A unified intelligent control strategy synthesizing multi-constrained guidance and avoidance penetration

<正>Satisfying various terminal constraints and achieving evasive flight in no-flying-zones (NFZs) are two important flight missions undertaken by high velocity vehicles. We studied an intelligent maneuvering guidance strategy using deep reinforcement learning (DRL) methods to achieve the coordination and unification of evasive flight and multi-constrained guidance with minimal energy loss.