Trajectory tracking control of underactuated UUVs with model uncertainties in presence of ocean currents

This paper investigates three-dimensional trajectory tracking of an underactuated unmanned underwater vehicle (UUV) subject to ocean currents, model uncertainties, and input constraints. A dual-loop cascade control method is proposed by integrating nonlinear model predictive control (NMPC), integral sliding mode control (ISMC), and radial basis function (RBF) neural networks. In the outer loop, NMPC suppresses position-tracking errors and generates desired velocity signals, which are transformed into continuous inputs for the inner loop. The inner loop employs an adaptive ISMC enhanced with RBF networks to compensate for uncertainties and improve robustness. To address unknown ocean currents, a PI observer optimized by the whale optimization algorithm (WOA) is designed. System stability is then established using Lyapunov theory. Unlike studies focusing on improving a single controller, this work integrates NMPC and ISMC within a dual-loop framework, which explicitly handles input constraints and enhances robustness against uncertainties and current disturbances. Furthermore, a UUV model including ocean current velocity is specifically established according to controller features and vehicle characteristics, and the combination of RBF networks with the WOA-PI observer further improves disturbance estimation and control accuracy. Simulation results demonstrate that, compared with single-control approaches, the proposed method achieves more accurate and robust trajectory tracking under parameter perturbations and ocean current disturbances.