WO2021135728A1 - 一种自动驾驶车辆碰撞预测判断方法及装置 - Google Patents

一种自动驾驶车辆碰撞预测判断方法及装置 Download PDF

Info

Publication number
WO2021135728A1
WO2021135728A1 PCT/CN2020/131002 CN2020131002W WO2021135728A1 WO 2021135728 A1 WO2021135728 A1 WO 2021135728A1 CN 2020131002 W CN2020131002 W CN 2020131002W WO 2021135728 A1 WO2021135728 A1 WO 2021135728A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
collision
obstacle
autonomous vehicle
distance
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2020/131002
Other languages
English (en)
French (fr)
Inventor
左帅
朱敏
李兴佳
蔡礼松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhengzhou Yutong Bus Co Ltd
Original Assignee
Zhengzhou Yutong Bus Co Ltd
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 Zhengzhou Yutong Bus Co Ltd filed Critical Zhengzhou Yutong Bus Co Ltd
Priority to EP20909407.7A priority Critical patent/EP4082848B1/en
Publication of WO2021135728A1 publication Critical patent/WO2021135728A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

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
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0027—Planning or execution of driving tasks using trajectory prediction for other traffic participants
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • 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
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
    • 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
    • B60W2720/00—Output or target parameters relating to overall vehicle dynamics
    • B60W2720/10—Longitudinal speed

