Change search
ReferencesLink to record
Permanent link

Direct link
Rao-Blackwellised particle methods for inference and identification
Linköping University, Department of Electrical Engineering, Automatic Control. Linköping University, The Institute of Technology.
2011 (English)Licentiate thesis, monograph (Other academic)
Abstract [en]

We consider the two related problems of state inference in nonlinear dynamical systems and nonlinear system identification. More precisely, based on noisy observations from some (in general) nonlinear and/or non-Gaussian dynamical system, we seek to estimate the system state as well as possible unknown static parameters of the system. We consider two different aspects of the state inference problem, filtering and smoothing, with the emphasis on the latter. To address the filtering and smoothing problems, we employ sequential Monte Carlo (SMC) methods, commonly referred to as particle filters (PF) and particle smoothers (PS).

Many nonlinear models encountered in practice contain some tractable substructure. If this is the case, a natural idea is to try to exploit this substructure to obtain more accurate estimates than what is provided by a standard particle method. For the filtering problem, this can be done by using the well-known Rao-Blackwellised particle filter (RBPF). In this thesis, we analyse the RBPF and provide explicit expressions for the variance reduction that is obtained from Rao-Blackwellisation. Furthermore, we address the smoothing problem and develop a novel Rao-Blackwellised particle smoother (RBPS), designed to exploit a certain tractable substructure in the model.

Based on the RBPF and the RBPS we propose two different methods for nonlinear system identification. The first is a recursive method referred to as the Rao-Blackwellised marginal particle filter (RBMPF). By augmenting the state variable with the unknown parameters, a nonlinear filter can be applied to address the parameter estimation problem. However, if the model under study has poor mixing properties, which is the case if the state variable contains some static parameter, SMC filters such as the PF and the RBPF are known to degenerate. To circumvent this we introduce a so called “mixing” stage in the RBMPF, which makes it more suitable for models with poor mixing properties.

The second identification method is referred to as RBPS-EM and is designed for maximum likelihood parameter estimation in a type of mixed linear/nonlinear Gaussian statespace models. The method combines the expectation maximisation (EM) algorithm with the RBPS mentioned above, resulting in an identification method designed to exploit the tractable substructure present in the model.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press , 2011. , 172 p.
Linköping Studies in Science and Technology. Thesis, ISSN 0280-7971 ; 1480
National Category
Engineering and Technology
URN: urn:nbn:se:liu:diva-68078Local ID: LiU-TEK-LIC-2011:19ISBN: 978-91-7393-173-1 (print)OAI: diva2:416071
2011-06-01, Visionen, Hus B, Campus Valla, Linköpings universitet, Linköping, 13:15 (Swedish)
Available from: 2011-05-10 Created: 2011-05-10 Last updated: 2011-05-10Bibliographically approved

Open Access in DiVA

Rao-Blackwellised particle methods for inference and identification(5375 kB)1647 downloads
File information
File name FULLTEXT01.pdfFile size 5375 kBChecksum SHA-512
Type fulltextMimetype application/pdf
cover(163 kB)22 downloads
File information
File name COVER01.pdfFile size 163 kBChecksum SHA-512
Type coverMimetype application/pdf

Search in DiVA

By author/editor
Lindsten, Fredrik
By organisation
Automatic ControlThe Institute of Technology
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 1647 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 9103 hits
ReferencesLink to record
Permanent link

Direct link