Terminal iterative learning control for discrete-time nonlinear system based on neural networks

The terminal iterative learning control (ILC) is designed for discrete-time nonlinear system based on neural networks. A terminal output tracking error model is derived by using a system input and output algebraic function as well as the differential mean value theorem. The weight is updated by optimizing an optimal objective function, and then is used for the input design. The technical convergence analysis and numerical simulations are given for the fixed input case. Further discussions on time-varying input case and random iteration-varying initial condition are also given in illustrative simulations.