Title :
Real-time neural network control of a free gyro stabilized mirror system
Author :
Ge, S.S. ; Lee, T.H. ; Zhao, Q.
Author_Institution :
Dept. of Electr. Eng., Nat. Univ. of Singapore, Singapore
Abstract :
Neural network modelling and control of a free gyro stabilised mirror system are presented. Two kinds of radial basis function networks, Gaussian and Hardy´s multiquadric, are investigated. Asymptotic tracking performance is guaranteed, and at the same time uniformly stable adaptation is achieved. Real-time experimental results demonstrate the effectiveness of the adaptive controller
Keywords :
electro-optical devices; feedforward neural nets; gyroscopes; mirrors; neurocontrollers; optical sensors; real-time systems; stability; Gaussian nets; Hardy multiquadric nets; asymptotic tracking performance; free gyro stabilized mirror system; radial basis function networks; real-time neural network control; uniformly stable adaptation; Adaptive control; Adaptive systems; Control systems; Electronic mail; Flywheels; Mirrors; Neural networks; Programmable control; Real time systems; Sensor systems;
Conference_Titel :
American Control Conference, 1997. Proceedings of the 1997
Conference_Location :
Albuquerque, NM
Print_ISBN :
0-7803-3832-4
DOI :
10.1109/ACC.1997.609697
