Abstract: This paper proposes a novel Hierarchical Deep Reinforcement Learning (HRL) framework for wake homing torpedo guidance, applying the Discrete Event System Specification (DEVS) formalism to ...