On adaptive iterative learning from tracking tasks with different magnitude and time scales

Abstract

In this paper, we introduce a new adaptive iterative learning control (AILC) scheme based on a time scaling factor, which enables learning from control tasks with different magnitude and time scales. The proposed AILC scheme overcomes the limitation of traditional ILC that the target trajectory must be identical in all iterations. In addition, the requirement on classic ILC that every trial must repeat in a fixed time duration is removed. For nonlinear systems with time-invariant and time-varying parametric uncertainties, the new learning algorithm works effectively to nullify the tracking error. It is shown that the new AILC is capable of fully utilizing all the learned knowledge despite the iteratively varying tracking tasks. In the end, an illustrative example is presented to demonstrate the performance and the effectiveness of the proposed AILC scheme.