Definitions

  • the invention belongs to the technical field of automatic driving, and specifically relates to a method and a device for predicting and judging collision of an automatic driving vehicle.
  • a Chinese invention patent application with application publication number CN109960261A discloses a dynamic obstacle avoidance method based on collision detection.
  • the method constructs a vehicle's desired trajectory through preliminary planning, and predicts the trajectory and the vehicle's expectations based on the received dynamic obstacles. Trajectory, collision detection, re-planning of the vehicle's speed and/or path, to obtain vehicle collision-free speed and/or path data, reconstruct the desired trajectory of the vehicle, avoid dynamic obstacles, and realize autonomous vehicle avoidance of dynamic obstacles Things.
  • This method makes collision prediction based on the expected trajectory of the vehicle, the predicted trajectory of the obstacle, and the outline size of the vehicle and the obstacle (such as a two-dimensional or three-dimensional rectangular frame).
  • the trajectory point sequence reconstructs the respective convex bodies (frames). You can imagine the trajectory as a track, and the vehicle or obstacle as a train. According to the sliding along the respective trajectory at different times, it is judged whether the two will collide.
  • the detection method is complicated and computationally intensive. Moreover, this method only calculates the shortest distance between the current trajectory and the dynamic obstacle, but cannot give the time and location of the collision. The information provided is not enough to support the vehicle to avoid reasonably, and the vehicle will slow down and stop based on the current trajectory. The efficiency is too low if the vehicle decelerates and stops.
  • the invention provides a method and a device for predicting and judging collisions of an automatic driving vehicle, which are used to solve the problems of complicated detection and large calculation amount when collisions with obstacles occur in the prior art.
  • An automatic driving vehicle collision prediction and judgment method of the present invention includes the following steps:
  • each obstacle prediction path has an intersection point with the respective vehicle prediction path: if there is an intersection point, determine the location of each intersection point, the predicted path of the vehicle where the intersection point and the corresponding edge point of the vehicle, and select each intersection point at the current moment and its corresponding
  • the shortest predicted path length between the edge points of the self-vehicle is the minimum distance, the time when the first intersection exists as the initial time of the collision, and the time when the last intersection exists as the end time of the collision;
  • the present invention no longer regards the obstacle and the self-vehicle as a point for corresponding path planning, but regards both the obstacle and the self-vehicle as a polygon with a convex two-dimensional level, and according to the obstacle
  • the driving motion direction and the driving motion direction of the self-vehicle determine the points on the edge of the polygon that are likely to collide to carry out the corresponding obstacle prediction path planning and the self-vehicle prediction path planning, and determine the predicted path of each obstacle and the predicted path of each vehicle In the case of an intersection, determine whether the vehicle collides with the obstacle based on the driving distance of the vehicle, the set safety distance and the minimum distance, and then calculate the time of the collision, and accurately determine whether the vehicle is on the convex edge
  • the specific location of the collision with the obstacle is more accurate in predicting whether the vehicle will collide with the obstacle.
  • the self-vehicle and the obstacle are both regarded as a point for path planning, because the convex edge of the self-vehicle is not considered, the judgment accuracy is very low when judging whether a collision occurs.
  • the method is simple and less computational. Moreover, this method is suitable for structured roads and unstructured roads such as straight roads, curved roads, and intersections, and no distinction processing is required.
  • step 3 in order to accurately determine whether the collision with the obstacle occurs, it is determined whether the first difference between the driving distance of the vehicle at the initial time of the collision and the set safety distance is greater than the minimum distance If the first difference is greater than the minimum distance, it is determined that the vehicle does not collide with the obstacle.
  • the initial moment that is, the earliest moment when the vehicle may collide with an obstacle. If the vehicle's driving distance before the earliest possible collision moment is greater than the earliest possible collision point (greater than and beyond the corresponding set safety distance), that is, the vehicle The speed is fast enough that it has exceeded the intersection of the path with the obstacle and is far away, the vehicle will no longer collide with the obstacle, so the current acceleration or increase the acceleration will not collide again.
  • the initial moment corresponds to the earliest possible collision path intersection
  • the end moment corresponds to the latest possible collision path intersection.
  • the initial and ending moments can be calculated based on the obstacle moving speed and the distance to the corresponding intersection.
  • step 3 in order to accurately determine whether a collision with an obstacle occurs, determine whether the distance between the driving distance of the vehicle at any time and the set safety distance is between the initial time of the collision and the end time of the collision Whether the second difference of is less than or equal to the minimum distance, if the second difference is less than or equal to the minimum distance, it is determined that the vehicle does not collide with the obstacle.
  • the vehicle's driving distance is less than the minimum distance of the distance from the vehicle to the collision point (and leave enough set safety distance), then the vehicle maintains the current Acceleration or reduction of acceleration will not cause a collision.
  • step 2) if there is no intersection point, it is determined that the vehicle does not collide with the obstacle, and the vehicle is controlled to reduce acceleration, increase acceleration or maintain current acceleration to travel.
  • step 3 in order to achieve the purposes of avoiding collisions, improving traffic efficiency, and improving speed smoothness, if the first difference is greater than the minimum distance, control the vehicle to maintain current acceleration or increase acceleration.
  • step 3 in order to achieve the purposes of avoiding collisions, improving traffic efficiency, and improving speed smoothness, if the second difference is less than or equal to the minimum distance, control the vehicle to maintain the current acceleration to travel.
  • the vehicle in order to achieve the purposes of avoiding collisions, improving traffic efficiency, and improving speed comfort, if the first difference is less than or equal to the minimum distance, it is determined that the vehicle will collide with the obstacle, and the vehicle is controlled to reduce acceleration Driving.
  • the second difference is greater than the minimum distance, it is determined that the vehicle will collide with the obstacle, and the vehicle will be controlled to reduce acceleration. .
  • the whisker algorithm is used to plan the predicted path of the self-vehicle at each point of the edge of the self-vehicle.
  • the present invention also provides an automatic driving vehicle collision prediction and judgment device, including a memory and a processor, the processor is used to execute instructions stored in the memory to implement the above automatic driving vehicle collision prediction and judgment method, and achieve the same as the method effect.
  • Figure 1 is a schematic diagram of the overall scheme of a method embodiment of the present invention.
  • Fig. 2 is a schematic diagram of the collision prediction principle of the method embodiment of the present invention.
  • This embodiment provides a method for predicting and judging an autonomous vehicle collision.
  • the method combines the movement of a moving obstacle (hereinafter referred to as the obstacle) and the movement of an autonomous vehicle (hereinafter referred to as the self-driving vehicle) to predict a period of time in the future Whether the obstacle will collide with the self-vehicle, and then correspondingly control the speed of the self-vehicle, so as to achieve the purpose of avoiding collision, improving traffic efficiency, and improving speed smoothness.
  • external inputs include:
  • Obstacle information at the initial time, the two-dimensional polygon information of the obstacles around the vehicle in the vehicle coordinate system (that is, the position coordinates of each edge point on the polygon relative to the vehicle), movement information (lateral velocity v x , longitudinal velocity v y , Lateral acceleration a x and longitudinal acceleration a y ).
  • Self-vehicle motion information the speed v car and longitudinal acceleration a car of the self-vehicle at the current moment.
  • Self-car planning path information refers to the path information that the self-car will travel within a certain distance in the future at the current moment, which is the discrete coordinate point information in the vehicle coordinate system, and the distance between two adjacent points is fixed at L 0 .
  • Specific treatment methods include:
  • Obstacle collision prediction According to the input obstacle information, self-vehicle planning path information and self-vehicle movement information, predict whether the vehicle will collide with the obstacle.
  • Vehicle speed control strategy According to the predicted collision situation, the vehicle is controlled to accelerate and decelerate.
  • the specific processing method is the automatic driving vehicle collision prediction and judgment method of this embodiment, and the method will be described in detail below.
  • Step 1 Regard the self-car as a two-dimensional polygonal object, and use the whisker algorithm to plan the predicted path of the self-car at each point on the edge of the polygon according to the movement of the self-car. For specific points to plan the path, the situation of the self-vehicle needs to be considered. For example, as shown in Figure 2, the self-vehicle is overlooked and regarded as a rectangle. The top of the rectangle is the front part of the self-vehicle (that is, the forward direction of the self-vehicle).
  • the predicted path of the three points in the forward direction of the self-vehicle is planned, corresponding to the three dashed lines on the rightmost of the four dashed lines from the bottom right to the top left in the figure (the dotted line 123 of the self-vehicle planning), and then plan
  • the predicted path of a point on the side of the vehicle corresponds to the leftmost dashed line of the four dashed lines from the bottom right to the top left in the figure (the dashed line 4 of the self-vehicle planning), so that the area swept by the vehicle body can be obtained .
  • each predicted path of the self-vehicle is discretized into multiple points, each point has corresponding coordinates, and the distance between the two points is the same, which is L 0 .
  • Step 2 The obstacle is also regarded as a two-dimensional polygonal object, and the prediction time length is set to T pre and the discrete time interval is T 0.
  • the obstacle prediction path at each point of the polygon edge is planned. That is, the position information of each point on the edge of the polygon at each time ( interval T 0 ) in the future T pre time.
  • the front of the obstacle (the front part of the obstacle) is planned.
  • the predicted path of the point corresponds to a dashed line from the bottom left to the top right in the figure.
  • the specific formula is as follows:
  • n is a certain point
  • t is a certain time in the future
  • x n0 and y n0 are the initial positions of n points
  • x nt and y nt are the predicted positions at time t
  • v xt and v yt are the predicted velocities at time t
  • a x and a y are the lateral and longitudinal accelerations of the obstacle at the current moment
  • v x and v y are the lateral and longitudinal accelerations of the obstacle at the current moment, respectively.
  • the specific points on the edge in steps 1 and 2 for path planning need to be determined according to the driving direction of the obstacle and the driving direction of the vehicle. For example, there is no need to determine the right edge point of the vehicle in Figure 2. The obstacle predicts the path.
  • Step 3 Determine whether there is an intersection point between the vehicle's respective predicted path of the vehicle and each obstacle predicted path of the obstacle: if there is no intersection point, perform step four; if there is an intersection point, perform step five.
  • Step 4 There is no intersection between the predicted path of each obstacle and the predicted path of the respective vehicle, indicating that the obstacle and the vehicle will not collide within T pre . At this time, there is no restriction on the speed of the vehicle, and it can be accelerated or decelerated. The original speed can also be maintained.
  • Step 5 There is an intersection point between the predicted path of each obstacle and the predicted path of each vehicle, indicating that there is a possibility of collision between the obstacle and the self-vehicle within the time T pre.
  • the length is taken as the minimum distance L tmin
  • the time when the first intersection exists is taken as the initial time t start of the collision
  • the time when the last intersection exists is taken as the end time t end of the collision.
  • the initial time t start and the end time t end can be calculated according to the speed of the obstacle and the distance to the corresponding intersection.
  • Step 6 traverse the driving distance of the vehicle from t start to t end, the formula is as follows:
  • St and v tcar are the travel distance and speed of the vehicle at time t , respectively, and v car and a car are the initial speed and acceleration of the vehicle at the initial time, respectively.
  • Step 7 Set the forward collision safety distance S lon , set the side collision safety distance S lat and the minimum distance L tmin according to the driving distance S t of the own vehicle (refers to the closest distance that the own vehicle will collide with the obstacle) , To judge whether the vehicle collides with the obstacle:
  • the initial moment that is, the earliest moment when the vehicle may collide with an obstacle. If the vehicle's driving distance before the earliest possible collision moment is greater than the earliest possible collision point (greater than and beyond the corresponding set safety distance), that is, the vehicle The speed is fast enough that it has exceeded the intersection of the path with the obstacle and is far away, and the vehicle will no longer collide with the obstacle, so it will not collide with the current acceleration or increase the acceleration. If the driving distance of the vehicle before the earliest possible collision moment is not greater than the earliest possible collision point (or there is no sufficient safety distance left), there is a possibility of collision, so acceleration should be reduced to prevent collision.
  • the vehicle's driving distance is less than the minimum distance of the distance from the vehicle to the collision point (and leave enough set safety distance), then the vehicle maintains the current Acceleration or reduction of acceleration will not cause a collision. If the driving distance of the vehicle is greater than or equal to the minimum distance of the distance from the vehicle to the point of collision (or insufficient safety distance is not set aside), there is a possibility of collision, so acceleration should be reduced to prevent collision.
  • this method offsets the expected or planned trajectory of the vehicle along the outer contour of the car body frame (projection along the driving direction).
  • Each trajectory means the trajectory that the corresponding car body contour point will traverse, similar to a lot of generated from the car body.
  • Tentacle to detect whether there are obstacles in the area to be driven, just like a cat's tentacles; and the prediction of obstacles, according to its current position, movement state, and frame size, to predict the position of the frame at a certain time in the future ; In this way, according to the position of the vehicle's tentacles and the position of the obstacle frame, it is judged which point is in the frame, so that it is judged whether there is interference, and then according to the safety distance condition, it is further judged whether there will be a collision for the control of the self-vehicle.
  • this method can obtain the time, distance, and location of the collision, and the obtained information is more comprehensive, so that the self-vehicle can be decelerated and avoided (including yielding) or running normally (including constant speed or acceleration) based on the above collision information.
  • This method uses the whisker algorithm for collision detection, and the detection method is simple and easy to implement, and the amount of calculation is small.
  • This embodiment provides a device for predicting and judging an automatic driving vehicle collision.
  • the device includes a memory and a processor, and the memory and the processor are directly or indirectly electrically connected to realize data transmission or interaction.
  • the processor here can be a vehicle controller or a processor added to the entire vehicle.
  • the processor added to the entire vehicle can be a general-purpose processor, such as a central processing unit (CPU), or other programmable logic devices.
  • the digital signal processor DSP is used to execute the instructions stored in the memory to implement the method for predicting and judging the collision of an automatic driving vehicle introduced in the method embodiment. Since the method has been described in detail in the method embodiment, I won't repeat it here.

