Concurrent Detection and Tracking using Multiple, Flying, Sensors

Abstract

We develop a probabilistic technique for performing multiple target detection and tracking based on data from multiple, flying, sensors. Multiple sensors can facilitate detecting and discriminating low signal-to-clutter targets by allowing correlation between different sensor types and/or different aspect angles. However, the data association problem can cause the computational complexity of standard trackers to become prohibitively high when combining too much data - a problem which will be exacerbated when including data from multiple sensors. Dynamic logic (DL) is a probabilistic technique for performing data association, based upon maximum likelihood parameter estimation of mixture models, which does not suffer from a computational explosion with increasing amounts of data. Previously, a DL-based tracker was developed for the relatively simple case incorporating data from a stationary sensor platform. In this paper we expand the framework to incorporate multiple, moving, sensor platforms, which requires a revision in the parameter estimation equations. The framework is general enough to be valid for different sensor types, for example radar or electro-optical. Sample results from synthetic data are presented