This paper explores the groundbreaking potential of reinforcement learning (RL) in revolutionizing missile guidance systems. We delve into three distinct scenarios where RL’s innovative application is particularly promising: firstly, in enhancing a missile’s ability to evade interception during midcourse flight; secondly, in navigating complex terrains and avoiding maritime obstacles, crucial for anti-shipping purposes and evading air defense systems; and thirdly, in its capability to perform effectively in unknown environments, demonstrating improved guidance over traditional methods. Through these scenarios, we illustrate how RL can be a game changer in the field of missile guidance, offering advanced adaptability, precision, and effectiveness. The aim of this review is to highlight the potential of RL to transform missile guidance technologies, ushering in a new era of intelligent and autonomous missile systems.
Yang Chaojie, Wu Jiang, Liu Guoqing, Zhang Yuncan. Ballistic Missile Maneuver Penetration Based on Reinforcement Learning. 2018 IEEE CSAA Guidance, Navigation and Control Conference (CGNCC). 2018. IEEE. [Cross Ref]
Hong Daseon, Park Sungsu. Avoiding Obstacles via Missile Real-Time Inference by Reinforcement Learning. Applied Sciences. Vol. 12(9)2022. MDPI AG. [Cross Ref]
Gaudet Brian, Furfaro Roberto. Missile Homing-Phase Guidance Law Design Using Reinforcement Learning. AIAA Guidance, Navigation, and Control Conference. 2012. American Institute of Aeronautics and Astronautics. [Cross Ref]