Informativity conditions for data-driven control based on input-state data and polyhedral cross-covariance noise bounds

Modeling and control of dynamical systems rely on measured data, which contains information about the system. Finite data measurements typically lead to a set system models that are unfalsified, i.e., explain data. The problem data-informativity for stabilization or with quadratic performance is concerned existence controller stabilizes all unfalsified achieves desired performance. Recent results in literature provide informativity conditions based input-state ellipsoidal noise bounds, such as energy magnitude bounds. In this paper, we consider where bounds defined through cross-covariance respect an instrumental variable; were introduced originally characterization parameter bounding identification. considered by finite number hyperplanes, induce (possibly unbounded) polyhedral systems. We H2/H∞ vertex/half-space representations