WO2023010854A1 - 路径跟踪方法、装置、车辆及存储介质 - Google Patents

路径跟踪方法、装置、车辆及存储介质 Download PDF

Info

Publication number
WO2023010854A1
WO2023010854A1 PCT/CN2022/080995 CN2022080995W WO2023010854A1 WO 2023010854 A1 WO2023010854 A1 WO 2023010854A1 CN 2022080995 W CN2022080995 W CN 2022080995W WO 2023010854 A1 WO2023010854 A1 WO 2023010854A1
Authority
WO
WIPO (PCT)
Prior art keywords
path
vehicle
target
obstacle
acceleration
Prior art date
Application number
PCT/CN2022/080995
Other languages
English (en)
French (fr)
Inventor
罗文�
李广南
吴祖亮
冼伯明
伍家胜
林驿
覃海勇
覃安之
Original Assignee
东风柳州汽车有限公司
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 东风柳州汽车有限公司 filed Critical 东风柳州汽车有限公司
Publication of WO2023010854A1 publication Critical patent/WO2023010854A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0217Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with energy consumption, time reduction or distance reduction criteria
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0223Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means

Definitions

  • the present application relates to the technical field of path planning, and in particular to a path tracking method, device, vehicle and storage medium.
  • the main purpose of the present application is to provide a path tracking method, device, vehicle and storage medium, aiming at solving the technical problem of how to improve the accuracy of automatic driving path planning and the tracking accuracy of the planned path in the prior art.
  • the present application provides a path tracking method, the method includes the following steps:
  • the vehicle tracks the target planning path based on longitudinal and lateral control.
  • the step of generating a local planned path includes:
  • the step of judging the type of the obstacle further includes:
  • the obstacle is a dynamic obstacle, determine the geometric size and speed information of the obstacle;
  • a local optimal path is generated according to the local planning path.
  • the step of generating candidate paths includes:
  • the first arc length is the arc length on the reference line where the closest point of the vehicle to the reference line is located, and the second arc length is the corresponding to the end of the candidate path on the reference line the arc length;
  • a candidate path is generated according to the lateral offset.
  • the step of controlling the vehicle to track the target planning path based on longitudinal and lateral control includes:
  • the vehicle is controlled to track the target planned path according to the steering angle and the target speed.
  • the step of controlling the vehicle to track the target planning path according to the steering angle and the target speed it may further include:
  • a target speed is determined based on the tracking speed.
  • the step of determining the target speed according to the tracking speed includes:
  • a target speed is determined according to the current speed, the target acceleration, and the target jerk.
  • a path tracking device which includes:
  • the acquisition module is configured to acquire the global planning path
  • a generation module configured to generate a local planning path when an obstacle is detected ahead
  • the generation module is further configured to generate a target planning path according to the global planning path and the local planning path;
  • the tracking module is configured to control the vehicle to track the target planning path based on longitudinal and lateral control.
  • the present application also proposes a vehicle, the vehicle includes: a memory, a processor, and a path tracking program stored in the memory and operable on the processor, the path tracking program Configured to implement the steps of the path tracing method as described above.
  • the present application also proposes a storage medium, on which a path tracing program is stored, and when the path tracing program is executed by a processor, the steps of the path tracing method as described above are implemented.
  • the present application obtains the global planning path; when an obstacle is detected ahead, generates a local planning path; generates a target planning path according to the global planning path and the local planning path; and tracks the target planning based on longitudinal and lateral control of the vehicle path.
  • the global planning path of the vehicle and the local planning path when the vehicle encounters an obstacle are obtained to determine the target planning path, thereby improving the accuracy of path planning, and controlling the vehicle tracking target planning path based on the longitudinal and lateral directions of the vehicle, thereby improving The accuracy of tracking the planned path is improved.
  • FIG. 1 is a schematic structural diagram of a path tracking device in a hardware operating environment involved in the solution of the embodiment of the present application;
  • FIG. 2 is a schematic flow chart of the first embodiment of the path tracking method of the present application.
  • FIG. 3 is a structural block diagram of the first embodiment of the path tracking device of the present application.
  • FIG. 1 is a schematic structural diagram of a path tracking device in a hardware operating environment involved in an embodiment of the present application.
  • the path tracing device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005.
  • the communication bus 1002 is configured to realize connection and communication between these components.
  • the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
  • the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface).
  • Wi-Fi Wireless-Fidelity
  • the memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory.
  • RAM Random Access Memory
  • NVM Non-Volatile Memory
  • the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .
  • FIG. 1 does not constitute a limitation to the vehicle, and may include more or less components than those shown in the illustration, or combine some components, or arrange different components.
  • the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a path tracking program.
  • the network interface 1004 is mainly configured to communicate data with the network server;
  • the user interface 1003 is mainly configured to perform data interaction with the user;
  • the processor 1001 and the memory 1005 in the vehicle of the present application can be set in In the path tracking device, the vehicle calls the path tracking program stored in the memory 1005 through the processor 1001, and executes the path tracking method provided in the embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a first embodiment of a path tracing method of the present application.
  • the path tracking method includes the following steps:
  • Step S10 Acquire the global planning path.
  • the executive body of this embodiment is an unmanned vehicle, and the vehicle is equipped with a camera for capturing the front and surrounding environment of the vehicle, a Global Positioning System (Global Positioning System, GPS) for real-time positioning, and receiving and processing Domain controller for vehicle camera and GPS information.
  • GPS Global Positioning System
  • the global positioning system can also be replaced by BeiDou Navigation Satellite System (BDS), which is not limited in this embodiment.
  • BDS BeiDou Navigation Satellite System
  • a x , b x , c x , d x , a y , by y , c y , d y are fitting parameters
  • s is the arc length of each line segment
  • s i is the ith waypoint
  • x 0 and y 0 are the x and y coordinates of the global path point in the earth Cartesian coordinate system
  • D(x) is the global planning path.
  • Step S20 When an obstacle is detected ahead, generate a local planning path.
  • the unmanned vehicle needs to plan a local path according to the obstacles on the current road when driving along the global planning path.
  • the step S20 includes: when an obstacle is detected ahead, generating a candidate path; judging the type of the obstacle; when the obstacle is a static obstacle, determining the radius and influence of the obstacle margin; determine the resultant force acting on the vehicle according to the radius and the influence margin; generate a local planned path according to the resultant force and the candidate path.
  • obstacles on the actual road generally include dynamic obstacles and static obstacles, dynamic obstacles such as pedestrians and moving vehicles, etc., and static obstacles include guardrails and the like.
  • the vehicle detects an obstacle in front of the vehicle through lidar, millimeter-wave radar or camera, based on the above-mentioned global planning path, when the vehicle drives to an obstacle in front, it needs to generate a local path (that is, a candidate path) that avoids the obstacle.
  • the step of generating candidate paths includes: obtaining a first arc length and a second arc length, wherein the first arc length is the arc length on the reference line where the closest point of the vehicle to the reference line is located, and the second arc length is The length is an arc length corresponding to the end of the candidate path on the baseline; a lateral offset is generated according to the first arc length, the second arc length, and preset boundary conditions; and a candidate path is generated according to the lateral offset.
  • s start is the arc length on the baseline where the closest point of the vehicle to the baseline is located, that is, the first arc length
  • s end is the arc length corresponding to the end of the candidate path on the baseline, that is, the second arc length; through boundary conditions
  • the coefficients a, b, and c are obtained, with boundary conditions:
  • a set of finite parameter values of the lateral offset d end can be designed to obtain a set of multiple candidate paths with different coefficients a, b, and c.
  • the unmanned vehicle detects that the position of the obstacle in front of it does not change through the sensor, it can be determined that the type of obstacle is a static obstacle at this time.
  • the artificial potential field method is used. (Artificial Potential Field, APF)-the path planning method of risk theory. Obtain the radius of the static obstacle according to the monitoring data of the sensor, take the center point of the static obstacle as the origin, the center point and the farthest point from the center point are the radius of the static obstacle, and calculate the influence of the static obstacle according to the monitoring data Margin, so that the influence radius is obtained according to the radius and the influence margin:
  • ⁇ 0 is the influence radius
  • r is the radius of the obstacle
  • d0 is the influence margin
  • an unmanned vehicle is represented by a circumscribed circle with a radius r a .
  • d is the distance between the center point of the unmanned vehicle and the center point of the static obstacle.
  • U att (q) is the gravitational field value
  • is the gravitational gain constant
  • is the distance from the obstacle.
  • U rep (q) is the numerical value of the repulsive force field
  • is the constant of the repulsive force gain
  • the resultant force acting on the driverless car can be obtained by adding the gravitational and repulsive forces:
  • the resultant force is calculated by initializing the parameters, the position of the next step is calculated by the resultant force, the resultant force is recalculated according to the parameters of the next step, the cycle is repeated, and the local path planning for static obstacles is obtained by combining the candidate paths.
