Dynamic neural networks for output feedback control

Author

Hovakimyan, Naira ; Rysdyk, Rolf ; Calise, Anthony J.

Author_Institution

Sch. of Aerosp. Eng., Georgia Inst. of Technol., Atlanta, GA, USA

Volume

2

fYear

1999

fDate

6/21/1905 12:00:00 AM

Firstpage

1685

Abstract

A dynamic neural network is designed to estimate velocities from displacement measurements for a nonlinear system. A linear-in-parameters NN is used for state reconstruction. Conditions are provided under which the estimation error is guaranteed to be ultimately bounded. Subsequently, this observer is integrated into an adaptive controller architecture. The controller is based on model inversion and is augmented with a second learning-while-controlling neural network. Conditions are derived which guarantee ultimate boundedness of all the errors in the combined observer-controller feedback system. Open loop and closed loop simulations for a Van Der Pol oscillator are used to illustrate the results

Keywords

adaptive control; closed loop systems; displacement measurement; feedback; neural nets; nonlinear control systems; relaxation oscillators; Van Der Pol oscillator; adaptive controller architecture; closed loop simulations; displacement measurements; dynamic neural networks; learning-while-controlling neural network; linear-in-parameters neural net; model inversion; nonlinear system; open loop simulations; output feedback control; state reconstruction; Adaptive control; Displacement measurement; Estimation error; Neural networks; Nonlinear dynamical systems; Nonlinear systems; Observers; Open loop systems; Output feedback; Programmable control;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 1999. Proceedings of the 38th IEEE Conference on

Conference_Location

Phoenix, AZ

ISSN

0191-2216

Print_ISBN

0-7803-5250-5

Type

conf

DOI

10.1109/CDC.1999.830266

Filename

830266