DocumentCode :
3269465
Title :
On a problem of time-varying learning rate influence on the adaptive system dynamics
Author :
Tyukin, Ivan ; Van Leeuwen, Cees ; Prokhorov, Danil ; Terekhov, Valery
Author_Institution :
Lab. for Preceptual Dynamics, RIKEN Brain Sci. Inst., Saitama, Japan
Volume :
4
fYear :
2002
fDate :
10-13 Dec. 2002
Firstpage :
4718
Abstract :
We consider a special class of nonlinear adaptive systems with full-state feedback and time-varying learning or adaptation rate. For the given class of systems we introduce sufficient conditions which guarantee boundedness of system trajectories and an additional assumption to guarantee asymptotic reaching of the control goal. These conditions extend the certainty-equivalence or attainability condition for gradient adaptive rules.
Keywords :
adaptive systems; asymptotic stability; learning systems; nonlinear systems; state feedback; time-varying systems; adaptive system dynamics; certainty equivalence; full state feedback; gradient adaptive rules; nonlinear systems; system trajectories boundedness; time varying learning rate; Adaptive systems; Automatic control; Chaos; Control systems; Laboratories; Nonlinear dynamical systems; Stability; State feedback; Sufficient conditions; Time varying systems;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Decision and Control, 2002, Proceedings of the 41st IEEE Conference on
ISSN :
0191-2216
Print_ISBN :
0-7803-7516-5
Type :
conf
DOI :
10.1109/CDC.2002.1185123
Filename :
1185123
Link To Document :
بازگشت