the pre-changing vehicle, and the target lane in the pre-change The first car behind the road vehicle;
  • the willingness degree exceeds the willingness threshold value, it is predicted that the pre-lane change vehicle will change lanes to behind the vehicle in front of the target lane at a future time.
  • further including calculating the willingness ⁇ includes:
  • the driving speed of the first car is lower than the traveling speed of the first car in front of the lane where the pre-changing vehicle is located, and the target lane is in front of the pre-changing vehicle and the first car is in the pre-changing lane
  • the willingness of the pre-lane-changing vehicle to change lanes to the target lane is zero.
  • further including calculating the willingness ⁇ also includes:
  • it further includes calculating the willingness ⁇ 1 according to formula 1,
  • ⁇ 1 represents the degree of congestion in front of the lane where the pre-changing vehicle is located
  • ⁇ 2 represents the degree of congestion in front of the first car behind the pre-changing vehicle in the target lane.
  • the method further includes calculating the willingness degree ⁇ 2 according to formula 2,
  • t represents the duration of a certain fixed value for the lateral speed v y of the pre-lane change vehicle in the direction of the target lane;
  • t1 and t2 respectively represent two fixed time values on the time axis where t is located.
  • further including calculating the willingness ⁇ 3 includes:
  • A represents the distance that the pre-lane change vehicle needs to maintain from the first car in front of the lane where the pre-lane change vehicle is traveling at a driving speed of V2;
  • B represents the driving distance of the pre-lane change vehicle decelerating from the driving speed V1 to the driving speed V2 at a deceleration a x;
  • L1 is the initial distance between the pre-lane change vehicle and the first vehicle in front of the lane where the pre-lane change vehicle is located;
  • V1 is the traveling speed of the pre-lane change vehicle
  • V2 is the driving speed of the first car in front of the lane where the pre-changing vehicle is located
  • S1 is the braking distance from pre-lane change vehicle braking to stop
  • S2 is the braking distance from the braking to the stop of the first car in front of the lane where the pre-changing vehicle is located;
  • t 1 is the braking reaction time of the pre-lane change vehicle
  • D1 is the safe stopping distance between the pre-lane change vehicle and the first vehicle in front of the lane where the pre-lane change vehicle is braked to a stop;
  • a represents the difference between the preset comfortable braking deceleration a Thres of the pre-lane change vehicle and the deceleration a x ;
  • k 1 represents the linear slope of the functional relationship between the difference between the deceleration a Thres and the deceleration a x and the willingness ⁇ 3;
  • a 1 represents a fixed deceleration value on the deceleration axis where the difference between the deceleration a Thres and the deceleration a x is located.
  • the method further includes calculating the willingness ⁇ 4 according to formula 4,
  • A represents the actual acceleration of the pre-lane change vehicle
  • k 2 represents the linear slope of the functional relationship between the actual acceleration of the pre-lane change vehicle and the willingness ⁇ 4;
  • a 1 represents a fixed acceleration value on the acceleration axis where the actual acceleration of the pre-lane change vehicle is located
  • ⁇ represents the increase in willingness that the actual acceleration of the pre-lane change vehicle exceeds the acceleration threshold for a certain period of time.
  • further including calculating the willingness ⁇ 5 includes:
  • A represents the distance that the pre-lane change vehicle needs to keep from the first car in front of the lane where the pre-lane change vehicle is traveling at a driving speed of V2;
  • V1 is the traveling speed of the pre-lane change vehicle
  • V2 is the driving speed of the first car in front of the lane where the pre-changing vehicle is located
  • S1 is the braking distance from pre-lane change vehicle braking to stop
  • S2 is the braking distance from the braking to the stop of the first car in front of the lane where the pre-changing vehicle is located;
  • t 1 is the braking reaction time of the pre-lane change vehicle
  • D1 is the safe stopping distance between the pre-lane change vehicle and the first vehicle in front of the lane where the pre-lane change vehicle is braked to a stop;
  • L1 is the initial distance between the pre-lane change vehicle and the first car in front of the lane where the pre-lane change vehicle is located;
  • ⁇ 1 represents a fixed distance value
  • X represents the distance along the lane line between the first car in front of the lane where the pre-changing vehicle is located and the first vehicle in the target lane in front of the pre-changing vehicle;
  • k 3 represents the linear slope of the functional relationship between the distance along the lane line and the willingness ⁇ 5 between the first vehicle in front of the lane where the pre-lane change vehicle is located and the first vehicle in the target lane in front of the pre-lane change vehicle;
  • X1 and X2 represent two fixed distance values on the distance axis along the lane line between the first car in front of the lane where the pre-changing vehicle is located and the first car in the target lane in front of the pre-changing vehicle.
  • further including calculating the willingness ⁇ 6 includes:
  • A represents the distance that the pre-lane change vehicle needs to maintain from the first car in front of the lane where the pre-lane change vehicle is traveling at a driving speed of V2;
  • V1 is the traveling speed of the pre-lane change vehicle
  • V2 is the driving speed of the first car in front of the lane where the pre-changing vehicle is located
  • S1 is the braking distance from pre-lane change vehicle braking to stop
  • S2 is the braking distance from the braking to the stop of the first car in front of the lane where the pre-changing vehicle is located;
  • t 1 is the braking reaction time of the pre-lane change vehicle
  • D1 is the safe stopping distance between the pre-lane change vehicle and the first vehicle in front of the lane where the pre-lane change vehicle is braked to a stop;
  • L1 is the initial distance between the pre-lane change vehicle and the first car in front of the lane where the pre-lane change vehicle is located;
  • ⁇ 2 represents a fixed distance value, which is a variable that can be calibrated
  • E represents the ratio of the driving speed of the first car in front of the target lane of the pre-changing vehicle to the driving speed of the first car in front of the pre-changing vehicle;
