Error-free tracking control of discrete-time systems by combining data-driven control with model-based ILC

For discrete-time linear systems, by applying the advantages of data-driven methods and iterative learning control, this paper proposes three data-driven learning control algorithms based on the linearity of the system. For iterative learning control, if the initial state shift is zero in each iteration and the system matrix is known, the algorithm presented in this paper can ensure that the system achieve complete tracking in the entire interval through a finite number of iterations. If the initial state shift is arbitrary in each iteration and the system matrix is unknown, the algorithm proposed in this paper can ensure that the system achieve complete tracking at all points except the initial point through finite iterations. Even for systems with non-repetitive processes, if the system matrix is unknown, the method in this paper can ensure that the system achieves complete tracking at all points except a finite number of points in the initial stage. Finally, the effectiveness of the three algorithms is verified by three examples, and the superiority of the proposed algorithms is demonstrated by comparison.