CN118210314B - A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle - Google Patents
A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle Download PDFInfo
- Publication number
- CN118210314B CN118210314B CN202410627162.1A CN202410627162A CN118210314B CN 118210314 B CN118210314 B CN 118210314B CN 202410627162 A CN202410627162 A CN 202410627162A CN 118210314 B CN118210314 B CN 118210314B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- obstacle
- time
- representing
- state
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000005457 optimization Methods 0.000 claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims description 29
- 230000006870 function Effects 0.000 claims description 25
- 239000013598 vector Substances 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 8
- 238000012546 transfer Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 206010048669 Terminal state Diseases 0.000 claims description 4
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims 1
- 238000012300 Sequence Analysis Methods 0.000 abstract description 2
- 230000009977 dual effect Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- MZWGYEJOZNRLQE-KXQOOQHDSA-N 1-stearoyl-2-myristoyl-sn-glycero-3-phosphocholine Chemical compound CCCCCCCCCCCCCCCCCC(=O)OC[C@H](COP([O-])(=O)OCC[N+](C)(C)C)OC(=O)CCCCCCCCCCCCC MZWGYEJOZNRLQE-KXQOOQHDSA-N 0.000 description 1
- 101001121408 Homo sapiens L-amino-acid oxidase Proteins 0.000 description 1
- 102100026388 L-amino-acid oxidase Human genes 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 101100012902 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) FIG2 gene Proteins 0.000 description 1
- 101100233916 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR5 gene Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/40—Control within particular dimensions
- G05D1/43—Control of position or course in two dimensions
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/60—Intended control result
- G05D1/617—Safety or protection, e.g. defining protection zones around obstacles or avoiding hazards
- G05D1/622—Obstacle avoidance
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
Description
技术领域Technical Field
本发明涉及自动驾驶技术领域,特别是涉及一种自动驾驶车辆轨迹规划和动态避障方法、设备及介质。The present invention relates to the field of autonomous driving technology, and in particular to an autonomous driving vehicle trajectory planning and dynamic obstacle avoidance method, device and medium.
背景技术Background technique
随着自动驾驶技术的迅猛发展,安全避障成为无人驾驶系统设计中的关键问题之一。在实际道路上行驶时,动态障碍物的运动不确定性和复杂性给自动驾驶系统带来了挑战。障碍物可能以各种速度、方向和轨迹运动,这些不确定性使得规划层难以规划出安全的行驶轨迹,甚至引发车辆碰撞等交通事故。因此,研究考虑障碍物运动不确定性的自动驾驶车辆安全轨迹规划,以提高行驶安全性,已成为一个亟待解决的问题。With the rapid development of autonomous driving technology, safe obstacle avoidance has become one of the key issues in the design of unmanned driving systems. When driving on actual roads, the uncertainty and complexity of the motion of dynamic obstacles pose challenges to autonomous driving systems. Obstacles may move at various speeds, directions, and trajectories. These uncertainties make it difficult for the planning layer to plan a safe driving trajectory, and may even cause traffic accidents such as vehicle collisions. Therefore, it has become an urgent issue to study the safe trajectory planning of autonomous driving vehicles considering the uncertainty of obstacle motion in order to improve driving safety.
发明内容Summary of the invention
本发明的目的是提供一种自动驾驶车辆轨迹规划和动态避障方法、设备及介质,以提高自动驾驶车辆行驶的安全性。The purpose of the present invention is to provide a method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle to improve the safety of the autonomous driving vehicle.
为实现上述目的,本发明提供了如下方案:To achieve the above object, the present invention provides the following solutions:
在一方面,本发明提供一种自动驾驶车辆轨迹规划和动态避障方法,所述方法包括如下步骤。In one aspect, the present invention provides a method for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle, the method comprising the following steps.
检测障碍物车辆的位姿和速度信息。Detect the position and speed information of the obstacle vehicle.
根据所述位姿和速度信息,利用障碍物车辆的运动模型和运动不确定性,预测障碍物车辆的状态估计序列和运动不确定性序列。According to the position and speed information, the state estimation sequence and motion uncertainty sequence of the obstacle vehicle are predicted by utilizing the motion model and motion uncertainty of the obstacle vehicle.
基于障碍物车辆的状态估计序列和运动不确定性序列,构建障碍物车辆的运动不确定性椭圆和安全可行域。Based on the state estimation sequence and motion uncertainty sequence of the obstacle vehicle, the motion uncertainty ellipse and safe feasible region of the obstacle vehicle are constructed.
根据障碍物车辆的运动不确定性椭圆和安全可行域,构建自动驾驶车辆的无碰撞约束。According to the motion uncertainty ellipse and safe feasible region of the obstacle vehicle, the collision-free constraints of the autonomous driving vehicle are constructed.
构建自动驾驶车辆的轨迹跟踪代价函数和约束条件,所述约束条件至少包括无碰撞约束。A trajectory tracking cost function and constraints for an autonomous driving vehicle are constructed, wherein the constraints include at least a no-collision constraint.
基于所述约束条件,求解使所述轨迹跟踪代价函数最优的规划轨迹和优化控制序列。Based on the constraints, a planning trajectory and an optimized control sequence are solved to optimize the trajectory tracking cost function.
在另一方面,本发明提供一种计算机设备,包括:存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序以实现上述方法的步骤。In another aspect, the present invention provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.
在另一方面,本发明提供一种计算机可读存储介质,其上存储有计算机程序,该计算机程序被处理器执行时实现上述方法的步骤。In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.
根据本发明提供的具体实施例,本发明公开了以下技术效果。According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects.