  • the step of judging the type of the obstacle also includes: when the obstacle is a dynamic obstacle, determining the geometric size and speed information of the obstacle; according to the geometric size and the Determine the distribution range of the risk field based on velocity information; determine the risk distribution model according to the source coordinates of the risk field and the distribution range of the risk field; determine the dynamic field according to the risk distribution model and the position of the obstacle; determine the dynamic field according to the risk distribution model and The candidate path determines a target path; and a local planning path is generated according to the dynamic field and the target path.
  • the risk distribution model is generated through the collision risk assessment RISK local path planning algorithm:
  • (u x , u y ) represent the source coordinates of the risk field in the geodetic coordinate system, that is, the coordinates of obstacles, and ⁇ xg , ⁇ yg represent the distribution factors of obstacles along the x-direction and y-direction of the geodetic coordinate system, respectively.
  • the geometric size and speed information of the obstacle are calculated through the data collected by the sensor, and the speed information includes the speed and acceleration of the obstacle. And the geometric size and speed information determine the risk field distribution range of obstacles:
  • Z represents the geometric size of the obstacle
  • N and ⁇ 0 represent undetermined constants
  • ⁇ t is the sampling interval
  • a x (-k ⁇ t) represents the acceleration of the obstacle
  • g represents the time penalty factor
  • V x and V y represent the obstacles respectively Velocity components along the x g , y g directions.
  • the distribution factors of obstacles along the x-direction and y-direction of the geodetic coordinate system are first calculated by formula ten, the risk distribution model is obtained according to the distribution factors, and the dynamic field is determined according to the risk distribution model and the position of the obstacle:
  • (x 0 , y 0 ) is the position of the obstacle
  • the dynamic field E s (x, y) is generated at (x, y) around the position of the obstacle
  • is the clip between r and the velocity v of the obstacle horn.
  • a candidate path with the lowest risk is determined, that is, the target path.
  • a local planning path is generated according to the risk distribution model and the target path.
  • an unmanned vehicle when on the road, it may detect static obstacles and dynamic obstacles at the same time, and the unmanned vehicle will generate the final local optimal path based on the local planning path generated for the two obstacles.
  • Step S30 Generate a target planned path according to the global planned path and the local planned path.
  • Step S40 Control the vehicle to track the target planned path based on longitudinal and lateral control.
  • step S40 includes: obtaining the orientation deviation and the lateral deviation of the vehicle; determining the fusion deviation according to the weight coefficient, the orientation deviation and the lateral deviation; establishing a sliding mode function according to the fusion deviation; determining according to the sliding mode function The steering angle of the steering wheels of the vehicle; controlling the vehicle to track the target planning path according to the steering angle and the target speed.
  • c 1 is a constant, e L , are the fusion bias and its first and second derivatives, respectively.
  • u * is the output required front wheel angle, that is, the steering angle of the steering wheel, and it is fed back to the unmanned vehicle.
  • the unmanned vehicle adjusts the angle by turning the steering wheel, so as to realize Path tracking.
  • the step of controlling the vehicle to track the target planning path according to the steering angle and target speed it also includes: acquiring the target acceleration of the vehicle and the vehicle Determine the longitudinal acceleration according to the target acceleration and the desired acceleration; determine the discrete state equation according to the longitudinal acceleration; determine the tracking speed of the vehicle according to the discrete state equation; determine the target speed according to the tracking speed .
  • the target acceleration of the unmanned vehicle is calculated through the vehicle positioning data, that is, the current acceleration of the unmanned vehicle, and the longitudinal direction and speed are obtained according to the expected acceleration and target acceleration.
  • the expected acceleration refers to the vehicle tracking target planning Optimal acceleration while on route.
  • the formula for calculating longitudinal acceleration is as follows:
  • a k and B k are the state matrix and the control input matrix respectively, and the matrices are expressed as, k is the current sampling time, k+1 is the next sampling time, and T s is the sampling period.
  • the system output is the tracking speed of the unmanned vehicle:
  • the control goal of unmanned vehicles is to ensure the speed tracking accuracy under the premise of ensuring that the unmanned vehicles do not experience excessive acceleration and jerk rate. Therefore, according to the The step of determining the target speed according to the tracking speed includes: determining the predicted acceleration according to the current speed of the vehicle and the tracking speed; determining the predicted jerk rate according to the predicted acceleration; constraining the predicted acceleration to obtain the target acceleration and the target acceleration change rate; determine the target speed according to the current speed, the target acceleration and the target jerk rate.
  • the performance evaluation function is defined as:
  • t-1 is the last sampling time
  • N p is the prediction step size
  • N c is the control step size
  • k) is the control output prediction value
  • k) is Control output variable reference value
  • the weight system matrix, R is the weight system matrix of the system control increment.
  • acceleration rate constraint is:
  • u min and u max are the longitudinal acceleration thresholds
  • ⁇ u min and ⁇ u max are the longitudinal acceleration change thresholds
  • u(k+i) and ⁇ u(k+i) are the control input and control input increment at k+i time, respectively .
  • the system completes the solution to the optimization problem in each cycle, and obtains a series of optimal solution control input increments ⁇ U t in each cycle, and takes the first control increment as the actual output variation of the system, and adds it to the system .
  • the system re-predicts a series of control increments at the next moment according to the system state, and continuously optimizes online rolling until the control process is completed.
  • m is the mass of the vehicle
  • a thre is the resistance demand acceleration
  • F roll is the rolling resistance
  • F aero is the air resistance
  • F grade is the gradient resistance.
  • the rolling resistance is:
  • Cr is the coefficient of rolling resistance.
  • the air resistance is:
  • C w is the air resistance coefficient
  • ⁇ ⁇ is the air density, which can be 1.29kg ⁇ m -3 for normal dry air
  • S is the windward area of the vehicle.
  • ⁇ thdes is the desired throttle opening
  • P bdes is the desired brake master cylinder pressure
  • the control amount is used as the actuator control input to control the intelligent vehicle to track the target speed.
  • the global planning path is obtained; when an obstacle is detected ahead, a local planning path is generated; a target planning path is generated according to the global planning path and the local planning path; the vehicle is tracked based on longitudinal and lateral control of the target Plan your path.
  • the embodiment of the present application also proposes a storage medium, on which a path tracing program is stored, and when the path tracing program is executed by a processor, the steps of the path tracing method as described above are implemented.
  • the storage medium adopts all the technical solutions of all the above-mentioned embodiments, it at least has all the functions brought by the technical solutions of the above-mentioned embodiments, which will not be repeated here.
  • FIG. 3 is a structural block diagram of the first embodiment of the path tracking device of the present application.
  • the path tracking device proposed in the embodiment of the present application includes:
  • the acquiring module 10 is configured to acquire the global planning path.
  • the generation module 20 is configured to generate a local planned path when an obstacle is detected ahead.
  • the generating module 20 is further configured to generate a target planned path according to the global planned path and the local planned path.
  • the tracking module 30 is configured to track the target planning path based on longitudinal and lateral control of the vehicle.
  • the global planning path is obtained; when an obstacle is detected ahead, a local planning path is generated; a target planning path is generated according to the global planning path and the local planning path; the vehicle is tracked based on longitudinal and lateral control of the target Plan your path.
  • the generation module 20 is further configured to generate a candidate path when an obstacle is detected ahead;
  • the generation module 20 is further configured to determine the geometric size and speed information of the obstacle when the obstacle is a dynamic obstacle;
  • a local optimal path is generated according to the local planning path.
  • the generation module 20 is further configured to obtain the first arc length and the second arc length, wherein the first arc length is the arc length on the reference line where the closest point of the vehicle to the reference line is located , the second arc length is the arc length corresponding to the end of the candidate path on the baseline;
  • a candidate path is generated according to the lateral offset.
  • the tracking module 30 is further configured to obtain the orientation deviation and the lateral deviation of the vehicle;
  • the vehicle is controlled to track the target planned path according to the steering angle and the target speed.
  • the tracking module 30 is further configured to acquire the target acceleration of the vehicle
  • a target speed is determined based on the tracking speed.
  • the tracking module 30 is further configured to determine a predicted acceleration according to the current speed of the vehicle and the tracking speed;
  • a target speed is determined according to the current speed, the target acceleration, and the target jerk.
  • the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation.
  • the technical solution of the present application can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the computer software product is stored in a storage medium (such as a read-only memory (Read Only Memory) , ROM)/RAM, magnetic disk, optical disk), including several instructions to make a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the methods described in various embodiments of the present application.
  • a storage medium such as a read-only memory (Read Only Memory) , ROM)/RAM, magnetic disk, optical disk
  • a terminal device which can be a mobile phone, computer, server, or network device, etc.

