Robust iterative learning control based on neural network for a class of uncertain robotic systems

This paper studies the problem of adaptive robust iterative learning control for trajectory-tracked task of a class of robotic systems with both structured and unstructured uncertainties. A composite control scheme is proposed in which the periodic uncertainties are approached by the learning controller, while the effect of non-periodic uncertainties on system performances is attenuated by the robust controller. In particular, by employing neural network the cone-bounded assumption on uncertain dynamics is removed. The simulation results are included