Landscapes

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

Abstract

一种自动驾驶车辆碰撞预测判断方法及装置，将障碍物和自车均看作是具有凸起的二维层面的多边形，依据障碍物的行驶运动方向与自车的行驶运动方向确定一些很可能发生碰撞的多边形边缘上的点进行对应的障碍物预测路径规划和自车预测路径规划，并在判断各障碍物预测路径与各自车预测路径存在交点的情况下，根据自车行驶距离、设定安全距离和最小距离，判断自车与障碍物是否发生碰撞，进而可计算出发生碰撞的时刻，并准确判断自车车身发生碰撞的具体位置，对自车与障碍物是否发生碰撞的预测更加准确，该方法适用于直行道路、弯曲道路、交叉路口等结构化道路以及非结构化道路，无需区分处理。

Description

一种自动驾驶车辆碰撞预测判断方法及装置 技术领域
本发明属于自动驾驶技术领域，具体涉及一种自动驾驶车辆碰撞预测判断方法及装置。
背景技术
随着科技的发展，无人驾驶技术也逐渐走入人们的生活。随着无人驾驶技术的快速发展，原先仅仅停留在实验室里的科学研究正在逐步转向商业化应用，要想实现完全且安全的无人驾驶，无人驾驶车辆的路径规划和速度规划是其核心技术之一。
在路径规划中，需要对环境进行感知，在对环境进行感知时，检测无人驾驶车辆周围的障碍物是环境感知的一个重要部分。在无障碍物的情况下，无人驾驶车辆会优先沿着所在车道的车道中心线行驶；在有障碍物的情形下，无人驾驶车辆需要在原有规划的路径下进行合理有效的避障行为。
例如，申请公布号为CN109960261A的中国发明专利申请公开了一种基于碰撞检测的动态障碍物避让方法，该方法通过初规划构造车辆期望轨迹，并根据接收的动态障碍物预测轨迹和所述车辆期望轨迹，进行碰撞检测，对车辆的行驶速度和/或路径进行重新规划，得到车辆无碰撞行驶速度和/或路径数据，重新构造车辆期望轨迹，避让动态障碍物，实现无人车辆自主避让动态障碍物。该方法根据车辆的期望的轨迹、障碍物的预测轨迹，以及车辆、障碍物的外廓尺寸(如，二维或三维矩形框体)，做碰撞预测，需要根据障碍物预测轨迹、自车期望轨迹点序列重构各自凸体(框体)，可以将轨迹想象成是轨道，车辆或者障碍物想象成火车，根据不同时刻沿各自轨迹滑移，判断两者是否会发生碰撞。该检测方法复杂、计算量大。而且，该方法仅计算出当前所处轨迹与动态障碍物的最近距离，而不能给出发生碰撞的时刻、位置，所给信息不足以支撑车辆合理避让，且基于当前轨迹减速停车，而所有情况均减速停车的话效率太低。
发明内容
本发明提供了一种自动驾驶车辆碰撞预测判断方法及装置，用以解决现有技术中对与障碍物是否发生碰撞时检测复杂、计算量大的问题。
为解决上述技术问题，本发明的技术方案和有益效果为：
本发明的一种自动驾驶车辆碰撞预测判断方法，包括如下步骤：
1)根据障碍物运动情况，规划障碍物边缘各点的障碍物预测路径，根据自车的运 动情况，规划自车边缘各点的自车预测路径；
2)判断各障碍物预测路径与各自车预测路径是否存在交点：若存在交点，确定各交点的位置、交点所在的自车预测路径以及对应的自车边缘点，选取当前时刻下各交点与其对应的自车边缘点之间最短的自车预测路径长度作为最小距离，以第一次存在交点的时刻作为发生碰撞的初始时刻，以最后一次存在交点的时刻作为发生碰撞的结束时刻；
3)根据自车行驶距离、设定安全距离和所述最小距离，判断自车与障碍物是否发生碰撞。
其有益效果：本发明不再将障碍物和自车当做一个点来进行对应的路径规划，而是将障碍物和自车均看作是具有凸起的二维层面的多边形，依据障碍物的行驶运动方向与自车的行驶运动方向确定一些很可能发生碰撞的多边形边缘上的点进行对应的障碍物预测路径规划和自车预测路径规划，并在判断各障碍物预测路径与各自车预测路径存在交点的情况下，根据自车行驶距离、设定安全距离和所述最小距离，判断自车与障碍物是否发生碰撞，进而可计算出发生碰撞的时刻，并准确判断出自车凸起边缘上与障碍物发生碰撞的具体位置，对自车与障碍物是否发生碰撞的预测更加准确。若把自车和障碍物均看作一个点来进行路径规划，因未考虑自车的凸起边缘，则在判断是否发生碰撞时判断精度很低。该方法简单、计算量少。而且，该方法适用于直行道路、弯曲道路、交叉路口等结构化道路以及非结构化道路，无需区分处理。
作为方法的进一步改进，步骤3)中，为了准确判断与障碍物是否发生碰撞，判断发生碰撞的初始时刻自车的行驶距离与设定安全距离之间的第一差值是否大于所述最小距离，若第一差值大于所述最小距离，则判定自车与障碍物不发生碰撞。
初始时刻，即车辆可能与障碍物发生碰撞的最早时刻，若车辆在该可能碰撞的最早时刻之前的行驶距离已经大于最早可能碰撞的碰撞点(大于且超出对应的设定安全距离)，即车辆速度足够快，已经超过与障碍物的路径交汇处并远离，车辆不会再与障碍物碰撞，因此维持当前加速度或提高加速度都不会再发生碰撞。
初始时刻对应最早可能发生碰撞的路径交汇点，结束时刻对应最晚可能发生碰撞的路径交汇点，初始时刻和结束时刻可以根据障碍物移动速度和到对应交汇点的距离计算得出。
作为方法的进一步改进，步骤3)中，为了准确判断与障碍物是否发生碰撞，判断发生碰撞的初始时刻与发生碰撞的结束时刻之间任一时刻自车的行驶距离与设定安全距离之间的第二差值是否小于等于所述最小距离，若第二差值小于等于所述最小距离，则 判定自车与障碍物不发生碰撞。
可能发生碰撞的初始时刻到可能发生碰撞的最终时刻中任意时刻之前，车辆的行驶距离都小于车辆到碰撞点的距离中的最小距离(且留出足够的设定安全距离)，则车辆维持当前加速度或降低加速度都不会发生碰撞。
作为方法的进一步改进，步骤2)中，若不存在交点，则判定自车与障碍物不发生碰撞，并控制自车降低加速度、提高加速度或保持当前加速度行驶。
作为方法的进一步改进，步骤3)中，为了达到避免碰撞、提高通行效率、改善速度平顺性的目的，若第一差值大于所述最小距离，控制自车保持当前加速度行驶或提高加速度行驶。
作为方法的进一步改进，步骤3)中，为了达到避免碰撞、提高通行效率、改善速度平顺性的目的，若第二差值小于等于所述最小距离，控制自车保持当前加速度行驶。
作为方法的进一步改进，为了达到避免碰撞、提高通行效率、改善速度平顺性的目的，若第一差值小于等于所述最小距离，则判定自车与障碍物会发生碰撞，控制自车降低加速度行驶。
作为方法的进一步改进，为了达到避免碰撞、提高通行效率、改善速度平顺性的目的，若第二差值大于所述最小距离，则判定自车与障碍物会发生碰撞，控制自车降低加速度行驶。
作为方法的进一步改进，为了准确预测得到步骤1)中，根据自车的运动情况，采用触须算法规划自车边缘各点的自车预测路径。
本发明还提供了一种自动驾驶车辆碰撞预测判断装置，包括存储器与处理器，所述处理器用于执行存储在存储器中的指令以实现上述自动驾驶车辆碰撞预测判断方法，并达到与方法相同的效果。
附图说明
图1是本发明的方法实施例的整体方案示意图；
图2是本发明的方法实施例的碰撞预测原理示意图。
具体实施方式
方法实施例：
该实施例提供了一种自动驾驶车辆碰撞预测判断方法，该方法结合运动障碍物(下简称障碍物)的运动情况，以及自动驾驶车辆(下简称自车)的运动情况，预测未来一段时间内障碍物与自车是否会发生碰撞，进而对自车的速度进行相应的控制，以达到避免碰 撞、提高通行效率、改善速度平顺性的目的。
该方法可同时适用于直行道路、弯曲道路、交叉路口等结构化道路以及非结构化道路，无需区分处理。为了实施该方法，整体方案由外部输入和处理方法两部分组成。如图1所示，外部输入包括：
