Research on collision avoidance route planning based on A-star algorithm

Abstract

The increasing complexity of maritime traffic necessitates advanced collision avoidance systems for autonomous navigation. Traditional path planning algorithms often struggle with dynamic obstacle prediction and adherence to navigational regulations. To address these challenges, this study introduces a collision avoidance framework that integrates the Obstacle Zone by Target (OZT) method with an improved A-star algorithm. This paper proposes a combination of the OZT method and an improved A-star algorithm, which incorporates a decision-making module and segmented path planning to generate COLREGs-compliant avoidance routes, optimizing route safety and smoothness. The effectiveness of this system was verified through simulations, and the results demonstrate the system’s ability to generate collision-free paths in multi-ship scenarios while considering and complying with COLREGs. Compared to the traditional A-star algorithm, the proposed approach reduced path length by an average of 5.92% and decreased turning points by 79.44% in give-way situations, significantly improving navigation efficiency and path smoothness.