Artificial Neural Network-Based Sliding Mode Position Tracking Control for Maglev System

The magnetic levitation (maglev) system faces challenges from uncertainties in model parameters and unknown external disturbances, particularly during levitation in unfavourable environmental conditions. In this paper, a radial basis function-based higher-order sliding mode (RBF-HOSM) controller is discussed to improve the performance of a highly nonlinear and unstable maglev system. The conventional sliding mode control (SMC) exhibits undesired high-frequency oscillations leading to the phenomena called chattering. The design of the SMC also needs perfect information on the model parameters. However, on many occasions, the model parameter’s uncertainties are not known. To address the issue of parameter uncertainties, a RBF-based HOSM controller is employed for the position control of the maglev ball. To validate the proposed controller, a simulation is carried out on a given reference position trajectory with and without disturbances. Finally, the simulation results are compared with the convention integer-order proportional-integral-derivative (IOPID) and HOSM control approaches. The proposed controller shows effectiveness in position tracking along with disturbance rejection capability.