1)障碍物信息：在初始时刻、车辆坐标系下车辆周围障碍物的二维多边形信息(即多边形上各边缘点相对于车辆的位置坐标)、运动信息(横向速度v x、纵向速度v y、横向加速度a x和纵向加速度a y)。
2)自车运动信息：自车在当前时刻下的速度v car和纵向加速度a car。
3)自车规划路径信息：是指在当前时刻下未来一段距离内自车将要行驶的路径信息，为车辆坐标系下离散的坐标点信息，相邻两点之间的距离间隔固定为L 0。
具体的处理方法包括：
1)障碍物碰撞预测：根据输入的障碍物信息、自车规划路径信息及自车运动信息，预测车辆与障碍物是否会发生碰撞。
2)车辆速度控制策略：根据预测的碰撞情况，对车辆进行加、减速控制。
具体的处理方法即为该实施例的自动驾驶车辆碰撞预测判断方法，下面对该方法做详细的说明。
步骤一，将自车看作是一个二维多边形的物体，根据自车的运动情况，采用触须算法，规划多边形边缘各点的自车预测路径。具体对哪些点进行路径规划，需考虑自车的情况，例如，如图2所示，俯视自车，将其看作是一个矩形，矩形的上方为自车车头部位(即自车的正向)，在该图中，规划了自车正向三个点的预测路径，对应图中从右下方至左上方的四条虚线中最右侧的三条虚线(自车规划的虚线①②③)，又规划了自车侧向的一个点的预测路径，对应图中从右下方至左上方的四条虚线中的最左侧的一条虚线(自车规划的虚线④)，这样便可得到车身扫过的区域。在此基础上，将每条自车预测路径离散为多个点，每个点都有相应坐标，两点之间距离间隔相同，均为L 0。
步骤二，同样将障碍物看作是一个二维多边形的物体，设定预测时间长度为T pre，离散时间间隔为T 0，根据障碍物运动情况，规划多边形边缘各点的障碍物预测路径，即多边形边缘各点在未来T pre时间内每个时刻(以T 0为间隔)的位置信息。例如，如图2所 示，俯视障碍物，将其也看作是一个矩形，矩形的右上方为障碍物前方部位，在该图中，规划了障碍物正向(障碍物车头部位)的一个点的预测路径，对应图中从左下方至右上方的一条虚线。具体的公式如下：
Figure PCTCN2020131002-appb-000001
Figure PCTCN2020131002-appb-000002
v xt＝v x+a xt
v yt＝v y+a yt
式中，n为某点，t为未来某一时刻，x n0、y n0为n点的初始位置，x nt、y nt为t时刻的预测位置，v xt、v yt为t时刻的预测测速，a x、a y分别为障碍物当前时刻横向、纵向加速度，v x、v y分别为障碍物当前时刻横向、纵向速度。
需说明的是，步骤一和步骤二中具体对边缘哪些点进行路径规划，需根据障碍物的行驶方向和自车的行驶方向来相应确定，例如，图2中无需确定自车右侧边缘点的障碍物预测路径。
步骤三，根据自车的各自车预测路径以及障碍物的各障碍物预测路径，判断两者之间是否存在交点：若不存在交点，则执行步骤四；若存在交点，则执行步骤五。
步骤四，各障碍物预测路径与各自车预测路径之间不存在交点，说明障碍物与自车在T pre时间内不会发生碰撞，此时对自车速度无限制，可加速，可减速，也可保持原来的车速。
步骤五，各障碍物预测路径与各自车预测路径之间存在交点，说明障碍物与自车的在T pre时间内存在碰撞的可能。此时，确定各交点的位置、交点所在的自车预测路径以及对应的自车边缘点，选取T pre时间内，当前时刻下各交点与其对应的自车边缘点之间最短的自车预测路径长度作为最小距离L tmin，并以第一次存在交点的时刻作为发生碰撞的初始时刻t start，以最后一次存在交点的时刻作为发生碰撞的结束时刻t end。初始时刻t start和结束时刻t end可以根据障碍物移动速度和到对应交汇点的距离计算得出。
其中，发生碰撞的某时刻t可能发生多条自车预测路径同时与障碍物发生碰撞的 情况，距离L mt＝i mL 0，m表示自车预测路径的序号，通过m可以判断出发生碰撞的车身位置(如正面、侧面)，i m表示自车预测路径m上的离散点的个数。具体公式为：
L tmin＝min(i 0,i 1,...,i m)·L 0
步骤六，遍历t start到t end时间内自车的行驶距离，公式如下：
Figure PCTCN2020131002-appb-000003
v tcar＝v car+a cart
式中，S t、v tcar分别为自车在t时刻的自车行驶距离及速度，v car、a car分别为自车在初始时刻的初速度和加速度。
步骤七，根据自车行驶距离S t、设定正向碰撞安全距离S lon、设定侧向碰撞安全距离S lat和最小距离L tmin(指自车将与障碍物发生碰撞的最近的距离)，来判断自车与障碍物是否发生碰撞：
1)判断是否满足S t-S lon≤L tmin：若满足，则判定自车与障碍物不会发生碰撞，控制自车正常行驶，保持当前加速度行驶；否则，判定自车与障碍物会发生碰撞，自车应低于当前加速度行驶。
初始时刻，即车辆可能与障碍物发生碰撞的最早时刻，若车辆在该可能碰撞的最早时刻之前的行驶距离已经大于最早可能碰撞的碰撞点(大于且超出对应的设定安全距离)，即车辆速度足够快，已经超过与障碍物的路径交汇处并远离，车辆不会再与障碍物碰撞，因此维持当前加速度或提高加速度都不会在发生碰撞。若车辆在该可能碰撞的最早时刻之前的行驶距离没有大于最早可能碰撞的碰撞点(或者未留出足够的安全距离)，则存在碰撞可能，因此应当降低加速度，防止碰撞。
2)判断是否满足S t-S lat＞L tmin，且t＝t start：若满足，则判定自车与障碍物不会发生碰撞且自车优先通行，自车可保持或高于当前加速度行驶；否则，判定自车与障碍物会发生碰撞，自车应低于当前加速度行驶。
可能发生碰撞的初始时刻到可能发生碰撞的最终时刻中任意时刻之前，车辆的行驶距离都小于车辆到碰撞点的距离中的最小距离(且留出足够的设定安全距离)，则车辆维持当前加速度或降低加速度都不会发生碰撞。若车辆的行驶距离大于或等于车辆到碰撞 点的距离中的最小距离(或者未留出足够的安全距离)，则存在碰撞可能，因此应当降低加速度，防止碰撞。
整体上看，该方法将车辆期望或规划的轨迹沿车身框体外廓(沿行驶方向的投影)做偏移，每条轨迹意味着相应车身外廓点将要走过的轨迹，类似从车身生成许多触须，去探测将要行驶的区域是否有障碍物，就像猫的触须一样；而障碍物的预测，根据其当前位置、运动状态、框体尺寸，来预测未来某时刻其框体所处的位置；如此根据自车触须点位置、障碍物框体位置，判断哪个点在框体里，如此判断是否干涉，进而根据安全距离条件，进一步判断对于自车控制而言是否会发生碰撞。
而且，该方法可以得出发生碰撞的时刻、距离、位置，得到的信息较为全面，从而可根据以上碰撞信息对自车进行减速避让(包括让行)或正常行驶(包括匀速或加速)。该方法采用触须算法进行碰撞检测，检测方法简单易实现、计算量小。
装置实施例：
该实施例提供了一种自动驾驶车辆碰撞预测判断装置，该装置包括存储器和处理器，存储器和处理器之间直接或间接地电性连接以实现数据的传输或交互。这里的处理器可以整车控制器，还可以是增加进整车的处理器，该增加进整车的处理器可以是通用处理器，例如中央处理器CPU，也可以是其他可编程逻辑器件，例如数字信号处理器DSP，处理器用于执行存储在存储器中的指令以实现方法实施例中介绍的一种自动驾驶车辆碰撞预测判断方法，由于在方法实施例中已对该方法做了详细说明，这里不再赘述。
尽管本发明的内容已经通过上述优选实施例作了详细介绍，但应当认识到上述的描述不应被认为是对本发明的限制。在本领域技术人员阅读了上述内容后，对于本发明的多种修改和替代都将是显而易见的。因此，本发明的保护范围应由所附的权利要求来限定。

