About
Model-based control system design has been the dominant paradigm in control system education and design. Nearly all courses in control theory deal with different aspects of model-based analysis and design techniques. However, in the past two decades, there has been a tremendous interest in data-driven control systems. The exponentially increasing number of research papers in this field and the growing number of courses offered in universities worldwide on the subject show this trend.
Data-driven control only refers to a closed-loop control in which the starting point and destination are both data. Data-based control is then a more general term in that controllers are designed without directly making use of parametric models, but based on knowledge of the plant input-output data. Sorted according to the relationship between the control strategy and the measurements, data-based control can be summarized as four types: post-identification control, direct data-driven control, learning control, and observer-integrated control.
Prerequisite: Linear Control, Modern Control, MATLAB and Simulink
