Trackability-Based Tracking Control for Stochastic Learning Systems: A Two-Dimensional System Method

This paper focuses on exploring a novel trackability-based framework for the iterative learning control (ILC) systems subject to stochastic disturbances by using a two-dimensional (2-D) system method. By examining the fundamental trackability property of the stochastic ILC systems, the trackability-based stochastic ILC design and analysis framework is developed, eliminating the need for the common realizability assumption. Under this framework, thanks to the 2-D system method with the Roesser systems, the convergence results for both the output and input errors can be established under a unified condition, regardless of the full column or row rank of the input-output coupling matrix. A simulation example is included to demonstrate the validity of our proposed stochastic ILC design framework for ILC systems.