An Anomaly Detection Method for AIS Trajectory Based on Kinematic Interpolation

With the enormous amount of information provided by ship Automatic Identification System (AIS), AIS is now playing a significant role in maritime transport system-related research and development. Many kinds industrial applications are based on trajectory extracted from raw data. However, due to issues equipment, transmission environment, human factors, data inevitably contain abnormal messages, which have hindered utilization such practice. Thus, this paper, an anomaly detection method that focuses proposed, making comprehensive use kinematic The employs three steps obtain non-error trajectories: (1) preprocessing, (2) estimation, (3) error clustering. It should be noted involved iterative process determine all A case study then conducted test proposed real-world data, followed comparison between rule-based method. As processed trajectories show fewer features, results indicate improves performance can accurately detect as much possible.