Title :
Stable, controllable neural control of affine uncertain systems
Author :
Mears, Mark J. ; Polycarpou, Marios M.
Author_Institution :
VACA, AFRL, Wright-Patterson AFB, OH, USA
Abstract :
This paper describes an approach to using neural networks as part of a control architecture that allows tracking performance to improve as the network assimilates the dynamic characteristics of the plant. Stability is guaranteed by using Lyapunov analysis and controllability is insured as the network learns. The results are shown for a scalar, affine plant where uncertainties exist in the functions representing the dynamics of the plant
Keywords :
Lyapunov methods; controllability; neurocontrollers; stability; Lyapunov analysis; affine plant; affine uncertain systems; control architecture tracking performance; controllability; controllable neural control; dynamic characteristics; stability; stable control; Adaptive systems; Control nonlinearities; Control systems; Controllability; Function approximation; Neural networks; Sliding mode control; Trajectory; Uncertain systems; Uncertainty;
Conference_Titel :
American Control Conference, 2001. Proceedings of the 2001
Conference_Location :
Arlington, VA
Print_ISBN :
0-7803-6495-3
DOI :
10.1109/ACC.2001.946182
