DocumentCode
3226707
Title
Robust NLq neural control theory
Author
Suykens, J.A.K. ; De Moor, B. ; Vandewalle, J.
Author_Institution
Dept. of Electr. Eng., Katholieke Univ., Leuven, Belgium
Volume
4
fYear
1997
fDate
9-12 Jun 1997
Firstpage
2396
Abstract
We present sufficient conditions for global asymptotic stability and I/O stability (finite L2-gain) of multilayer recurrent neural networks, with parametric uncertainties upon the interconnection matrices. This is done by considering perturbed NLq systems. NLqs are discrete time nonlinear dynamical systems in state space form, containing q layers with alternating linear and static nonlinear operators that satisfy a sector condition. Within the present framework, linear fractional transformations with real diagonal uncertainty block are interpreted as perturbed NLqs with q=1. While in μ robust control theory uncertainties upon nominal linear models are investigated, uncertainties upon nominal nonlinear models can be studied in NLq neural control theory
Keywords
asymptotic stability; discrete time systems; input-output stability; multilayer perceptrons; neurocontrollers; nonlinear dynamical systems; recurrent neural nets; robust control; state-space methods; I/O stability; discrete time nonlinear dynamical systems; finite L2-gain; global asymptotic stability; interconnection matrices; linear fractional transformations; multilayer recurrent neural networks; nominal nonlinear models; parametric uncertainties; real diagonal uncertainty block; robust NLq neural control theory; state space form; sufficient conditions; Asymptotic stability; Control theory; Multi-layer neural network; Nonlinear dynamical systems; Recurrent neural networks; Robust control; Robustness; State-space methods; Sufficient conditions; Uncertainty;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks,1997., International Conference on
Conference_Location
Houston, TX
Print_ISBN
0-7803-4122-8
Type
conf
DOI
10.1109/ICNN.1997.614443
Filename
614443
Link To Document