Research on the decision-making method for autonomous navigation for the ocean-going ship in the ships' routeing waters

Ships' routeing, designed to regulate and guide navigation, significantly mitigates maritime accidents and is widely implemented worldwide. Research into autonomous navigation decision-making in such waters is critical for advancing maritime intelligence and ensuring safe and efficient operations. This paper presents a novel autonomous collision avoidance framework, structured around the perception-decision-execution sequence. Through the digital twin traffic environment, the environment information and ship state are detected. Then an enhanced velocity obstacle (EVO) algorithm is proposed specifically developed for collision avoidance in ships' routeing waters, which incorporates ship manoeuvrability and channel boundary constraints. Route tracking method based on navigation practice is utilized considering the characteristics of the environment. To validate the effectiveness of the decision-making system, two realistic scenarios involving the ships' routeing system in the Yangtze River Estuary, a region known for its high traffic density, are presented. The results demonstrate that the proposed system is highly effective in ensuring safe navigation, even in complex, multi-ship environments, and provides a robust solution for collision avoidance in these challenging waters.