Unmanned surface vessel cluster cooperative replenishment task scheduling strategy via GA-CFG algorithm

The current trend in unmanned surface vessel (USV) is towards diversified and heterogeneous cluster development. This paper studies the resupply strategy for USV when the number of vessels to be resupplied greatly exceeds the number of supply ships. A cooperative algorithm combining genetic algorithms with an improved alliance formation game is proposed. Firstly, a cost function is designed based on the mobility of USV and the duration of berthing for resupply, considering the Euclidean distance with the supply platform. Genetic algorithms are used to generate an initial optimized scheduling plan according to the resupply task requirements. Secondly, an improved alliance formation rule is designed to adjust the members within the alliances formed by the genetic algorithm, achieving a Nash stable state for each alliance. Simulation experiments show the feasibility and superiority of the proposed strategy, providing a reasonable and efficient resupply scheduling plan for the resupply allocation process in heterogeneous vessel clusters.