Title :
Quasi-invariant stable multivariable CMAC variable structure neurocontrol
Author :
Bordon, Maurício José ; Bottura, Celso Pascoli ; Teixeira, Marcelo Carvalho Minhoto
Author_Institution :
DMCSI-FEEC-UNICAMP, Brazil
Abstract :
A quasi-invariant intelligent control method for partially unknown dynamical systems stabilization is proposed. The ability of a CMAC network in the reproduction of arbitrary functions is used to online approximation of plant parameters related information in order to satisfy the existence conditions of a sliding mode regime in a set of discontinuity surfaces without the knowledge of plant model other than system order and structure
Keywords :
cerebellar model arithmetic computers; intelligent control; multivariable control systems; neurocontrollers; parameter estimation; stability; uncertain systems; variable structure systems; CMAC network; VSC; VSS; arbitrary function reproduction; discontinuity surface set; existence conditions; intelligent control method; online approximation; partially unknown dynamical systems stabilization; plant parameters; quasi-invariant stable multivariable CMAC variable structure neurocontrol; sliding mode regime; Chemical processes; Function approximation; Intelligent control; Learning; Noise generators; Nonlinear control systems; Quantization; Robot kinematics; Sliding mode control; State estimation;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.832730
