A novel constraint?tightening approach for robust data?driven predictive control

In this paper, we present a data-driven model predictive control (MPC) scheme that is capable of stabilizing unknown linear time-invariant systems under the influence process disturbances. To end, Willems' lemma used to predict future behavior system. This allows entire be set up using only priori measured data and knowledge an upper bound on system order. First, develop state-feedback MPC scheme, based input-state data, which guarantees closed-loop practical exponential stability recursive feasibility as well constraint satisfaction. The extended by suitable tightening, can also constructed data. order unstable systems, presented contains prestabilizing controller associated input tightening. We first proposed for case full state measurements, provide extensions obtaining similar in output feedback. applied numerical example.