Abstract: Abstract：Objectives To tackle the control challenges of autonomous underwater vehicle (AUV) docking with a dynamic base under external disturbances and model uncertainties, a high-performance dual-loop control strategy is proposed to achieve fast and stable pose alignment. Methods Using the White Dolphin 100 docking system as the research platform, an AUV motion model was developed, and the dynamic base docking problem was formulated. A dual-loop controller, integrating kinematic and dynamic control, was designed based on a fast nonsingular integral terminal sliding mode surface. An adaptive lumped disturbance estimation approach was employed to mitigate the effects of external disturbances and model uncertainties. The finite-time convergence of the controller was proven using Lyapunov theory. Simulations were conducted based on the characteristics of the Baiji 100 system to validate the approach. Results Simulation results demonstrate that the proposed dual-loop control method (AFNITSMC) achieves rapid pose convergence of the AUV with the dynamic base within 10 seconds, maintaining effective docking control under time-varying external disturbances with 20% thrust saturation and 20% model uncertainties. The steady-state mean absolute error (MAE) is 0.142 cm, 0.103 cm, 0.0397 cm for position and 0.012°, 0.054° for attitude. Compared to the baseline method (NITSMC, with position errors of 0.585 cm, 0.834 cm, 0.850 cm and attitude errors of 0.343°, 0.143°), position errors are reduced by 75.7%, 87.6%, and 95.3%, and attitude errors by 96.5% and 62.2%, respectively.Conclusions The proposed adaptive fast nonsingular integral terminal sliding mode double-loop control method exhibits superior control performance and strong engineering applicability for AUV dynamic base docking under external disturbances and model uncertainties.