Abstract

本申请属于路径规划技术领域,公开了一种路径跟踪方法、装置、车辆及存储介质。该方法包括:获取全局规划路径;当检测到前方存在障碍物时,生成局部规划路径;根据所述全局规划路径以及所述局部规划路径生成目标规划路径;基于纵向以及横向控制车辆跟踪所述目标规划路径。通过上述方式,获取车辆的全局规划路径以及车辆遇见障碍物时的局部规划路径确定目标规划路径,从而提升路径规划时的准确度,并基于车辆的纵向以及横向控制车辆跟踪目标规划路径,从而提升了跟踪规划路径的精度。

Description

路径跟踪方法、装置、车辆及存储介质
相关申请
本申请要求于2021年8月4号申请的、申请号为202110893744.0的中国专利申请的优先权,其全部内容通过引用结合于此。
技术领域
本申请涉及路径规划技术领域,尤其涉及一种路径跟踪方法、装置、车辆及存储介质。
背景技术
在目前的路径规划中,通常需要接收云服务平台基于定位信息及目的地信息,而在信号发送、接收不好的地方,车辆会出现信号弱或者信号不好的现象,导致车辆无法接收实时的局部路径信息。当车辆行驶速度较快或者环境变化较快时,接收的局部路径规划信息跟现实情况会出现不一致现象。
上述内容仅用于辅助理解本申请的技术方案,并不代表承认上述内容是现有技术。
申请内容
本申请的主要目的在于提供一种路径跟踪方法、装置、车辆及存储介质,旨在解决现有技术如何提升自动驾驶路径规划的准确度以及规划路径的跟踪精度的技术问题。
为实现上述目的,本申请提供了一种路径跟踪方法,所述方法包括以下步骤:
获取全局规划路径;
当检测到前方存在障碍物时,生成局部规划路径;
根据所述全局规划路径以及所述局部规划路径生成目标规划路径;
基于纵向以及横向控制车辆跟踪所述目标规划路径。
在一实施方式中,所述当检测到前方存在障碍物时,生成局部规划路径的步骤,包括:
当检测到前方存在障碍物时,生成候选路径;
判断所述障碍物的类型;
当所述障碍物为静态障碍物时,确定所述障碍物的半径以及影响边距;
根据所述半径以及所述影响边距确定作用在所述车辆的合力;
根据所述合力以及所述候选路径,生成局部规划路径。
在一实施方式中,所述判断所述障碍物的类型的步骤之后,还包括:
当所述障碍物为动态障碍物时,确定所述障碍物的几何尺寸以及速度信息;
根据所述几何尺寸以及所述速度信息确定风险场分布范围;
根据风险场场源坐标以及所述风险场分布范围确定风险分布模型;
根据所述风险分布模型以及所述障碍物的位置确定动态场;
根据所述动态场以及所述候选路径确定目标路径;
根据所述风险分布模型以及所述目标路径生成局部规划路径;
根据所述局部规划路径生成局部最优路径。
在一实施方式中,所述生成候选路径的步骤,包括:
获取第一弧长以及第二弧长,其中,所述第一弧长为车辆距离基准线的最近点所在基准线上的弧长,所述第二弧长为基准线上的候选路径末端对应的弧长;
根据所述第一弧长、所述第二弧长以及预设边界条件生成横向偏移量;
根据所述横向偏移量生成候选路径。
在一实施方式中,所述基于纵向以及横向控制车辆跟踪所述目标规划路径的步骤,包括:
获取车辆的方位偏差以及横向偏差;
根据权重系数、所述方位偏差以及所述横向偏差确定融合偏差;
根据所述融合偏差建立滑模函数;
根据所述滑模函数确定所述车辆的转向轮的转向角度;
根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径。
在一实施方式中,所述根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径的步骤之前,还包括:
获取所述车辆的目标加速度;
获取所述车辆的期望加速度;
根据所述目标加速度以及所述期望加速度确定纵向加速度;
根据所述纵向加速度确定离散状态方程;
根据所述离散状态方程确定所述车辆的跟踪速度;
根据所述跟踪速度确定目标速度。
在一实施方式中,所述根据所述跟踪速度确定目标速度的步骤,包括:
根据所述车辆的当前速度以及所述跟踪速度确定预测加速度;
根据所述预测加速度确定预测加速度变化率;
约束所述预测加速度得到目标加速度以及目标加速度变化率;
根据所述当前速度、所述目标加速度以及所述目标加速度变化率确定目标速度。
此外,为实现上述目的,本申请还提出一种路径跟踪装置,所述路径跟踪装置包括:
获取模块,被配置为获取全局规划路径;
生成模块,被配置为当检测到前方存在障碍物时,生成局部规划路径;
所述生成模块,还被配置为根据所述全局规划路径以及所述局部规划路径生成目标规划路径;
跟踪模块,被配置为基于纵向以及横向控制车辆跟踪所述目标规划路径。
此外,为实现上述目的,本申请还提出一种车辆,所述车辆包括:存储器、处理器及存储在所述存储器上并可在所述处理器上运行的路径跟踪程序,所述路径跟踪程序配置为实现如上文所述的路径跟踪方法的步骤。
此外,为实现上述目的,本申请还提出一种存储介质,所述存储介质上存储有路径跟踪程序,所述路径跟踪程序被处理器执行时实现如上文所述的路径跟踪方法的步骤。
本申请通过获取全局规划路径;当检测到前方存在障碍物时,生成局部规划路径;根据所述全局规划路径以及所述局部规划路径生成目标规划路径;基于纵向以及横向控制车辆跟踪所述目标规划路径。通过上述方式,获取车辆的全局规划路径以及车辆遇见障碍物时的局部规划路径确定目标规划路径,从而提升路径规划时的准确度,并基于车辆的纵向以及横向控制车辆跟 踪目标规划路径,从而提升了跟踪规划路径的精度。
附图说明
图1是本申请实施例方案涉及的硬件运行环境的路径跟踪设备的结构示意图;
图2为本申请路径跟踪方法第一实施例的流程示意图;
图3为本申请路径跟踪装置第一实施例的结构框图。
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
具体实施方式
应当理解,此处所描述的具体实施例仅用以解释本申请,并不用于限定本申请。
参照图1,图1为本申请实施例方案涉及的硬件运行环境的路径跟踪设备的结构示意图。
如图1所示,该路径跟踪设备可以包括:处理器1001,例如中央处理器(Central Processing Unit,CPU),通信总线1002、用户接口1003,网络接口1004,存储器1005。其中,通信总线1002被配置为实现这些组件之间的连接通信。用户接口1003可以包括显示屏(Display)、输入单元比如键盘(Keyboard),可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口(如无线保真(Wireless-Fidelity,Wi-Fi)接口)。存储器1005可以是高速的随机存取存储器(Random Access Memory,RAM),也可以是稳定的非易失性存储器(Non-Volatile Memory,NVM),例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
本领域技术人员可以理解,图1中示出的结构并不构成对车辆的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置。
如图1所示,作为一种存储介质的存储器1005中可以包括操作系统、网络通信模块、用户接口模块以及路径跟踪程序。
