Behavioral detection in the maritime domain

Abstract

The maritime domain is important to the security, prosperity and vital interests of the global community. In order to protect these interests, governments require capabilities that provide situational awareness of the maritime domain. In [11] a spatiotemporal analysis approach is proposed that autonomously analyzes and classifies ship movement and possible intent at sea. The analysis focuses on detection of vessels of interest that exhibit one behavior, paralleling or following behavior. In this paper, we extend this approach by proposing a generalized semantic method that enables consideration of other behaviors of interest. Additionally we conduct a series of simulations using simulated and real AIS data to assess the performance of the algorithm to variation in behavior thresholds.