Adaptive Kriging control of discrete-time nonlinear systems

Author

Wu, H. ; Sun, F.

Author_Institution

Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing

Volume

1

Issue

3

fYear

2007

fDate

5/1/2007 12:00:00 AM

Firstpage

646

Lastpage

656

Abstract

A novel adaptive Kriging controller for a class of discrete-time nonlinear systems with known control gains is proposed. The dynamic models of the systems may involve uncertainties that can be described as realizations of Gaussian random functions. The Kriging system is used to approximate the uncertainties, and an adaptive control law is established so that the tracking error remains bounded. To eliminate the infinite increase of the data size in the Kriging system, which usually occurs in non-parametric estimator-based adaptive controllers, a recursive Kriging estimation algorithm is proposed. In addition, the application of the proposed control scheme is extended to the case of uncertain control gain. Finally, some numerical examples are given to show the viability and effectiveness of the proposed control approach

Keywords

Gaussian processes; adaptive control; discrete time systems; nonlinear control systems; random processes; recursive estimation; statistical analysis; uncertain systems; Gaussian random functions; adaptive Kriging controller; adaptive control law; control gains; discrete-time nonlinear systems; dynamic models; nonparametric estimator-based adaptive controllers; recursive Kriging estimation algorithm; tracking error; uncertainties;

fLanguage

English

Journal_Title

Control Theory & Applications, IET

Publisher

iet

ISSN

1751-8644

Type

jour

Filename

4205000