Data-Driven Control Design With LMIs and Dynamic Programming

The goal of this paper is to study model-free data-driven control evaluation and design strategies for discrete-time linear time-invariant systems, where the system model unknown. In particular, our main contribution twofold: 1) new state-input exploration data collection schemes from experiences; 2) matrix inequalities dynamic programming methods stabilization optimal problems. proposed theoretically guarantee acquire sufficient information system’s trajectories that can solve underlying We prove under mild assumptions, as more accumulated, collected problems with higher probability along algorithms.