Tracking of multiple surface vessels based on passive acoustic underwater arrays

This paper introduces an approach to localize and track an unknown number of non-cooperative surface vessels based on passive acoustic sensing of their noise radiated underwater. Time-Difference of Arrival (TDOA) measurements are extracted from pairs of hydrophones, and a recently introduced Bayesian framework for multi-object tracking is employed to detect and track vessels from TDOA measurements. Results based on data from a three-dimensional compact hydrophone array towed by an autonomous vehicle confirm that non-cooperative surface vessels can be detected and tracked.