Research on Path Planning of Railway Freight Car Bottom Inspection Robot Arm Based on Improved RRT

Abstract

Aiming at the problems of densely distributed parts, long detection time and low path efficiency of traditional path planning RRT algorithm (Rapidly Exploring Random Tree) in detecting the bottom of railway freight car, an improved RRT path planning method is proposed to improve detection efficiency. Firstly, the end of the manipulator is kept a certain distance from the target to be detected by improving the random point extension time choice strategy, so as to ensure the safety of the detection process so as to make the detection processes safer; Then, on the basis of adding target bias, the path search method at the starting and ending points of the random tree was optimized, so that the path search process was concentrated in the center of the path; Finally, redundant nodes are removed from the searched path to shorten the planned path. Simulations were conducted on MATLAB and VREP platforms, and the results showed that the improved RRT path planning algorithm was more helpful than traditional RRT algorithms and could effectively complete detection tasks.