A discrete-time adaptive iterative learning from different reference trajectory for linear time-varying systems

An adaptive ILC approach is presented for a linear MIMO discrete-time system to track different reference trajectory. The presented adaptive ILC is a data-driven approach in nature and the controller does not contain any modeling information of the plant. The unknown Markov parameter of the plant is iteratively estimated only using the I/O data of the plant. The rigorous analysis shows the asymptotic convergence of the presented method for iteration-varying reference trajectory.