一种基于随机路径规划的实时避障搜寻路径算法  

A Search Path Algorithm for Real-time Obstacle Avoidance Based on Randomized Path Planning

作　　者：王慧斌 金向阳[1] 丁刚[2] 于信鑫 高北辰 WANG Huibin;JIN Xiangyang;DING Gang;YU Xinxin;GAO Beichen(School of Light Industry,Harbin University of Commerce,Harbin 150028,China;School of Mechanical and Electrical Engineering,Harbin Institute of Technology,Harbin 150006,China)

机构地区：[1]哈尔滨商业大学轻工学院,哈尔滨150028 [2]哈尔滨工业大学机电工程学院,哈尔滨150006

出　　处：《机械工程师》2025年第3期42-47,共6页Mechanical Engineer

摘　　要：在未知环境搜寻任务的路径规划中,现有算法无法自动搜寻并且寻找最短路径。为了解决这一问题,基于快速随机树(Rapidly-exploring Random Tree Star,RRT*)算法提出了快速随机树改进(Rapidly-exploring Random Trees Progress,RRTpro)算法。RRTpro算法选取最外层节点成为父节点,同时对失败的节点进行记录,构建出概率函数,优化算法逻辑,对算法进行应用化处理。通过虚拟试验,RRTpro生成树的节点分布更均匀,在计算速度上RRTpro耗时为RRT*的51.16%,在生成路径质量上RRTpro生成的路径长度为RRT*生成路径的96.1%。在测试小车仿真应用中,RRTpro在测试环境中自动搜寻并找到了目标点,并生成了最优的路径。In path planning for search tasks in unknown environments,existing algorithms cannot automatically search and find shortcuts.To solve this problem,this paper proposes the Rapidly-exploring Random Trees Progress(RRTpro)algorithm based on the Rapidly-exploring Random Trees Star(RRT)algorithm.The RRTpro algorithm selects the outermost nodes as parent nodes while recording failed nodes and constructing a probability function to optimize the algorithm's logic and apply it in practice.Through virtual experiments,the tree generated by RRTpro has a more uniform node distribution,with a computation time that is 51.16%of RRT and a path length that is 96.1%of the path generated by RRT.In a simulated vehicle application,RRTpro automatically searches for and finds the target point in the test environment,generating the optimal path.

关 键 词：路径规划 RRT*算法 RRTpro算法 实时检测 目标搜寻 

分 类 号：TP391.7[自动化与计算机技术—计算机应用技术]