在图1所示的车辆中,网络接口1004主要被配置为与网络服务器进行数据通信;用户接口1003主要被配置为与用户进行数据交互;本申请车辆中的处理器1001、存储器1005可以设置在路径跟踪设备中,所述车辆通过处理器1001调用存储器1005中存储的路径跟踪程序,并执行本申请实施例提供的路径跟踪方法。
本申请实施例提供了一种路径跟踪方法,参照图2,图2为本申请一种路径跟踪方法第一实施例的流程示意图。
本实施例中,所述路径跟踪方法包括以下步骤:
步骤S10:获取全局规划路径。
需要说明的是,本实施例的执行主体为无人驾驶车辆,车辆上安装有用于摄取车辆前方及周边环境的摄像头、用于实时定位的全球定位系统(Global Positioning System,GPS)以及接收和处理车载摄像头及GPS信息的域控制器。全球定位系统还可以更换为北斗卫星导航系统(BeiDou Navigation Satellite System,BDS),本实施例不加以限制。
可以理解的是,基于当前车辆的定位数据以及目的地的定位数据,从而计算得到全局规划路径:
D(x)=(x 0(s),y 0(s))    公式一;
Figure PCTCN2022080995-appb-000001
其中,a x、b x、c x、d x、a y、b y、c y、d y为拟合参数,s为每个线段的弧长,s i为第i个路点,x 0和y 0为全局路径点在大地笛卡尔坐标系的x、y坐标,D(x)为全局规划路径。
步骤S20:当检测到前方存在障碍物时,生成局部规划路径。
需要说明的是,由于路况在实时变化,因此无人驾驶车辆在沿着全局规划路径行驶时需要根据当前道路的障碍物情况规划局部路径。
进一步地,所述步骤S20包括:当检测到前方存在障碍物时,生成候选路径;判断所述障碍物的类型;当所述障碍物为静态障碍物时,确定所述障碍物的半径以及影响边距;根据所述半径以及所述影响边距确定作用在所述车辆的合力;根据所述合力以及所述候选路径,生成局部规划路径。
可以理解的是,实际道路上的障碍物通常包括动态障碍物以及静态障碍物,动态障碍物例如行人、行驶的车辆等,静态障碍物包括护栏等。当车辆通过激光雷达、毫米波雷达或摄像头检测到车辆前方出现障碍物时,基于上述全局规划路径,当车辆行驶至前方有障碍物时,需要生成规避障碍物的局部路径(即候选路径)。
进一步地,生成候选路径的步骤包括:获取第一弧长以及第二弧长,其中,所述第一弧长为车辆距离基准线的最近点所在基准线上的弧长,所述第二弧长为基准线上的候选路径末端对应的弧长;根据所述第一弧长、所述第二弧长以及预设边界条件生成横向偏移量;根据所述横向偏移量生成候选路径。
需要说明的是,生成候选路径的公式如下:
Figure PCTCN2022080995-appb-000002
其中,s start为车辆距离基准线的最近点所在基准线上的弧长,即第一弧长;s end为基准线上的候选路径末端对应的弧长,即第二弧长;通过边界条件得出系数a、b和c,有边界条件:
Figure PCTCN2022080995-appb-000003
因此可以设计出一组有限数量的横向偏移量d end参数值,从而求解得到一组不同系数a、b和c的多条候选路径。
可以理解的是,当无人驾驶车辆通过传感器检测到前方的障碍物位置不变时,此时则可以判定障碍物的类型为静态障碍物,针对静态障碍物规划路径时,采用人工势场法(Artificial Potential Field,APF)-风险理论的路径规划方法。根据传感器的监测数据得到静态障碍物的半径,以静态障碍物的中心点为原点,中心点与离中心点最远点为此静态障碍物的半径,并根据监测数据计算得到静态障碍物的影响边距,从而根据半径以及影响边距得到影响半径:
ρ 0=r+d 0    公式五;
其中,ρ 0为影响半径,r为障碍物的半径,d 0为影响边距。
需要说明的是,无人驾驶车辆采用外接圆表示,其半径为r a。当d<r a0时,无人驾驶车辆开始局部路径规划,d为无人驾驶车辆中心点与静态障碍物中心点之间的距离。
在具体实现中,首先计算引力场:
Figure PCTCN2022080995-appb-000004
其中,U att(q)为引力场数值,ε为引力增益常数,ρ为与障碍物的距离。
进一步地,斥力场计算公式如下:
Figure PCTCN2022080995-appb-000005
其中,U rep(q)为斥力场数值,η为斥力增益常数。
引力和斥力相加可得作用在无人驾驶车上的合力:
U(q)=U att(q)+U rep(q)    公式八;
其中,U(q)为合力。
最后,通过初始化参数计算合力,通过合力计算下一步的位置,根据下一步的参数重新计算合力,循环往复,结合候选路径,得到针对静态障碍物的局部路径规划。
进一步地,所述判断所述障碍物的类型的步骤之后,还包括:当所述障碍物为动态障碍物时,确定所述障碍物的几何尺寸以及速度信息;根据所述几何尺寸以及所述速度信息确定风险场分布范围;根据风险场场源坐标以及所述风险场分布范围确定风险分布模型;根据所述风险分布模型以及所述障碍物的位置确定动态场;根据所述风险分布模型以及所述候选路径确定目标路径;根据所述动态场以及所述目标路径生成局部规划路径。
需要说明的是,当无人驾驶车联的传感器检测到前方的障碍物的类型为动态障碍物时,通过碰撞风险评估RISK的局部路径规划算法,生成风险分布模型:
Figure PCTCN2022080995-appb-000006
其中,(u x,u y)表示大地坐标系下风险场场源坐标,即障碍物的坐标, σ xg、σ yg分别表示障碍物分别沿大地坐标系x方向和y方向的分布因子。
需要说明的是,在无人驾驶车辆在检测到障碍物为动态障碍物时,通过传感器采集的数据计算障碍物几何尺寸以及速度信息,速度信息中包括障碍物的速度以及加速度等。并几何尺寸以及速度信息确定障碍物的风险场分布范围:
Figure PCTCN2022080995-appb-000007
其中,Z表示障碍物的几何尺寸,N和σ 0表示待定常数,Δt为采样间隔,a x(-kΔt)表示障碍物的加速度;g表示时间惩罚因子,V x、V y表示障碍物分别沿x g、y g方向的速度分量。
可以理解的是,先通过公式十计算得到障碍物分别沿大地坐标系x方向和y方向的分布因子,根据分布因子获得风险分布模型,并根据风险分布模型以及障碍物的位置确定动态场:
Figure PCTCN2022080995-appb-000008
其中,(x 0,y 0)为障碍物的位置,动态场E s(x,y)为在障碍物位置周围(x,y)处产生,θ为r与障碍物速度v之间的夹角。
并根据动态场以及候选路径确定风险最低的一条候选路径,即目标路径。最后根据风险分布模型以及目标路径生成局部规划路径。
需要说明的是,无人驾驶车辆上路时,可能会同时检测到静态障碍物以及动态障碍物,无人驾驶车辆会根据针对两种障碍物生成的局部规划路径生成最终的局部最优路径。
步骤S30:根据所述全局规划路径以及所述局部规划路径生成目标规划路径。
可以理解的是,当无人驾驶车辆前方没有障碍物时,则按照全局规划路径行驶,若出现障碍物则实时生成局部规划路径,从而生成最终的目标规划 路径。
步骤S40:基于纵向以及横向控制车辆跟踪所述目标规划路径。
