快速搜索随机树(rapidly-exploring random tree,RRT)算法是智能汽车路径规划的常用方法,但传统RRT和RRT^*算法存在路径抖动大、易陷入局部区域和计算效率低等缺点。针对这些问题,本文中结合实车数据提出了一种基于安全场改进RRT^*算法...快速搜索随机树(rapidly-exploring random tree,RRT)算法是智能汽车路径规划的常用方法,但传统RRT和RRT^*算法存在路径抖动大、易陷入局部区域和计算效率低等缺点。针对这些问题,本文中结合实车数据提出了一种基于安全场改进RRT^*算法的智能汽车路径规划方法。首先,建立了基于安全距离模型的安全场,通过驾驶数据采集试验对模型关键参数进行了提取;在此基础上,提出了具备安全场引导和角度约束等策略的改进RRT^*算法;最后,通过仿真对算法进行了验证。结果表明,本文提出的路径规划方法能计算出满足车辆轨迹曲率约束的有效路径,同时具有较快的搜索速度和更高的成功率。展开更多
In order to solve the problem of path planning of mobile robots in a dynamic environment,an improved rapidly-exploring random tree^(*)(RRT^(*))algorithm is proposed in this paper.First,the target bias sampling is intr...In order to solve the problem of path planning of mobile robots in a dynamic environment,an improved rapidly-exploring random tree^(*)(RRT^(*))algorithm is proposed in this paper.First,the target bias sampling is introduced to reduce the randomness of the RRT^(*)algorithm,and then the initial path planning is carried out in a static environment.Secondly,apply the path in a dynamic environment,and use the initially planned path as the path cache.When a new obstacle appears in the path,the invalid path is clipped and the path is replanned.At this time,there is a certain probability to select the point in the path cache as the new node,so that the new path maintains the trend of the original path to a greater extent.Finally,MATLAB is used to carry out simulation experiments for the initial planning and replanning algorithms,respectively.More specifically,compared with the original RRT^(*)algorithm,the simulation results show that the number of nodes used by the new improved algorithm is reduced by 43.19%on average.展开更多
文摘快速搜索随机树(rapidly-exploring random tree,RRT)算法是智能汽车路径规划的常用方法,但传统RRT和RRT^*算法存在路径抖动大、易陷入局部区域和计算效率低等缺点。针对这些问题,本文中结合实车数据提出了一种基于安全场改进RRT^*算法的智能汽车路径规划方法。首先,建立了基于安全距离模型的安全场,通过驾驶数据采集试验对模型关键参数进行了提取;在此基础上,提出了具备安全场引导和角度约束等策略的改进RRT^*算法;最后,通过仿真对算法进行了验证。结果表明,本文提出的路径规划方法能计算出满足车辆轨迹曲率约束的有效路径,同时具有较快的搜索速度和更高的成功率。
基金National Natural Science Foundation of China(No.61903291)。
文摘In order to solve the problem of path planning of mobile robots in a dynamic environment,an improved rapidly-exploring random tree^(*)(RRT^(*))algorithm is proposed in this paper.First,the target bias sampling is introduced to reduce the randomness of the RRT^(*)algorithm,and then the initial path planning is carried out in a static environment.Secondly,apply the path in a dynamic environment,and use the initially planned path as the path cache.When a new obstacle appears in the path,the invalid path is clipped and the path is replanned.At this time,there is a certain probability to select the point in the path cache as the new node,so that the new path maintains the trend of the original path to a greater extent.Finally,MATLAB is used to carry out simulation experiments for the initial planning and replanning algorithms,respectively.More specifically,compared with the original RRT^(*)algorithm,the simulation results show that the number of nodes used by the new improved algorithm is reduced by 43.19%on average.
