DocumentCode
2972037
Title
Remarks on neural network controller which uses genetic algorithm
Author
Yamada, Takayuki ; Yabuta, Tetsuro
Author_Institution
R&D Center, NTT Telecommunication Field Syst., Ibaraki-Ken, Japan
Volume
3
fYear
1993
fDate
25-29 Oct. 1993
Firstpage
2783
Abstract
There is potential for constructing nonlinear servo controllers with a learning capability by using neural network control systems. However, such systems have an undesirable local minimum problem. This paper proposes a neural network servo controller which uses a genetic algorithm to solve this problem. Simulation results using a second order discrete time dynamic system confirm the usefulness of this servo controller.
Keywords
discrete time systems; genetic algorithms; learning (artificial intelligence); neurocontrollers; nonlinear control systems; servomechanisms; genetic algorithm; learning capability; neural network controller; nonlinear servo controllers; undesirable local minimum problem; Control systems; Cost function; Equations; Genetic algorithms; Genetic mutations; Neural networks; Neurons; Nonlinear control systems; Servomechanisms; Telecommunication control;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN
0-7803-1421-2
Type
conf
DOI
10.1109/IJCNN.1993.714301
Filename
714301
Link To Document