An Optimization Method for Collaborative Radar Antijamming Based on Multi-Agent Reinforcement Learning

Attacking a naval vessel with multiple missiles is an important way to improve the hit rate of missiles. Missile-borne radars need complete detection and antijamming tasks guide missiles, but communication between these often difficult. In this paper, optimization method based on multi-agent reinforcement learning proposed for collaborative against one vessel. We consider as player make their confrontation two-person zero-sum game. With temporal constraints radar’s jammer’s recognition preparation interval, game focuses taking favorable position at end confrontation. It assumed total jamming capability shipborne jammer constant limited, allocates in direction according radar threat assessment result its probability successful detection. The work collaboratively through prior centralized training obtain good performance by decentralized execution. can collaborate detect vessel, rather than only considering each itself. Experimental results show that paper effective, improving winning 10% 25% two-radar four-radar scenarios, respectively.