本发明实施例的一种自动驾驶车辆轨迹规划和动态避障方法、设备及介质,所述方法包括:检测障碍物车辆的位姿和速度信息;根据所述位姿和速度信息,利用障碍物车辆的运动模型和运动不确定性,预测障碍物车辆的状态估计序列和运动不确定性序列;基于障碍物车辆的状态估计序列和运动不确定性序列,构建障碍物车辆的运动不确定性椭圆和安全可行域;根据障碍物车辆的运动不确定性椭圆和安全可行域,构建自动驾驶车辆的无碰撞约束;构建自动驾驶车辆的轨迹跟踪代价函数和约束条件,所述约束条件至少包括无碰撞约束;基于所述约束条件,求解使所述轨迹跟踪代价函数最优的规划轨迹和优化控制序列。本发明通过考虑动态障碍物的运动不确定性,对障碍物进行运动行为估计和不确定性序列分析,获取障碍物车辆的状态估计序列和运动不确定性序列,并将基于障碍物车辆的状态估计序列和运动不确定性序列获取的无碰撞约束添加至轨迹优化求解的约束条件中,以规划出安全的行驶轨迹,提高了自动驾驶车辆行驶的安全性。The present invention discloses a method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle, the method comprising: detecting the position and speed information of an obstacle vehicle; predicting the state estimation sequence and motion uncertainty sequence of the obstacle vehicle based on the position and speed information and the motion model and motion uncertainty of the obstacle vehicle; constructing the motion uncertainty ellipse and safe feasible domain of the obstacle vehicle based on the state estimation sequence and motion uncertainty sequence of the obstacle vehicle; constructing the collision-free constraint of the autonomous driving vehicle based on the motion uncertainty ellipse and safe feasible domain of the obstacle vehicle; constructing the trajectory tracking cost function and constraint conditions of the autonomous driving vehicle, the constraint conditions at least including the collision-free constraint; solving the planning trajectory and optimization control sequence that optimizes the trajectory tracking cost function based on the constraint conditions. The present invention considers the motion uncertainty of the dynamic obstacle, estimates the motion behavior and analyzes the uncertainty sequence of the obstacle, obtains the state estimation sequence and motion uncertainty sequence of the obstacle vehicle, and adds the collision-free constraint obtained based on the state estimation sequence and motion uncertainty sequence of the obstacle vehicle to the constraint conditions of the trajectory optimization solution, so as to plan a safe driving trajectory, thereby improving the driving safety of the autonomous driving vehicle.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
图1为本发明实施例提供的一种自动驾驶车辆轨迹规划和动态避障方法的流程图。FIG1 is a flow chart of a method for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle provided in an embodiment of the present invention.
图2为本发明实施例提供的一种自动驾驶车辆轨迹规划和动态避障方法的原理框架图。FIG2 is a principle framework diagram of an autonomous driving vehicle trajectory planning and dynamic obstacle avoidance method provided in an embodiment of the present invention.
图3为本发明实施例提供的安全可行区域构建的示意图。FIG3 is a schematic diagram of constructing a safe and feasible area provided by an embodiment of the present invention.
图4为本发明实施例提供的EV与OV的SFR之间的距离表示的示意图。FIG. 4 is a schematic diagram showing the distance between the SFRs of the EV and the OV provided in an embodiment of the present invention.
图5为本发明实施例提供的车辆动力学模型的示意图。FIG. 5 is a schematic diagram of a vehicle dynamics model provided by an embodiment of the present invention.
图6为本发明实施例提供的系统状态紧缩约束的示意图。FIG. 6 is a schematic diagram of a system state tightening constraint provided by an embodiment of the present invention.
图7为本发明实施例提供的计算机设备的内部结构图。FIG. 7 is a diagram showing the internal structure of a computer device provided in an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
本发明的目的是提供一种自动驾驶车辆轨迹规划和动态避障方法、设备及介质,以提高自动驾驶车辆行驶的安全性。The purpose of the present invention is to provide a method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle to improve the safety of the autonomous driving vehicle.
为使本发明的上述目的、特征和优点能够更加明显易懂,下面结合附图和具体实施方式对本发明作进一步详细的说明。In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
实施例1Example 1
本发明实施例1提供一种自动驾驶车辆轨迹规划和动态避障方法,该方法为一种考虑障碍物运动不确定性的自动驾驶车辆安全轨迹规划方法,首先,基于预测层预测的动态障碍物在一段时间内的动作行为信息,获取障碍物运动状态估计和不确定性序列;然后,构建运动不确定性椭圆和动态障碍物安全可行域(Safety-critical Feasible Region,SFR),建立安全避障约束,即无碰撞约束;然后,基于随机模型预测控制(Stochastic ModelPredictive Control,SMPC)方法,构建障碍物具有运动不确定性下的安全避障轨迹规划模型,安全避障轨迹规划模型包括轨迹跟踪代价函数和约束条件。Embodiment 1 of the present invention provides a method for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle. The method is a safe trajectory planning method for an autonomous driving vehicle that takes into account the uncertainty of obstacle motion. First, based on the action behavior information of a dynamic obstacle predicted by a prediction layer over a period of time, an obstacle motion state estimation and an uncertainty sequence are obtained; then, a motion uncertainty ellipse and a dynamic obstacle safety feasible region (Safety-critical Feasible Region, SFR) are constructed to establish a safe obstacle avoidance constraint, that is, a no-collision constraint; then, based on a stochastic model predictive control (Stochastic Model Predictive Control, SMPC) method, a safe obstacle avoidance trajectory planning model is constructed under the condition that the obstacle has motion uncertainty. The safe obstacle avoidance trajectory planning model includes a trajectory tracking cost function and constraints.
