基于改进RRT^(*)算法的机械臂避障路径规划  

Path Planning for Manipulator Obstacle Avoidance Based on Improved RRT^(*)Algorithm

作　　者：李丁 张宇[1] 金皓 邓竣碧 李泰泉 LI Ding;ZHANG Yu;JIN Hao;DENG Junbi;LI Taiquan(Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学机电工程学院,昆明650500

出　　处：《组合机床与自动化加工技术》2024年第8期6-12,共7页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：云南省重点研发计划项目(2018BA070)。

摘　　要：针对渐进最优快速扩展随机树(RRT^(*))算法在机械臂避障路径规划中存在随机性较大、效率低、路径不光滑等缺点,提出了一种基于目标导向策略并结合双向扩展的改进RRT^(*)算法。在传统RRT^(*)算法基础上,添加一个目标偏置函数,增加目标点的采样概率,同时引入双向扩展机制,加速扩展过程。在扩展新节点时,进行重复性检测,删除重复节点,首次找到路径后采用椭球子空间采样策略,缩小采样空间,最后采用缩短路径策略和B样条优化路径。在MATLAB中仿真结果表明,相比RRT^(*)算法,搜索时间提升76.9%,规划路径缩短5.6%,采样节点数减少86.87%,平均路径节点数减少45.45%,机械臂顺利平滑避障,且运动过程中关节处参数无突变。在实体机械臂进行避障实验,进一步证实了该算法在实际应用中的可行性。Aiming at the problems of high randomness,low efficiency,and non-smooth paths in path planning for manipulator obstacle avoidance using the rapidly exploring random tree^(*)(RRT^(*))algorithm,an improved RRT^(*)algorithm based on a goal-directed strategy and incorporating bidirectional expansion is proposed.A target bias function was incorporated on the basis of the traditional RRT^(*)algorithm to increase the sampling probability of the target point.Simultaneously,a bidirectional expansion mechanism was introduced to accelerate the expansion process.During the expansion of new nodes,duplicate node detection and removal are performed.After finding the path for the first time,an ellipsoidal subspace sampling strategy is employed to reduce the sampling space.Finally,a path shortening strategy and B-spline optimization are applied.Simulation results in MATLAB demonstrate that compared to the RRT^(*)algorithm,the proposed approach achieves a 76.9%improvement in search time,a 5.6%reduction in planned path length,an 86.87%decrease in sampled nodes,and a 45.45%decrease in average path nodes.The robotic arm successfully achieves smooth obstacle avoidance,with no sudden changes in joint parameters during motion.Importantly,the proposed algorithm ensures successful obstacle avoidance for the manipulator,maintains smooth joint parameters during motion.Further validates its feasibility through physical experiments with a real manipulator.

关 键 词：机械臂 路径规划 椭球采样 RRT 目标导向 

分 类 号：TH112[机械工程—机械设计及理论] TG659[金属学及工艺—金属切削加工及机床]