Heterogeneous information fusion for multitarget tracking using the sum-product algorithm

Abstract

The sum-product algorithm (SPA) was recently shown to provide a scalable methodology for multitarget tracking (MTT) using multiple sensors. Here, we focus on another advantage of the SPA frame-work, namely, its capacity for Bayesian fusion of heterogeneous data sources and auxiliary information. We develop extensions of the SPA-based multisensor MTT algorithm that integrate data from an auxiliary surveillance system and geographic information about standard target routes. The effectiveness of our approach is demonstrated for a simulated scenario and for a real maritime scenario.