Claims (10)

  1. 一种自动驾驶车辆碰撞预测判断方法，其特征在于，包括如下步骤：
    1)根据障碍物运动情况，规划障碍物边缘各点的障碍物预测路径，根据自车的运动情况，规划自车边缘各点的自车预测路径；
    2)判断各障碍物预测路径与各自车预测路径是否存在交点：若存在交点，确定各交点的位置、交点所在的自车预测路径以及对应的自车边缘点，选取当前时刻下各交点与其对应的自车边缘点之间最短的自车预测路径长度作为最小距离，以第一次存在交点的时刻作为发生碰撞的初始时刻，以最后一次存在交点的时刻作为发生碰撞的结束时刻；
    3)根据自车行驶距离、设定安全距离和所述最小距离，判断自车与障碍物是否发生碰撞；所述自车行驶距离为自车在当前时刻到可能碰撞时间段内任意时刻之间所行驶的距离，所述可能碰撞时间段从初始时刻开始到结束时刻为止。
  2. 根据权利要求1所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤3)中，判断发生碰撞的初始时刻自车的行驶距离与设定安全距离之间的第一差值是否大于所述最小距离，若第一差值大于所述最小距离，则判定自车与障碍物不发生碰撞。
  3. 根据权利要求1或2所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤3)中，判断发生碰撞的初始时刻与发生碰撞的结束时刻之间任一时刻自车的行驶距离与设定安全距离之间的第二差值是否小于等于所述最小距离，若第二差值小于等于所述最小距离，则判定自车与障碍物不发生碰撞。
  4. 根据权利要求1所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤2)中，若不存在交点，则判定自车与障碍物不发生碰撞，并控制自车降低加速度、提高加速度或保持当前加速度行驶。
  5. 根据权利要求2所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤3)中，若第一差值大于所述最小距离，控制自车保持当前加速度行驶或提高加速度行驶。
  6. 根据权利要求3所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤3)中，若第二差值小于等于所述最小距离，控制自车保持当前加速度行驶。
  7. 根据权利要求2所述的自动驾驶车辆碰撞预测判断方法，其特征在于，若第一差值小于等于所述最小距离，则判定自车与障碍物会发生碰撞，控制自车降低加速度行驶。
  8. 根据权利要求3所述的自动驾驶车辆碰撞预测判断方法，其特征在于，若第二差值大于所述最小距离，则判定自车与障碍物会发生碰撞，控制自车降低加速度行驶。
  9. 根据权利要求1所述的自动驾驶车辆碰撞预测判断方法，其特征在于，步骤1)中， 根据自车的运动情况，采用触须算法规划自车边缘各点的自车预测路径。
  10. 一种自动驾驶车辆碰撞预测判断装置，其特征在于，包括存储器与处理器，所述处理器用于执行存储在存储器中的指令以实现如权利要求1～9任一项所述的自动驾驶车辆碰撞预测判断方法。
PCT/CN2020/131002 2019-12-30 2020-11-24 一种自动驾驶车辆碰撞预测判断方法及装置 Ceased WO2021135728A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP20909407.7A EP4082848B1 (en) 2019-12-30 2020-11-24 Determination device for collision prediction of autonomous vehicle

