− In this paper a development of an adaptive Kalman filter through a fuzzy inference system (FIS ) is outlined. The adaptation is concerned with the impo sition of conditions under which the filter measurement noise covariance matrix R or the process noise covariance matrix Q are estimated. The adaptive adjustment is carried o ut using a FIS based on the whiteness of the filter innovation sequence (IS) and employing the covariance-matching techniqu e. If a statistical analysis of the IS shows discrepancies with its expected statistics then the FIS adjusts a factor t hr ugh which the matrices R or Q are estimated. This fuzzy adaptive Kalman filter is tested on a numerical example. The results are compared with these obtained using a convention al Kalman filter and a traditionally adapted Kalman fi lter. The fuzzy-adapted Kalman filter showed better results t han its traditional counterparts.
[1]
R. Lynn Kirlin,et al.
Robust Adaptive Kalman Filtering with Unknown Inputs
,
1986,
1986 American Control Conference.
[2]
A. H. Mohamed,et al.
Adaptive Kalman Filtering for INS/GPS
,
1999
.
[3]
J. H. Oh,et al.
Gain fusion algorithm for decentralised parallel Kalman filters
,
2000
.
[4]
N. F. Toda,et al.
Divergence in the Kalman Filter
,
1967
.
[5]
Hsiao Chao-Yin.
Modified Gain Fuzzy Kalman Filtering Algorithm.
,
1999
.
[6]
R. Mehra.
On the identification of variances and adaptive Kalman filtering
,
1970
.
[7]
T. Başar,et al.
A New Approach to Linear Filtering and Prediction Problems
,
2001
.
[8]
Raman K. Mehra,et al.
Approaches to adaptive filtering
,
1970
.