在具体实现中,为了更精准地控制无人驾驶车辆在目标规划路径上行驶,需要对无人驾驶车辆横向以及纵向进行控制。
进一步地,步骤S40包括:获取车辆的方位偏差以及横向偏差;根据权重系数、所述方位偏差以及所述横向偏差确定融合偏差;根据所述融合偏差建立滑模函数;根据所述滑模函数确定所述车辆的转向轮的转向角度;根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径。
需要说明的是,在对无人驾驶车辆进行横向控制时,首先根据目标规划路径确定当前的道路类型,生成预瞄模型:
Figure PCTCN2022080995-appb-000009
其中,
Figure PCTCN2022080995-appb-000010
为方位偏差,
Figure PCTCN2022080995-appb-000011
表示横向偏差,v x、v y分别表示纵向车速和横向车速,ω r表示横摆角速度,ρ为道路曲率;L s为预瞄距离。
对公式十二的横向偏差以及方位偏差进行无量纲化处理,有:
Figure PCTCN2022080995-appb-000012
其中,
Figure PCTCN2022080995-appb-000013
为无量纲化处理后的横向偏差,
Figure PCTCN2022080995-appb-000014
为无量纲化处理后的方位偏差。对无量纲化处理后的横向偏差和方位偏差分别加入权重系数,从而分配两者的比重:
Figure PCTCN2022080995-appb-000015
其中,η 1、η 2均为大于0的权重系数,且η 12=1,e L即为融合偏差。
选取融合偏差来建立本文控制系统所需的滑模函数s,形式如下:
Figure PCTCN2022080995-appb-000016
其变化率表达式如下:
Figure PCTCN2022080995-appb-000017
上述公式十五、十六中,c 1为常数,e L
Figure PCTCN2022080995-appb-000018
分别为融合偏差及其一阶导数和二阶导数。
需要说明的是,为了克服了抖振现象,有模糊控制器:选择双输入的二维控制器,输入变量为滑模变结构的切换函数s及其变化率
Figure PCTCN2022080995-appb-000019
输出变量为车辆前轮转角δ f
首先进行模糊化,将三个变量的量化等级均设定为7级对模糊空间进行分割,7级对应如下:NB=负大,NM=负中,NS=负小,ZO=零,PS=正小,PM=正中,PB=正大,定义s、
Figure PCTCN2022080995-appb-000020
和δ f的模糊子集的变量分别为:
Figure PCTCN2022080995-appb-000021
设定论域集X、Y、Z,均设为[-1,1],然后将变量s、
Figure PCTCN2022080995-appb-000022
和δ f分别映射到对应论域中,并进行如下离散化处理:
Figure PCTCN2022080995-appb-000023
之后选取变量的隶属度函数,指定合适的模糊控制规则。
采用重心法完成反模糊化操作,有:
Figure PCTCN2022080995-appb-000024
因此,得到模糊控制器反模糊化后u *,即为输出所需的前轮转角,即转向轮的转向角度,并反馈给无人驾驶车辆,无人驾驶车辆通过转动方向盘调整转角,从而实现路径的跟踪。
进一步地,为了更精准的纵向控制无人驾驶车辆,根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径的步骤之前,还包括:获取所述车辆的目标加速度和所述车辆的期望加速度;根据所述目标加速度以及 所述期望加速度确定纵向加速度;根据所述纵向加速度确定离散状态方程;根据所述离散状态方程确定所述车辆的跟踪速度;根据所述跟踪速度确定目标速度。
可以理解的是,首先通过车辆的定位数据计算得到无人驾驶车辆的目标加速度,即无人驾驶车辆的当前加速度,并根据期望加速度以及目标加速度得到纵向及速度,期望加速度是指车辆跟踪目标规划路径时的最佳加速度。纵向加速度计算公式如下:
Figure PCTCN2022080995-appb-000025
其中,
Figure PCTCN2022080995-appb-000026
为纵向加速度,a des为期望加速度,K为系统增益,τ d为时间常数。
考虑速度与加速度之间的关系,速度跟踪控制用连续系统的状态方程表示:
Figure PCTCN2022080995-appb-000027
其中,x=[v a] T为系统纵向状态变量,u=a des为系统控制输入。
将公式二十一通过向前欧拉法进行离散化处理,得系统的离散状态方程:
Figure PCTCN2022080995-appb-000028
其中,A k、B k分别为状态矩阵和控制输入矩阵,矩阵分别表示为,k为当前采样时刻,k+1为下一采样时刻,T s为采样周期。
系统输出量为无人驾驶车辆的跟踪速度:
y(k)=C x(k)    公式二十三;
其中C=[1 0],y(k)为跟踪速度。
进一步地,在无人驾驶车辆纵向控制中,考虑无人驾驶车辆的控制目标是在保证无人驾驶车辆不发生过于剧烈的加速度和加速度变化率的前提下,保证速度跟踪精度,因此,根据所述跟踪速度确定目标速度的步骤,包括:根据所述车辆的当前速度以及所述跟踪速度确定预测加速度;根据所述预测加速度确定预测加速度变化率;约束所述预测加速度得到目标加速度以及目标加速度变化率;根据所述当前速度、所述目标加速度以及所述目标加速度变化率确定目标速度。
首先将性能评价函数定义为:
Figure PCTCN2022080995-appb-000029
其中,t-1为上一采样时刻,N p为预测步长,N c为控制步长,y p(k+i|k)为控制输出预测值,y ref(k+i|k)为控制输出变量参考值,(k+i|k)表示根据k采样时刻的状态信息来预测第k+1时刻的信息,其中i=1,2,..,N p,Q为系统输出量的权重系统矩阵,R为系统控制增量的权重系统矩阵。
在控制器跟踪速度过程中,需要考虑添加主动约束,即控制过程中的预测加速度约束及其变化率约束,保证其约束在合理的范围之内从而得到目标加速度以及目标加速度变化率,最终得到目标速度,其不等式表达如下:
u min≤u(k+i)≤u max,i=0,1,...,N c-1公式二十五;
加速度变化率约束表达形式为:
Δu min≤Δu(k+i)≤Δu max,i=0,1,...,N c-1公式二十六;
其中,u min和u max为纵向加速度阈值,Δu min和Δu max为纵向加速度变化量阈值,u(k+i)和Δu(k+i)分别是k+i时刻控制输入和控制输入增量。
系统在每个周期完成对优化问题的求解,得出每个周期一系列的最优解控制输入增量ΔU t,并将第一个控制增量作为系统的实际输出变化量,加入至系统中。在新的时刻,系统根据系统状态,重新预测下一时刻的一系列控制增量,不断在线滚动优化,直到完成控制过程。
需要说明的是,在最后根据目标速度控制车辆行驶时,需要考虑空气阻力、滚动阻力以及坡度阻力,将车辆阻力需求加速度方程用以下公式表示:
ma thre=F res=F roll+F aero+F grade    公式二十七;
其中,m为车辆质量,a thre为阻力需求加速度,F roll为滚动阻力,F aero为空气阻力,F grade为坡度阻力。
坡度阻力为汽车重力沿坡道的分力,考虑坡度为i的直线道路,i=tanα,因此考虑坡度较小时,坡度阻力:
F grade=mgi    公式二十八;
滚动阻力为:
F roll=mg cosαC r    公式二十九;
其中,C r为滚动阻力系数。
空气阻力为:
Figure PCTCN2022080995-appb-000030
其中,C w为空气阻力系数,ρ α为空气密度,正常的干燥空气可取1.29kg·m -3,S为车辆迎风面积。假设车辆在行驶过程中,以上参数均不发生变化。
在设计切换逻辑过程中,要保证制动和驱动模式不能同时起作用的同时,也要根据现实情况,不能频繁的切换制动和驱动模式。因此,根据期望的加速度定义以下逻辑切换:
Figure PCTCN2022080995-appb-000031
Figure PCTCN2022080995-appb-000032
其中,α thdes为期望节气门开度,P bdes为期望制动主缸压力。
通过比例-积分控制,得到期望节气门开度α thdes和期望制动主缸压力P bdes后,将控制量作为执行器控制输入,从而控制智能车辆跟踪目标速度。
本实施例通过获取全局规划路径;当检测到前方存在障碍物时,生成局部规划路径;根据所述全局规划路径以及所述局部规划路径生成目标规划路径;基于纵向以及横向控制车辆跟踪所述目标规划路径。通过上述方式,获取车辆的全局规划路径以及车辆遇见障碍物时的局部规划路径确定目标规划路径,从而提升路径规划时的准确度,并基于车辆的纵向以及横向控制车辆跟踪目标规划路径,从而提升了跟踪规划路径的精度。
此外,本申请实施例还提出一种存储介质,所述存储介质上存储有路径跟踪程序,所述路径跟踪程序被处理器执行时实现如上文所述的路径跟踪方法的步骤。
由于本存储介质采用了上述所有实施例的全部技术方案,因此至少具有上述实施例的技术方案所带来的所有功能,在此不再一一赘述。
参照图3,图3为本申请路径跟踪装置第一实施例的结构框图。
如图3所示,本申请实施例提出的路径跟踪装置包括:
获取模块10,被配置为获取全局规划路径。
生成模块20,被配置为当检测到前方存在障碍物时,生成局部规划路径。
所述生成模块20,还被配置为根据所述全局规划路径以及所述局部规划路径生成目标规划路径。
跟踪模块30,被配置为基于纵向以及横向控制车辆跟踪所述目标规划路径。
应当理解的是,以上仅为举例说明,对本申请的技术方案并不构成任何限定,在具体应用中,本领域的技术人员可以根据需要进行设置,本申请对此不做限制。
本实施例通过获取全局规划路径;当检测到前方存在障碍物时,生成局部规划路径;根据所述全局规划路径以及所述局部规划路径生成目标规划路径;基于纵向以及横向控制车辆跟踪所述目标规划路径。通过上述方式,获取车辆的全局规划路径以及车辆遇见障碍物时的局部规划路径确定目标规划路径,从而提升路径规划使得准确度,并基于车辆的纵向以及横向控制车辆 跟踪目标规划路径,从而提升了跟踪规划路径的精度。
在一实施例中,所述生成模块20,还被配置为当检测到前方存在障碍物时,生成候选路径;
判断所述障碍物的类型;
当所述障碍物为静态障碍物时,确定所述障碍物的半径以及影响边距;
根据所述半径以及所述影响边距确定作用在所述车辆的合力;
根据所述合力以及所述候选路径,生成局部规划路径。
在一实施例中,所述生成模块20,还被配置为当所述障碍物为动态障碍物时,确定所述障碍物的几何尺寸以及速度信息;
根据所述几何尺寸以及所述速度信息确定风险场分布范围;
根据风险场场源坐标以及所述风险场分布范围确定风险分布模型;
根据所述风险分布模型以及所述障碍物的位置确定动态场;
根据所述动态场以及所述候选路径确定目标路径;
根据所述风险分布模型以及所述目标路径生成局部规划路径;
根据所述局部规划路径生成局部最优路径。
在一实施例中,所述生成模块20,还被配置为获取第一弧长以及第二弧长,其中,所述第一弧长为车辆距离基准线的最近点所在基准线上的弧长,所述第二弧长为基准线上的候选路径末端对应的弧长;
根据所述第一弧长、所述第二弧长以及预设边界条件生成横向偏移量;
根据所述横向偏移量生成候选路径。
在一实施例中,所述跟踪模块30,还被配置为获取车辆的方位偏差以及横向偏差;
根据权重系数、所述方位偏差以及所述横向偏差确定融合偏差;
根据所述融合偏差建立滑模函数;
根据所述滑模函数确定所述车辆的转向轮的转向角度;
根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径。
在一实施例中,所述跟踪模块30,还被配置为获取所述车辆的目标加速度;
获取所述车辆的期望加速度;
根据所述目标加速度以及所述期望加速度确定纵向加速度;
根据所述纵向加速度确定离散状态方程;
根据所述离散状态方程确定所述车辆的跟踪速度;
根据所述跟踪速度确定目标速度。
在一实施例中,所述跟踪模块30,还被配置为根据所述车辆的当前速度以及所述跟踪速度确定预测加速度;
根据所述预测加速度确定预测加速度变化率;
约束所述预测加速度得到目标加速度以及目标加速度变化率;
根据所述当前速度、所述目标加速度以及所述目标加速度变化率确定目标速度。
需要说明的是,以上所描述的工作流程仅仅是示意性的,并不对本申请的保护范围构成限定,在实际应用中,本领域的技术人员可以根据实际的需要选择其中的部分或者全部来实现本实施例方案的目的,此处不做限制。
另外,未在本实施例中详尽描述的技术细节,可参见本申请任意实施例所提供的路径跟踪方法,此处不再赘述。
此外,需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者系统不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者系统所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者系统中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述,不代表实施例的优劣。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述 实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的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(如只读存储器(Read Only Memory,ROM)/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端设备(可以是手机,计算机,服务器,或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的可选实施例,并非因此限制本申请的专利范围,凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本申请的专利保护范围内。

