Cooperative obstacle avoidance for multiple dance robots based on improved DWA and leader-follower method

The demand for cooperative obstacle avoidance among multiple dance robots is increasing rapidly. However, traditional methods rely on fixed rules, which fail to respond to dynamic obstacles in real time and overlook the nonlinear coupling of movements, leading to insufficient coordination. This paper proposes a cooperative obstacle avoidance method for multiple dance robots based on an improved Dynamic Window Approach algorithm and the Leader-Follower method. The proposed model efficiently adapts to dynamic environments and enhances both safety and motion smoothness during cooperative obstacle avoidance. By integrating the advantages of dynamic path planning from the improved Dynamic Window Approach algorithm and coordinated control from the Leader-Follower method, the model enables real-time path generation and motion synchronization to achieve precise multi-robot obstacle avoidance. Experimental results show that the proposed model achieves a minimum obstacle avoidance success rate of 87.3% and a minimum multi-robot coordination success rate of 90.0%. In practical scenarios, it reaches a maximum memory usage of only 47.1% and a maximum delay of just 32.7 ms, outperforming comparison models. Additionally, it predicts movement timing conflicts and sensor failure risks with an accuracy of up to 99.0%, demonstrating strong robustness and practical applicability. These findings indicate that the proposed model performs exceptionally well in cooperative obstacle avoidance for multiple dance robots. It effectively addresses the limitations of delayed response and poor coordination found in existing methods, offering an innovative solution and technical foundation for intelligent obstacle avoidance, while enhancing the fluency and safety of robot dance performances.