Consensus-based multiple-model Bayesian filtering for distributed tracking

This study addresses distributed state estimation of jump Markovian systems and its application to tracking of a manoeuvring target by means of a network of heterogeneous sensors and communication nodes. Two novel consensus-based multiple-model filters are presented. Simulation experiments in a tracking case study, involving a strongly manoeuvring target and a sensor network characterised by weak connectivity, demonstrate the superiority of the proposed distributed multiple-mode filters with respect to existing solutions.