Robust tracking of iteration-dependent references for 2D systems with dual compensation technique

Most of the existing iterative learning control (ILC) methods for 2D discrete systems are concerned with normal communication network, uncoupled input and iteration-independent trajectory. This paper first investigates the robust/convergent tracking problem for 2D linear discrete systems (LDSs) with random data dropouts, coupled input, iteration-dependent reference trajectories (IDRTs), and nonrepetitive uncertainties from boundary states and disturbances. The IDRTs is generated by a stable high-order internal model (HOIM). By using the signal of IDRTs, the random data dropout occurred at the measurement side is compensated; With the tracking error with one-step forward difference, the coupled input is compensated. A HOIM-based networked ILC law with dual compensation technique is proposed. Consequently, it is theoretically proved that the ILC tracking error in the sense of mathematical expectation robustly converge to a bounded range, the bound of which depends continuously on the bounds of nonrepetitive uncertainties and probability of the random data dropouts. If these nonrepetitive uncertainties are progressively convergent along the iteration domain, a perfectly convergent result can be obtained with mathematical expectation sense. Additionally, an improved ILC law with dual compensation technique is extended to the 2D LDSs with zero relative degree. The effectiveness of the presented HOIM-based networked ILC law is illustrated by a practical thermal process and some comparative results are provided.