Adaptive Dynamic Sliding-Mode Fuzzy CMAC for Voice Coil Motor Using Asymmetric Gaussian Membership Function

When the precise model of a controlled system is difficult to obtain, a model-free control method is suitable for control system design. The design goal of this paper is to propose a more efficient control method to deal with control systems with unknown system dynamic models and to achieve favorable chattering-free trajectory tracking performance. The cerebellar model articulation controller (CMAC) is an efficient neural network that can be applied for model-free control systems. This study proposes an adaptive dynamic sliding-mode fuzzy CMAC (ADSFC) system, which is comprised of a fuzzy CMAC and a fuzzy compensator. A fuzzy CMAC using an asymmetric Gaussian membership function is the main controller and the fuzzy compensator can compensate the approximation error introduced by a fuzzy CMAC. Moreover, a proportional-integral-type adaptation learning algorithm is developed to speed up the parameter learning. Finally, the proposed ADSFC system is applied to control a voice coil motor (VCM). Finally, the experimental results demonstrate the effectiveness of the proposed ADSFC scheme.