Summary
International Symposium on Nonlinear Theory and its Applications
2005
Session Number:3-4-4
Session:
Number:3-4-4-4
System Identification using Constrained Kalman Filters
David M. Walker,
pp.658-661
Publication Date:2005/10/18
Online ISSN:2188-5079
Summary:
We suggest incorporating dynamical information such as locations of fixed points into parameter estimation algorithms in order to improve the method of reconstructing dynamics from time series. We show how the process of reconstruction using the extended or iterated Kalman filter can be easily modified to include the additional information. We demonstrate the methods using data from the Chua circuit operating in the chaotic regime. We find the models reconstructed by using constraints can better approximate the unstable fixed point structure of the underlying systems.