On data-driven control theory: the state of the art and perspective

Abstract

The definitions and concepts of data-driven control theories and designs are described in this paper. The motivation and necessity for data-driven control are discussed from the aspects of theory, applications and the history of system control. The state-of-art of the existing data-driven control methods are presented with appropriate classifications and insights, and the differences among these methods and the application scopes are also highlighted. Finally, the perspective of data-driven control and associated research topics are also briefly explored and discussed.