如图1和图2所示,本发明实施例1提供的自动驾驶车辆轨迹规划和动态避障方法包括如下步骤。As shown in FIG. 1 and FIG. 2 , the autonomous driving vehicle trajectory planning and dynamic obstacle avoidance method provided in Example 1 of the present invention includes the following steps.
步骤101,检测障碍物车辆的位姿和速度信息。Step 101, detecting the position and speed information of the obstacle vehicle.
步骤102,根据所述位姿和速度信息,利用障碍物车辆的运动模型和运动不确定性,预测障碍物车辆的状态估计序列和运动不确定性序列。Step 102: predicting a state estimation sequence and a motion uncertainty sequence of the obstacle vehicle based on the position and velocity information and using the motion model and motion uncertainty of the obstacle vehicle.
本发明实施例中步骤101和步骤102的具体实现步骤如下。The specific implementation steps of step 101 and step 102 in the embodiment of the present invention are as follows.
假设自动驾驶预测模块给出障碍物车辆(Obstacle Vehicle,OV)的动作序列为:,N为预测时域长度,、和分别为0时刻、1时刻和N-1时刻障碍物车辆的动作,采样时间步长为0.1s。障碍物车辆的运动模型可以表示为公式(1)。Assume that the autonomous driving prediction module gives the action sequence of the obstacle vehicle (OV) as follows: , N is the prediction time domain length, , and These are the actions of the obstacle vehicle at time 0, time 1, and time N-1, respectively, and the sampling time step is 0.1s. The motion model of the obstacle vehicle can be expressed as formula (1).
其中,和分别为k时刻和k+1时刻障碍物车辆的状态向量,为k时刻障碍物车辆的动作,表示障碍物车辆的运动模型的状态传递方程。in, and are the state vectors of the obstacle vehicle at time k and time k+1 respectively, is the action of the obstacle vehicle at time k, State transfer equations representing the motion model of the obstacle vehicle.
其中,状态向量为,包括障碍物车辆的位姿和速度信息。动作(即控制输入)包括加速度和前轮转角,和分别为障碍物车辆在水平方向上的位置变量和垂直方向上的位置变量,和分别为障碍物车辆的偏航角和速度。The state vector is , including the position and speed information of the obstacle vehicle. Action (i.e. control input) Including acceleration and front wheel angle , and are the position variables of the obstacle vehicle in the horizontal direction and the vertical direction, respectively. and are the yaw angle and speed of the obstacle vehicle respectively.
将运动不确定性引入到障碍物车辆的运动模型的状态传递方程和输出方程中,得到公式(2)和公式(3)。Introducing motion uncertainty into the state transfer equation of the motion model of the obstacle vehicle And the output equation In this paper, we obtain formula (2) and formula (3).
其中,和分别表示k时刻的运动不确定性和输出不确定性,服从0均值的高斯分布,对应的k时刻的协方差矩阵为和,表示k+1时刻障碍物车辆的输出状态。in, and They represent the motion uncertainty and output uncertainty at time k, respectively, and obey the Gaussian distribution with zero mean. The corresponding covariance matrix at time k is and , Represents the output state of the obstacle vehicle at time k+1.
障碍物车辆的先验状态和相应的运动不确定性先验协方差矩阵可以表示为公式(4)和公式(5)。The prior state of the obstacle vehicle and the corresponding motion uncertainty prior covariance matrix can be expressed as formula (4) and formula (5).
其中,为k+1时刻障碍物车辆的先验状态,为k时刻障碍物车辆的先验状态,为k时刻障碍物车辆的动作,表示障碍物车辆的运动模型的状态传递方程;为k+1时刻障碍物车辆的运动不确定性先验协方差矩阵,表示k时刻运动不确定性后验协方差矩阵,表示时刻状态转移函数关于状态量的雅克比矩阵,表示k时刻障碍物车辆的运动不确定性,上标T表示转置。in, is the prior state of the obstacle vehicle at time k+1, is the prior state of the obstacle vehicle at time k, is the action of the obstacle vehicle at time k, The state transfer equations representing the motion model of the obstacle vehicle; is the motion uncertainty prior covariance matrix of the obstacle vehicle at time k+1, represents the posterior covariance matrix of motion uncertainty at time k, express The Jacobian matrix of the state transfer function at the moment about the state quantity, represents the motion uncertainty of the obstacle vehicle at time k, and the superscript T represents the transpose.
结合预测模块关于先验状态的预测信息,可以获取更加准确的后验状态和后验不确定性协方差矩阵,如式(6)-式(8)所示。Combined with the prediction information of the prediction module about the prior state, a more accurate posterior state and posterior uncertainty covariance matrix can be obtained, as shown in Equations (6) to (8).
其中,为k+1时刻的卡尔曼增益系数,为k+1时刻输出函数关于状态量的雅克比矩阵,为时刻障碍物车辆的输出不确定性,为k+1时刻障碍物车辆的后验状态,为k+1时刻障碍物车辆的输出状态,表示障碍物车辆的运动模型的输出方程,表示k+1时刻障碍物车辆的运动不确定性后验协方差矩阵,表示单位矩阵。in, is the Kalman gain coefficient at time k+1, is the Jacobian matrix of the output function at time k+1 with respect to the state quantity, for The output uncertainty of the obstacle vehicle at each moment, is the posterior state of the obstacle vehicle at time k+1, is the output state of the obstacle vehicle at time k+1, The output equation representing the motion model of the obstacle vehicle, represents the posterior covariance matrix of the motion uncertainty of the obstacle vehicle at time k+1, Represents the identity matrix.
