A two-stage ship path planning method based on trial maneuvers

Ship path planning is a core technology in achieving fully autonomous ship navigation. This paper proposes a novel path planning method based on ship trial maneuvers for nearshore waters. The method consists of two main stages: global route planning and local path planning. For global route planning, ocean environment numerical models are employed to forecast nearshore hydrodynamic environments. Then, the improved Dijkstra algorithm is developed to plan the shortest travel-time routes, followed by key waypoint extraction using Douglas-Peucker (DP) algorithm. For local path planning, collision avoidance strategy in close-quarters single-ship encounters is primary focus. The predefined trial maneuver strategy is established considering International Regulations for Preventing Collisions at Sea (COLREGs). By integrating hydrodynamic environmental data with a ship motion model, trajectories with different maneuvers strategies are simulated. These trajectories are assessed for collision risk to select the safest maneuver. The key advantage of this method is its integration of critical dynamic constraints for ship navigation in real sea, which is often neglected in traditional methods. Case studies demonstrate the effectiveness of this method, showing that the proposed method effectively plans feasible paths for ships in nearshore waters, offering important support for navigation safety and autonomous ship development.