Claims (10)

  1. 一种路径跟踪方法,其中,所述路径跟踪方法包括:
    获取全局规划路径;
    当检测到前方存在障碍物时,生成局部规划路径;
    根据所述全局规划路径以及所述局部规划路径生成目标规划路径;
    基于纵向以及横向控制车辆跟踪所述目标规划路径。
  2. 如权利要求1所述的方法,其中,所述当检测到前方存在障碍物时,生成局部规划路径的步骤,包括:
    当检测到前方存在障碍物时,生成候选路径;
    判断所述障碍物的类型;
    当所述障碍物为静态障碍物时,确定所述障碍物的半径以及影响边距;
    根据所述半径以及所述影响边距确定作用在所述车辆的合力;
    根据所述合力以及所述候选路径,生成局部规划路径。
  3. 如权利要求2所述的方法,其中,所述判断所述障碍物的类型的步骤之后,还包括:
    当所述障碍物为动态障碍物时,确定所述障碍物的几何尺寸以及速度信息;
    根据所述几何尺寸以及所述速度信息确定风险场分布范围;
    根据风险场场源坐标以及所述风险场分布范围确定风险分布模型;
    根据所述风险分布模型以及所述障碍物的位置确定动态场;
    根据所述动态场以及所述候选路径确定目标路径;
    根据所述风险分布模型以及所述目标路径生成局部规划路径;
    根据所述局部规划路径生成局部最优路径。
  4. 如权利要求2所述的方法,其中,所述生成候选路径的步骤,包括:
    获取第一弧长以及第二弧长,其中,所述第一弧长为车辆距离基准线的最近点所在基准线上的弧长,所述第二弧长为基准线上的候选路径末端对应 的弧长;
    根据所述第一弧长、所述第二弧长以及预设边界条件生成横向偏移量;
    根据所述横向偏移量生成候选路径。
  5. 如权利要求1所述的方法,其中,所述基于纵向以及横向控制车辆跟踪所述目标规划路径的步骤,包括:
    获取车辆的方位偏差以及横向偏差;
    根据权重系数、所述方位偏差以及所述横向偏差确定融合偏差;
    根据所述融合偏差建立滑模函数;
    根据所述滑模函数确定所述车辆的转向轮的转向角度;
    根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径。
  6. 如权利要求5所述的方法,其中,所述根据所述转向角度以及目标速度控制所述车辆跟踪所述目标规划路径的步骤之前,还包括:
    获取所述车辆的目标加速度;
    获取所述车辆的期望加速度;
    根据所述目标加速度以及所述期望加速度确定纵向加速度;
    根据所述纵向加速度确定离散状态方程;
    根据所述离散状态方程确定所述车辆的跟踪速度;
    根据所述跟踪速度确定目标速度。
  7. 如权利要求6所述的方法,其中,所述根据所述跟踪速度确定目标速度的步骤,包括:
    根据所述车辆的当前速度以及所述跟踪速度确定预测加速度;
    根据所述预测加速度确定预测加速度变化率;
    约束所述预测加速度得到目标加速度以及目标加速度变化率;
    根据所述当前速度、所述目标加速度以及所述目标加速度变化率确定目标速度。
  8. 一种路径跟踪装置,其中,所述路径跟踪装置包括:
    获取模块,被配置为获取全局规划路径;
    生成模块,被配置为当检测到前方存在障碍物时,生成局部规划路径;
    所述生成模块,还被配置为根据所述全局规划路径以及所述局部规划路径生成目标规划路径;
    跟踪模块,被配置为基于纵向以及横向控制车辆跟踪所述目标规划路径。
  9. 一种车辆,其中,所述车辆包括:存储器、处理器及存储在所述存储器上并可在所述处理器上运行的路径跟踪程序,所述路径跟踪程序配置为实现如权利要求1至7中任一项所述的路径跟踪方法。
  10. 一种存储介质,其中,所述存储介质上存储有路径跟踪程序,所述路径跟踪程序被处理器执行时实现如权利要求1至7任一项所述的路径跟踪方法。
PCT/CN2022/080995 2021-08-04 2022-03-15 路径跟踪方法、装置、车辆及存储介质 WO2023010854A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110893744.0A CN113655789A (zh) 2021-08-04 2021-08-04 路径跟踪方法、装置、车辆及存储介质
CN202110893744.0 2021-08-04

Publications (1)

Publication Number Publication Date
WO2023010854A1 true WO2023010854A1 (zh) 2023-02-09

Family

ID=78478410

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/080995 WO2023010854A1 (zh) 2021-08-04 2022-03-15 路径跟踪方法、装置、车辆及存储介质

