Path Planning of Mobile Robot Based on the Improved RRT-Connect Algorithm

摘要

To address the issues of traditional RRT-Connect algorithms in global path planning, such as low exploration efficiency, slow convergence speed, and excessive path waypoints. In this paper, an improved RRT-Connect path planning algorithm is proposed. To enhance exploration efficiency, the algorithm constructs a generalized Voronoi diagram (GVD) to analyze obstacle distribution in the environment, which in turn generates multiple intermediate guide nodes to improve exploration efficiency. In addition, the algorithm combines the goal bias strategy and the concept of attractive field to assist the generation of new nodes to improve the path convergence speed. To ensure smoothness of the path, a two-way pruning strategy is used to eliminate redundant nodes on the initial path. Ultimately, the paths are trajectory optimized using the cubic spline interpolation algorithm so that they satisfy the kinematic constraints of the robot. The algorithm proposed in this paper will be compared with the traditional RRT-Connect algorithm with, RRT algorithm, and RRT* algorithm in two different environments. The results show a significant improvement over the traditional RRT-Connect algorithm.