关键词: 机械臂避障, 路径规划, 快速路径规划, 双树RRT算法, 引斥力模型

Abstract: In the process of automatic operation of robot arm,static or dynamic obstacles usually appear,and real-time obstacle avoidance path planning has become one of the urgent problems to be solved. Aiming at the problems such as Rapidly exploring Random Tree (RRT) algorithm long time comsumption,uncontrollable heading to the target point,and poor executability of path planning results,according to the idea that branches always face the sun and avoid obstacles in the real world,an obstacle avoidance path planning algorithm for RRT manipulator with gravitation and repulsive force is presented. According to the environment,the step size is changed adaptively,the gravitation and repulsive force model is integrated,and the stem stop growth strategy is added,so that the path planning calculation time of the robot arm is significantly shortened in the actual scene. In addition,the cost function is used to constrain the growth direction of the random tree to further improve the feasibility of the path. Then,the two-tree connection strategy is used to constrain the heading to the target point. Finally,a joint experiment of a robotic arm is carried out to verify the effectiveness of the proposed algorithm.

Key words: mechanical arm obstacle avoidance, path planning, fast path planning, double tree RRT algorithm, gravitation and repulsive force model