Title :
Discrete-time CMAC neural networks for control applications
Author :
Commuri, S. ; Lewis, F.L. ; Jagannathan, S.
Author_Institution :
Autom. & Robotics Res. Inst., Texas Univ., Fort Worth, TX, USA
Abstract :
The cerebellar model articulation controller (CMAC) neural network (NN) has advantages over fully-connected NNs due to its increased structure. Since almost all the implementations of CMAC NNs are done on a digital computer, the discrete-time implementation of CMACs is of great importance. This paper attempts to provide a comprehensive treatment of digital implementation of CMAC NNs in closed-loop control applications. The function approximation capabilities of the CMAC NN are first established and novel weight-update laws derived that guarantee the stability of the closed-loop system. The passivity properties of the CMAC under the specified tuning laws are examined and the relationship between passivity and closed-loop stability is derived. The utility of the CMAC NN in controlling a nonlinear system with unknown dynamics is demonstrated through numerical examples
Keywords :
cerebellar model arithmetic computers; closed loop systems; discrete time systems; function approximation; neurocontrollers; stability; cerebellar model articulation controller; closed-loop control; digital implementation; discrete-time CMAC neural networks; function approximation; nonlinear system; stability; Associative memory; Control systems; Electronic mail; Function approximation; Neural networks; Nonlinear control systems; Nonlinear dynamical systems; Nonlinear systems; Robots; Stability;
Conference_Titel :
Decision and Control, 1995., Proceedings of the 34th IEEE Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-2685-7
DOI :
10.1109/CDC.1995.478453
