Robust path following control of underactuated AUV subject to multiple uncertainties

This paper proposes a robust controller to address a path following problem for underactuated autonomous underwater vehicle (AUV) subject to multiple uncertainties in the horizontal plane. The controller synthesizing the guidance layer and AUV dynamic controller layer is proposed to guarantee both stability and tracking performance along a predefined path. The guidance layer is mainly used to generate the desired virtual command for the dynamic controller and an LPV based dynamic robust controller is proposed to track the virtual command. To simplify dynamic controller design, the AUV dynamic model is firstly transformed to a generalized LPV model and the multiple uncertainties including internal model parameters perturbations, external environmental disturbances and unmodeled dynamics are viewed as lumped uncertainties. Sensor measurement errors are also considered in controller design. Finally, comparative simulations have demonstrated the effectiveness and superiority of the proposed approach.