Machine Learning Techniques for Enhancing Maritime Surveillance Based on GMTI Radar and AIS

Abstract

Classical maritime surveillance systems are enhanced with disruptive elements comingf om big data and machine learning. Available receiver networks deliver a huge amount of worldwide maritime traffc data. The information includes the position as well as signifcant attributes of all vessels, which are equipped with AIS. The processing of this data lake with modern machine learning and big data techniques offer improved decision support for the user. This is especially the case, when AIS is not available and only sensor information, e.g., GMTI is gathered. New design concepts – e.g. the lambda architecture offer the modular integration of these new assets within existing surveillance systems.