基于改进型P-RRT^(*)算法的机器臂路径优化  

Path Planning Based on Improved P-RRT^(*)Robot Arm

作　　者：潘振钊 王建青[1] 祝锡晶[1] 王文超 PAN Zhen-zhao;WANG Jian-qing;ZHU Xi-jing;WANG Wen-chao(Shanxi Key Laboratory of Advanced Manufacturing Technology,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学山西省先进制造技术重点实验室,太原030051

出　　处：《组合机床与自动化加工技术》2023年第3期21-24,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然基金(51975540)。

摘　　要：针对机械臂在路径搜索过程中存在逆解计算效率低、扩展无方向和路径冗长的问题,提出了一种改进的P-RRT^(*)的运动规划算法。该算法利用利用位姿分离法对求逆的过程进行优化;再其次,在节点避过障碍物后接近目标点时,设计一种概率向的目标检测来减少算法无效的搜索,缩短路径长度;在搜索完成后使用优化剪枝策略进行重联接,删除冗余的采样点和进一步缩短路径长度。为了验证算法的优势,首先在二维环境下分别对已有和改进P-RRT^(*)算法进行仿真实验和性能分析,证明了改进算法能够快速有效地到达目标点,缩短了路径长度;最后利用机械臂开展了实验探究,证明了改进算法在机械臂上应用的优势。Aiming at the problems of low inverse calculation efficiency,no direction of extension and lengthy path in the path search process of manipulator,an improved motion planning algorithm of P-RRT^(*)is proposed.The algorithm optimizes the inverse process by using the posive separation method.Secondly,when the node approaches the target point after avoiding obstacles,a probabilistic target detection is designed to reduce the search for ineffective algorithms and shorten the path length.Rejoin after the search is complete,using an optimized pruning strategy,removing redundant sample points and further reducing the path length.In order to verify the advantages of the algorithm,the simulation experiments and performance analysis of the existing and improved P-RRT^(*)algorithms are carried out in the two-dimensional environment,which proves that the improved algorithm can reach the target point quickly and effectively,and shorten the path length.Finally,the robotic arm is used to carry out experimental exploration,which proves the advantages of the improved algorithm in the application of the robotic arm.

关 键 词：机械臂 运动学 路径规划 避障 RRT 

分 类 号：TH112[机械工程—机械设计及理论] TG506[金属学及工艺—金属切削加工及机床]