The Vessels Trajectory Clustering And Its Applications Based On AIS

Under the situation of the globalization and integration of the world economy, shipping industry is undergoing tremendous changes. The big trend of standardization in management, trade, loading and unloading, triggers a constant change of ship building technology, bigger and bigger of ship building, and more and more ship types. However, it also leads more burdens in port, waterway and sea route. Because of the ships own problems (such as failure in automatic rudder, misalignment in compass and radar, etc.) and human factor (such as no safe speed, the ship retrograde, etc.) maritime accidents are increasing, which result in huge economic losses. So exploring, discovering and solving abnormal problems of shipping, become an important research subject for current shipping industry. As clustering is a base to solve the anomaly, cluster analysis is performed on the ship trajectory. The main work of this paper include:1. Decoding and processing of AIS data. On the introduction of AIS information types, information content, update time basis and analysis of components of original AIS information and the main sentence structure, we decode the AIS information. Deal with the four common situations after decoding, and get the clean AIS information.2. Clustering ships’ tracks according to clustering algorithm based on distance track structure. Form the track structure distance, based on evaluation and analysis of various spatial data under the method of measurement. Realize part of the ships trajectory clustering of Qiongzhou Strait, by clustering algorithm based on distance track structure3. Apply the trajectory clustering in the fields of ship dynamic monitoring, anomaly detection, and route planning. In ECDIS simulator, do the dynamic monitoring simulation of the speed, position and course of the ships in the clustering area; and introduce applications of trajectory clustering in the anomaly detection and route planning.