Robust Particle Filtering via Bayesian Nonparametric Outlier Modeling

TL;DR

This paper introduces a robust particle filtering method that models outliers with a Bayesian nonparametric approach, improving state estimation accuracy in noisy systems with frequent outliers.

Contribution

It proposes a novel DPM-based particle filter that adaptively models outliers without prior assumptions, enhancing robustness in nonlinear dynamic systems.

Findings

Outperforms existing methods in presence of frequent outliers

Adapts model complexity to data automatically

Demonstrates significant accuracy improvements in simulations

Abstract

This paper is concerned with the online estimation of a nonlinear dynamic system from a series of noisy measurements. The focus is on cases wherein outliers are present in-between normal noises. We assume that the outliers follow an unknown generating mechanism which deviates from that of normal noises, and then model the outliers using a Bayesian nonparametric model called Dirichlet process mixture (DPM). A sequential particle-based algorithm is derived for posterior inference for the outlier model as well as the state of the system to be estimated. The resulting algorithm is termed DPM based robust PF (DPM-RPF). The nonparametric feature makes this algorithm allow the data to "speak for itself" to determine the complexity and structure of the outlier model. Simulation results show that it performs remarkably better than two state-of-the-art methods especially when outliers appear frequently along time.