Double Recurrent Perturbation Fuzzy Neural Network Fractional-Order Sliding Mode Control of Micro Gyroscope

In this paper, a fractional-order sliding mode control method based on a double recurrent perturbation fuzzy neural network (DRPFNN) is proposed for a micro gyroscope system with parameter uncertainty and external disturbance. The DRPFNN is used to realize the adaptive estimation of the unknown part, which ensures that the controller is independent of the accurate mathematical model. The sine-cosine perturbation membership function is used to deal with the uncertainty of rules in neural network, increasing the accuracy, reduce the calculation load, and simplifying computational complexity. In addition, double recurrent links are added to transmit more information with superior dynamic ability. Then, the proposed control system is composed of a DRPFNN estimator and a fractional-order sliding mode controller (FSMC). The fractional calculus operator is introduced into the sliding surface to improve the controller flexibility. The parameter adaptive laws in the neural network are designed to be adaptively stabilized to the optimal value. Finally, simulation studies verify the effectiveness of the proposed control method, showing it can obtain higher control accuracy and enhance robustness than the traditional sliding mode control method.