Shape-Based Vessel Trajectory Similarity Computing and Clustering: A Brief Review

With the widespread application of the Automatic Identification System (AIS) in maritime management, a large amount of vessel motion information is recorded. The AIS data has become the key to maintaining vessel navigation safety and traffic supervision, and has promoted the extensive research of data mining technology applied to vessel trajectory analysis. As an unsupervised data analysis method, clustering method can effectively extract the habitual path of vessels from the AIS data and find abnormal trajectory. It is of great significance in vessel collision avoidance, behavior analysis and traffic control and so on. In this paper, the development of vessel trajectory clustering methods in recent years is reviewed, and the classic shape-based trajectory similarity computing methods are summarized. Based on two clustering algorithms, Spectral clustering and Hierarchical clustering, the AIS data in the south channel of Shanghai port is analyzed. In order to obtain the best clustering results, the performance of clustering results obtained by different methods is compared by using the clustering metrics.