In this paper, we consider the design problem of making the convergence of the bounded-input, multi-input iterative learning controller presented in our previous work robust to errors in the model-based value of the ...