A comparative study of nonlinear observers applied to a DC servo motor

This paper studies and compares three nonlinear observers (Nonlinear Lyapunov Observer (NLO), Lipschitz Observer (LIO) and Partial Lipschitz Observer (PLIO)) applied to nonlinear model of the DC servo motor. The considered criteria of computations for white noise is the amplitude of the residual and the estimated shape of residual and error probability density functions (PDF) which is estimated by Kernel Density Function for additive fault. The same disturbance and additive fault are assumed for each case separately. According to the simulation results, it has been concluded that PLIO is the best robust observer against the disturbance. and NLO is very powerful with additive faults. LIO and PLIO are based on the gain matrix which are found by pole placement. Therefore, continuous tuning of the optimal gain matrix is needed to improve the performance of observer.