Path Planning for Wheeled Robots Based on the Fusion of Improved A* and TEB Algorithms

Abstract

In this paper, we propose a hierarchical path planning method to address the issues of unsmooth path curves in autonomous navigation for wheeled robots using the original A * algorithm, and the adverse impact on robots caused by jumping changes in output commands of the original TEB algorithm. Firstly, we evaluate the curvature using the curvature smoothing function in the A * algorithm, followed by introducing interpolation to replace some of the unsmoothed path nodes to enhance path smoothness. Secondly, we add jerk constraints on multiple nodes in the TEB algorithm to reduce the impact of acceleration jumps in the output on the mechanical structure of the robot body, thereby improving the operation efficiency of the local path planning algorithm. Finally, we integrate the improved A * algorithm with the TEB algorithm to enable the robot to follow global path points better while bypassing obstacles on time. Simulation experiments demonstrate that the proposed method significantly improves global path smoothing, output speed changes more smoothly, and the wheeled robot's motion smoothness is enhanced.