Control Strategies for Target-Attacker-Defender Games of USVs

This paper presents two strategies for the target-attacker-defender (TAD) game of unmanned surface vessels (USVs) with bounded velocity and angular velocity. We use nonlinear model predictive control (NMPC) to design strategies which minimize the effort for each agent to win the game. A novel R-C-S trajectory framework is proposed to evaluate the time for USVs to reach the prearranged coordinates, and is applied to both sides' strategies of the TAD game. It is assumed that strategies of both sides are unknown to each other. The attacker's strategy is based on the dynamic artificial potential field method, which guides the attacker to evasive actions according to the threat level of the defender. The strategy for the defender and the target guides the two agents to cooperate for the overall interest of the team. The performance of the proposed algorithms is tested in numerical simulations, and it turns out that the the strategies perform better than traditional methods.