Learning-based adaptive control with an accelerated iterative adaptive law

A novel adaptive control framework equipped with an accelerated iterative learning update mechanism is developed to handle time-varying uncertainties, based on the combination of standard adaptive control architecture and Heavy-ball optimization algorithm. The stability analysis shows that the tracking error and the estimated weight error are both bounded, and the closed-loop system is exponentially stable. The momentum term, introduced in the accelerated iterative adaptive law, makes the proposed learning-based adaptive control possess a faster convergence rate. The proposed learning-based adaptive control is applied to aircraft control to show that the proposed framework can handle time-varying uncertain parameters.