WSEAS Transactions on Mathematics
Print ISSN: 1109-2769, E-ISSN: 2224-2880
Volume 13, 2014
Ridge-Type Kalman Filter and Its Algorithm
Authors: , , , ,
Abstract: Kalman filter is one of common ways of dealing with dynamic data under linear minimum variance estimation. However, in practice, the observation matrix maybe has multicollinearity. Therefore, the impact of multicollinearity on Kalman filter is studied in this paper. Firstly, by analyzing the normal equation of Kalman filter, we proposed sufficient conditions about how the multicollinearity affects the estimator. Secondly, a ridge- type Kalman filter algorithm is designed under the mean square error (MSE), and then the characters of the new algorithm are analyzed. Thirdly, six specific methods for determining the ridge parameters of ridge-type Kalman filter are proposed based on the canonical form of normal equation. Finally, examples illustrate the new algorithms can overcome the influence of the ill-condition on Kalman filter effectively, which improves the accuracy of the estimates of parameters.
Search Articles
Pages: 852-862
WSEAS Transactions on Mathematics, ISSN / E-ISSN: 1109-2769 / 2224-2880, Volume 13, 2014, Art. #83