  • E1 and E2 represent the ratio of the speed of the first car in front of the pre-change vehicle to the speed of the first car in the lane where the target lane is located. Two fixed values on the travel speed ratio axis;
  • k 4 characterizes the linear slope of the functional relationship between the ratio of the speed of the first vehicle in front of the pre-change vehicle to the speed of the first vehicle in the lane where the target lane is located and the degree of willingness ⁇ 6;
  • ⁇ * represents a fixed value greater than zero and less than or equal to 1, which is a variable that can be calibrated.
  • the prediction method for a vehicle in the automatic driving of the present invention to change lanes to the target lane after the vehicle in front combines the pre-lane change vehicle, the first vehicle in front of the lane where the pre-lane vehicle is located, and the first vehicle in the target lane in front of the pre-change vehicle
  • the driving state data of the four cars and the first car behind the pre-changing vehicle in the target lane calculate the willingness of the pre-changing vehicle to change lanes to the vehicle in front of the target lane, and predict the pre-change lane based on the willingness Whether the vehicle will change lanes to the target lane in the future, effectively eliminate the factors that interfere with the decision of the vehicle by surrounding vehicles, improve the confidence of the decision of the vehicle; at the same time, effectively reduce the amount of data analysis and calculation of the prediction module in the automatic driving system. Reduce the operating difficulty and operating cost of the automatic driving system, and effectively improve the lag time of decision-making.
  • Fig. 1 is a flowchart of a prediction method in a preferred embodiment of the present invention
  • FIG. 2 is a schematic diagram of road conditions for calculating the willingness degree of the predicted vehicle after changing lanes to the target lane in the preferred embodiment of the present invention
  • Fig. 4 is a graph of calculating willingness ⁇ 3 in a preferred embodiment of the present invention.
  • Fig. 5a is a graph of calculating willingness ⁇ 4 in a preferred embodiment of the present invention.
  • Fig. 5b is a graph of calculating willingness increment ⁇ in a preferred embodiment of the present invention.
  • Fig. 6 is a graph of calculating willingness ⁇ 5 in a preferred embodiment of the present invention.
  • Fig. 7 is a graph of calculating willingness ⁇ 6 in a preferred embodiment of the present invention.
  • This embodiment discloses a method for predicting a vehicle's lane change to a target lane after the vehicle in front of the target lane in automatic driving. As shown in FIG. 1, the method includes the following steps:
  • Pre-changing vehicle 3 the first car in front of the lane where the pre-changing vehicle is located 8
  • the first car in the target lane in front of the pre-changing vehicle 4 in the target lane
  • vehicle 3 represents a pre-lane change vehicle (hereinafter referred to as "vehicle 3")
  • vehicle 8 represents the first vehicle in front of the lane where the pre-lane change vehicle is located (hereinafter referred to as “vehicle 8")
  • vehicle 4 represents the target The first vehicle in the lane in front of the pre-change vehicle (hereinafter referred to as “vehicle 4")
  • vehicle 0 represents the first vehicle in the target lane behind the pre-change vehicle (hereinafter referred to as "vehicle 0").
  • the driving state data of each car includes the driving speed, the lateral speed, the distance between the cars along the lane line direction (or “longitudinal distance"), acceleration, deceleration, lateral acceleration and deceleration, etc.
  • the distance related to the car is considered to be infinite.
  • the willingness threshold is preset according to the characteristics of the driver, for example, the willingness threshold is set to 0.6, or 0.65, or 0.7.
  • Calculating the above-mentioned willingness ⁇ includes excluding the following two negative items:
  • Negative item 1 The lateral speed v y of the vehicle 3 in the direction of the target lane is less than or equal to zero for a certain period of time, and the willingness of the vehicle 3 to change lanes to the lane where the vehicle 0 is located is zero. For example, when the lateral speed v y of the vehicle 3 in the direction of the target lane is less than or equal to zero and lasts for 2 s, the willingness of the vehicle 3 to change lanes to the lane where the vehicle 0 is located is zero.
  • the duration is the standard amount that can be adjusted. It should be noted that: the lateral speed v y is greater than 0, which indicates that the vehicle 3 is deviating from the target lane; the lateral velocity v y is less than 0, which indicates that the vehicle 3 is deviating from the target lane.
  • Negative item 2 When the vehicle 4 is located behind the vehicle 8 and the traveling speed of the vehicle 4 is lower than the traveling speed of the vehicle 8, the willingness of the vehicle 3 to change lanes to the lane where the vehicle 0 is located is zero.
  • the willingness degree ⁇ is calculated by the following method:
  • the willingness degrees ⁇ 1, ⁇ 2, ⁇ 3, ⁇ 4, ⁇ 5, and ⁇ 6 are preferably calculated in the following manner:
  • ⁇ 1 represents the degree of congestion in front of the lane where the vehicle 3 is located
  • ⁇ 2 represents the degree of congestion in front of the lane where vehicle 0 is located.
  • the congestion level When calculating the congestion level here, it is characterized by the number of cars within a certain distance. For example, if there are 6 cars within 100m ahead, the calculated congestion level is
  • the lateral speed v y is greater than 0, which indicates that the vehicle 3 is deviating from the target lane; the lateral velocity v y is less than 0, which indicates that the vehicle 3 is deviating from the target lane.
  • the willingness ⁇ 2 and the lateral velocity v y are functions of the duration of a certain value. The relationship is:
  • t represents the duration of a certain fixed value for the lateral speed v y of the pre-lane change vehicle in the direction of the target lane;
  • t1 and t2 respectively represent two fixed time values on the abscissa axis, for example, t1 takes 1s and t2 takes 2s.
  • t1 and t2 are both standard amounts that can be adjusted.
  • Vertex abscissa value line shown in FIG. 3 the ordinate value, the calibration parameters are able to adjust the slope of the line, and the lateral velocity v y related to the size, y greater lateral velocity v, corresponding to a vertical in FIG. 3
  • A characterizes the distance that the vehicle 3 needs to keep from the vehicle 8 when it is running at the driving speed V2;
