Neural-Fuzzy Inference Engine for Maritime Anomaly Detection

In an era of heightened security concerns, a vast§amount of data concerning vessel activity is§collected. However, the ability for aggregation and§real-time analysis for every vessel in a port or§waterway is limited by availability of personnel and§human capability for real-time analysis. An automated§decision tool can alleviate some of the processing§burden by providing maritime personnel with detection§alerts that indicate unusual vessel activity and§direct personnel focus to problem areas. The system§is referred to as VASE (Vessel Anticipatory§Situation-Awareness Engine).VASE provides an§automated detection tool that uses spatial and§kinematics indicators to detect unusual vessel§activity. The system overlays machine learning§techniques on maritime data streams to provide alerts§of unusual indicators usable by maritime personnel.§These indicators include unusual changes in speed,§course, or proximity to other maritime objects. With§these alerts, maritime personnel would be able to§determine the appropriate course of action, such as§contacting a vessel, allocating commandable sensors§to a vessel of interest, or performing further§investigation with other sources.