Country Status (2)

Country Link
CN (1) CN113655789A (zh)
WO (1) WO2023010854A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116540723A (zh) * 2023-05-30 2023-08-04 南通大学 一种基于人工势场的水下机器人滑模轨迹跟踪控制方法
CN116719229A (zh) * 2023-03-09 2023-09-08 哈尔滨理工大学 基于势函数的动力定位船避碰容错控制方法
CN117724504A (zh) * 2024-02-07 2024-03-19 北京科技大学 一种基于投影面积的无人车路径跟踪控制方法及装置

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113655789A (zh) * 2021-08-04 2021-11-16 东风柳州汽车有限公司 路径跟踪方法、装置、车辆及存储介质
CN114557640B (zh) * 2022-02-21 2023-08-01 广州宝乐软件科技有限公司 清洁机器人的数据处理方法、装置及清洁机器人

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109709945A (zh) * 2017-10-26 2019-05-03 深圳市优必选科技有限公司 一种基于障碍物分类的路径规划方法、装置及机器人
CN109855639A (zh) * 2019-01-15 2019-06-07 天津大学 基于障碍物预测与mpc算法的无人驾驶轨迹规划方法
CN109933945A (zh) * 2019-03-28 2019-06-25 北京智行者科技有限公司 交通环境建模方法
CN110147105A (zh) * 2019-05-27 2019-08-20 安徽江淮汽车集团股份有限公司 无人驾驶车辆的路径控制方法、设备、存储介质及装置
US20200174475A1 (en) * 2018-11-29 2020-06-04 Electronics And Telecommunications Research Institute Autonomous driving method and system
CN112269389A (zh) * 2020-11-18 2021-01-26 武汉理工大学 一种多功能乘务服务智能机器车系统及其控制方法
CN112327836A (zh) * 2020-10-27 2021-02-05 南宁市第一人民医院 基于5g技术的轮椅自动驾驶控制方法及自动驾驶轮椅
CN113655789A (zh) * 2021-08-04 2021-11-16 东风柳州汽车有限公司 路径跟踪方法、装置、车辆及存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103439884B (zh) * 2013-07-19 2015-12-23 大连理工大学 一种基于模糊滑模的智能汽车横向控制方法
CN111409641B (zh) * 2020-03-10 2023-07-25 厦门理工学院 一种基于纵横向协调的轨迹跟踪控制方法及系统
CN111781948A (zh) * 2020-06-18 2020-10-16 南京非空航空科技有限公司 一种无人机编队队形变换与在线动态避障方法
CN115257743A (zh) * 2021-03-19 2022-11-01 浙江亚太智能网联汽车创新中心有限公司 一种双向行驶摆渡车变道时扭矩计算方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109709945A (zh) * 2017-10-26 2019-05-03 深圳市优必选科技有限公司 一种基于障碍物分类的路径规划方法、装置及机器人
US20200174475A1 (en) * 2018-11-29 2020-06-04 Electronics And Telecommunications Research Institute Autonomous driving method and system
CN109855639A (zh) * 2019-01-15 2019-06-07 天津大学 基于障碍物预测与mpc算法的无人驾驶轨迹规划方法
CN109933945A (zh) * 2019-03-28 2019-06-25 北京智行者科技有限公司 交通环境建模方法
CN110147105A (zh) * 2019-05-27 2019-08-20 安徽江淮汽车集团股份有限公司 无人驾驶车辆的路径控制方法、设备、存储介质及装置
CN112327836A (zh) * 2020-10-27 2021-02-05 南宁市第一人民医院 基于5g技术的轮椅自动驾驶控制方法及自动驾驶轮椅
CN112269389A (zh) * 2020-11-18 2021-01-26 武汉理工大学 一种多功能乘务服务智能机器车系统及其控制方法
CN113655789A (zh) * 2021-08-04 2021-11-16 东风柳州汽车有限公司 路径跟踪方法、装置、车辆及存储介质

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116719229A (zh) * 2023-03-09 2023-09-08 哈尔滨理工大学 基于势函数的动力定位船避碰容错控制方法
CN116719229B (zh) * 2023-03-09 2024-01-05 哈尔滨理工大学 基于势函数的动力定位船避碰容错控制方法
CN116540723A (zh) * 2023-05-30 2023-08-04 南通大学 一种基于人工势场的水下机器人滑模轨迹跟踪控制方法
CN116540723B (zh) * 2023-05-30 2024-04-12 南通大学 一种基于人工势场的水下机器人滑模轨迹跟踪控制方法
CN117724504A (zh) * 2024-02-07 2024-03-19 北京科技大学 一种基于投影面积的无人车路径跟踪控制方法及装置
CN117724504B (zh) * 2024-02-07 2024-04-16 北京科技大学 一种基于投影面积的无人车路径跟踪控制方法及装置

Also Published As

Publication number Publication date
CN113655789A (zh) 2021-11-16

Similar Documents

Publication Publication Date Title
WO2023010854A1 (zh) 路径跟踪方法、装置、车辆及存储介质
CN112389427B (zh) 车辆轨迹优化方法、装置、电子设备和存储介质
Zhou et al. Multiobjective optimization of lane-changing strategy for intelligent vehicles in complex driving environments
WO2020143288A1 (zh) 一种复杂工况下自动驾驶车辆决策系统及其轨迹规划方法
Liu et al. Dynamic lane-changing trajectory planning for autonomous vehicles based on discrete global trajectory
WO2021073079A1 (zh) 一种自动驾驶车辆路径与速度高度耦合的轨迹规划方法
CN107298103B (zh) 一种智能电动汽车自动换道分层控制系统及方法
JP6715899B2 (ja) 衝突回避装置
CN105676643B (zh) 一种智能汽车转向和制动自适应协调控制方法
CN109799828B (zh) 自动驾驶车辆横纵向协调控制方法
CN111289978A (zh) 一种车辆无人驾驶行为决策的方法和系统
CN112068445B (zh) 自动驾驶车辆路径规划与路径跟踪集成控制方法及系统
US20230419824A1 (en) Method and device for determining traffic stream information, electronic equipment and storage medium
CN114771563A (zh) 一种自动驾驶车辆轨迹规划控制实现方法
Li et al. State-of-the-art trajectory tracking of autonomous vehicles
CN112859838B (zh) 一种自动驾驶控制方法、装置、设备和介质
WO2023024914A1 (zh) 车辆避让方法、装置、计算机设备和存储介质
Hima et al. Controller design for trajectory tracking of autonomous passenger vehicles
Liu et al. Handling and stability integrated control of AFS and DYC for distributed drive electric vehicles based on risk assessment and prediction
Lee et al. Automated driving control in safe driving envelope based on probabilistic prediction of surrounding vehicle behaviors
Peng et al. Lane-change model and tracking control for autonomous vehicles on curved highway sections in rainy weather
Xing et al. Lane change strategy for autonomous vehicle
Duan et al. Implementing trajectory tracking control algorithm for autonomous vehicles
CN113370980B (zh) 极限工况下车辆失稳与碰撞综合风险评估方法及装置
Li et al. Lane‐keeping system design considering driver's nervousness via scene analysis

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: 22851579

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE