Bounded-gain continuous sliding-mode tracking of a 6-DOF ROV with a reduced-tuning super-twisting disturbance observer

Accurate motion control of remotely operated vehicles (ROVs) is difficult in the presence of coupled 6-DOF dynamics, hydrodynamic uncertainty, tether loads, actuator limits, and imperfect measurements. This paper develops an observer-assisted continuous sliding-mode tracking scheme for an inspection-class ROV. A bounded state-dependent nonlinear gain limits the switching authority, while an auxiliary finite-horizon dynamics shapes the reaching transient without claiming strict prescribed-time stability after the shaping interval. A reduced-tuning super-twisting disturbance observer (STDO), constructed from the nonlinear ROV model, reconstructs the lumped generalized disturbance for feedforward compensation. Explicit disturbance and disturbance-derivative assumptions, damping dissipation, bounded-gain conditions and a weighted composite Lyapunov function are used to establish uniform ultimate boundedness of the interconnected observer–controller loop. Validation comprises four original 6-DOF simulation scenarios, a dedicated observer study and water-tank experiments on a tethered ROV. The observer study evaluates convergence, estimation bandwidth, noise sensitivity, and rapidly varying disturbances against two recent observers. The physical experiments demonstrate depth and heading regulation under tether manipulation, manual pushing and water agitation. The results show that the proposed method improves disturbance reconstruction and attitude regulation while retaining bounded, continuous control commands, with the limitations imposed by actuator authority and observer bandwidth stated explicitly.