Abstract
A vessel must navigate along designated routes within a harbor area to ensure navigation safety. The impact of strong currents is one of the most dangerous factors in coastal navigation. However, it is challenging to determine the deviation of a ship in advance from the ship’s position data in the case of a marine accident. In this study, to support the decision-making of ship navigators and vessel traffic service (VTS) operators in track monitoring tasks, tracks were classified according to the tidal stream, and the track distribution was analyzed according to the tidal current situations. Marine accident analysis was performed to investigate the tidal influence on ship tracks. Track data were collected for 12 months from a VTS center in Korea, and tidal information was collected through a meteorological observation buoy. Representative tracks were extracted from the track data using the support vector regression (SVR) seaway model. K-fold cross-validation and a grid search were performed to determine the optimal parameters. The ship tracks appeared in specific patterns according to the forces and directions of tidal currents, and specific deviation patterns were observed. This study is expected to contribute to the reduction of marine accidents by predicting ship trajectories according to the tidal situations in advance.