Decentralized, Adaptive, Look-Ahead Particle Filtering

TL;DR

This paper enhances the decentralized particle filter by introducing a look-ahead approach with Monte Carlo approximation and adaptive configuration using bandit algorithms, leading to improved efficiency and broader applicability.

Contribution

It develops a more efficient look-ahead DPF with Monte Carlo approximation of the optimal proposal and introduces bandit algorithms for automatic state space decomposition.

Findings

Look-ahead DPF outperforms standard particle filter on a single machine.

Monte Carlo approximation improves proposal distribution efficiency.

Bandit algorithms effectively automate state space decomposition.

Abstract

The decentralized particle filter (DPF) was proposed recently to increase the level of parallelism of particle filtering. Given a decomposition of the state space into two nested sets of variables, the DPF uses a particle filter to sample the first set and then conditions on this sample to generate a set of samples for the second set of variables. The DPF can be understood as a variant of the popular Rao-Blackwellized particle filter (RBPF), where the second step is carried out using Monte Carlo approximations instead of analytical inference. As a result, the range of applications of the DPF is broader than the one for the RBPF. In this paper, we improve the DPF in two ways. First, we derive a Monte Carlo approximation of the optimal proposal distribution and, consequently, design and implement a more efficient look-ahead DPF. Although the decentralized filters were initially designed to capitalize on parallel implementation, we show that the look-ahead DPF can outperform the standard particle filter even on a single machine. Second, we propose the use of bandit algorithms to automatically configure the state space decomposition of the DPF.