Fuzzy-sliding model reference learning control of inverted pendulum with big bang — Big crunch optimization method

In this paper, a fuzzy-sliding model reference learning controller is proposed in which optimal scaling factors are assigned for the fuzzy sliding mode controllers. As the name of this study suggest the method is a breeding or hybrid combination of the fuzzy-sliding mode control (FSMC) and fuzzy model reference learning control (FMRLC) which inherits the benefits of these two methods. The main advantage of the proposed controller is that the number of rules has been reduced dramatically in comparison with the traditional FMRLC since fuzzy-sliding mode controllers are invoked in place of standard fuzzy logic controllers. The input and output scaling factors of fuzzy sliding mode controllers are adjusted using big bang - big crunch optimization method to provide an optimal result. The simulations for the proposed method are done on the inverted pendulum system. The results of these simulations demonstrate that the FS-MRLC achieves a robust performance with minimum number of fuzzy rules.