USV path planning algorithm based on 20D-A∗ and IDWA algorithms under COLREGs

Abstract

The path planning of Unmanned Surface Vehicles(USV) represents a pivotal technology for autonomous navigation. However, prevailing USV path planning methodologies frequently overlook the constraints imposed by the International Regulations for Preventing Collisions at Sea (COLREGs), resulting in certain deficiencies in the corresponding planned paths. To address this issue, a hybrid path planning algorithm integrating 20D-A∗ and improved dynamic window approach(IDWA) is proposed. The proposed algorithm enables USVs to comply with COLREGs requirements for collision avoidance when encountering other vessels. The 20D-A∗ algorithm performs an initial path planning based on the starting and ending points. When the USV encounters other vessels while navigating along the planned path, IDWA autonomously executes collision avoidance actions in compliance with COLREGs. This involves locally adjusting the planned path and resuming the original path once the avoidance actions have been completed. The experimental results demonstrate that the proposed method demonstrates high efficiency and obstacle avoidance performance in complex environments. It effectively addresses the impact of static obstacles and navigating vessels on the USV while satisfying the collision avoidance requirements specified by COLREGs. This research offers a novel solution for USV path planning in real-world navigation scenarios, showcasing promising application prospects.