1: \begin{abstract}

2: This article presents a complete framework for multiple estimation

3: models in the discrete-time Adaptive Iterative Learning Control (ILC)

4: problem. The use of multiple models improves transient response and

5: error convergence in adaptive systems and has not been previously

6: explored in the context of Adaptive ILC. A new control and

7: identification scheme with a single estimation model is

8: presented. This scheme enables the extension to multiple models,

9: unlike existing Adaptive ILC strategies. Next, the fundamentals of the

10: Multiple Models, Switching and Tuning (MMST) methodology are used for

11: multiple estimation models in Adaptive ILC. Two switching strategies

12: are outlined, and convergence is proved for all techniques using the

13: properties of square-summable sequences. Simulation results

14: demonstrate the efficacy of the proposed strategies, with improved

15: convergence using multiple estimation models.

16: \end{abstract}