DocumentCode :
285225
Title :
Stochastic approximation for neural network weight estimation in the control of uncertain nonlinear systems
Author :
Spall, James C. ; Cristion, John A.
Author_Institution :
Appl. Phys. Lab., Johns Hopkins Univ., Laurel, MD, USA
Volume :
3
fYear :
1992
fDate :
7-11 Jun 1992
Firstpage :
930
Abstract :
The use of neural networks for controlling a nonlinear system with unknown process equations is considered. To make such an approach practical, it is necessary that connection weights in the neural network be estimated. The use of a new stochastic approximation algorithm for this weight estimation that is based on a simultaneous perturbation gradient approximation is considered. It is shown that this algorithm can greatly improve on the efficiency of more standard stochastic approximation algorithms based on finite-difference gradient approximations
Keywords :
approximation theory; finite difference methods; neural nets; nonlinear control systems; stochastic processes; connection weights; finite-difference gradient approximations; neural network weight estimation; simultaneous perturbation gradient approximation; stochastic approximation; uncertain nonlinear systems; unknown process equations; Adaptive control; Approximation algorithms; Control systems; Intelligent networks; Neural networks; Nonlinear control systems; Nonlinear equations; Nonlinear systems; Physics; Stochastic processes;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1992. IJCNN., International Joint Conference on
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-0559-0
Type :
conf
DOI :
10.1109/IJCNN.1992.227080
Filename :
227080
Link To Document :
بازگشت