基于改进RRT的关节机械臂路径规划  

Joint Robotic Arm Path Planning Based on Improved RRT

作　　者：张婷婷 柳林燕[1] 汪惠芬[1] ZHANG Ting-ting;LIU Lin-yan;WANG Hui-fen(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing,210094,China)

机构地区：[1]南京理工大学机械工程学院,南京210094

出　　处：《组合机床与自动化加工技术》2022年第5期11-14,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(51705256)。

摘　　要：针对传统的快速随机扩展树(RRT)全局规划算法在搜索路径时间长,最终路径不平滑等问题,提出了一种基于目标导向和冗余检测的搜索算法。改进的算法采用生成的随机点更加偏向于目标点,避免产生更多不必要的随机树,再通过冗余检测,剔除不必要的节点。使用B样条曲线平滑路径,让机械臂可以快速、准确且平稳地沿着最佳路径运动到达目标点。仿真表明,目标导向处理后平均时间减少44%,冗余检测处理后平均距离缩短18.5%。In view of the problems such as the long search path time and the smoothness of the final path of the traditional Rapidly-Exploring Random Tree(RRT)global planning algorithm,a search algorithm based on target-oriented and redundant detection is proposed.The improved algorithm uses the resulting random points to be more biased towards the target point,avoiding the creation of more unnecessary random trees,and then through redundant detection,eliminating unnecessary nodes.Smooth paths with B-splines allow the robot arm to move quickly,accurately,and smoothly along the optimal path to reach the target point.Simulation showed that the average time after target-oriented processing was reduced by 44%and the average distance after redundant detection processing was reduced by 18.5%.

关 键 词：机械臂 RRT 目标导向 冗余检测 路径平滑 

分 类 号：TH112[机械工程—机械设计及理论] TG502[金属学及工艺—金属切削加工及机床]