Adaptive neural integral sliding mode control using neural compensator for MEMS gyroscope

In this paper, an adaptive neural integral sliding mode controller using neural compensator is presented to deal with the strong nonlinearity and parameter uncertainty for a Micro-Electro-Mechanical Systems (MEMS) vibratory gyroscope. The proposed control scheme is designed by two kinds of RBF neural networks, where the self-adaptive RBF NN controller is implemented to online approximate and compensate the uncertain gyroscope system dynamics. Another RBF neural network is built to learn the upper bound of modeling error and external disturbances so that the prior knowledge of model uncertainties is not required. Moreover, incorporating the integral control into the adaptive sliding neural controller decreases the tracking error. Numerical simulations for the MEMS gyroscope are investigated to demonstrate the validity of the proposed controller.