Title :
Dynamic neural network-based robust identification and control of a class of nonlinear systems
Author :
Dinh, H. ; Bhasin, S. ; Dixon, W.E.
Author_Institution :
Dept. of Mech. & Aerosp. Eng., Univ. of Florida, Gainesville, FL, USA
Abstract :
A methodology for dynamic neural network (DNN) identification-based control of nonlinear systems is proposed. The multi-layer DNN structure is modified by the addition of a sliding mode term in order to robustly account for exogenous disturbances and DNN reconstruction errors. New weight update laws for the DNN are proposed which guarantee asymptotic regulation of the identification error to zero. The DNN identifier is used in conjunction with a continuous RISE feedback term for asymptotic tracking of a desired trajectory. Both the identifier and the controller operate simultaneously in real time.
Keywords :
identification; neurocontrollers; nonlinear control systems; tracking; uncertain systems; DNN identifier; DNN reconstruction errors; asymptotic regulation; asymptotic tracking; complex uncertain nonlinear systems; continuous RISE feedback; dynamic neural network-based robust identification; exogenous disturbances; identification error; multilayer DNN structure; Approximation methods; Artificial neural networks; Nonlinear systems; Robustness; Stability analysis; Trajectory;
Conference_Titel :
Decision and Control (CDC), 2010 49th IEEE Conference on
Conference_Location :
Atlanta, GA
Print_ISBN :
978-1-4244-7745-6
DOI :
10.1109/CDC.2010.5717445