Reachability-Informed Reinforcement Learning for Multi-Impulse Interplanetary Transfers
Reachability Analysis-Informed Reinforcement Learning (RARL) is a method that uses local geometry to guide decisions in multi-impulse interplanetary transfers. It combines learned waypoint selection with classical astrodynamics to minimize total fuel use while respecting impulse limits. The approach achieves high performance and policy reuse across different departure conditions.