Global sliding mode control of MEMS gyroscope based on neural network

In this paper, global sliding mode control (GSMC) is utilized to eliminate the arriving time of sliding mode stage, and to overcome the shortcoming of non-robustness of reaching stage in traditional sliding mode control by designing a dynamic non-linear sliding surface. In order to eliminate the existing chattering, neural networks are incorporated into global sliding mode control to approximate the unknown model uncertainties and external disturbances, thereby reducing the chattering Furthermore, the parameters of sliding mode control can be adjusted by neural networks based on Lyapunov analysis. Simulation results show that the control system can improve the dynamic characteristics of the micro-gyroscope and robustness to the external disturbance.