Linear quadratic tracking for noisy signal with state space recursive least squares noise rejection

In many real life problems, related to closed loop control systems the reference signal is corrupted by additive noise. The noisy reference signal leads to inferior tracking by the plant. However tracking performance can further improved if noise is removed from the reference signal prior applying to the control system. In this paper, we present a linear quadratic regulator (LQR) based control scheme that incorporates state space recursive least squares (SSRLS) method for cleaning the noisy reference signal. The proposed closed loop structure provides an optimal tracking of a reference signal while minimizing the effect of external disturbance acting on the plant. The prior knowledge about the external disturbance is utilized by the control scheme. Functioning of the proposed algorithm is demonstrated with the help of computer simulations with a practical application of third order system of grid tie converters. The result shows significant improvement in tracking performance as compared to the tracking of a noisy reference signal applied directly to the control system.