Backstepping Adaptive Sliding Mode Control for ROV Under Multi-Source Time-Varying Disturbances

To address the reduced motion control accuracy of underwater cleaning remotely operated vehicles (ROVs) caused by strongly coupled nonlinear dynamics and multisource time-varying disturbances, this study proposes a backstepping adaptive sliding mode control (B-ASMC) strategy. The proposed method combines the systematic design framework of backstepping control with the strong robustness of sliding mode control. An adaptive law is introduced to estimate the lumped system disturbance online and to compensate for it in real time, thereby avoiding dependence on an accurate disturbance observer or an exact system model. In addition, a hybrid thrust allocation strategy that integrates the pseudo-inverse method with the active-set method is developed to improve computational efficiency while suppressing thrust saturation. The global asymptotic stability of the closed-loop system is rigorously proved using Lyapunov stability theory. Simulations and tank experiments are conducted using a self-developed ROV. Simulation analyses under two representative operating conditions, namely depth holding and bow heading holding, show that B-ASMC achieves higher tracking accuracy, faster dynamic response, and stronger robustness than PID under time-varying disturbances. The tank experiments further verify that the proposed method exhibits control accuracy, disturbance rejection, and engineering applicability.