基于Informed-RRT 的苹果采摘机械臂路径规划研究

Path Planning of Apple Picking Robot Arm Based on Informed RRT

针对非结构化复杂环境下采摘机器人成功率低、规划时间长等问题,提出了一种基于informed-RRT的改进采摘运动规划算法。在改进算法中,采用P概率采样取代随机采样,提高采样的目标性,动态步长生成子节点。改进算法提高了Informed-RRT算法探索未知空间的速度和灵活性,提高最优路径的收敛速度。二维仿真实验表明,与Informed-RRT相比,改进算法可将初始路径查询更短,成功率更高。通过三维仿真实验可以看出,提出的改进采摘机械臂规划算法,实现了快速的路径查询,提高了规划查询率,降低了索引的盲目性,验证了该算法的有效性与优越性。

Aiming at the problems of low success rate and long planning time of harvesting robots in unstructured and complex environments,an improved harvesting motion planning algorithm based on informed RRT*is proposed.In the improved algorithm,P probability sampling is used instead of random sampling to improve the targeting of sampling and generate sub nodes with dynamic step size.The improved algorithm has increased the speed and flexibility of the Informed RRT algorithm in exploring unknown spaces,and improved the convergence speed of the optimal path.Two dimensional simulation experiments show that compared with Informed RRT*,the improved algorithm can shorten the initial path query and have a higher success rate.Through 3D simulation experiments,it can be seen that the improved harvesting robotic arm planning algorithm proposed in this paper achieves fast path queries,improves planning query rates,reduces blind indexing,and verifies the effectiveness and superiority of the algorithm.