An Improved Feedback-feedforward Model-free Adaptive Iterative Learning Control with High-order Estimation

Abstract

In this paper, an improved feedback-feedforward model-free adaptive iterative learning control with high-order estimation (FF-HOE-MFAILC) is proposed for the discrete-time nonlinear system. A novel pseudo-partial derivatives (PPD) estimation algorithm is derived based on the high-order optimization input criterion function with a proof derivation. Then, the model-free adaptive control with PPD high-order estimation is used as the feedback control term. Moreover, to improve the rapidity and accuracy of convergence, a P-type iterative learning control is employed into the proposed improved MFAC as the feedforward control term. This algorithm is essential a data-driven control method involved the repetitive iterative learning ability for nonlinear system. Numerical simulations are conducted to demonstrate the effectiveness of the proposed FF-HOE-MFAILC. Compared with the existing improved-MFAC (i-MFAC) and MFAC-based feedback-feedforward ILC (MFAC+ILC), our FF-HOE-MFAILC is proven by some comparative analyses that, the convergent speed is increased and the tracking error is decreased significantly.