获得障碍物车辆在未来一段时间内的状态估计序列{}和运动不确定性序列{},其中,和分别为0时刻和N-1时刻障碍物车辆的后验状态,和分别为0时刻和N-1时刻障碍物车辆的运动不确定性后验协方差矩阵。Obtain the state estimation sequence of the obstacle vehicle in the future period of time { } and motion uncertainty sequence { },in, and are the posterior states of the obstacle vehicle at time 0 and time N-1 respectively, and are the posterior covariance matrices of the motion uncertainty of the obstacle vehicle at time 0 and time N-1 respectively.
步骤103,基于障碍物车辆的状态估计序列和运动不确定性序列,构建障碍物车辆的运动不确定性椭圆和安全可行域。Step 103: construct the motion uncertainty ellipse and safe feasible region of the obstacle vehicle based on the state estimation sequence and motion uncertainty sequence of the obstacle vehicle.
步骤104,根据障碍物车辆的运动不确定性椭圆和安全可行域,构建自动驾驶车辆的无碰撞约束。Step 104, constructing a collision-free constraint for the autonomous driving vehicle based on the motion uncertainty ellipse and the safe feasible region of the obstacle vehicle.
步骤103和步骤104的具体实现过程如下。The specific implementation process of step 103 and step 104 is as follows.
考虑障碍物车辆的横向和纵向的耦合运动不确定性,遵循二元高斯分布,其概率密度函数如公式(9)所示。Considering the lateral and longitudinal coupled motion uncertainty of the obstacle vehicle, it follows a bivariate Gaussian distribution , and its probability density function is shown in formula (9).
其中,为随机变量的维度,为障碍物车辆在水平方向上的位置变量,为障碍物车辆在垂直方向上的位置变量。和表示的均值和正定协方差矩阵,可以从状态估计序列和运动不确定性序列中获取。马氏距离平方的标量,马氏距离平方的标量表明了和的距离,可以表示为:。in, is a random variable The dimension of is the horizontal position variable of the obstacle vehicle, is the position variable of the obstacle vehicle in the vertical direction. and express The mean and positive covariance matrix of can be obtained from the state estimation sequence and the motion uncertainty sequence. The scalar of the squared Mahalanobis distance , a scalar of the squared Mahalanobis distance shows and The distance can be expressed as: .
进一步量化障碍物车辆的运动不确定性如公式(10)所示。The motion uncertainty of the obstacle vehicle is further quantified as shown in formula (10).
其中,是具有两个自由度的分布中使得累积概率为α的上分位数。是置信水平,即随机变量的置信度。in, It has two degrees of freedom The upper quantile of the distribution such that the cumulative probability is α. is the confidence level, i.e., the random variable confidence level.
运动不确定性的标准偏差椭圆(Standard Deviational Ellipse,SDE)可以通过选择安全置信度来确定,如公式(11)所示。The standard deviation ellipse (SDE) of motion uncertainty can be determined by selecting the safety confidence level, as shown in formula (11).
其中,为障碍物车辆在水平方向上的位置变量,为障碍物车辆在水平方向上的平均位置,为障碍物车辆在垂直方向上的位置变量,为障碍物车辆在垂直方向上的平均位置,为具有两个自由度的分布中使得累积概率为α的上分位数,的值通过查阅卡方分布表获取,为置信水平,和是水平方向协方差矩阵和垂直方向协方差矩阵的特征值,、、和均基于障碍物车辆的状态估计序列和运动不确定性序列获取得到。in, is the horizontal position variable of the obstacle vehicle, is the average horizontal position of the obstacle vehicle, is the position variable of the obstacle vehicle in the vertical direction, is the average vertical position of the obstacle vehicle, For a system with two degrees of freedom The upper quantile of the distribution with cumulative probability α, The value of is obtained by consulting the chi-square distribution table. is the confidence level, and is the horizontal covariance matrix and the vertical covariance matrix The characteristic value of , , and They are all obtained based on the state estimation sequence and motion uncertainty sequence of the obstacle vehicle.
如图3所示,使用表示的障碍物车辆的全维几何模型,代表障碍物车辆的运动不确定性椭圆,SDE,根据和采用Minkowski Sum(多边形求和)方法重构障碍物车辆的安全可行区域,如公式(12)所示。As shown in Figure 3, use The full-dimensional geometric model of the obstacle vehicle represented by The uncertainty ellipse, SDE, of the motion of the obstacle vehicle is represented by and The Minkowski Sum method is used to reconstruct the safe and feasible area of the vehicle with obstacles. , as shown in formula (12).
其中,为中的元素,为中的元素,为中的元素。in, for The elements in for The elements in for The elements in .
障碍物车辆的全维几何模型和运动不确定性椭圆均为凸包,从而安全可行域的凸包可表示为公式(13)。Full-dimensional geometric model of obstacle vehicle and the motion uncertainty ellipse are all convex hulls, so the convex hull of the safe feasible region can be expressed as formula (13).
其中,表示向量卷积运算。in, Represents a vector convolution operation.
采用基于符号距离构建安全避障约束,约束要保证自车(Ego Vehicle, EV,即自动驾驶车辆)和障碍物车辆之间的距离大于安全阈值,如公式(14)所示。The safety obstacle avoidance constraint is constructed based on the signed distance. The constraint must ensure that the distance between the ego vehicle (EV) and the obstacle vehicle is greater than the safety threshold, as shown in formula (14).
其中,表示集合之间的距离,表示最小安全距离阈值,和分别表示自动驾驶车辆的全维模型的集合和第个障碍物车辆的安全可行区域的集合,集合之间的距离表示如图4所示。in, represents the distance between sets, Indicates the minimum safety distance threshold, and Represents the set of full-dimensional models of autonomous vehicles and the The safe and feasible areas of the vehicles with obstacles are represented by the distance between the sets as shown in Figure 4.
