Application of Improved Rapidly-exploring Random Trees (RRT) algorithm for Obstacle Avoidance of Snake-like Manipulator

摘要

With the development of science and technology, more and more redundant manipulators have been applied in various fields. Aim to realize autonomous obstacle avoidance in unstructured environment for snake-like manipulator, an improved Rapidly-exploring Random Trees (RRT) algorithm is proposed. Establish the Kinematic model of snake-like manipulator and obstacle. Collision detection is also performed. In order to improve the efficiency of the algorithm, build together multiple random trees and generate a valid path without collision. Next the path is optimized to make the path smoother. Finally, simulation is performed in MATLAB environment to verify the effectiveness and practicability of the algorithm. This method has important practical significance for the redundant manipulator.