Bayesian Fault Detection Method for Linear Systems with Outliers

Henri Pesonen; Robert Piche

doi:10.1109/UPINLBS.2012.6409777

Bayesian Fault Detection Method for Linear Systems with Outliers

Henri Pesonen, Robert Piche

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

1 Citation (Scopus)

13 Downloads (Pure)

Abstract

A novel approach for monitoring the accuracy of the Bayesian estimate of linear Gaussian state-space model is introduced, based on the monitoring of the propagation of the errors in the Kalman filter algorithm. The effect of the sensor errors on the Kalman filter estimate is explicitly computed and compensated for. Marginalized particle filter is used to compute the posterior distribution of the sensor errors and using a target tracking simulation it is shown that the proposed method has improved performance over the standard DIA method

Translated title of the contribution	Bayesian Fault Detection Method for Linear Systems with Outliers
Original language	English
Title of host publication	Ubiquitous Positioning, Indoor Navigation and Location-Based Services, UPINLBS, 3-4 October 2012, Helsinki
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	1-5
Number of pages	5
ISBN (Electronic)	978-1-4673-1909-6
ISBN (Print)	978-1-4673-1908-9
DOIs	https://doi.org/10.1109/UPINLBS.2012.6409777
Publication status	Published - 2012
Publication type	A4 Article in conference proceedings

Publication series

Name	Ubiquitous Positioning, Indoor Navigation and Location-Based Services

Publication forum classification

No publication forum level

Access to Document

10.1109/UPINLBS.2012.6409777

Pesonen piche bayesian fault detection method
© 2012 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Cite this

@inproceedings{76b5cc71102e4e7097bc58f6b08d4e6f,

title = "Bayesian Fault Detection Method for Linear Systems with Outliers",

abstract = "A novel approach for monitoring the accuracy of the Bayesian estimate of linear Gaussian state-space model is introduced, based on the monitoring of the propagation of the errors in the Kalman filter algorithm. The effect of the sensor errors on the Kalman filter estimate is explicitly computed and compensated for. Marginalized particle filter is used to compute the posterior distribution of the sensor errors and using a target tracking simulation it is shown that the proposed method has improved performance over the standard DIA method",

author = "Henri Pesonen and Robert Piche",

note = "ei ut-numeroa 28.8.2013<br/>Contribution: organisation=mat,FACT1=1<br/>Publisher name: IEEE",

year = "2012",

doi = "10.1109/UPINLBS.2012.6409777",

language = "English",

isbn = "978-1-4673-1908-9",

series = "Ubiquitous Positioning, Indoor Navigation and Location-Based Services",

publisher = "IEEE",

pages = "1--5",

booktitle = "Ubiquitous Positioning, Indoor Navigation and Location-Based Services, UPINLBS, 3-4 October 2012, Helsinki",

address = "United States",

}

Bayesian Fault Detection Method for Linear Systems with Outliers. / Pesonen, Henri; Piche, Robert.
Ubiquitous Positioning, Indoor Navigation and Location-Based Services, UPINLBS, 3-4 October 2012, Helsinki. Piscataway, NJ: IEEE, 2012. p. 1-5 (Ubiquitous Positioning, Indoor Navigation and Location-Based Services).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Bayesian Fault Detection Method for Linear Systems with Outliers

AU - Pesonen, Henri

AU - Piche, Robert

N1 - ei ut-numeroa 28.8.2013<br/>Contribution: organisation=mat,FACT1=1<br/>Publisher name: IEEE

PY - 2012

Y1 - 2012

N2 - A novel approach for monitoring the accuracy of the Bayesian estimate of linear Gaussian state-space model is introduced, based on the monitoring of the propagation of the errors in the Kalman filter algorithm. The effect of the sensor errors on the Kalman filter estimate is explicitly computed and compensated for. Marginalized particle filter is used to compute the posterior distribution of the sensor errors and using a target tracking simulation it is shown that the proposed method has improved performance over the standard DIA method

AB - A novel approach for monitoring the accuracy of the Bayesian estimate of linear Gaussian state-space model is introduced, based on the monitoring of the propagation of the errors in the Kalman filter algorithm. The effect of the sensor errors on the Kalman filter estimate is explicitly computed and compensated for. Marginalized particle filter is used to compute the posterior distribution of the sensor errors and using a target tracking simulation it is shown that the proposed method has improved performance over the standard DIA method

U2 - 10.1109/UPINLBS.2012.6409777

DO - 10.1109/UPINLBS.2012.6409777

M3 - Conference contribution

SN - 978-1-4673-1908-9

T3 - Ubiquitous Positioning, Indoor Navigation and Location-Based Services

SP - 1

EP - 5

BT - Ubiquitous Positioning, Indoor Navigation and Location-Based Services, UPINLBS, 3-4 October 2012, Helsinki

PB - IEEE

CY - Piscataway, NJ

ER -

Bayesian Fault Detection Method for Linear Systems with Outliers

Abstract

Publication series

Publication forum classification

Access to Document

Fingerprint

Cite this