集合之间的距离可进一步表示为公式(15)。The distance between sets can be further expressed as formula (15).
其中,表示中的元素,表示中的元素。in, express The elements in express The elements in .
求解集合之间的距离可以转化为一个优化问题,如公式(16)-公式(18)所示。Solving the distance between sets can be transformed into an optimization problem, as shown in Formula (16)-Formula (18).
其中,为自动驾驶车辆的安全补偿系数矩阵,为第i个障碍物车辆的安全补偿系数矩阵,表示描述自动驾驶车辆外形轮廓的常数向量,表示描述第i个障碍物车辆外形轮廓的常数向量。in, is the safety compensation coefficient matrix of the autonomous driving vehicle, is the safety compensation coefficient matrix of the i-th obstacle vehicle, represents a constant vector describing the contour of the autonomous vehicle, Represents a constant vector describing the outline of the i-th obstacle vehicle.
上面优化问题的对偶优化问题如公式(19)所示。The dual optimization problem of the above optimization problem is shown in formula (19).
其中,表示自动驾驶车辆的正定对偶可行解,表示第i个障碍物车辆的正定对偶可行解,表示最小安全距离阈值。in, represents the positive definite dual feasible solution for the autonomous vehicle, represents the positive definite dual feasible solution of the i-th obstacle vehicle, Indicates the minimum safety distance threshold.
通过求解优化问题获得最优对偶变量。从而构建了无碰撞约束,如公式(20)所示。Obtaining the optimal dual variables by solving the optimization problem . Thus, a collision-free constraint is constructed, as shown in formula (20).
其中,表示自动驾驶车辆的正定对偶可行解,表示描述自动驾驶车辆在状态时的外形轮廓的常数向量,表示车辆状态变量,表示第i个障碍物车辆的正定对偶可行解,表示描述第i个障碍物车辆外形轮廓的常数向量,表示第i个障碍物车辆的避障软约束的松弛变量,表示最小安全距离阈值。in, represents the positive definite dual feasible solution for the autonomous vehicle, Describes the state of the autonomous vehicle The constant vector of the contour when , represents the vehicle state variable, represents the positive definite dual feasible solution of the i-th obstacle vehicle, represents a constant vector describing the outline of the i-th obstacle vehicle, represents the slack variable of the obstacle avoidance soft constraint of the i-th obstacle vehicle, Indicates the minimum safety distance threshold.
步骤105,构建自动驾驶车辆的轨迹跟踪代价函数和约束条件,所述约束条件至少包括无碰撞约束,具体步骤如下所示。Step 105, constructing a trajectory tracking cost function and constraint conditions of the autonomous driving vehicle, wherein the constraint conditions at least include a no-collision constraint. The specific steps are as follows.
1、优化问题:基于MPC的轨迹规划通过结合系统模型、约束和目标来制定优化问题,求解最佳规划轨迹。该问题通常以离散时间形式设计,如公式(21)-(23)所示。1. Optimization problem: MPC-based trajectory planning formulates an optimization problem by combining system models, constraints, and objectives to solve the optimal planning trajectory. This problem is usually designed in discrete time form, as shown in formulas (21)-(23).
其中,和分别表示k时刻自动驾驶车辆的系统状态和控制量。是终端状态。N是预测时域长度。和分别表示阶段成本函数和终端成本函数。是系统动态。和表示系统状态和输入的约束空间。in, and They represent the system state and control quantity of the autonomous driving vehicle at time k respectively. is the terminal state. N is the prediction time domain length. and They represent the stage cost function and terminal cost function respectively. It is system dynamics. and Represents the constraint space of system states and inputs.
2、系统动态模型:采用车辆动力学模型去描述动态。假设车辆是一个刚体,忽略垂向运动。动力学分析如图5所示。2. System dynamic model: The vehicle dynamics model is used to describe the dynamics. Assume that the vehicle is a rigid body and ignore vertical motion. The dynamic analysis is shown in Figure 5.
连续的车辆动力学模型可以表示为公式(24)-公式(31)。The continuous vehicle dynamics model can be expressed as Equation (24)-Equation (31).
其中,是自动驾驶车辆的系统状态,自动驾驶车辆的动作(即控制输入)包括自动驾驶车辆的加速度和前轮转角,是自动驾驶车辆全局坐标位置,和分别表示自动驾驶车辆的纵向位置和横向位置,是自动驾驶车辆的偏航角,和分别表示自动驾驶车辆的纵向速度和横向速度,是自动驾驶车辆的偏航率,表示参考路径的偏航率,和表示自动驾驶车辆与参考路径之间的航向误差和横向误差,和是自动驾驶车辆的质量和转动惯量,和分别是自动驾驶车辆前后轮到质心的距离,和分别表示自动驾驶车辆前后轮胎侧向力。in, is the system state of the autonomous vehicle, the actions of the autonomous vehicle (i.e., control inputs) Including the acceleration of autonomous vehicles and front wheel angle , is the global coordinate position of the autonomous driving vehicle, and Respectively represent the longitudinal position and lateral position of the autonomous driving vehicle, is the yaw angle of the autonomous vehicle, and They represent the longitudinal speed and lateral speed of the autonomous vehicle respectively, is the yaw rate of the autonomous vehicle, represents the yaw rate of the reference path, and represents the heading error and lateral error between the autonomous vehicle and the reference path, and is the mass and moment of inertia of the autonomous vehicle, and are the distances from the front and rear wheels of the autonomous driving vehicle to the center of mass, and Respectively represent the lateral forces of the front and rear tires of the autonomous driving vehicle.
