A data-driven iterative learning control for I/O constrained nonlinear systems

In this paper, a new data-driven ILC method is proposed for I/O constrained nonlinear systems. An iterative dynamic linearization is introduced for the controlled nonlinear systems. All of the constraints on the system inputs and outputs are reformulated with a linear matrix inequality. The learning control law is then developed by minimizing a predesigned cost function subjected to the linear matrix inequality constraint. Simulation results illustrate the effectiveness of the proposed approach.