×

Distributed optimal fusion steady-state Kalman filter for systems with coloured measurement noises. (English) Zbl 1061.93093

Summary: Based on the optimal fusion criterion weighted by matrices in the linear minimum variance sense, an optimal information fusion steady-state Kalman filter is given for the discrete time-invariant linear stochastic control system measured by multiple sensors with coloured measurement noises, which is equivalent to an optimal information fusion steady-state Kalman predictor with a two-layer fusion structure for the system with correlated noises. Furthermore, the steady-state optimal fusion predictor can be obtained only by fusing once after all local subsystems entered the steady-state predictions. The solution of the steady-state prediction error cross-covariance matrix between any two subsystems can be obtained by iteration with an arbitratry initial value, whose convergence is proved. Applying this method to a tracking system with three sensors shows its effectiveness.

MSC:

93E11 Filtering in stochastic control theory
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Anderson BDO, Optimal Filtering (1979)
[2] Bar-Shalom Y, IEEE Transactions on Automatic Control 26 pp pp. 571–572– (1981) · Zbl 0487.93050
[3] DOI: 10.1109/7.106130 · doi:10.1109/7.106130
[4] DOI: 10.1109/TAES.2003.1238742 · doi:10.1109/TAES.2003.1238742
[5] Deng ZL, Chinese Science Abstracts 6 pp pp. 183–184– (2000)
[6] Hashmipour HR, IEEE Transactions on Automatic Control 33 pp pp. 88–93– (1988)
[7] Kim KH, Proceedings of the American Control Conference pp pp. 1037–1041– (1994)
[8] Roy S, IEEE Transactions on Aerospace and Electronic Systems 27 pp pp. 276–288– (1991)
[9] DOI: 10.1109/7.640278 · doi:10.1109/7.640278
[10] Sun SL, Automatica 40 pp pp. 1017–1023– (2004)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.