Affiliation:
1. Heibei University of Architecture
Abstract
The generalized Kalman filter (GKF) is a method which is based on a generalized linear model. The coefficients of the linear model are fixed, and are not needed to be distinguished on-line or predefined. The GKF can be used to estimate the real-time estate of the continuous system, especially; the GKF can be used to estimate the estate of the mobile objective. The GKF is based on the minimum mean-square error. The recursive formula of the generalized kalman filter was also presented. Different from the EKF published on the magazines, the GKF doesn’t need to build the system model, so its ability of calculate and accuracy is improved.
Publisher
Trans Tech Publications, Ltd.
Reference4 articles.
1. Benveniste, A. Design of adaptive algorithms for the tracking of time-varying systems [J]. Int. J. Adaptive Control Signal Proc. 1987. 1.
2. Bitmad, R. R. Convergence properties of LMS adaptive estimators with unbounded dependent inputs [J]. IEEE Trans. Autom. Control, 1984, AC-29.
3. Boray, G. K., and M. D. Srinath. Conjugate techniques for adaptive filtering [J]. IEEE Trans. Circuits Syst., Fundam. 1992, Appli, 39.
4. Butterweck, H. J. A wave theory of long adaptive filters [J]. IEEE Transactions on Circuits and System-1: Fundamental Theory and Applications. 2001, 48.