  • B represents the travel distance of the vehicle 3 decelerating from the travel speed V1 to the travel speed V2 at a deceleration a x;
  • L1 is the initial distance between vehicle 3 and vehicle 8;
  • V1 is the traveling speed of vehicle 3;
  • V2 is the traveling speed of vehicle 8
  • S1 is the braking distance from vehicle 3 to stop
  • S2 is the braking distance from the braking to the stop of the vehicle 8;
  • t 1 is the braking reaction time of vehicle 3
  • D1 is the safe stopping distance between vehicle 3 and vehicle 8 when they both brake to a stop
  • the comfortable braking deceleration a Thres is preset according to the driver's characteristics. As shown in Figure 4, the difference between the deceleration a Thres and the deceleration a x is the abscissa, and the willingness ⁇ 3 is the ordinate to establish a coordinate system.
  • the functional relationship between ⁇ 3 and the difference between deceleration a Thres and deceleration a x is:
  • a represents the difference between the preset comfortable braking deceleration a Thres of the pre-lane change vehicle and the deceleration a x ;
  • k 1 represents the linear slope of the functional relationship between the difference between the deceleration a Thres and the deceleration a x and the willingness ⁇ 3;
  • a 1 represents a fixed deceleration value on the abscissa axis, for example, a 1 takes 0.5g. It is a standard amount that can be adjusted.
  • the willingness ⁇ 4 is the sum of the willingness and willingness increments corresponding to the actual acceleration A, but the sum of the two should be less than 1.
  • A represents the actual acceleration of the vehicle 3
  • k 2 represents the linear slope of the functional relationship between the actual acceleration of the vehicle 3 and the willingness ⁇ 4;
  • a 1 represents a fixed acceleration value on the abscissa axis, for example, A 1 takes 0.1g.
  • a 1 is the standard amount that can be adjusted.
  • ⁇ represents the increase in willingness that the actual acceleration of the pre-lane change vehicle exceeds the acceleration threshold for a certain period of time
  • ⁇ T is the calibrated quantity
  • V1 is the traveling speed of the pre-lane change vehicle
  • V2 is the speed of the first car in front of the lane where the pre-changing vehicle is located
  • L1 is the initial distance between vehicle 3 and vehicle 8.
  • the willingness increment ⁇ is the ordinate axis to establish a coordinate system, and the time that the actual acceleration of the vehicle 3 exceeds the acceleration threshold is related to The functional relationship between the willingness increment ⁇ is shown in Fig. 5(b).
  • A characterizes the distance that the vehicle 3 needs to keep from the vehicle 8 when it is running at the driving speed V2;
  • V1 is the traveling speed of vehicle 3;
  • V2 is the traveling speed of vehicle 8
  • S1 is the braking distance from vehicle 3 to stop
  • S2 is the braking distance from the braking to the stop of the vehicle 8;
  • t 1 is the braking reaction time of vehicle 3
  • D1 is the safe stopping distance between vehicle 3 and vehicle 8 when they both brake to a stop
  • L1 is the initial distance between vehicle 3 and vehicle 8;
  • ⁇ 1 represents a fixed distance value, which is a standard amount that can be adjusted.
  • the degree of willingness ⁇ 5 is related to the distance along the lane line between vehicle 8 and vehicle 4.
  • the distance between vehicle 8 and vehicle 4 along the lane line is the abscissa, and the degree of willingness ⁇ 5 is the ordinate to establish the coordinates.
  • System the functional relationship between the willingness ⁇ 5 and the distance along the lane line between the vehicle 8 and the vehicle 4 is shown in Fig. 6,
  • X represents the distance along the lane line between the vehicle 8 and the vehicle 4;
  • k 3 represents the linear slope of the functional relationship between the distance along the lane line between the vehicle 8 and the vehicle 4 and the willingness ⁇ 5;
  • X1 and X2 are two fixed distance values on the abscissa axis, for example, X1 is 2m, and X2 is 0.5s*V1; X1 and X2 are both calibrated quantities that can be adjusted.
  • A characterizes the distance that the vehicle 3 needs to keep from the vehicle 8 when it is running at the driving speed V2;
  • V1 is the traveling speed of vehicle 3;
  • V2 is the traveling speed of vehicle 8
  • V3 is the traveling speed of vehicle 4
  • S1 is the braking distance from vehicle 3 to stop
  • S2 is the braking distance from the braking to the stop of the vehicle 8;
  • t 1 is the braking reaction time of vehicle 3
  • D1 is the safe stopping distance between vehicle 3 and vehicle 8 when they both brake to a stop
  • L1 is the initial distance between vehicle 3 and vehicle 8;
  • ⁇ 2 represents a fixed distance value, which is a standard amount that can be adjusted
  • the willingness ⁇ 6 is related to the speed ratio E of the vehicle 4 to the vehicle 8.
  • the coordinate system is established with the speed ratio of the vehicle 4 and the vehicle 8 as the abscissa and the willingness ⁇ 6 as the ordinate.
  • the functional relationship between the willingness ⁇ 6 and the speed ratio E of the vehicle 4 to the vehicle 8 is:
  • E represents the speed ratio of vehicle 4 to vehicle 8;
  • E1 and E2 represent two fixed values on the abscissa, for example, E1 is 1, and E2 is 1.5.
  • both E1 and E2 are standard amounts that can be adjusted.
  • k 4 represents the linear slope of the functional relationship between the ratio of the running speed of the vehicle 4 to the running speed of the vehicle 8 and the degree of willingness ⁇ 6;
  • ⁇ * represents a fixed value greater than zero and less than or equal to 1, which is a standard amount that can be adjusted.
  • the prediction method of a vehicle changing lane to the target lane after the vehicle in front of the target lane in automatic driving combines the pre-lane vehicle, the first car in front of the lane where the pre-change vehicle is located, and the first vehicle in the target lane in front of the pre-change vehicle
  • the driving state data of the four cars and the first car behind the pre-change vehicle in the target lane calculates the willingness of the pre-change vehicle to change lanes to the vehicle in front of the target lane, and predicts the pre-change lane based on the willingness Whether the vehicle will change lanes to the target lane in the future, effectively eliminate the factors that interfere with the decision of the vehicle by surrounding vehicles, improve the confidence of the decision of the vehicle; at the same time, effectively reduce the amount of data analysis and calculation of the prediction module in the automatic driving system. Reduce the operating difficulty and operating cost of the automatic driving system, and effectively improve the lag time of decision-making.

