Data-driven control of continuous-time systems: A synthesis-operator approach

Abstract

This paper addresses data-driven control of continuous-time systems. We develop a framework based on synthesis operators associated with state and input trajectories. A key advantage of the proposed method is that it does not require the state derivative and uses continuous-time data directly without sampling or filtering. First, systems consistent with the data are represented in terms of synthesis operators, into which the data trajectories are embedded. Next, we characterize data informativity properties for system identification and for stabilization in the noise-free case. Finally, we establish a necessary and sufficient condition for noisy data to be informative for quadratic stabilization. All these informativity characterizations are formulated in terms of finite-dimensional matrices, by leveraging the finite-rank structure of the synthesis operators.