Autonomous Formation Obstacle Avoidance Strategy for Multiple Mobile Robots Based on Improved DWA
Gang Hu, Yulian Jiang, Shenquan Wang, Huadi Shan · 2024
This work studies the autonomous formation obstacle avoidance problem of multi-mobile robots in complex environment. An improved dynamic window approach (DWA) algorithm is proposed, aiming at the defects of traditional DWA algorithm, such as unreasonable mechanism, serious vibration and easy to fall into local optimal. Three existing evaluation functions are modified, and two new evaluation functions are added to enhance the adaptability of the algorithm. Further, the obstacle avoidance strategies for leader and followers are designed respectively. Hence, the robots' ability to navigate towards the target and avoid obstacles are strengthened. A distributed formation controller for multiple mobile robots is designed by combining backstepping with virtual leader-follower method, and its stability is proved by Lyapunov stability theorem. Finally, the effectiveness of the proposed strategy is verified by simulation experiments. Results show that the improved DWA algorithm can solve the existing defects and improve the traffic efficiency and global search ability in the complex obstacle environment. Formation control algorithm and obstacle avoidance strategy can realize stable formation maintenance and flexible obstacle avoidance.