关键词: 高超声速再入滑翔飞行器, 协同制导, 元学习, 强化学习, 近端策略优化

Abstract: For the cooperative guidance problem of high-hypersonic re-entry gliding vehicles to hit targets at a specified angle in a complex environment, a cooperative guidance law based on meta-learning and reinforcement learning algorithms is proposed. Considering the interference of complex combat environments, a Markov decision model for the cooperative guidance problem is established, taking gliding vehicles’ motion status and proportional guidance factor as the state space and action space. A reward function is designed by comprehensively considering the vehicle-target distance, remaining flight time difference, and overload situation for multiple gliding vehicles attacking the target. Based on meta-learning theory and reinforcement learning algorithms, this study combines proximal policy optimization algorithms with gated recurrent units to learn the common features of similar cooperative guidance tasks. This approach enhances the accuracy of cooperative guidance strategies in complex interference environments, achieving constraints on attack angles and attack timing, while also improving the adaptability of cooperative guidance strategies to different combat scenarios. Simulation results indicate that this cooperative guidance law enables multiple aircraft to simultaneously attack targets at specified attack angles in complex battlefield environments and quickly adapts to new cooperative guidance tasks, maintaining good performance even when the cooperative combat scenario changes.

Key words: hypersonic re-entry gliding vehicle, cooperative guidance, meta learning, reinforcement learning, proximal policy optimization