Robotic Arm Obstacle Avoidance Path Planning Based on Improved RRT algorithm

Abstract In order to solve the problems of low sampling efficiency, poor adaptability to fixed-step-size environments, and poor path quality of the traditional Rapidly-Exploring Random Tree (RRT) algorithm in three-dimensional environments, an enhanced RRT algorithm is put forth. Firstly, a differential sampling mechanism is suggested as a solution to the RRT algorithm’s poor sampling efficiency issue by minimizing the number of times spatial regions are sampled; secondly, an adaptive target bias and adaptive step size mechanism is proposed to adaptively adjust the target bias probability and sampling step size in combination with the spatial occupancy of the obstacles and the progress of the path planning, so as to solve the problems of poor adaptability of the fixed target bias probability and the fixed-step-size environment, to Further improve the path planning efficiency; finally, using greedy pruning strategy and B spline curve optimization strategy, the initial path is pruned and optimized, which removes the redundant nodes, smoothes the path, and improves the path quality; simulation experiments are analyzed in three-dimensional obstacle environments, and the outcomes demonstrate that the improved RRT algorithm in this work can successfully raise the quality of the path and search efficiency. The improved algorithm is applied to the robotic arm physical platform. The robotic arm can move smoothly, accurately, and quickly to the desired location., which further demonstrates the algorithm’s viability and efficacy.