Dynamic Modeling of Water Depth and Navigation Decision-Making Methods in Shallow Waters Considering Tidal Effects
With the rapid advancement of intelligent shipping, autonomous navigation in complex and confined waters has become a critical challenge. This study aims to develop a robust autonomous navigation decision-making method to address the combined effects of tidal variations, water depth gradients, and restricted maneuvering ability in shallow waters. A digital traffic environment is constructed by fusing real-time automatic identification system (AIS) data with electronic chart display and information system (ECDIS) information and incorporating tidal effects, thereby enabling spatiotemporal situational awareness for autonomous navigation decision-making. The methodology quantitatively interprets collision avoidance rules and navigational best practices to determine optimal maneuvering thresholds for typical encounter scenarios in restricted waters. By coupling ship kinematic characteristics with bathymetric features, a three-dimensional ship domain model is developed, incorporating squat effects and under-keel clearance requirements, whereas a risk quantification algorithm accounts for water depth gradient transitions. The experimental results show that this method performs reliably in complex shallow waters. The proposed perception-decision-execution-feedback framework enables rapid information updates and allows the system to adapt to uncoordinated actions of target ships, handle residual errors, maintain a safe distance between ships, and reduce potential collision risk. A virtual-real integrated scenario based on AIS and ECDIS data is established to systematically validate the proposed method. It provides reliable theoretical and methodological support for the theoretical research and engineering application of autonomous navigation technology in complex shallow waters.