Research On Forecasting Ship Sailed Track Behavioral Abnormalities Algorithm Based On Kalman Filter

With the application and popularity of modern information equipments, such as Automatic Identification System (AIS), Electronic Chart Display and Information System (ECDIS), Global Positioning System (GPS) in maritime field and maritime transporting control department, there are more and more studies focusing on digging effective information from the mass data. For marine transporting control department, detecting and forecasting ship abnormal behavior is particularly important. This thesis researches on ship behavioral abnormality forecasting algorithm based on sailing trajectory and describes details of experimental platform, clustering of ship track, detecting and forecasting abnormal behaviors.In aspect of basic experimental platform, this thesis chooses international electronic chart display and information system based on S-57platform to finish the research. This section elaborates development technology of experimental platform, module classification, platform design and implementation.In aspect of digging historical track, the data source is received from data decoding process of ship automatic identification system. This thesis chooses DBSCAN algorithm as the theoretical foundation and VS2008combining with SQL Server2005as the simulation environment to dig ship history track map. The normal ship flight path map can be used as the density clustering criterion to determine whether the ship sailing is suitable.In aspect of forecasting ship behavioral abnormality, the track map which infers from density clustering, and combining Kalman Filter algorithm to forecast the ship track. The algorithm is integrated into the ECDIS to predict abnormal ship and give real-time warning. Finally, the algorithm which is used to predict ship behavioral abnormality and give real-time alarm can help the maritime transporting control department to control abnormal ships intelligently.