Robust trajectory tracking for fully-input-bounded actuated unmanned surface vessel with stochastic disturbances: An approach by the homogeneous nonlinear extended state observer and dynamic surface control

Abstract

Nowadays, academia, industry, and the government have increased their interests and found the necessities of developing unmanned surface vessels (USV) for maritime shipping, resource exploration, and scientific research. To completely address the USV trajectory tracking problem, a robust trajectory tracking controller based on the homogeneous nonlinear extended state observer (HNESO) and the dynamic surface control (DSC) is introduced in this study, which is capable to deal with saturations both for the control inputs and their change rates, as well as stochastic lumped disturbances that include the model uncertainties, unmeasured disturbances, and noises. With strict mathematical derivation of the HNESO, the estimated stochastic lumped disturbances can be further regarded as compensations during the DSC controller design, such that, the robustness of the proposed controller would be enhanced. The Lyapunov stability analysis proves that all signals in the closed-loop tracking system are bounded. Compared with the Conventional Dynamics Surface Control (CDSC), Nonlinear Model Predictive Control (NMPC), and Adaptive Sliding Mode Control (ASMC), simulation results show that the proposed controller offers better tracking performances, which ensures that the USV precisely tracks the desired trajectory in different working scenarios under conventional or intense stochastic lumped disturbances. Finally, we also show that the HNESO has better estimation results for the stochastic lumped disturbances than the linear extended state observer (LESO).