Title: System Identification and Model-based Robust Nonlinear Disturbance Rejection Control

Abstract: The robust disturbance rejection controller has been the subject of intensive research due to its undeniable importance for automation. Modern control theory tends to use model-based approaches versus model-free approaches, especially when it comes to highly modern applications. The backbone of the dissertation is based on the systematic modeling of dynamic systems and the development of advanced control methods. Accordingly, the dissertation begins with the investigation of nonlinearities in dynamic systems. The extension of classic subspace algorithms for linear systems in the frequency domain is tackled using the new local polynomial approach. Next, the problem of disturbance control is addressed, namely modeling of uncertainties and non-modeled high-order dynamics, fragility of the controller and observer systems, and the non-linearities are analyzed separately.