Grey FCMAC controller design for robotic manipulators

Abstract

A design method for the Cerebella Model Articulation Controller (CMAC) with theories of the grey and the fuzzy systems is proposed in this paper. The concept of sliding mode control is also used to determine the weighting values of CMAC. It can be easily realized in the requirement of real-time control cases. However, the overall performance of control system depends on the trained CMAC; usually, it deeply depends on the structure of CMAC. Therefore, to improve the performance without learning process maybe is a good idea in designing the CMAC. In addition, a constructed compensation rule mechanism basing on the information of estimation via grey prediction is used to enhance the characteristic of CMAC. Finally, a simulation example of controlling a nonlinear system is employed to illustrate the performance and applicability of the proposed control scheme.