Abnormal Behavior Detection and Risk Analysis of Ships in the Yangtze River Estuary

The Yangtze River Estuary is one of the shipping hubs in China. In order to ensure the safety of ship navigation, it is necessary to study the detection of abnormal behavior of ships. In this paper, the abnormal behavior of ships is identified through the study of ship behavior, and the potential accident risk factors are analyzed. Firstly, the route is identified by trajectory compression and clustering algorithm. Then, according to the distribution of the ship’s lateral position, the abnormal behavior of the ship’s offset is detected. Finally, the risk model is established by using the dynamic Bayesian network, and the influence of the detected abnormal behavior on the potential accident risk is analyzed. According to the AIS data of ships in the Yangtze River Estuary of China, it is found that the Nangang Shuidao of the Yangtze River Estuary is prone to the risk of abnormal behavior of ships. The results show that the course over ground and ship heading have a great influence on the risk of the abnormal behavior of the ship’s lateral deviation, and the improper control of the ship’s speed causes the abnormal behavior of the ship’s lateral deviation.