Abstract

本发明公开了一种自动驾驶中车辆变道至目标车道前车之后的预测方法，获取以下车辆当前采样周期的行驶状态数据：预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道车辆前方的第一辆车、目标车道内位于预变道车辆后方的第一辆车；结合获取的各车辆的行驶状态数据以及车道环境数据计算出预变道车辆变道至目标车道前车之后的意愿度；当意愿度超过意愿度阈值时，预测出预变道车辆将在未来时刻变道至目标车道的前车之后。本发明的预测方法有效排除周围车辆干扰本车决策的因素，提高本车决策的确信度；同时，有效降低自动驾驶系统中预测模块的数据分析计算量，降低自动驾驶系统的运行难度和运行成本，有效改善决策的滞后延时。

Description

自动驾驶中车辆变道至目标车道前车之后的预测方法 技术领域
本发明涉及自动驾驶技术领域，具体涉及一种自动驾驶中车辆变道至目标车道前车之后的预测方法。
背景技术
自动驾驶包括预测、决策、规划和控制四大模块，预测模块根据各车道的车辆行驶状态数据和车道环境数据预测车辆在未来时刻是否会变道，以及变道情况下的变道轨迹。变道包括左变道、右变道，左、右变道又包括变道至目标车道前车之前和变道至目标车道前车之后。决策模块根据预测模块的预测输出结果、环境信息、导航信息、本车驾驶员行为模块、本车动力学模型等计算本车的下一个期望状态；规划模块根据当前状态以及下一个期望状态规划本车行驶轨迹；控制模块根据规划的行驶轨迹计算相应的油门、制动和转向。
目前，一些互联网企业和汽车厂商的自动驾驶系统中，将预测出的所有感兴趣车辆(定义为可能对本车行驶带来干扰的周围车辆)均默认为在未来时刻会发生变道。这些车辆中不排除有部分车辆变道的意愿度较低的车辆，这里的意愿度为变道的意愿程度。这些变道意愿度较低的周围感兴趣车辆会干扰本车决策，还会增加本车对各车行驶状态时数据分析的计算量，增加自动驾驶系统运行难度和运行成本。
发明内容
本发明提供一种自动驾驶中车辆变道至目标车道前车之后的预测方法，结合预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道 车辆前方的第一辆车和目标车道内位于预变道车辆后方的第一辆车这四辆车的行驶状态数据计算出预变道车辆变道至目标车道前车之后的意愿度，根据意愿度来预测预变道车辆在未来时刻是否会变道至目标车道前车之后，有效排除周围车辆干扰本车决策的因素，提高本车决策的确信度；同时，有效降低自动驾驶系统中预测模块的数据分析计算量，降低自动驾驶系统的运行难度和运行成本，有效改善决策的滞后延时。
为了解决上述技术问题，本发明提供了一种自动驾驶中预测车辆变道至目标车道前车之后的方法，包括以下步骤，
获取以下车辆当前采样周期的行驶状态数据：预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道车辆前方的第一辆车、目标车道内位于预变道车辆后方的第一辆车；
结合获取的各车辆的所述行驶状态数据以及车道环境数据计算出所述预变道车辆变道至所述目标车道前车之后的意愿度λ，0≤λ≤1；
当所述意愿度超过意愿度阈值时，预测出所述预变道车辆将在未来时刻变道至所述目标车道的前车之后。
本发明一个较佳实施例中，进一步包括计算所述意愿度λ包括，
所述预变道车辆向目标车道方向的横向速度小于零、且维持一定时间时，所述预变道车辆变道至目标车道的意愿度为零；
或者，目标车道位于预变道车辆前方第一辆车的行驶速度小于预变道车辆所在车道前方第一辆车的行驶速度、且目标车道位于预变道车辆前方第一辆车位于预变道车辆所在车道前方第一辆车的后方时，所述预变道车辆变道至目标车道的意愿度为零。
本发明一个较佳实施例中，进一步包括计算所述意愿度λ还包括，
计算车道拥堵程度影响下的意愿度λ1；
满足预变道车辆向目标车道方向的横向速度大于等于零的条件下计算预变道车辆向目标车道方向的横向速度v y影响下的意愿度λ2；
计算预变道车辆与预变道车辆所在车道前方第一辆车之间纵向距离影响下的意愿度λ3；
计算预变道车辆加速度影响下的意愿度λ4；
计算预变道车辆所在车道前方第一辆车和目标车道位于预变道车辆前方第一辆车之间纵向距离影响下的意愿度λ5；
计算目标车道位于预变道车辆前方第一辆车与预变道车辆所在车道前方第一辆车的速度比影响下的意愿度λ6；
λ＝max{λ1，λ2，λ3，λ4，λ5，λ6}   (式一)；
根据式一计算所述预变道车辆变道至目标车道前车之后的意愿度λ。
本发明一个较佳实施例中，进一步包括根据公式一计算所述意愿度λ1，
Figure PCTCN2019123927-appb-000001
其中，ρ1表征所述预变道车辆所在车道前方拥堵程度；
ρ2表征目标车道位于预变道车辆后方第一辆车的前方拥堵程度。
本发明一个较佳实施例中，进一步包括根据公式二计算所述意愿度λ2，
Figure PCTCN2019123927-appb-000002
其中，t表征预变道车辆向目标车道方向的横向速度v y为某个定值持续的时间；
t1和t2分别表征t所在时间轴上的两个固定时间值。
本发明一个较佳实施例中，进一步包括计算所述意愿度λ3包括，
计算L1＝A+B、A＝S1-S2+V1*t1+D1；
A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方第一辆车保持的距离；
B表征预变道车辆以减速度a x从行驶速度V1减速至行驶速度V2的行驶距离；
L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
V1为预变道车辆的行驶速度；
V2为预变道车辆所在车道前方第一辆车的行驶速度；
S1为预变道车辆制动至停止的制动距离；
S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
t 1为预变道车辆的制动反应时间；
D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
当B＝(L1-A)≤0时，意愿度λ3＝1；
当B＝(L1-A)＞0，且V1≤V2时，意愿度λ3＝0；
当B＝(L1-A)＞0，且V1＞V2时，根据式二计算预变道车辆的减速度a x：
Figure PCTCN2019123927-appb-000003
根据公式三计算所述意愿度λ3：
Figure PCTCN2019123927-appb-000004
其中，a表征预变道车辆的预设舒适制动减速度a Thres与减速度a x的差值；
k 1表征减速度a Thres与减速度a x的差值与意愿度λ3之间函数关系的直线斜率；
a 1表征减速度a Thres与减速度a x的差值所在减速度轴上的一个固定减速度值。
本发明一个较佳实施例中，进一步包括根据公式四计算所述意愿度λ4，
Figure PCTCN2019123927-appb-000005
A表征预变道车辆的实际加速度；
k 2表征预变道车辆实际加速度与意愿度λ4之间函数关系的直线斜率；
A 1表征预变道车辆实际加速度所在加速度轴上的一个固定加速度值；
Δλ表征预变道车辆的实际加速度超出加速度阈值持续一定时间的意愿度增量。
本发明一个较佳实施例中，进一步包括计算所述意愿度λ5包括，
计算A＝S1-S2+V1*t1+D1；
A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方 第一辆车保持的距离；
V1为预变道车辆的行驶速度；
V2为预变道车辆所在车道前方第一辆车的行驶速度；
S1为预变道车辆制动至停止的制动距离；
S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
