Adaptive iterative learning control of nonparametric systems based on inverse deadzone model

This work studies the angle tracking problem of robot manipulators with input deadzone and nonzero initial errors. An robust initial-rectification adaptive iterative learning control scheme is proposed to solve this problem. First, the rectification reference trajectory is constructed for dealing with nonzero initial errors during ILC design. Then, based on system parameterization of robot manipulators, by developing Lyapunov function candidate, together with reasonable dealing with the deadzone nonlinearity, a novel robust adaptive iterative learning control scheme is proposed for uncertain robotic system with unknown input deadzone. All closed closed-loop signals are proved to be bounded, with the desired tracking performance achieved. At the end, a numerical simulation is carried out to verify the effective of the proposed robust adaptive iterative learning control scheme.