Abstract
This article proposes a Rao-Blackwellized particle filter (RBPF) for fully mixing state-space models that replace the Kalman filter within the RBPF method with a noise-adaptive Kalman filter. This extension aims to deal with unknown time-varying measurement variances. Consequently, a variational Bayesian (VB) adaptive Kalman filter estimates the conditionally linear states and the measurement noise variances whereas the nonlinear (or latent) states are handled by sequential Monte Carlo (MC) sampling. Thus, by modifying the underlying mathematical framework of RBPF, we construct the Monte Carlo Variational Bayesian (MCVB) filter. A stopping criterion for VB approximations is proposed by employing Tikhonov regularization. Additionally, an analysis of the numerical stability of the proposed filtering mechanism is presented. The performance of the MCVB filter is illustrated in simulations and mobile robot tracking experiments in the presence of measurement model uncertainties.
| Original language | English |
|---|---|
| Pages (from-to) | 1-11 |
| Number of pages | 11 |
| Journal | IEEE Transactions on Aerospace and Electronic Systems |
| DOIs | |
| Publication status | E-pub ahead of print - 5 Jun 2024 |
| MoE publication type | A1 Journal article-refereed |
Keywords
- Adaptation models
- Adaptive kalman filter
- Bayes methods
- Kalman filters
- monte carlo methods
- Monte Carlo methods
- Noise
- particle filter
- rao–blackwellization
- State-space methods
- tikhonov regularization
- tracking
- variational bayesian methods
- Vectors