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911398952.2A CN113119964B (zh) 2019-12-30 2019-12-30 一种自动驾驶车辆碰撞预测判断方法及装置
CN201911398952.2 2019-12-30

Publications (1)

Publication Number Publication Date
WO2021135728A1 true WO2021135728A1 (zh) 2021-07-08

Family

ID=76687273

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/131002 Ceased WO2021135728A1 (zh) 2019-12-30 2020-11-24 一种自动驾驶车辆碰撞预测判断方法及装置

Country Status (3)

Country Link
EP (1) EP4082848B1 (zh)
CN (1) CN113119964B (zh)
WO (1) WO2021135728A1 (zh)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113734163A (zh) * 2021-09-07 2021-12-03 北京三快在线科技有限公司 无人车的控制方法、装置、存储介质及电子设备
CN113776550A (zh) * 2021-09-08 2021-12-10 京东鲲鹏(江苏)科技有限公司 一种路线规划方法和装置
CN113885525A (zh) * 2021-10-30 2022-01-04 重庆长安汽车股份有限公司 一种自动驾驶车辆脱困的路径规划方法、系统、车辆及存储介质
CN114137964A (zh) * 2021-11-09 2022-03-04 长沙中联重科环境产业有限公司 车辆姿态及扫盘与路沿跟随控制方法、装置、设备及专用车
CN114235441A (zh) * 2021-12-31 2022-03-25 深兰人工智能(深圳)有限公司 无人车实时碰撞检测方法及相关装置
CN114237229A (zh) * 2021-11-26 2022-03-25 青岛德智汽车科技有限公司 基于经验路径拟合的非结构化道路作业车辆路径规划方法
CN114537438A (zh) * 2022-03-07 2022-05-27 京东鲲鹏(江苏)科技有限公司 速度控制方法、装置、控制设备及存储介质
CN114802251A (zh) * 2022-05-11 2022-07-29 阿波罗智能技术(北京)有限公司 自动驾驶车辆的控制方法、装置、电子设备以及存储介质
CN114882706A (zh) * 2022-05-31 2022-08-09 华中科技大学 一种基于非结构性路面的周向防碰撞预警方法及系统
CN115027464A (zh) * 2022-07-29 2022-09-09 西安电子科技大学芜湖研究院 一种基于方向包围盒的自动驾驶碰撞检测方法
CN115092130A (zh) * 2022-06-16 2022-09-23 阿波罗智能技术(北京)有限公司 车辆碰撞预测方法、装置、电子设备、介质和车辆
CN115092136A (zh) * 2022-07-27 2022-09-23 广州小鹏自动驾驶科技有限公司 车速规划方法、装置、车辆及存储介质
CN115214724A (zh) * 2022-09-20 2022-10-21 毫末智行科技有限公司 一种轨迹预测的方法、装置、电子设备及存储介质
CN115355916A (zh) * 2022-10-24 2022-11-18 北京智行者科技股份有限公司 移动工具的轨迹规划方法、设备和计算机可读存储介质
CN115407774A (zh) * 2022-08-25 2022-11-29 魔门塔(苏州)科技有限公司 行驶路径规划方法及装置、车辆、存储介质、终端
CN115534943A (zh) * 2022-10-10 2022-12-30 中国第一汽车股份有限公司 车辆与待监测车辆碰撞预测的方法和装置
CN115620554A (zh) * 2022-07-27 2023-01-17 潍柴智能科技有限公司 基于路测设备的交叉路口自动驾驶车辆行驶方法及系统
CN115907245A (zh) * 2022-10-24 2023-04-04 劢微机器人科技(深圳)有限公司 用于agv调度的碰撞检测方法、装置、设备及存储介质
CN116279500A (zh) * 2023-05-24 2023-06-23 深圳联友科技有限公司 一种车辆碰撞识别方法
CN116409339A (zh) * 2021-12-30 2023-07-11 魔门塔(苏州)科技有限公司 一种基于前方目标的自动驾驶速度规划方法、装置及介质
CN116572994A (zh) * 2023-07-10 2023-08-11 之江实验室 一种车辆速度规划方法、装置及计算机可读介质
CN116612660A (zh) * 2023-04-10 2023-08-18 山东高速建设管理集团有限公司 一种高速公路改扩建施工区车辆冲突预警方法及设备
CN116719318A (zh) * 2023-06-05 2023-09-08 库卡机器人(广东)有限公司 碰撞检测方法及装置、自主移动装置、存储介质
CN117148848A (zh) * 2023-10-27 2023-12-01 上海伯镭智能科技有限公司 一种无人驾驶车辆智能避障方法及系统
CN118046926A (zh) * 2024-04-02 2024-05-17 重庆大学 一种用于自动驾驶系统的车辆行驶轨迹预测方法

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102021109661B4 (de) * 2021-04-16 2023-03-09 Avl Software And Functions Gmbh Verfahren zur geometrischen Darstellung einer Fahrzeugfläche eines Fahrzeugs zur Kollisionserkennung
CN115691152A (zh) * 2021-07-29 2023-02-03 北京万集科技股份有限公司 匝道口交通调度的方法及设备
CN113442924A (zh) * 2021-08-03 2021-09-28 浙江吉利控股集团有限公司 一种车辆轨迹的规划方法及规划系统
CN113335312B (zh) * 2021-08-06 2021-10-19 新石器慧通（北京）科技有限公司 一种绕障行驶方法、装置、电子设备及介质
CN113485383B (zh) * 2021-09-02 2021-12-31 北京三快在线科技有限公司 一种无人驾驶设备的控制方法及控制装置
CN115230731B (zh) * 2021-09-13 2025-11-25 上海仙途智能科技有限公司 行驶路径确定方法、装置、终端及介质
CN114103893A (zh) * 2021-11-26 2022-03-01 河北春玖智能科技有限公司 一种无人车的轨迹预测防撞方法
CN115292796B (zh) * 2021-11-26 2023-06-02 上海仙途智能科技有限公司 一种碰撞检测方法及装置
CN114030486B (zh) * 2021-12-21 2024-05-28 阿波罗智联(北京)科技有限公司 动态障碍物的轨迹预测方法、装置、电子设备和存储介质
CN116409342A (zh) * 2021-12-30 2023-07-11 魔门塔(苏州)科技有限公司 一种自动泊车速度规划方法、装置及介质
CN114407919B (zh) * 2021-12-31 2023-06-02 武汉中海庭数据技术有限公司 一种基于自动驾驶的碰撞检测方法及系统
CN114187764B (zh) * 2022-02-16 2022-05-13 深圳佑驾创新科技有限公司 一种针对vru穿行场景的碰撞危险度快速检测方法
CN114510053B (zh) * 2022-02-17 2025-03-21 北京京东乾石科技有限公司 机器人规划路径校验方法、装置、存储介质及电子设备
CN114537439B (zh) * 2022-03-14 2026-03-24 广州小鹏自动驾驶科技有限公司 自动驾驶速度规划方法、电子设备、车辆及存储介质
CN115167426B (zh) * 2022-07-15 2025-07-22 深圳市正浩创新科技股份有限公司 多自移动设备的路径规划方法、装置、自移动设备
DE102022123474A1 (de) * 2022-09-14 2024-03-14 Valeo Schalter Und Sensoren Gmbh Verfahren zum Berechnen einer prognostizierten Zeit bis zu einer möglichen Kollision zweier Kraftfahrzeuge, Computerprogramm und Fahrerassistenzsystem
CN115309168B (zh) * 2022-10-11 2023-01-10 天地科技股份有限公司 一种井下无人车控制方法及装置
CN116552515B (zh) * 2023-05-09 2025-10-10 联创汽车电子有限公司 车辆碰撞时间计算方法、系统及存储介质
CN118636882A (zh) * 2024-06-28 2024-09-13 上海安亭地平线智能交通技术有限公司 车辆碰撞时间的确定方法及装置、设备、存储介质
CN119335559A (zh) * 2024-09-30 2025-01-21 东风商用车有限公司 基于激光雷达的碰撞检测方法及装置
CN119428759A (zh) * 2024-11-27 2025-02-14 易控智驾科技有限公司 自动驾驶车辆控制方法、装置和系统及自动驾驶车辆

