Sigma Point Belief Propagation

TL;DR

This paper introduces Sigma Point Belief Propagation (SPBP), a low-complexity method extending sigma point filters to loopy factor graphs for efficient decentralized Bayesian inference.

Contribution

It proposes SPBP as a novel approximation of belief propagation that is computationally efficient and suitable for decentralized inference in complex networks.

Findings

SPBP outperforms particle-based BP in sensor localization tasks.

SPBP requires less computation and communication than traditional BP methods.

SPBP effectively approximates posterior marginals in loopy graphs.

Abstract

The sigma point (SP) filter, also known as unscented Kalman filter, is an attractive alternative to the extended Kalman filter and the particle filter. Here, we extend the SP filter to nonsequential Bayesian inference corresponding to loopy factor graphs. We propose sigma point belief propagation (SPBP) as a low-complexity approximation of the belief propagation (BP) message passing scheme. SPBP achieves approximate marginalizations of posterior distributions corresponding to (generally) loopy factor graphs. It is well suited for decentralized inference because of its low communication requirements. For a decentralized, dynamic sensor localization problem, we demonstrate that SPBP can outperform nonparametric (particle-based) BP while requiring significantly less computations and communications.