基于GB＿RRT算法的机械臂路径规划

Path Planning of Manipulators Based on GB＿RRT Algorithm

针对五自由度机械臂路径规划问题,提出一种基于快速扩展随机树(rapidly-exploring random tree,RRT)优化算法—GB_RRT算法。为弥补因基本RRT算法采样盲目性导致的效率低下的缺陷,GB_RRT算法采用高斯采样的方法进行启发式采样,同时结合贪婪扩展算法来提高随机树的局部扩展速度。为进一步缩短规划路径,该算法采用双向同时剪枝取最优的策略来删除不必要的采样节点。最后对机械臂进行了仿真实验和样机实验。实验结果表明,高斯采样法结合贪婪策略不仅降低了采样的盲目性,而且能够提高扩展树的扩展速度,更好地规避开障碍物;双向剪枝取最优的策略也在一定程度上缩短了规划路径的长度。

A certain optimization rapidly-exploring random tree is proposed to solve the path planning for five DOFs manipulators-GB_RRT algorithm. The GB_RRT algorithm adopts heuristic sampling method to make up for the shortcomings of low efficiency caused by blindness of the standard RRT algorithm. It also combines with the greedy expansion algorithm to improve the local expansion speed of the random tree. And in order to further shorten the planning path, a two-direction pruning strategy is introduced to remove the unwanted sampling nodes. Finally, the simulation experiment and prototype experiment of the manipulator are carried out. And Experimental results show that, the Gaussian sampling method combined with greedy strategy not only reduces the blindness of sampling, but also improves the expansion speed and avoids the obstacles better;the two-direction pruning strategy shortens the length of planning path in a certain degree.