Detection and Route Estimation of Ship Vessels using Linear Filtering and ARMA Model from AIS Data

This paper focuses on estimating the route of a vessel and detecting abnormal behaviors based on AIS data. Linear Filtering using ARMA models is proposed in order to create trajectory forecasts. Furthermore, various alert criteria are set expressing different abnormalities. The software applica-tion created to test the proposed methodology is presented. AIS data collected from Saronikos Bay were used and the results show that the program could successfully monitor the bay in real time and report abnormalities in vessel behavior.