High-altitude satellites range scheduling for urgent request utilizing reinforcement learning

High-altitude satellites are visible to more ground station antennas for longer periods of time, its requests often specify an antenna set and optional service windows, consequently leaving huge scheduling search space. The exploitation of reinforcement learning techniques provides a novel approach to the problem of high-altitude orbit satellite range scheduling. Upper sliding bound of request pass was calculated, combining customized scheduling strategy with overall antenna effectiveness, a frame of satellite range scheduling for urgent request using reinforcement learning was proposed. Simulations based on practical circumstances demonstrate the validity of the proposed method.