对连续的动力学模型进行泰勒展开线性化和零阶保持离散化,获得离散的线性动力学模型,如公式(32)所示。The continuous dynamic model is linearized by Taylor expansion and discretized by zero-order hold to obtain a discrete linear dynamic model, as shown in formula (32).
其中,和分别表示k+1时刻和k时刻自动驾驶车辆的系统状态,、和均为自动驾驶车辆的离散线性动力学模型的系数矩阵。in, and They represent the system status of the autonomous driving vehicle at time k+1 and time k respectively, , and are the coefficient matrices of the discrete linear dynamics model of the autonomous vehicle.
3、系统状态紧缩约束:车辆行驶过程中,由于外界扰动、传感器噪声等,给系统状态带来了不确定性。因此,系统状态约束可以表述为机会约束的形式,如公式(33)所示。3. System state tight constraints: During vehicle driving, external disturbances, sensor noise, etc., bring uncertainty to the system state. Therefore, the system state constraints It can be expressed in the form of a chance constraint, as shown in formula (33).
其中,表示概率约束,为违反约束的概率值,表示自动驾驶车辆的系统状态的约束空间。in, represents the probability constraint, is the probability value of violating the constraint, Represents the constraint space of the system state of the autonomous vehicle.
使用泰勒近似获取系统不确定性传播序列{},系统状态的紧缩约束如图6所示,其中,、和分别表示对k时刻的正定协方差矩阵使用泰勒近似获得的k+1时刻、k+2时刻和k+N时刻的正定协方差矩阵。Use Taylor approximation to obtain the system uncertainty propagation sequence { }, the tightening constraints of the system state are shown in Figure 6, where , and They represent the positive definite covariance matrices at time k+1, time k+2, and time k+N obtained by using Taylor approximation to the positive definite covariance matrix at time k.
利用切比雪夫不等式原理,将概率约束转化为线性约束形式,如公式(34)所示。Using the Chebyshev inequality principle, the probability constraint is transformed into a linear constraint form, as shown in formula (34).
其中,和分别表示不等式约束的系数矩阵和常数向量,是标准正态分布的分位数函数,表示正定协方差矩阵。in, and denote the coefficient matrix and constant vector of the inequality constraints, respectively. is the quantile function of the standard normal distribution, represents the positive definite covariance matrix.
4、其他约束:轨迹规划优化问题的约束还包括式(20)中避障约束,以及对控制量和控制变化率的约束,如公式(35)和(36)所示。4. Other constraints: The constraints of the trajectory planning optimization problem also include the obstacle avoidance constraint in equation (20), as well as constraints on the control quantity and control change rate, as shown in equations (35) and (36).
其中,和分别为k时刻自动驾驶车辆的控制量的下限和上限,为k时刻自动驾驶车辆的控制量,和分别为k时刻自动驾驶车辆的控制变化率的下限和上限,为k时刻自动驾驶车辆的控制变化率。in, and are the lower and upper limits of the control amount of the autonomous driving vehicle at time k, is the control amount of the autonomous driving vehicle at time k, and are the lower and upper limits of the control change rate of the autonomous driving vehicle at time k, is the control change rate of the autonomous driving vehicle at time k.
5、代价函数:构建考虑跟踪性能、舒适度和安全性等多目标的轨迹跟踪代价函数,代价函数的二次形式可以表示为公式(37)。5. Cost function: Construct a trajectory tracking cost function that takes into account multiple objectives such as tracking performance, comfort and safety. The quadratic form of the cost function can be expressed as formula (37).
其中,为轨迹跟踪代价函数,为系统状态量的均值,为系统控制输入的均值,表示避障约束的松弛变量,表示预测时域长度,为系统终端状态的均值,、和分别表示系统状态、控制输入和终端状态的权重矩阵,是松弛变量的惩罚因子,和分别为k时刻和N时刻参考的系统状态,表示障碍物车辆数量,表示k时刻的系统状态量,表示k时刻的系统控制输入,表示k时刻的第i个障碍物车辆的避障软约束的松弛变量。in, is the trajectory tracking cost function, is the mean value of the system state quantity, is the mean value of the system control input, represents the slack variable of the obstacle avoidance constraint, represents the prediction time domain length, is the mean of the system terminal states, , and The weight matrices representing the system state, control input and terminal state, respectively, is the penalty factor for the slack variable, and are the system states referenced at time k and time N respectively, Indicates the number of vehicles in the way. represents the system state at time k, represents the system control input at time k, Represents the slack variable of the obstacle avoidance soft constraint of the i-th obstacle vehicle at time k.
6、基于MPC的避障轨迹优化模型:该框架能够考虑障碍物车辆运动不确定性和系统状态不确定性,实现安全有效的轨迹规划。避障轨迹优化模型可以表示为公式(38)-(46)。6. Obstacle avoidance trajectory optimization model based on MPC: This framework can take into account the uncertainty of obstacle vehicle motion and system state uncertainty to achieve safe and effective trajectory planning. The obstacle avoidance trajectory optimization model can be expressed as formulas (38)-(46).
其中,表示k+1时刻自动驾驶车辆的正定对偶可行解,表示k+1时刻描述自动驾驶车辆在状态时的外形轮廓的常数向量,表示k+1时刻第i个障碍物车辆的正定对偶可行解,表示k+1时刻第i个障碍物车辆外形轮廓的常数向量,表示k时刻第i个障碍物车辆的避障软约束的松弛变量;表示初始状态的均值;表示预测时域长度,表示障碍物车辆数量。in, represents the positive definite dual feasible solution of the autonomous driving vehicle at time k+1, Indicates the state of the autonomous driving vehicle at time k+1 The constant vector of the contour when , represents the positive definite dual feasible solution of the i-th obstacle vehicle at time k+1, A constant vector representing the outline of the i-th obstacle vehicle at time k+1, represents the slack variable of the obstacle avoidance soft constraint of the i-th obstacle vehicle at time k; represents the mean of the initial state; represents the prediction time domain length, Indicates the number of obstacle vehicles.
