Collision Avoidance Path Planning Method Combining Improved DWA and A* Algorithms in Complex Sea Environments

This paper proposes a collision avoidance path planning method that combines the improved DWA algorithm with the A* algorithm to address navigation challenges in complex sea environments. First, a simplified grid map is constructed based on electronic chart data and the start and end points of the path planning, while the physical information of the ship is determined. The improved DWA algorithm is then used for collision avoidance path planning. When the ship reaches a safe distance from an obstacle, the improved A* algorithm is triggered to plan a path, and redundant points in the returned path are extracted to obtain key waypoints. These waypoints are used as temporary target points for the DWA algorithm, and the improved DWA algorithm is further applied for collision avoidance path planning. The improvements in the DWA algorithm mainly focus on the trajectory evaluation function, while the enhancements in the A* algorithm focus on the heuristic function. Finally, by combining the proposed algorithm with rich multi-source marine environment data and target fusion data, the ship's intelligent collision avoidance decision-making system is able to generate the optimal collision avoidance path in complex water environments. Simulation results demonstrate that the proposed method can effectively plan a safe and efficient collision avoidance path.