Mission Planning for Collaborative Operations of Autonomous Marine Systems

Autonomous systems consisting of multiple surface and underwater vehicles are capable of performing long-term, high level tasks through cooperation. However, achieving reliable autonomy under substantial uncertaies and environmental changes requires a system that can recognize the situation and make high-level decisions. This study presents a mission planner developed using Planning Domain Definition Language (PDDL) with a replanning structure that incorporates three methods to enhance both optimality and robustness. The planner employs three methods to achieve this: the greedy iterative method improves the planner’s optimality, the model adaptive reduces the gap between the environment and the model to increase operational performance, and the reserved enhances mission success rates by allowing the planner to adapt to changes in the environment. We demonstrate that the proposed mission planner can provide an efficient and stable plan to complete a mission.