Diffusive particle filtering for distributed multisensor estimation

Abstract

The paper proposes an on-line distributed implementation of the particle filter (DPF) for applications, where the sensing and consensus time scales are the same. We are motivated by state estimation problems in large, geographically-distributed agent/sensor networks, where bandwidth constraints limit the number of information transfers between neighbouring nodes. As an alternative to consensus strategies often used by the DPF, we propose a diffusive framework to eliminate the need of running the consensus step. In our Monte Carlo simulations, the proposed diffusion based DPF (D/DPF) outperforms the state-of-the-art consensus based DPF approaches in environments with limited bandwidth or/and intermittent connectivity.

Compressed Monte Carlo with application in particle filtering

“…Therefore, if we are interested only in one specific integral involving π(x), we can obtain a perfect compression by choosing the summary particles as in Eq. (37). With the choice in Eq.…”

“…In this section, we consider L independent computational nodes where the Monte Carlo computation is performed in parallel. In the literature, specific techniques have been designed for providing a distributed or diffused inference depending on whether a central node is available or not, respectively [37], [14], [17]. Here, we focus on a centralized distributed framework, i.e., we consider a central node where the transmitted local information is properly combined, as represented in Figure 2.…”