步骤106,基于所述约束条件,求解使所述轨迹跟踪代价函数最优的规划轨迹和优化控制序列。Step 106 : based on the constraint conditions, solving the planned trajectory and optimized control sequence that optimizes the trajectory tracking cost function.
本发明实施例1提供自动驾驶车辆轨迹规划和动态避障方法的实施步骤可概括为如下步骤。The implementation steps of the autonomous driving vehicle trajectory planning and dynamic obstacle avoidance method provided in Example 1 of the present invention can be summarized as follows.
S1:初始化时间步。S1: Initialization time step .
S2:初始化自车位置。S2: Initialize the vehicle position .
S3:for。S3: for .
S4:检测障碍物车辆位姿和速度信息。S4: Detect obstacle vehicle position and speed information .
S5:结合预测模块信息,获取障碍物车辆状态估计序列和运动不确定性序列(6)-(8)。S5: Combine the prediction module information to obtain the obstacle vehicle state estimation sequence and motion uncertainty sequence (6)-(8).
S6:构建障碍物车辆运动不确定性椭圆(11)和安全可行域(13)。S6: Construct the obstacle vehicle motion uncertainty ellipse (11) and the safe feasible region (13).
S7:建立安全避障约束(20)。S7: Establish safety obstacle avoidance constraints (20).
S8:利用车辆动力学模型(32)描述系统动态。S8: Use the vehicle dynamics model (32) to describe the system dynamics.
S9:构建系统状态紧缩约束(34)和控制量相关约束(35)-(36)。S9: Construct system state tightening constraints (34) and control quantity related constraints (35)-(36).
S10:建立多优化目标代价函数(37)。S10: Establish a multi-optimization objective cost function (37).
S11:求解轨迹规划的优化问题(38)-(46)。S11: Solve the trajectory planning optimization problem (38)-(46).
S12:获得规划轨迹和优化控制序列。S12: Obtain the planned trajectory and optimized control sequence.
S13:更新车辆状态。S13: Update vehicle status.
S14:更新时间步。S14: Update time step.
S15:end for。S15: end for.
实施例2Example 2
本发明实施例2提供一种计算机设备,包括:存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序以实现实施例1方法的步骤。Embodiment 2 of the present invention provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method of embodiment 1.
该计算机设备可以是数据库,其内部结构图可以如图7所示。该计算机设备包括处理器、存储器、输入/输出接口(Input/Output,简称I/O)和通信接口。其中,处理器、存储器和输入/输出接口通过系统总线连接,通信接口通过输入/输出接口连接到系统总线。其中,该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质和内存储器。该非易失性存储介质存储有操作系统、计算机程序和数据库。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的数据库用于存储待处理事务。该计算机设备的输入/输出接口用于处理器与外部设备之间交换信息。该计算机设备的通信接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现实施例1中的方法。The computer device may be a database, and its internal structure diagram may be shown in FIG7. The computer device includes a processor, a memory, an input/output interface (Input/Output, referred to as I/O) and a communication interface. Among them, the processor, the memory and the input/output interface are connected via a system bus, and the communication interface is connected to the system bus via the input/output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The input/output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the method in Example 1 is implemented.
实施例3Example 3
一种计算机可读存储介质,其上存储有计算机程序,该计算机程序被处理器执行时实现实施例1中的方法的步骤。A computer-readable storage medium stores a computer program, which implements the steps of the method in embodiment 1 when executed by a processor.
本发明实施例提供的技术方案在高不确定性的复杂动态环境下,该方法在避障安全性和轨迹优化方面优于现有方法,具体的,该方案与现有方法相比具有以下优势:The technical solution provided by the embodiment of the present invention is superior to the existing methods in terms of obstacle avoidance safety and trajectory optimization in a complex dynamic environment with high uncertainty. Specifically, the solution has the following advantages over the existing methods:
(1)通过考虑动态障碍物的运动不确定性,对障碍物运动行为估计和不确定性序列分析,能够提升自车对障碍物运动的理解。构建运动不确定性椭圆和障碍物安全可行域,量化了障碍物车辆的运动不确定性,有助于系统更全面地考虑障碍物的运动范围,避免潜在的冲突和碰撞风险。基于对偶优化的避障约束方法加快了动态行驶环境下避障约束的构建,满足了自动驾驶车辆的实时性需求,减少了计算负担。(1) By considering the motion uncertainty of dynamic obstacles, the estimation of obstacle motion behavior and uncertainty sequence analysis can improve the vehicle's understanding of obstacle motion. The construction of the motion uncertainty ellipse and the obstacle safety feasible region quantifies the motion uncertainty of the obstacle vehicle, which helps the system to more comprehensively consider the motion range of the obstacle and avoid potential conflicts and collision risks. The obstacle avoidance constraint method based on dual optimization speeds up the construction of obstacle avoidance constraints in dynamic driving environments, meets the real-time requirements of autonomous driving vehicles, and reduces the computational burden.
