Iterative learning control algorithm for a class of discrete LTI system with batch-varying reference trajectories

In this paper, we present adaptive iterative learning control (AILC) schemes for batch-varying reference trajectories. In the general ILC, reference trajectory must be identical for all batches, but reference trajectories can be changed in dynamic systems such as robotics and chemical processes according to cycles or batches. The plant-model mismatch error must vanish to make outputs converge to different references in each batch. For this reason, Markov parameters of the system dynamics are identified at the end of each iteration in an iterative learning manner. ILC schemes for batch-varying reference trajectories are proposed in two forms, which are inverse of model-based ILC (I-ILC) and quadratic-criterion based ILC (Q-ILC). These control schemes are studied for discrete linear time-invariant (LTI) system. A numerical example is provided to demonstrate the performance of the proposed algorithm.