Online learning-based active disturbance rejection control of autonomous surface vehicles with HIL simulations

Abstract

Modeling and control of autonomous surface vehicles (ASVs) is a challenging task due to the requirement of real-time adaptation to internal hydrodynamic parameter variations and external ocean disturbances induced by wind, waves, and ocean currents. In this paper, an online learning-based active disturbance rejection control method is proposed by combining real-time model learning and anti-disturbance control. Specifically, a filtering learning extended state observer (FLESO) is proposed for online identifying the unknown hydrodynamic parameters, the unknown control input gain, and estimating system uncertainties composed of the unknown external disturbances and dynamic modeling errors, with guaranteed convergence. Then, a self-learning anti-disturbance control law is developed based on the FLESO, enabling self-learning anti-disturbance speed tracking without prior knowledge of model parameters. A salient feature of the proposed method is that the unknown hydrodynamic parameters, the unknown control input gain, and the unknown system uncertainties can be estimated online simultaneously. Lyapunov stability analysis is employed to verify the stability of the closed-loop system. The effectiveness of the proposed online learning-based active disturbance rejection control method is demonstrated through HIL simulations.