t 1为预变道车辆的制动反应时间；
D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
当L1＞A+Δ1时，λ5＝0；
当L1≤A+Δ1时，根据公式五计算所述意愿度λ5，
Figure PCTCN2019123927-appb-000006
其中，L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
Δ1表征一个固定的距离值；
X表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车辆前方第一辆车之间的沿车道线距离；
k 3表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车辆前方第一辆车之间的沿车道线距离与意愿度λ5之间函数关系的直线斜率；
X1和X2表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车 辆前方第一辆车之间的沿车道线距离所在距离轴上的两个固定距离值。
本发明一个较佳实施例中，进一步包括计算所述意愿度λ6包括，
计算A＝S1-S2+V1*t1+D1；
A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方第一辆车保持的距离；
V1为预变道车辆的行驶速度；
V2为预变道车辆所在车道前方第一辆车的行驶速度；
S1为预变道车辆制动至停止的制动距离；
S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
t 1为预变道车辆的制动反应时间；
D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
当L1＞A+Δ2时，λ6＝0；
当L1≤A+Δ2时，根据公式六计算所述意愿度λ6，
Figure PCTCN2019123927-appb-000007
其中，L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
Δ2表征一个固定的距离值，其为能够标定的变量；
E表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所 在车道前方第一辆车的行驶速度的比值；
E1和E2表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所在车道前方第一辆车的行驶速度的比值所在行驶速度比值轴上的两个固定值；
k 4表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所在车道前方第一辆车的行驶速度比值与意愿度λ6之间函数关系的直线斜率；
λ *表征大于零小于等于1的一个固定值，其为能够标定的变量。
本发明的有益效果：
本发明自动驾驶中车辆变道至目标车道前车之后的预测方法，结合预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道车辆前方的第一辆车和目标车道内位于预变道车辆后方的第一辆车这四辆车的行驶状态数据计算出预变道车辆变道至目标车道前车之后的意愿度，根据意愿度来预测预变道车辆在未来时刻是否会变道至目标车道前车之后，有效排除周围车辆干扰本车决策的因素，提高本车决策的确信度；同时，有效降低自动驾驶系统中预测模块的数据分析计算量，降低自动驾驶系统的运行难度和运行成本，有效改善决策的滞后延时。
附图说明
图1是本发明优选优选实施例中预测方法的流程图；
图2是本发明优选实施例中计算被预测车辆变道至目标车道前车之后意愿度的路况示意图；
图3a是本发明优选实施例中在v y＝1时计算意愿度λ2的曲线图；
图3b是本发明优选实施例中在v y＝2时计算意愿度λ2的曲线图；
图4是本发明优选实施例中计算意愿度λ3的曲线图；
图5a是本发明优选实施例中计算意愿度λ4的曲线图；
图5b是本发明优选实施例中计算意愿度增量Δλ的曲线图；
[根据细则91更正 06.05.2020]　
图6是本发明优选实施例中计算意愿度λ5的曲线图；
[根据细则91更正 06.05.2020]　
图7是本发明优选实施例中计算意愿度λ6的曲线图；
[根据细则91更正 06.05.2020]　
具体实施方式
下面结合附图和具体实施例对本发明作进一步说明，以使本领域的技术人员可以更好地理解本发明并能予以实施，但所举实施例不作为对本发明的限定。
实施例
本实施例公开一种自动驾驶中预测车辆变道至目标车道前车之后的方法，参照图1所示，包括以下步骤，
获取以下车辆当前采样周期的行驶状态数据：预变道车辆3、预变道车辆所在车道的前方第一辆车8、目标车道内位于预变道车辆前方的第一辆车4、目标车道内位于预变道车辆后方的第一辆车0。参照图2所示，车辆3表征预变道车辆(以下简称“车辆3”)，车辆8表征预变道车辆所在车道的前方第一辆车(以下简称“车辆8”)，车辆4表征目标车道内位于预变道车辆前方的第一辆车(以下简称“车辆4”)，车辆0表征目标车道内位于预变道车辆后方的第一辆车(以下简称“车辆0”)。各车的行驶状态数据包括行驶速度、横向速度、各车之间沿车道线方向的距离(或称“纵向距离”)、加速度、减速度、横向加减速度等。当图2中某辆车不存在时，认为与该车相关的距离为无穷远。
结合获取的各车辆的行驶状态数据以及车道环境数据计算出车辆3变道至车辆4之后的意愿度λ，0≤λ≤1；
当意愿度λ超过意愿度阈值时，预测出车辆3将在未来时刻变道至车辆4 之后。此处，根据驾驶员特性预设意愿度阈值，比如，设定意愿度阈值为0.6，或者0.65，或者0.7等。
计算上述意愿度λ包括排除以下两组否定项：
否定项1：车辆3向目标车道方向的横向速度v y小于等于零、且持续一定时间，车辆3变道至车辆0所在车道的意愿度为零。比如，车辆3向目标车道方向的横向速度v y小于等于零、且持续2s时，车辆3变道至车辆0所在车道的意愿度为零。此处的持续时间为能够调整的标定量。需要说明的有：横向速度v y大于0，表征车辆3偏向目标车道；横向速度v y小于0，表征车辆3偏离目标车道。
否定项2：车辆4位于车辆8的后方、且车辆4的行驶速度小于车辆8行驶速度时，车辆3变道至车辆0所在车道的意愿度为零。
当各车的行驶状态满足以上两组否定项之一或者同时满足以上两组否定项时，车辆3变道至车辆4所在车道、且位于车辆4之后的意愿度为0，且满足以上否定项时，不做后续计算。
当各车的行驶状态不满足以上两组否定项时，通过以下方式计算意愿度λ：
计算车道拥堵程度影响下的意愿度λ1；
满足车辆3向车辆0所在车道方向的横向速度v y大于等于零的条件下计算车辆3向车辆0所在车道方向的横向速度v y影响下的意愿度λ2；
计算车辆3与车辆4之间纵向距离影响下的意愿度λ3；
计算车辆3加速度影响下的意愿度λ4；
计算车辆8和车辆4之间纵向距离影响下的意愿度λ5；
计算车辆4与车辆8的速度比影响下的意愿度λ6；
λ＝max{λ1，λ2，λ3，λ4，λ5，λ6}   (式一)；
根据式一计算车辆3变道至车辆4与之后的意愿度λ。
本实施例技术方案中，优选通过以下方式计算获得意愿度λ1、λ2、λ3、λ4、λ5和λ6：
(1)根据公式一计算上述意愿度λ1，
Figure PCTCN2019123927-appb-000008
其中，ρ1表征车辆3所在车道前方拥堵程度；
ρ2表征车辆0所在车道前方拥堵程度。
此处计算拥堵程度时，以一定距离内的车数量表征，比如，前方100m内有6辆车，计算出来的拥堵程度为
Figure PCTCN2019123927-appb-000009
