Neural adaptive robust control for MEMS gyroscope with output constraints

A neural adaptive robust control method is proposed for the desired tracking of micro-electro-mechanical system (MEMS) triaxial gyroscope. In this work, input constraints and external disturbance are taken into account, and a barrier Lyapunov function (BLF) is used to ensure that the constraints are not violated and that tracking performance is achieved. In the presented control approach, RBFNNs with a non-zero parameter are employed to approximate the lumped uncertainties of the system, where the approximation precision can be modified online by the provided adaptive laws in the control strategy. All of the signals in the closed-loop system are uniformly finally bounded (UUB) thanks to the designed control mechanism, which may overcome the limitation of the finite universal approximation domain. The effectiveness of the suggested control is demonstrated by the comparison of simulation results.