Vessel trajectory prediction in curving channel of inland river

The water transportation plays an important role in the world recently. However, the maritime accidents are attracting public attention. The way to improve safety of navigation has become a prior task for the operations. The trajectory is useful to analyze the features of the traffic flow and helpful to simulate traffic flows. Therefore, to explore the law of navigation in curving waterway and the reliability prediction of trajectories could provide security for the navigation of ships, and also provide the decision-making for trajectory-planning and risk warning. This paper uses Markov Chain and Grey prediction, and improves the traditional Markov model. Based on this, it offers a prediction method for island ship in curving waterway on the foundation of Automatic Identification System (AIS) data. It can be showed that this method can effectively predict the trajectory of island ship.