Dynamic path planning based on the fusion of improved RRT and DWA algorithms

Abstract

A hybrid algorithm combining improved RRT algorithm and dynamic window method is proposed to solve the path planning problem of robots in the environment of dynamic or unknown obstacles. Firstly, aiming at the problems of many redundant nodes and low efficiency in the RRT algorithm, the target bias sampling and parent node reselection strategies are used to improve them. Secondly, for the problems of poor environmental adaptability and difficulty in taking into account speed and safety in the dynamic window method, a DWA algorithm with fuzzy control of parameters is proposed, which dynamically adjusts the weight of the evaluation function according to the distance of the robot from the target and obstacle. Finally, the above improved RRT algorithm and fuzzy DWA algorithm are mixed to realize dynamic path programming. By comparing the simulation results, the effectiveness of the hybrid algorithm in the complex dynamic environment is verified.