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101837782A (zh) * 2009-01-26 2010-09-22 通用汽车环球科技运作公司 用于碰撞预备系统的多目标融合模块
CN102806912A (zh) * 2011-06-01 2012-12-05 通用汽车环球科技运作有限责任公司 用于连接的自主车辆和手动车辆的快速碰撞检测技术
CN104882025A (zh) * 2015-05-13 2015-09-02 东华大学 一种基于车联网技术的碰撞检测预警方法
CN108062600A (zh) * 2017-12-18 2018-05-22 北京星云互联科技有限公司 一种基于矩形建模的车辆碰撞预警方法及装置
CN108536149A (zh) * 2018-04-25 2018-09-14 吉林大学 一种基于Dubins路径的无人驾驶车辆避障控制装置及控制方法
CN109814544A (zh) * 2017-11-22 2019-05-28 通用汽车环球科技运作有限责任公司 用于在自主车辆中操纵绕过障碍物的系统和方法
CN109960261A (zh) 2019-03-22 2019-07-02 北京理工大学 一种基于碰撞检测的动态障碍物避让方法
CN110221595A (zh) * 2018-03-02 2019-09-10 本田技研工业株式会社 车辆控制装置
US20190354781A1 (en) * 2018-05-17 2019-11-21 GM Global Technology Operations LLC Method and system for determining an object location by using map information

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002074594A (ja) * 2000-08-25 2002-03-15 Alpine Electronics Inc 障害物検知システム
JP2007279892A (ja) * 2006-04-04 2007-10-25 Honda Motor Co Ltd 衝突予知システムの制御装置、衝突予知方法および乗員保護システム
DE102010002105A1 (de) * 2010-02-18 2011-08-18 Robert Bosch GmbH, 70469 Verfahren zur Unterstützung eines Fahrers eines Fahrzeugs bei einem Fahrmanöver
DE102010051203B4 (de) * 2010-11-12 2022-07-28 Zf Active Safety Gmbh Verfahren zur Erkennung von kritischen Fahrsituationen von Lastkraft- oder Personenkraftwagen, insbesondere zur Vermeidung von Kollisionen
JP5754510B2 (ja) * 2011-09-26 2015-07-29 トヨタ自動車株式会社 車両の運転支援システム
CN103335658B (zh) * 2013-06-19 2016-09-14 华南农业大学 一种基于弧线路径生成的自主车辆避障方法
KR101604447B1 (ko) * 2015-05-26 2016-03-28 주식회사 피엘케이 테크놀로지 선행차량 추돌 경보 장치 및 방법
JP6308186B2 (ja) * 2015-08-28 2018-04-11 トヨタ自動車株式会社 衝突回避支援装置
CN107672588B (zh) * 2017-08-29 2020-05-08 广州小鹏汽车科技有限公司 一种自动泊车路径障碍物碰撞检测方法、装置及系统
CN109407673A (zh) * 2018-12-19 2019-03-01 清华大学苏州汽车研究院（吴江） 一种适用于结构化道路的自主换道控制方法及系统
CN110065494B (zh) * 2019-04-09 2020-07-31 魔视智能科技(上海)有限公司 一种基于车轮检测的车辆防碰撞方法
CN110550029B (zh) * 2019-08-12 2021-02-09 华为技术有限公司 障碍物避让方法及装置

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101837782A (zh) * 2009-01-26 2010-09-22 通用汽车环球科技运作公司 用于碰撞预备系统的多目标融合模块
CN102806912A (zh) * 2011-06-01 2012-12-05 通用汽车环球科技运作有限责任公司 用于连接的自主车辆和手动车辆的快速碰撞检测技术
CN104882025A (zh) * 2015-05-13 2015-09-02 东华大学 一种基于车联网技术的碰撞检测预警方法
CN109814544A (zh) * 2017-11-22 2019-05-28 通用汽车环球科技运作有限责任公司 用于在自主车辆中操纵绕过障碍物的系统和方法
CN108062600A (zh) * 2017-12-18 2018-05-22 北京星云互联科技有限公司 一种基于矩形建模的车辆碰撞预警方法及装置
CN110221595A (zh) * 2018-03-02 2019-09-10 本田技研工业株式会社 车辆控制装置
CN108536149A (zh) * 2018-04-25 2018-09-14 吉林大学 一种基于Dubins路径的无人驾驶车辆避障控制装置及控制方法
US20190354781A1 (en) * 2018-05-17 2019-11-21 GM Global Technology Operations LLC Method and system for determining an object location by using map information
CN109960261A (zh) 2019-03-22 2019-07-02 北京理工大学 一种基于碰撞检测的动态障碍物避让方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4082848A4

