Robust Distributed Fusion With Labeled Random Finite Sets

This paper considers the problem of the distributed fusion of multiobject posteriors in the labeled random finite set filtering framework, using a generalized covariance intersection (GCI) method. Our analysis shows that GCI fusion with labeled multiobject densities strongly relies on label consistencies between local multiobject posteriors at different sensor nodes, and hence suffers from a severe performance degradation when perfect label consistencies are violated. Moreover, we mathematically analyze this phenomenon from the perspective of the principle of minimum discrimination information and the so-called yes-object probability. Inspired by the analysis, we propose a novel and general solution for the distributed fusion with labeled multiobject densities that is robust to label inconsistencies between sensors. Specifically, the labeled multiobject posteriors are first marginalized to their unlabeled posteriors, which are then fused using the GCI method. We also introduce a principled method to construct the labeled fused density and produce tracks formally. Based on the developed theoretical framework, we present tractable algorithms for the family of generalized labeled multi-Bernoulli (GLMB) filters including δ-GLMB, marginalized δ-GLMB, and labeled multi-Bernoulli filters. The robustness and efficiency of the proposed distributed fusion algorithm are demonstrated in challenging tracking scenarios via numerical experiments.