Neural Network Based Robust Nonlinear Control for a Magnetic Levitation System

In this paper, a robust nonlinear control approach is presented for a magnetic levitated ball system with uncertain parameters and external disturbance. Gaussian basis RBF neural networks are used to approximate the nonlinear uncertainties, a highgain observer is used to estimate the ball velocity which cannot be measured. A fixed controller and an adaptive robust controller derived can guarantee that the closed-loop system is stable and robust; the desired position tracking performance is achieved when the system parameters change. Simulation results are provided to demonstrate the utility of the proposed method.