(2)若车辆3向目标车道方向的横向速度v y大于零，车辆3变道至车辆4与之后的意愿度λ2与车辆3向目标车道的横向速度v y和横向速度v y持续的时间有关。此处，横向速度v y大于0，表征车辆3偏向目标车道；横向速度v y小于0，表征车辆3偏离目标车道。参照图3所示，以横向速度v y为某一定值所持续的时间t为横坐标，意愿度λ2为纵坐标建立坐标系，意愿度λ2与横向速度v y为某一定值持续时间的函数关系为：
Figure PCTCN2019123927-appb-000010
其中，t表征预变道车辆向目标车道方向的横向速度v y为某个定值持续的时间；
t1和t2分别表征横坐标轴上的两个固定时间值，比如，t1取1s，t2取2s。此处，t1和t2均为能够调整的标定量。
参照图3(a)所示，横向速度v y等于1时的函数关系。参照图3(b)所示，横向速度v y等于2时的函数关系。
图3所示的直线的折点横坐标值、纵坐标值、直线斜率均为能够调整的标定参数，并且与横向速度v y的大小有关，横向速度v y越大，对应图3中的纵坐标值越大，横坐标值越小，直线斜率越大，但意愿度λ2最大值为1。
(3)计算意愿度λ3包括，
计算L1＝A+B、A＝S1-S2+V1*t1+D1；
A表征车辆3以行驶速度V2行驶时需要与车辆8需要保持的距离；
B表征车辆3以减速度a x从行驶速度V1减速至行驶速度V2的行驶距离；
L1为车辆3和车辆8之间的初始间距；
V1为车辆3的行驶速度；
V2为车辆8的行驶速度；
S1为车辆3制动至停止的制动距离；
S2为车辆8制动至停止的制动距离；
t 1为车辆3的制动反应时间；
D1为车辆3和车辆8均制动至停止时两者之间的安全停车距离；
当B＝(L1-A)≤0时，意愿度λ3＝1；
当B＝(L1-A)＞0，且V1≤V2时，意愿度λ3＝0；
当B＝(L1-A)＞0，且V1＞V2时，根据式二计算车辆3的减速度a x：
Figure PCTCN2019123927-appb-000011
根据驾驶员特性预设舒适制动的减速度a Thres，参照图4所示，以减速度a Thres和减速度a x的差值为横坐标，意愿度λ3为纵坐标建立坐标系，意愿度λ3与减速度a Thres和减速度a x的差值之间的函数关系为：
Figure PCTCN2019123927-appb-000012
其中，a表征预变道车辆的预设舒适制动减速度a Thres与减速度a x的差值；
k 1表征减速度a Thres与减速度a x的差值与意愿度λ3之间函数关系的直线斜率；
a 1表征横坐标轴上的一个固定减速度值，比如，a 1取0.5g。其为能够调整的标定量。
(4)若L1＞(V1-V2)*ΔT，则λ4＝0；
若L1≤(V1-V2)*ΔT否则，进行以下计算：
先根据车辆3当前的实际加速度A计算一个意愿度，然后根据加速度A持续的时间计算一个意愿度的增量Δλ，该意愿度增量随持续时间的增加变化，并且不同加速度A对应的增量曲线斜率不同，加速度A值越大，增量曲线斜率越大，意愿度λ4为实际加速度A对应的意愿度和意愿度增量之和，但两者之和应与1取小。
以车辆3当前的实际加速度A为横坐标，意愿度λ4为纵坐标建立坐标系，意愿度λ4与车辆3当前实际加速度A的函数关系参照图5(a)所示，
Figure PCTCN2019123927-appb-000013
A表征车辆3的实际加速度；
k 2表征车辆3实际加速度与意愿度λ4之间函数关系的直线斜率；
A 1表征横坐标轴上的一个固定加速度值，比如，A 1取0.1g。此处A 1为能够调整的标定量。
Δλ表征预变道车辆的实际加速度超出加速度阈值持续一定时间的意愿度增量；
ΔT为可标定量；
V1为预变道车辆的行驶速度；
V2为预变道车辆所在车道前方第一辆车的行驶速度；
L1为车辆3和车辆8之间的初始间距。
参照图5(b)所示，以车辆3的实际加速度超出加速度阈值持续的时间为横坐标，意愿度增量Δλ为纵坐标轴建立坐标系，车辆3的实际加速度超出加速度阈值持续的时间与意愿度增量Δλ之间的函数关系参照图5(b)所示。
图5(a)和图5(b)中，除纵轴坐标最大值为1外，直线的折点横坐标值、纵坐标值、直线斜率均为能够调整的标定参数。
(5)计算上述意愿度λ5包括，
计算A＝S1-S2+V1*t1+D1；
A表征车辆3以行驶速度V2行驶时需要与车辆8需要保持的距离；
V1为车辆3的行驶速度；
V2为车辆8的行驶速度；
S1为车辆3制动至停止的制动距离；
S2为车辆8制动至停止的制动距离；
t 1为车辆3的制动反应时间；
D1为车辆3和车辆8均制动至停止时两者之间的安全停车距离；
L1为车辆3和车辆8之间的初始间距；
当L1＞A+Δ1时，λ5＝0；此处，Δ1表征一个固定的距离值，其为能够调整的标定量。
当L1≤A+Δ1时，意愿度λ5与车辆8和车辆4之间的沿车道线距离相关，以车辆8和车辆4之间沿车道线距离为横坐标，意愿度λ5为纵坐标建立坐标系，意愿度λ5与车辆8和车辆4之间的沿车道线距离相关之间的函数关系参照图6所示，
Figure PCTCN2019123927-appb-000014
其中，X表征车辆8与车辆4之间的沿车道线距离；
k 3表征车辆8与车辆4之间的沿车道线距离与意愿度λ5之间函数关系的直线斜率；
X1和X2横坐标轴上的两个固定距离值，比如，X1取2m，X2取0.5s*V1； X1和X2均为能够能够调整的标定量。
图6中，除纵轴坐标最大值为1外，直线的折点横坐标值、纵坐标值、直线斜率均为能够调整的标定参数。
(6)计算上述意愿度λ6包括：
计算A＝S1-S2+V1*t1+D1；
A表征车辆3以行驶速度V2行驶时需要与车辆8需要保持的距离；
V1为车辆3的行驶速度；
V2为车辆8的行驶速度；
V3为车辆4的行驶速度；
S1为车辆3制动至停止的制动距离；
S2为车辆8制动至停止的制动距离；
t 1为车辆3的制动反应时间；
D1为车辆3和车辆8均制动至停止时两者之间的安全停车距离；
L1为车辆3和车辆8之间的初始间距；
当L1＞A+Δ2时，λ6＝0。此处，Δ2表征一个固定的距离值，其为能够调整的标定量
当L1≤A+Δ2时，意愿度λ6与车辆4与车辆8的行驶速度比E相关，以车辆4与车辆8的行驶速度比为横坐标，意愿度λ6为纵坐标建立坐标系，参照图7所示，意愿度λ6与车辆4与车辆8的行驶速度比E的函数关系为：
Figure PCTCN2019123927-appb-000015
其中，E表征车辆4与车辆8的行驶速度比；
E1和E2表征横坐标上的两个固定值，比如，E1取1，E2取1.5。此处，E1和E2均为能够调整的标定量。
k 4表征车辆4行驶速度与车辆8行驶速度比值与意愿度λ6之间函数关系的直线斜率；
λ *表征大于零小于等于1的一个固定值，其为能够调整的标定量。
图7中，除纵轴坐标最大值为1外，，直线的折点横坐标值、纵坐标值、直线斜率均为能够调整的标定参数。
以上，自动驾驶中车辆变道至目标车道前车之后的预测方法，结合预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道车辆前方的第一辆车和目标车道内位于预变道车辆后方的第一辆车这四辆车的行驶状态数据计算出预变道车辆变道至目标车道前车之后的意愿度，根据意愿度来预测预变道车辆在未来时刻是否会变道至目标车道前车之后，有效排除周围车辆干扰本车决策的因素，提高本车决策的确信度；同时，有效降低自动驾驶系统中预测模块的数据分析计算量，降低自动驾驶系统的运行难度和运行成本，有效改善决策的滞后延时。
以上所述实施例仅是为充分说明本发明而所举的较佳的实施例，本发明的保护范围不限于此。本技术领域的技术人员在本发明基础上所作的等同替代或变换，均在本发明的保护范围之内。本发明的保护范围以权利要求书为准。

