Yu Duan

This article mainly investigates the depth tracking problem for an underactuated autonomous underwater vehicle (AUV), considering system dynamic uncertainties and unknown external disturbances. First, a three-layer hierarchical control system is developed for the underactuated AUV, enhancing system flexibility and control performance. Second, an adaptive line-of-sight guidance law is presented, which estimates the attack angle and mitigates the influence of unknown disturbances on the guidance layer. Third, a finite-time control law based on second-order sliding-mode algorithm is proposed along with a modified sliding surface-based nonlinear observer for approximating complex uncertainties. This provides additional robustness to overcome intense transient impact. Furthermore, the sensitivity to precise modeling and the chattering phenomenon in actuators are relaxed. In addition, an adaptive saturation compensator is given to overcome the adverse effects caused by control input saturation constraints without singularities. Finally, the proposed hierarchical control system is validated in practice based on a lightweight AUV prototype. Comparative experiments are conducted under the influence of various types of complicated time-varying disturbances. The experimental results demonstrate the effectiveness and advantages of the proposed control system in practical scenarios.