Cited By (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113734163A (zh) * 2021-09-07 2021-12-03 北京三快在线科技有限公司 无人车的控制方法、装置、存储介质及电子设备
CN113776550A (zh) * 2021-09-08 2021-12-10 京东鲲鹏(江苏)科技有限公司 一种路线规划方法和装置
CN113776550B (zh) * 2021-09-08 2024-04-16 京东鲲鹏(江苏)科技有限公司 一种路线规划方法和装置
CN113885525A (zh) * 2021-10-30 2022-01-04 重庆长安汽车股份有限公司 一种自动驾驶车辆脱困的路径规划方法、系统、车辆及存储介质
CN114137964B (zh) * 2021-11-09 2024-05-31 长沙中联重科环境产业有限公司 车辆姿态及扫盘与路沿跟随控制方法、装置、设备及专用车
CN114137964A (zh) * 2021-11-09 2022-03-04 长沙中联重科环境产业有限公司 车辆姿态及扫盘与路沿跟随控制方法、装置、设备及专用车
CN114237229B (zh) * 2021-11-26 2023-06-13 青岛德智汽车科技有限公司 基于经验路径拟合的非结构化道路作业车辆路径规划方法
CN114237229A (zh) * 2021-11-26 2022-03-25 青岛德智汽车科技有限公司 基于经验路径拟合的非结构化道路作业车辆路径规划方法
CN116409339A (zh) * 2021-12-30 2023-07-11 魔门塔(苏州)科技有限公司 一种基于前方目标的自动驾驶速度规划方法、装置及介质
CN114235441A (zh) * 2021-12-31 2022-03-25 深兰人工智能(深圳)有限公司 无人车实时碰撞检测方法及相关装置
CN114537438A (zh) * 2022-03-07 2022-05-27 京东鲲鹏(江苏)科技有限公司 速度控制方法、装置、控制设备及存储介质
CN114802251A (zh) * 2022-05-11 2022-07-29 阿波罗智能技术(北京)有限公司 自动驾驶车辆的控制方法、装置、电子设备以及存储介质
CN114882706A (zh) * 2022-05-31 2022-08-09 华中科技大学 一种基于非结构性路面的周向防碰撞预警方法及系统
CN115092130A (zh) * 2022-06-16 2022-09-23 阿波罗智能技术(北京)有限公司 车辆碰撞预测方法、装置、电子设备、介质和车辆
CN115620554A (zh) * 2022-07-27 2023-01-17 潍柴智能科技有限公司 基于路测设备的交叉路口自动驾驶车辆行驶方法及系统
CN115092136A (zh) * 2022-07-27 2022-09-23 广州小鹏自动驾驶科技有限公司 车速规划方法、装置、车辆及存储介质
CN115092136B (zh) * 2022-07-27 2023-09-12 广州小鹏自动驾驶科技有限公司 车速规划方法、装置、车辆及存储介质
CN115027464A (zh) * 2022-07-29 2022-09-09 西安电子科技大学芜湖研究院 一种基于方向包围盒的自动驾驶碰撞检测方法
CN115407774A (zh) * 2022-08-25 2022-11-29 魔门塔(苏州)科技有限公司 行驶路径规划方法及装置、车辆、存储介质、终端
CN115214724B (zh) * 2022-09-20 2022-12-09 毫末智行科技有限公司 一种轨迹预测的方法、装置、电子设备及存储介质
CN115214724A (zh) * 2022-09-20 2022-10-21 毫末智行科技有限公司 一种轨迹预测的方法、装置、电子设备及存储介质
CN115534943A (zh) * 2022-10-10 2022-12-30 中国第一汽车股份有限公司 车辆与待监测车辆碰撞预测的方法和装置
CN115355916B (zh) * 2022-10-24 2023-03-03 北京智行者科技股份有限公司 移动工具的轨迹规划方法、设备和计算机可读存储介质
CN115355916A (zh) * 2022-10-24 2022-11-18 北京智行者科技股份有限公司 移动工具的轨迹规划方法、设备和计算机可读存储介质
CN115907245A (zh) * 2022-10-24 2023-04-04 劢微机器人科技(深圳)有限公司 用于agv调度的碰撞检测方法、装置、设备及存储介质
CN116612660A (zh) * 2023-04-10 2023-08-18 山东高速建设管理集团有限公司 一种高速公路改扩建施工区车辆冲突预警方法及设备
CN116612660B (zh) * 2023-04-10 2024-05-28 山东高速建设管理集团有限公司 一种高速公路改扩建施工区车辆冲突预警方法及设备
CN116279500A (zh) * 2023-05-24 2023-06-23 深圳联友科技有限公司 一种车辆碰撞识别方法
CN116279500B (zh) * 2023-05-24 2023-09-19 深圳联友科技有限公司 一种车辆碰撞识别方法
CN116719318A (zh) * 2023-06-05 2023-09-08 库卡机器人(广东)有限公司 碰撞检测方法及装置、自主移动装置、存储介质
CN116572994B (zh) * 2023-07-10 2023-09-22 之江实验室 一种车辆速度规划方法、装置及计算机可读介质
CN116572994A (zh) * 2023-07-10 2023-08-11 之江实验室 一种车辆速度规划方法、装置及计算机可读介质
CN117148848B (zh) * 2023-10-27 2024-01-26 上海伯镭智能科技有限公司 一种无人驾驶车辆智能避障方法及系统
CN117148848A (zh) * 2023-10-27 2023-12-01 上海伯镭智能科技有限公司 一种无人驾驶车辆智能避障方法及系统
CN118046926A (zh) * 2024-04-02 2024-05-17 重庆大学 一种用于自动驾驶系统的车辆行驶轨迹预测方法

Also Published As

Publication number Publication date
EP4082848A4 (en) 2023-07-19
CN113119964B (zh) 2022-08-02
EP4082848C0 (en) 2024-05-01
CN113119964A (zh) 2021-07-16
EP4082848B1 (en) 2024-05-01
EP4082848A1 (en) 2022-11-02

Similar Documents

Publication Publication Date Title
WO2021135728A1 (zh) 一种自动驾驶车辆碰撞预测判断方法及装置
JP7444874B2 (ja) 方法、非一時的コンピュータ可読媒体、および、車両
US11465619B2 (en) Vehicle collision avoidance based on perturbed object trajectories
US11875681B2 (en) Drive envelope determination
JP7295012B2 (ja) 車両制御システム、および、車両制御方法
JP6831420B2 (ja) 自動運転車の軌跡候補を評価するための方法
EP3911548B1 (en) Vehicle control
CN115571130B (zh) 一种基于车道参考线的车辆变道辅助方法及系统
CN105774806B (zh) 车辆行驶控制装置
US11299176B2 (en) Vehicle control device
WO2022053026A1 (zh) 一种自动驾驶会车场景处理方法及其装置、车辆、存储介质
CN112740299A (zh) 用于车辆的碰撞预测和规避
CN114761301B (zh) 车辆控制方法及车辆控制装置
CN108693878A (zh) 前进路线设定装置以及前进路线设定方法
CN111332286A (zh) 车辆行驶控制方法及利用该方法的自动驾驶车辆
JP7735264B2 (ja) 障害物回避アクション
JP7552918B2 (ja) 車両制御方法及び車両制御装置
CN115056787A (zh) 自动驾驶的速度决策规划系统、方法及自动驾驶车辆
CN114516329A (zh) 车辆自适应巡航控制系统、方法和计算机可读介质
JP7356892B2 (ja) 車両の走行環境推定方法、及び、走行環境推定システム
CN115376333A (zh) 一种交叉口车辆调度方法、装置及电子设备
JP2023115440A (ja) 車両制御装置、自律分散型交通管制システムおよび車両制御方法
JP7214076B2 (ja) 車両運転支援システム
US20240083419A1 (en) Processing method, processing system and storage medium for storing processing program
WO2025057361A1 (ja) 運転制御方法及び運転制御装置

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: 20909407

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020909407

Country of ref document: EP

Effective date: 20220801