Claims (9)

  1. 一种自动驾驶中预测车辆变道至目标车道前车之后的方法，其特征在于：包括以下步骤，
    获取以下车辆当前采样周期的行驶状态数据：预变道车辆、预变道车辆所在车道的前方第一辆车、目标车道内位于预变道车辆前方的第一辆车、目标车道内位于预变道车辆后方的第一辆车；
    结合获取的各车辆的所述行驶状态数据以及车道环境数据计算出所述预变道车辆变道至所述目标车道前车之后的意愿度λ，0≤λ≤1；
    当所述意愿度超过意愿度阈值时，预测出所述预变道车辆将在未来时刻变道至所述目标车道的前车之后。
  2. 如权利要求1所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：计算所述意愿度λ包括，
    所述预变道车辆向目标车道方向的横向速度小于零、且维持一定时间时，所述预变道车辆变道至目标车道的意愿度为零；
    或者，目标车道位于预变道车辆前方第一辆车的行驶速度小于预变道车辆所在车道前方第一辆车的行驶速度、且目标车道位于预变道车辆前方第一辆车位于预变道车辆所在车道前方第一辆车的后方时，所述预变道车辆变道至目标车道的意愿度为零。
  3. 如权利要求1所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：计算所述意愿度λ还包括，
    计算车道拥堵程度影响下的意愿度λ1；
    满足预变道车辆向目标车道方向的横向速度大于等于零的条件下计算预变 道车辆向目标车道方向的横向速度v y影响下的意愿度λ2；
    计算预变道车辆与预变道车辆所在车道前方第一辆车之间纵向距离影响下的意愿度λ3；
    计算预变道车辆加速度影响下的意愿度λ4；
    计算预变道车辆所在车道前方第一辆车和目标车道位于预变道车辆前方第一辆车之间纵向距离影响下的意愿度λ5；
    计算目标车道位于预变道车辆前方第一辆车与预变道车辆所在车道前方第一辆车的速度比影响下的意愿度λ6；
    λ＝max{λ1，λ2，λ3，λ4，λ5，λ6}  (式一)；
    根据式一计算所述预变道车辆变道至目标车道前车之后的意愿度λ。
  4. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：根据公式一计算所述意愿度λ1，
    Figure PCTCN2019123927-appb-100001
    其中，ρ1表征所述预变道车辆所在车道前方拥堵程度；
    ρ2表征目标车道位于预变道车辆后方第一辆车的前方拥堵程度。
  5. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：根据公式二计算所述意愿度λ2，
    Figure PCTCN2019123927-appb-100002
    其中，t表征预变道车辆向目标车道方向的横向速度v y为某个定值持续的时间；
    t1和t2分别表征t所在时间轴上的两个固定时间值。
  6. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：计算所述意愿度λ3包括，
    计算L1＝A+B、A＝S1-S2+V1*t1+D1；
    A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方第一辆车保持的距离；
    B表征预变道车辆以减速度a x从行驶速度V1减速至行驶速度V2的行驶距离；
    L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
    V1为预变道车辆的行驶速度；
    V2为预变道车辆所在车道前方第一辆车的行驶速度；
    S1为预变道车辆制动至停止的制动距离；
    S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
    t 1为预变道车辆的制动反应时间；
    D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
    当B＝(L1-A)≤0时，意愿度λ3＝1；
    当B＝(L1-A)＞0，且V1≤V2时，意愿度λ3＝0；
    当B＝(L1-A)＞0，且V1＞V2时，根据式二计算预变道车辆的减速度a x：
    Figure PCTCN2019123927-appb-100003
    根据公式三计算所述意愿度λ3：
    Figure PCTCN2019123927-appb-100004
    其中，a表征预变道车辆的预设舒适制动减速度a Thres与减速度a x的差值；
    k 1表征减速度a Thres与减速度a x的差值与意愿度λ3之间函数关系的直线斜率；
    a 1表征减速度a Thres与减速度a x的差值所在减速度轴上的一个固定减速度值。
  7. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：根据公式四计算所述意愿度λ4，
    Figure PCTCN2019123927-appb-100005
    A表征预变道车辆的实际加速度；
    k 2表征预变道车辆实际加速度与意愿度λ4之间函数关系的直线斜率；
    A 1表征预变道车辆实际加速度所在加速度轴上的一个固定加速度值；
    Δλ表征预变道车辆的实际加速度超出加速度阈值持续一定时间的意愿度增量。
  8. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方 法，其特征在于：计算所述意愿度λ5包括，
    计算A＝S1-S2+V1*t1+D1；
    A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方第一辆车保持的距离；
    V1为预变道车辆的行驶速度；
    V2为预变道车辆所在车道前方第一辆车的行驶速度；
    S1为预变道车辆制动至停止的制动距离；
    S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
    t 1为预变道车辆的制动反应时间；
    D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
    当L1＞A+Δ1时，λ5＝0；
    当L1≤A+Δ1时，根据公式五计算所述意愿度λ5，
    Figure PCTCN2019123927-appb-100006
    其中，L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
    Δ1表征一个固定的距离值；
    X表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车辆前方第一辆车之间的沿车道线距离；
    k 3表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车辆前方第一辆车之间的沿车道线距离与意愿度λ5之间函数关系的直线斜率；
    X1和X2表征预变道车辆所在车道前方第一辆车与目标车道位于预变道车辆前方第一辆车之间的沿车道线距离所在距离轴上的两个固定距离值。
  9. 如权利要求3所述的自动驾驶中车辆变道至目标车道前车之后的预测方法，其特征在于：计算所述意愿度λ6包括，
    计算A＝S1-S2+V1*t1+D1；
    A表征预变道车辆以行驶速度V2行驶时需要与预变道车辆所在车道前方第一辆车保持的距离；
    V1为预变道车辆的行驶速度；
    V2为预变道车辆所在车道前方第一辆车的行驶速度；
    S1为预变道车辆制动至停止的制动距离；
    S2为预变道车辆所在车道前方第一辆车制动至停止的制动距离；
    t 1为预变道车辆的制动反应时间；
    D1为预变道车辆和预变道车辆所在车道前方第一辆车均制动至停止时两者之间的安全停车距离；
    当L1＞A+Δ2时，λ6＝0；
    当L1≤A+Δ2时，根据公式六计算所述意愿度λ6，
    Figure PCTCN2019123927-appb-100007
    其中，L1为预变道车辆和预变道车辆所在车道前方第一辆车之间的初始间距；
    Δ2表征一个固定的距离值；
    E表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所在车道前方第一辆车的行驶速度的比值；
    E1和E2表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所在车道前方第一辆车的行驶速度的比值所在行驶速度比值轴上的两个固定值；
    k 4表征目标车道位于预变道车辆前方第一辆车的行驶速度与预变道车辆所在车道前方第一辆车的行驶速度比值与意愿度λ6之间函数关系的直线斜率；
    λ *表征大于零小于等于1的一个固定值。
PCT/CN2019/123927 2019-12-06 2019-12-09 自动驾驶中车辆变道至目标车道前车之后的预测方法 WO2021109164A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911242729.9A CN110920623B (zh) 2019-12-06 2019-12-06 自动驾驶中车辆变道至目标车道前车之后的预测方法
CN201911242729.9 2019-12-06

