Allocation Strategy Optimization Using Repulsion-Enhanced Quantum Particle Swarm Optimization for Multi-AUV Systems

In the context of multi-autonomous underwater vehicle (multi-AUV) operations, the target assignment is addressed as a multi-objective allocation (MOA) problem. The selection of strategy for multi-AUV target allocation is dependent on the current non-cooperative environment. This paper establishes a multi-AUV allocation situation advantage evaluation system to assess and quantify the non-cooperative environment. Based on this framework, a multi-AUV target allocation model using a bi-matrix game theory is developed, where multi-AUV target allocation strategies are considered as part of the strategic framework within the game. The payoff matrix is constructed based on factors including the situational context of multi-AUV operations, effectiveness, and AUV operational integrity. The Nash equilibrium derived from the game analysis serves as the optimal solution for resource distribution in multi-AUV non-cooperative scenarios. To address the challenge of finding the Nash equilibrium in bi-matrix games, this paper introduces a repulsion process quantum particle swarm optimization (RPQPSO) algorithm. This method not only resolves the complexities of Nash equilibrium computation but also overcomes the limitations of traditional optimization methods that often converge to local optima. A simulation experiment of multi-AUV operations is designed to validate the multi-AUV target allocation model, demonstrating that the RPQPSO algorithm performs effectively and is applicable to multi-AUV task scenarios.