基于低差异序列与快速扩展随机树融合算法的 机械臂路径规划

Manipulator path planning using fusion algorithm of low difference sequence and rapidly exploring random tree

中文关键词  机械臂  路径规划  低差异序列  快速扩展随机树  采样池

英文关键词  manipulator  path planning  low difference sequence  rapidly exploring random tree  sampling pool

中文摘要

      针对机械臂在高维关节空间下路径规划效率低的问题, 本文提出了一种基于低差异序列与快速扩展随机 树融合的路径规划算法. 该方法首次使用Sobol序列代替快速扩展随机树中的伪随机序列, 从而生成均匀差异采样 点, 且在采样过程中通过建立采样池对采样点进行优选, 提高了采样点质量和采样效率. 在此基础上, 为使规划的路 径变得光滑, 本文采用基于最小二乘法的多项式拟合方法对各关节角的离散点进行后处理. 实验部分首先在二维 空间中进行算法性能分析, 证明了本文改进的算法能够快速稳定的避开障碍物到达目标点; 最后以AUBO–i5机械 臂为原型开展了实验研究, 验证了所提算法在机械臂上应用的优势.

英文摘要

      To solve the problem of low path planning efficiency of manipulator in high-dimensional joint space, a path planning method is proposed based on Sobol sequence and rapidly exploring random tree (RRT) algorithm. The proposed method firstly adopts Sobol sequences to replace the pseudo-random sequences for generating uniformly different sampling points in RRT. Secondly, a sampling pool is built to obtain the optimal sampling points during the sampling process, which improves the sampling quality and efficiency. Aiming at smoothing the planned path, a least square based polynomial fitting method is employed to fit the discrete points of each joint angle. The performance of the proposed method is evaluated in the two-dimensional space, and the results indicate that the improved method can quickly and steadily avoid obstacles to reach the target point. Finally, using AUBO–i5 manipulator as the prototype, the experimental study is carried out to verify the advantages of the proposed method in the application of the manipulator.