Anticipatory iterative learning control for nonlinear systems with arbitrary relative degree

In this paper, the anticipatory iterative learning control is extended to a class of nonlinear continuous-time systems without restriction on relative degree. The learning algorithm calculates the required input action for the next operation cycle based on the pair of input action taken and its resultant variables. The tracking error convergence performance is examined under input saturation being taken into account. The learning algorithm is shown effective even if differentiation of any order from the tracking error is not used.