Online Data-Driven MPC for Unknown Switched Linear Systems: A Hybrid Sampling Approach

This paper is concerned with the design of online data-driven model predictive control (MPC) for switched linear systems with online-generated subsystems and unknowns witching signals. The online input-state data are used to detect switching instants, and update the MPC law directly to match the new subsystem bypassing identifying its model. Once a switching is detected, a fast-to slow excitation scheme will be executed until a suitable MPC law is attainable. Specifically, the fast-sampling excitation is applied for the stabilizing controller design, thereby shortening the inevitable phase of data deficiency. This is followed by a slow-sampling stage to collect sufficient data for updating MPC law. Compared with existing online ex citation strategies, the proposed method avoids excessive growth of state trajectories and degrades the risk of constraints violation. Based on set-theory-based techniques, it is revealed that the proposed online data-driven MPC approach theoretically ensures persistent feasibility and stability, provided that the switching is sufficiently slow. Illustrative examples including an aerospace application are provided to demonstrate the effectiveness and potential of the theoretical results.