New adaptive iterative learning control (AILC) for uncertain robot manipulators

Abstract

In this paper, we propose two simple adaptive iterative learning control (AILC) algorithms for trajectory tracking control problem of rigid robot manipulators that track the same control trajectory repeatedly over a finite time interval. The design comprises of a linear parameterization robot feedback control structure and a learning parametric adaptation law that iteratively updates unknown uncertain parameters based upon the use of a Lyapunov energy function. In contrast to other existing adaptive ILC schemes for robot manipulators, where large feedback and learning gains are required to get robustness against large modelling uncertainties and disturbances in the early stage of the operation, the proposed adaptive ILC schemes require small feedback gains. The presented scheme 2 is simpler in structure and easier to implement in the real-world operation in the sense that it requires less computational effort and computing power without any priori knowledge of robot dynamics. Owing to the robustness of the adaptation laws against large disturbances and modelling uncertainties in the early trials, a high-learning gain can be used in order to achieve fast learning convergence.