Output Feedback Control of an Active Magnetic Bearing System based on Adaptive Command Filtered Backstepping

This paper discusses the output feedback control problem of the active magnetic bearing (AMB) suspension system in absence of full-state feedback. Firstly, an exponential convergence state observer is designed to estimate the state variables only depending on output feedback. Secondly, the adaptive command filtered backstepping approach is adopted to facilitate the controller design which can avoid the complexity for calculating the analytic derivatives of the virtual control inherent in standard backstepping. Then, the compensation dynamic for filtered errors of virtual control signals is designed according to Lyapunov theory and the stability analysis shows that the compensated tracking errors can be proven asymptotic convergence. Finally, the simulation results are provided to illustrate the effectiveness of the proposed method.