Passive 3D multitarget tracking using multiple heterogeneous sensors

Abstract

In this paper we investigate a multi-hypothesis algorithm for tracking multiple underwater targets in the three dimensional (3D) space using two acoustic sensor arrays attached to the same underwater platform. One of the arrays measures the bearing to the target and the other array, consisting of three spatially distributed hydrophones, measures two time difference of arrivals of the acoustic wave from the target. The scenario we consider consists of an unknown number of targets with measurement origin uncertainty due to missed detections and false alarms. The objective is to estimate the number of targets in the surveillance region and their kinematic states, where the challenge is to accurately estimate the depths of the targets. The algorithm that we propose is a multiple hypothesis tracker based on the multidimensional assignment algorithm. The simulation results demonstrate that even when only vague prior information is available, the proposed algorithm accurately estimates the number of targets and their states including the depth.