A novel predefined-time criterion-based sliding mode control with adaptive disturbance observer for underactuated autonomous underwater vehicle

This paper addresses the challenging problem of high-precision three-dimensional (3D) trajectory tracking control for underactuated autonomous underwater vehicles (AUVs) operating under model uncertainties and time-varying external disturbances. The main contributions and novelty of this work are summarised as follows. First, a novel and more flexible predefined-time stability (PTS) criterion is proposed, featuring an additional tunable parameter that enables superior performance in convergence speed, control effort, and robustness compared to existing methods. Second, a practical predefined-time stability (PPTS) framework is developed, extending the applicability of PTS theory to more realistic engineering scenarios. Third, an innovative adaptive predefined-time disturbance observer (APtDO) is designed, which guarantees convergence of the estimation error within a user-defined time without requiring prior knowledge of disturbance bounds. Fourth, a new predefined-time fractional-order sliding mode control (PtFoSMC) strategy is proposed, incorporating an adaptive reaching law (ARL) and a fractional-order (FO) sliding surface to effectively avoid singularity and attenuate chattering. The integrated APtDO-PtFoSMC control scheme is validated through comparative simulations, demonstrating significant improvements in trajectory tracking performance for AUVs under complex operating conditions.