Skip to content

Publications

Detail of publication

Citation

Šimandl, M. and Duník, J. : Multi-step prediction and its application for estimation of state and measurement noise covariance matrices . p. 50, University of West Bohemia, Pilsen, 2007.

Abstract

Estimation of noise covariance matrices for linear or nonlinear stochastic dynamic systems is treated. The stress is laid on the case when the initial state mean and the initial state covariance matrix are exactly known. The properties of the innovation sequence of the Kalman Filter and the Extended Kalman Filter are discussed and the new method for estimation of the covariance matrices of the state and the measurement noise is designed. The proposed method is based on special choice of the filter gain allowing the significant simplification of relations for computation of the covariance matrices of the innovation sequence and it takes an advantage of the well-known standard relations from the area of state estimation techniques and least square method. The theoretical results are verified in numerical examples.

Detail of publication

Title: Multi-step prediction and its application for estimation of state and measurement noise covariance matrices
Author: Šimandl, M. ; Duník, J.
Language: English
Date of publication: 1 Jan 2007
Year: 2007
Type of publication: Research reports
Page: 50
Publisher: University of West Bohemia
Address: Pilsen
/ /

Keywords

stochastic systems, state estimation, Kalman filtering, estimation theory

BibTeX

@MISC{SimandlM_2007_Multi-stepprediction,
 author = {\v{S}imandl, M. and Dun\'{i}k, J.},
 title = {Multi-step prediction and its application for estimation of state and measurement noise covariance matrices},
 year = {2007},
 publisher = {University of West Bohemia},
 address = {Pilsen},
 pages = {50},
 url = {http://www.kky.zcu.cz/en/publications/SimandlM_2007_Multi-stepprediction},
}