A fuzzy inference system for ship-ship collision alert generation

Ship-ship collisions are rare events mainly caused by human errors. In an ever growing global trade world, the avoidance of such accidents is of great importance as they may cause significant human, environmental, and financial consequences. A collision warning system providing to Maritime related authorities (Vessel Traffic Services, Coast Guard authorities, Port authorities, other) real-time alerts for ship-ship encounter may significantly contribute to the reduction of collision accidents and the increase of navigation safety. In this paper a Fuzzy inference system (FIS) is proposed to estimate risk collision for pairs of ships receiving as input parameters calculated based on data obtained from the Automatic Identification System (AIS) which is installed onboard ships. The proposed system operates in real time examining pairs of ships that recently transmitted an AIS signal while discards, at an early processing stage, those pairs which do not satisfy specific conditions in order to reduce computational burden. Experiments have been carried out with real data and the results show that the proposed approach is effective in automatic alert generation for ships involved in near ship-ship situations.