改进RRT算法的机械臂路径规划  

Path planning of manipulator based on improved RRT algorithm

作　　者：谭薪兴 李光[1] 易静 薛晨慷 龙厚云 TAN Xinxing;LI Guang;YI Jing;XUE Chenkang;LONG Houyun(College of Mechanical Engineering,Hunan University of Technology,Zhuzhou 412007,China)

机构地区：[1]湖南工业大学机械工程学院,湖南株洲412007

出　　处：《计算机集成制造系统》2025年第3期1014-1023,共10页Computer Integrated Manufacturing Systems

基　　金：湖南省自然科学基金资助项目(2018JJ4079)。

摘　　要：为了解决面向快速扩展随机树(RRT)方法的路径规划存在环境探索能力不足、收敛速度慢、路径质量差的问题,提出一种适用于机械臂的全局自适应步长与节点拒绝RRT路径规划算法,首先提出一种全局自适应步长的方法,根据地图中障碍物的空间大小自适应地计算初始步长,同时在扩展过程中,利用收集到的环境信息,自适应调整当前步长,实现路径规划的全局自适应步长方法,增强了对地图的探索能力;然后使用一种节点拒绝的方法,避免探索重复区域,加快了算法的收敛速度;最后对路径进行冗余节点的去除操作,缩短了路径长度,并使用三次B样条曲线对路径进行平滑处理,使得机械臂的路径更加平滑。在机械臂上进行仿真实验,结果表明:改进RRT算法的探索能力高于标准RRT、GB-RRT、RRT-Connect算法,对环境的适应性强,提高了路径搜索效率,从而验证了该算法的优越性和可行性。For the problems of insufficient environment exploration ability,slow convergence speed and poor path quality in the path planning of the Rapidly expanding Random Tree(RRT)method,a global adaptive step size and node rejection RRT path planning algorithm for robotic arms was proposed.In this algorithm,a global adaptive step size method was proposed,which adaptively calculated the initial step size according to the spatial size of obstacles in the map.At the same time,during the expansion process,the current step length was adaptively adjusted using the collected environmental information to achieve a global adaptive step length method for path planning,which was more capable of exploring the map.Then,a node rejection method was used to avoid exploring duplicate regions,and the convergence of the algorithm was speeded up.The path was subjected to a redundant node removal operation to shorten the path length,and a three-time B spline curve was used to smoothing,making the path of the robotic arm smoother.Simulation experiments on the robotic arm showed that the improved RRT algorithm had a higher exploration capability than the standard RRT,GB-RRT and RRT-Connect algorithms,was more adaptable to the environment,improved the path search efficiency,thus verified the superiority and feasibility of the algorithm.

关 键 词：机械臂 快速搜索随机树算法 路径规划 全局自适应步长 节点拒绝 

分 类 号：TP241[自动化与计算机技术—检测技术与自动化装置] TP18[自动化与计算机技术—控制科学与工程]