(2)针对障碍物具有运动不确定性的轨迹规划问题,采用随机模型预测控制方法实现轨迹优化。综合考虑系统内部状态不确定性和外部环境不确定性,提高了规划器的鲁棒性,有助于自动驾驶车辆在复杂交通环境中的安全,高效的行驶。所提出的规划框架包含了多优化目标的代价函数,包括轨迹跟踪、舒适性和安全性等,相比于现有方法能够实现更全面的轨迹优化。这为实现更为安全和可靠的自动驾驶技术提供了创新的方法。(2) For the trajectory planning problem with motion uncertainty of obstacles, the stochastic model predictive control method is used to achieve trajectory optimization. Taking into account the uncertainty of the internal state of the system and the uncertainty of the external environment, the robustness of the planner is improved, which helps the autonomous driving vehicle to drive safely and efficiently in complex traffic environments. The proposed planning framework contains cost functions for multiple optimization objectives, including trajectory tracking, comfort, and safety, and can achieve more comprehensive trajectory optimization compared to existing methods. This provides an innovative approach to achieving safer and more reliable autonomous driving technology.
此外,上述的存储器中的计算机程序通过软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。In addition, when the computer program in the above-mentioned memory is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk.
本说明书中各个实施例采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似部分互相参见即可。The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
本发明中应用了具体个例对本发明的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处。综上所述,本说明书内容不应理解为对本发明的限制。The present invention uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims (7)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410627162.1A CN118210314B (en) | 2024-05-21 | 2024-05-21 | A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410627162.1A CN118210314B (en) | 2024-05-21 | 2024-05-21 | A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle |
Publications (2)
Publication Number | Publication Date |
---|---|
CN118210314A CN118210314A (en) | 2024-06-18 |
CN118210314B true CN118210314B (en) | 2024-07-16 |
Family
ID=91448881
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410627162.1A Active CN118210314B (en) | 2024-05-21 | 2024-05-21 | A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN118210314B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN119310986B (en) * | 2024-09-06 | 2025-06-06 | 东南大学 | Steering fault-tolerant control method for automatic driving vehicle with dynamic safety constraint |
CN119512129B (en) * | 2025-01-21 | 2025-05-06 | 新石器慧通(北京)科技有限公司 | Obstacle start-stop state detection method and system for autonomous driving |
CN119958597B (en) * | 2025-04-10 | 2025-07-15 | 北京理工大学 | Automatic driving vehicle path planning method and system under uncertain perception environment |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114995415A (en) * | 2022-05-25 | 2022-09-02 | 武汉理工大学 | Automatic driving automobile trajectory planning method based on space-time reachable set theory |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8380367B2 (en) * | 2009-03-26 | 2013-02-19 | The University Of North Dakota | Adaptive surveillance and guidance system for vehicle collision avoidance and interception |
CN116946188A (en) * | 2023-08-03 | 2023-10-27 | 西安交通大学 | A trajectory planning method and system for mixed environments |
CN117213497A (en) * | 2023-10-10 | 2023-12-12 | 北京理工大学 | AGV global path planning method based on deep reinforcement learning |
-
2024
- 2024-05-21 CN CN202410627162.1A patent/CN118210314B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114995415A (en) * | 2022-05-25 | 2022-09-02 | 武汉理工大学 | Automatic driving automobile trajectory planning method based on space-time reachable set theory |
Also Published As
Publication number | Publication date |
---|---|
CN118210314A (en) | 2024-06-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN118210314B (en) | A method, device and medium for trajectory planning and dynamic obstacle avoidance of an autonomous driving vehicle | |
Tian et al. | Safety assurances for human-robot interaction via confidence-aware game-theoretic human models | |
US11753023B2 (en) | Adaptive control of autonomous or semi-autonomous vehicle | |
CN109885891B (en) | Intelligent vehicle GPU parallel acceleration trajectory planning method | |
CN110015306B (en) | Driving trajectory acquisition method and device | |
Tang et al. | Driving environment uncertainty-aware motion planning for autonomous vehicles | |
US20240166196A1 (en) | Obstacle avoidance method, apparatus, electronic device and storage medium for vehicle | |
CN111862604B (en) | Unmanned vehicle control method and device, computer storage medium and electronic equipment | |
CN112429004B (en) | A kind of vehicle automatic lane change control method | |
CN114537435A (en) | Real-time whole vehicle track planning method in automatic driving | |
CN116182884A (en) | Local path planning method for intelligent vehicles based on frenet coordinate system horizontal and vertical decoupling | |
Zhu et al. | A Gaussian process model for opponent prediction in autonomous racing | |
Wu et al. | Continuous decision-making in lane changing and overtaking maneuvers for unmanned vehicles: A risk-aware reinforcement learning approach with task decomposition | |
Liu et al. | Refine: Reachability-based trajectory design using robust feedback linearization and zonotopes | |
CN118348998A (en) | A trajectory planning and tracking control method for dynamic obstacle avoidance of autonomous driving vehicles | |
Zhang et al. | Bi-layered synchronized optimization control with prescribed performance for vehicle platoon | |
CN120066033A (en) | Unmanned ship dynamic obstacle avoidance control system and method based on random nonlinear model predictive control | |
Xu et al. | Learning-based control barrier function with provably safe guarantees: Reducing conservatism with heading-aware safety margin | |
Zhang et al. | Control Barrier Function-Guided Deep Reinforcement Learning for Decision-Making of Autonomous Vehicle at On-Ramp Merging | |
Baumann et al. | Predictive spliner: Data-driven overtaking in autonomous racing using opponent trajectory prediction | |
Svensson et al. | Traction adaptive motion planning at the limits of handling | |
Fridovich-Keil et al. | Approximate solutions to a class of reachability games | |
CN117170357A (en) | RRT and DWA mixed path planning method based on Gaussian mixture model | |
Hu et al. | Anti-bullying Adaptive Cruise Control: A proactive right-of-way protection approach | |
Albarella et al. | A Hybrid Deep Reinforcement Learning and Optimal Control Architecture for Autonomous Highway Driving. Energies 2023, 16, 3490 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |