Abstract:

Aiming at the current problem of insufficient intelligence for unmanned surface vehicle (USV) in solving the problem of dense environment crossing based on the central control strategy, a method of autonomous decision-making of USV swarms dense environment crossing based on local perception is proposed. The proposed method adopts the idea of self-organizing collaboration, uses the local perception ability of the USV to obtain effective information of the surrounding waters, and hands over the self-organizing formation and autonomous obstacle avoidance decision-making power to the USV to promote the swarm self-organization cross the dense environment and achieve the formation evolution from disorder to order. Simulation results show that the process of swarm dense environment crossing using this method is obviously better than similar methods based on artificial potential field. The proposed method can self-organize real-time swarm self-organizing formation, autonomous obstacle avoidance, and adaptive navigation under the premise of ensuring survival rate, which is of great significance for improving swarm adaptability.

Key words: unmanned surface vehicle (USV) swarm, local perception, dense environment, self-organized formation, safe avoidance