An Estimator-Based Feedback Control Approach for Measurement Noise Rejection

This paper presents an estimator-based robust feedback control solution for linear time invariant systems in the presence of measurement noises. Two key features of this approach are: (i) a noise estimator (NE) is adopted for measurement noise rejection, and (ii) it is easy to derive explicit design conditions on the estimator, which ensure the closed-loop stability and arbitrarily small estimation error. A novel parameter mapping is then proposed to show a simple relationship between the parameters of NE and the ultimate bound of tracking error. The stability and convergence of single/double-integrator systems are analyzed, and numerical simulation results are demonstrated to verify the effectiveness of proposed approach and show the advantages over classic filter/observer-based approaches.