This paper introduces an intelligent subsea cable inspection technology utilizing an autonomous underwater vehicle (AUV), which enhances the safety and efficiency of traditional methods employing engineering vessels and remotely operated vehicles. A simultaneous localization, guidance, and control (SLGC) scheme is proposed for subsea cable inspection with an underactuated AUV. First, based on subsea cable electromagnetic characteristics, a spatial electromagnetic localization model was devised for an AUV equipped with a dual triaxial electromagnetic array, enabling real-time calculation of the AUV’s spatial relative location and routing angle concerning the cable. Second, based on the electromagnetic localization signals, an L 1 guidance law for the underactuated AUV was designed, integrating the cable electromagnetic detection system with the AUV navigation system to achieve perception of the spatial electromagnetic field and self-navigation. Third, considering the actuator response, AUV hydrodynamic complexities, and ocean current disturbances, an adaptive neural network and barrier Lyapunov function-based kinetic controller was constructed, achieving stable cable-tracking control by estimating tracking performance and uncertain factors. Finally, comprehensive simulation environments incorporating underwater electromagnetic fields and ocean current disturbances were established. Comparative studies against conventional PID control and data-driven model predictive control demonstrate the robustness, effectiveness, and practical applicability of the proposed SLGC framework for subsea cable inspection. • A simultaneous localization, guidance, and control (SLGC) architecture is proposed for underactuated AUV, which performs autonomous tracking and inspection for subsea cables with unknown routing. • Electromagnetic L 1 guidance is devised for underactuated AUV based on subsea cable localization, bridging the detection and control systems, and avoiding complex calculations on tracking errors and its’ accurate differentiation. • An adaptive cable-tracking kinetic controller is designed based on the neural network approximation and barrier Lyapunov function, which solves multiple uncertainties and nonlinearities.
Zhang et al. (2026) studied this question.