Rao-Blackwellized Gaussian Smoothing

Roland Hostettler; Simo Särkkä

doi:10.1109/TAC.2018.2828087

Rao-Blackwellized Gaussian Smoothing

Roland Hostettler, Simo Särkkä

Research output: Contribution to journal › Article › Scientific › peer-review

19 Citations (Scopus)

338 Downloads (Pure)

Abstract

In this paper, we consider Rao-Blackwellization of linear substructures in sigma-point-based Gaussian assumed density smoothers. We derive marginalized prediction, smoothing, and update steps for the mixed linear/nonlinear Gaussian state-space model as well as for a hierarchical model for both conventional and iterated posterior linearization Gaussian smoothers. The proposed method is evaluated in a numerical example and it is shown that the computational complexity is reduced considerably compared to non-Rao--Blackwellized Gaussian smoothers for systems with high-dimensional linear subspaces.

Original language	English
Article number	8340820
Pages (from-to)	305-312
Number of pages	8
Journal	IEEE Transactions on Automatic Control
Volume	64
Issue number	1
Early online date	18 Apr 2018
DOIs	https://doi.org/10.1109/TAC.2018.2828087
Publication status	Published - 1 Jan 2019
MoE publication type	A1 Journal article-refereed

Keywords

Computational modeling
Density measurement
Gaussian assumed density smoothing
Kalman filters
Linear regression
Noise measurement
Nonlinear smoothing
Nonlinear state estimation
Rao-Blackwellization
Smoothing methods
State-space methods

Access to Document

10.1109/TAC.2018.2828087

ELEC_Hostettler_Sarkka_Rao-Blackwellized_IEEETraAutCon_64_authoracceptedmanuscriptAccepted author manuscript, 326 KB

Cite this

@article{160b35c0ff5c4bb4802dac590da690dc,

title = "Rao-Blackwellized Gaussian Smoothing",

abstract = "In this paper, we consider Rao-Blackwellization of linear substructures in sigma-point-based Gaussian assumed density smoothers. We derive marginalized prediction, smoothing, and update steps for the mixed linear/nonlinear Gaussian state-space model as well as for a hierarchical model for both conventional and iterated posterior linearization Gaussian smoothers. The proposed method is evaluated in a numerical example and it is shown that the computational complexity is reduced considerably compared to non-Rao--Blackwellized Gaussian smoothers for systems with high-dimensional linear subspaces.",

keywords = "Computational modeling, Density measurement, Gaussian assumed density smoothing, Kalman filters, Linear regression, Noise measurement, Nonlinear smoothing, Nonlinear state estimation, Rao-Blackwellization, Smoothing methods, State-space methods",

author = "Roland Hostettler and Simo S{\"a}rkk{\"a}",

year = "2019",

month = jan,

day = "1",

doi = "10.1109/TAC.2018.2828087",

language = "English",

volume = "64",

pages = "305--312",

journal = "IEEE Transactions on Automatic Control",

issn = "0018-9286",

publisher = "IEEE",

number = "1",

}

TY - JOUR

T1 - Rao-Blackwellized Gaussian Smoothing

AU - Hostettler, Roland

AU - Särkkä, Simo

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we consider Rao-Blackwellization of linear substructures in sigma-point-based Gaussian assumed density smoothers. We derive marginalized prediction, smoothing, and update steps for the mixed linear/nonlinear Gaussian state-space model as well as for a hierarchical model for both conventional and iterated posterior linearization Gaussian smoothers. The proposed method is evaluated in a numerical example and it is shown that the computational complexity is reduced considerably compared to non-Rao--Blackwellized Gaussian smoothers for systems with high-dimensional linear subspaces.

AB - In this paper, we consider Rao-Blackwellization of linear substructures in sigma-point-based Gaussian assumed density smoothers. We derive marginalized prediction, smoothing, and update steps for the mixed linear/nonlinear Gaussian state-space model as well as for a hierarchical model for both conventional and iterated posterior linearization Gaussian smoothers. The proposed method is evaluated in a numerical example and it is shown that the computational complexity is reduced considerably compared to non-Rao--Blackwellized Gaussian smoothers for systems with high-dimensional linear subspaces.

KW - Computational modeling

KW - Density measurement

KW - Gaussian assumed density smoothing

KW - Kalman filters

KW - Linear regression

KW - Noise measurement

KW - Nonlinear smoothing

KW - Nonlinear state estimation

KW - Rao-Blackwellization

KW - Smoothing methods

KW - State-space methods

UR - http://www.scopus.com/inward/record.url?scp=85045769849&partnerID=8YFLogxK

U2 - 10.1109/TAC.2018.2828087

DO - 10.1109/TAC.2018.2828087

M3 - Article

AN - SCOPUS:85045769849

SN - 0018-9286

VL - 64

SP - 305

EP - 312

JO - IEEE Transactions on Automatic Control

JF - IEEE Transactions on Automatic Control

IS - 1

M1 - 8340820

ER -

Rao-Blackwellized Gaussian Smoothing

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Crowdsourced mapping of the environment- multimodal real-time SLAM via combinedinertial, optical, and magnetic sensoring

Sequential Monte Carlo Methods for State and Parameter Estimation in Stochastic Dynamic Systems

Cite this

Rao-Blackwellized Gaussian Smoothing

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Projects

Crowdsourced mapping of the environment- multimodal real-time SLAM via combinedinertial, optical, and magnetic sensoring

Sequential Monte Carlo Methods for State and Parameter Estimation in Stochastic Dynamic Systems

Cite this