Collaborative Hunting Strategy for Multi-Robots in Complicated Underwater Environments

Abstract

Aiming at the problems of limited communication, turbulence interference, and obstacle avoidance in multiple underwater robots ‘ cooperative hunting tasks in dynamic ocean environments, a new hunting framework combining improved Artificial Potential Field (APF) and distributed consensus protocol was proposed. Firstly, the APF model was designed and optimized, which could effectively realize the path planning and obstacle avoidance of the robot. Secondly, an event-triggered dynamic topology consensus algorithm was developed, the adjacency matrix was used to adaptively adjust the communication topology, and the position matrix was formed to constrain the point position and Angle between robots. Combining the advantages and disadvantages of artificial potential field and first-order consistency theory, a more efficient hunting method was proposed, which could hunt the target faster without collision. Experimental results showed that compared with the traditional APF method, the proposed method reduced the completion time of rounding by 18.8%. This paper provides a solution with both theoretical rigor and engineering practicability for underwater swarm intelligence systems in complex ocean scenes.