Publications (1)

Publication Number Publication Date
WO2021109164A1 true WO2021109164A1 (zh) 2021-06-10

Family

ID=69857378

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/123927 WO2021109164A1 (zh) 2019-12-06 2019-12-09 自动驾驶中车辆变道至目标车道前车之后的预测方法

Country Status (2)

Country Link
CN (1) CN110920623B (zh)
WO (1) WO2021109164A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4134288B1 (en) * 2020-04-06 2024-02-28 Nissan Motor Co., Ltd. Vehicle behavior estimation method, vehicle control method, and vehicle behavior estimation device
WO2023168630A1 (zh) * 2022-03-09 2023-09-14 华为技术有限公司 一种控车方法及相关装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060161331A1 (en) * 2005-01-14 2006-07-20 Denso Corporation Drive control system for automotive vehicle
CN104670091A (zh) * 2013-12-02 2015-06-03 现代摩比斯株式会社 利用投影单元的增强现实车道变更辅助系统
CN107644551A (zh) * 2017-09-04 2018-01-30 维沃移动通信有限公司 一种超车预判方法及终端
CN108534792A (zh) * 2017-03-01 2018-09-14 本田技研工业株式会社 车道变更推定装置、车道变更推定方法及存储介质
US20190077398A1 (en) * 2017-09-14 2019-03-14 Toyota Motor Engineering & Manufacturing North America, Inc. System and method for vehicle lane change prediction using structural recurrent neural networks
CN110267856A (zh) * 2017-03-01 2019-09-20 本田技研工业株式会社 车辆控制装置、车辆控制方法及程序
JP2019211830A (ja) * 2018-05-31 2019-12-12 マツダ株式会社 車線変更推定装置および車線変更推定方法と、車両制御装置および車両制御方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060161331A1 (en) * 2005-01-14 2006-07-20 Denso Corporation Drive control system for automotive vehicle
CN104670091A (zh) * 2013-12-02 2015-06-03 现代摩比斯株式会社 利用投影单元的增强现实车道变更辅助系统
CN108534792A (zh) * 2017-03-01 2018-09-14 本田技研工业株式会社 车道变更推定装置、车道变更推定方法及存储介质
CN110267856A (zh) * 2017-03-01 2019-09-20 本田技研工业株式会社 车辆控制装置、车辆控制方法及程序
CN107644551A (zh) * 2017-09-04 2018-01-30 维沃移动通信有限公司 一种超车预判方法及终端
US20190077398A1 (en) * 2017-09-14 2019-03-14 Toyota Motor Engineering & Manufacturing North America, Inc. System and method for vehicle lane change prediction using structural recurrent neural networks
JP2019211830A (ja) * 2018-05-31 2019-12-12 マツダ株式会社 車線変更推定装置および車線変更推定方法と、車両制御装置および車両制御方法

Also Published As

Publication number Publication date
CN110920623B (zh) 2021-02-02
CN110920623A (zh) 2020-03-27

Similar Documents

Publication Publication Date Title
US10179589B2 (en) System and method for optimizing fuel economy using predictive environment and driver behavior information
JP5494332B2 (ja) 車両制御システム
US9428187B2 (en) Lane change path planning algorithm for autonomous driving vehicle
JP5195929B2 (ja) 隊列走行制御システム及び車両
US9457807B2 (en) Unified motion planning algorithm for autonomous driving vehicle in obstacle avoidance maneuver
CN108919795A (zh) 一种自动驾驶汽车换道决策方法及装置
JP3777970B2 (ja) 先行車追従制御装置
JP7151179B2 (ja) 車線変更推定装置および車線変更推定方法と、車両制御装置および車両制御方法
CN112242069B (zh) 一种确定车速的方法和装置
JP4453217B2 (ja) 車間距離制御装置
CN108032858B (zh) 基于旁车行驶路径预测的自适应巡航控制方法及系统
CN108569287B (zh) 生成车辆控制命令的方法和装置、车辆控制器、存储介质
US10703373B2 (en) Vehicle velocity control
WO2015047177A1 (en) Method and system for a common driving strategy for vehicle platoons
WO2015047174A1 (en) Method and system for managing obstacles for vehicle platoons
US20190227560A1 (en) Route Setting Method and Route Setting Device
US20150291162A1 (en) Vehicle spacing control
CN113548050A (zh) 车辆行驶控制方法、装置、系统和存储介质
WO2021109164A1 (zh) 自动驾驶中车辆变道至目标车道前车之后的预测方法
US11618473B2 (en) Vehicle control system
Wang et al. Predictive safety control for road vehicles after a tire blowout
CN113195323A (zh) 经在检测到目标对象的情况下保障安全的自主驾驶
Park et al. A vehicle speed harmonization strategy for minimizing inter-vehicle crash risks
JP2019127081A (ja) 車両の自動運転制御装置及び自動運転制御方法
US11479220B2 (en) Adaptive cruise control

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: 19955133

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19955133

Country of ref document: EP

Kind code of ref document: A1