Intelligent Pursuit–Evasion Game Based on Deep Reinforcement Learning for Hypersonic Vehicles

As defense technology develops, it is essential to study the pursuit–evasion (PE) game problem in hypersonic vehicles, especially situation where a head-on scenario created. Under situation, vehicle’s speed advantage offset. This paper, therefore, establishes and model for two sides of attack defense, using twin delayed deep deterministic (TD3) gradient strategy, which has faster convergence reduces over-estimation. In view flight state–action value function, decision framework escape control based on actor–critic method constructed, solution reinforcement learning TD3 network presented. Simulation results show that proposed strategy enables vehicle evade successfully, even under an adverse scene. Moreover, programmed maneuver improved, transforming into intelligent strategy.