Title :
Experimental studies in neural network control
Author :
Mistry, Sanjay I. ; Outangoun, Sermsak ; Nair, Satish S.
Author_Institution :
Dept. of Mech. & Aerosp. Eng., Missouri Univ., Columbia, MO, USA
Abstract :
Control algorithms are designed using neural networks and experimentally implemented on an electromechanical system. Such algorithms require very little knowledge about the systems which, together with their capability of learning, make them attractive for the control of nonlinear partially known dynamic systems. Issues pertaining to such algorithms are investigated, and preliminary results are presented. The controllers perform well compared to conventional designs
Keywords :
controllers; neural nets; nonlinear systems; control algorithms; electromechanical system; neural network control; nonlinear dynamic systems; Adaptive control; Aerodynamics; Algorithm design and analysis; Control systems; Electromechanical systems; Intelligent networks; Multi-layer neural network; Neural networks; Nonlinear control systems; Nonlinear dynamical systems; Programmable control;
Conference_Titel :
Decision and Control, 1992., Proceedings of the 31st IEEE Conference on
Conference_Location :
Tucson, AZ
Print_ISBN :
0-7803-0872-7
DOI :
10.1109/CDC.1992.371208