复杂环境下六自由度机械臂路径规划的Biased-RRT修正算法

Biased-RRT modified algorithm for path planning of a 6-DOF manipulator in complex environment

为解决复杂环境下六自由度机械臂的路径规划问题,提出一种基于采样规则目标导向设计、父节点重选的Biased-RRT修正算法.该算法在原目标偏置策略的基础上对随机采样点的选取规则进行重新设定,引导算法搜索树在尽可能向目标区域扩展的同时有效避开复杂障碍物.在节点扩展方面,依据新节点距离目标点的远近采用变步长扩展方式,即在距离远时选用大步长,加快搜索树扩展;进入目标区域后选用小步长,防止节点扩展陷入局部死循环.在路径优化方面,本算法通过引入基于路径代价最小的重选父节点操作及多余路径节点剔除操作,使规划出的路径相对优化.最后,利用3次样条插值技术为机械臂各关节规划出一条光滑、连续且无障的运动曲线.仿真结果表明,本算法可有效缩短路径规划时间、减少路径长度,较好地完成了复杂环境下六自由度机械臂的预期路径规划任务.

To solve the path planning problem of a 6-DOF manipulator in complex environment,a Biased-RRT modified algorithm based on the target-oriented design of the sampling rule and the reselection of parent node was proposed.On the basis of the original target bias strategy,the selection rules of random sampling points were reset in the algorithm to guide the search tree to expand to the target area as much as possible while effectively avoiding complex obstacles.In terms of node expansion,the variable step length expansion method was adopted according to the distance between the new node and the target point,that is,when the distance was long,a large step length was selected to speed up the expansion of the search tree.A small step length was selected after entering the target area to prevent the node expansion from falling into local death cycle.In terms of path optimization,the planned path was relatively optimized in the proposed algorithm by reselecting the parent node based on the minimum path cost and the removal of redundant path nodes.Finally,the cubic spline interpolation technique was used to plan a smooth,continuous and obstacle-free motion curve for each joint of the manipulator.The simulation results show that the proposed algorithm can effectively shorten the path planning time,reduce the path length,and better complete the expected path planning task of the 6-DOF manipulator in complex environment.