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 $\delta$-GLMB, marginalized $\delta$-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.

This paper addresses multi-agent multi-object tracking with labeled random finite sets via Generalized Covariance Intersection (GCI) fusion. While standard GCI fusion of Labeled Multi-Object (LMO) densities is labelwise and hence fully parallelizable, previous work unfortunately revealed that its fusion performance is highly sensitive to the unavoidable label inconsistencies among different agents. In order to overcome the label inconsistency sensitivity problem, we present a novel approach for the GCI fusion of LMO densities that is both robust to label inconsistencies andcomputationally efficient. Thenovel approach consists of: (1) first finding the best matching between labels of different agents by minimization of a suitable label inconsistency indicator; (2) then performing GCI fusion labelwise according to the obtained label matching. Further, it is shown how the label matching problem, which is at the core of the proposed method, can be formulated as a linear assignment problem of finite length (efficiently solvable in polynomial time by the Hungarian algorithm), exactly for Labeled Multi-Bernoulli (LMB) densities and approximately for arbitrary LMO densities. Simulation experiments are carried out to demonstrate the robustness and effectiveness of the proposed approach in challenging tracking scenarios.

The joint task of detecting attacks and securely monitoring the state of a cyber-physical system is addressed over a cluster-based network wherein multiple fusion nodes collect data from sensors and cooperate in a neighborwise fashion in order to accomplish the task. The attack detection-state estimation problem is formulated in the context of random set theory by representing joint information on the attack presence/absence, on the system state and on the attack signal in terms of a Hybrid Bernoulli Random Set (HBRS) density. Then, combining previous results on HBRS recursive Bayesian filtering with novel results on Kullback-Leibler averaging of HBRSs, a novel distributed HBRS filter is developed and its effectiveness is tested on a case-study concerning wide-area monitoring of a power network.

This paper presents an exact Bayesian filtering solution for the multiobject tracking problem with the generic observation model. The proposed solution is designed in the labeled random finite set framework, using the product styled representation of labeled multiobject densities, with the standard multiobject transition kernel and no particular simplifying assumptions on the multiobject likelihood. Computationally tractable solutions are also devised by applying a principled approximation involving the replacement of the full multiobject density with a labeled multi-Bernoulli density that minimizes the Kullback-Leibler divergence and preserves the first-order moment. To achieve the fast performance, a dynamic-grouping-procedure-based implementation is presented with a step-by-step algorithm. The performance of the proposed filter and its tractable implementations are verified and compared with the state of the art in numerical experiments.