Hongtao Pan, Xinbin Li, Song Han, Junzhi Yu, Tongwei Zhang
Autonomous underwater vehicles (AUVs) are indispensable for ocean exploration; however, achieving accurate trajectory tracking control suffers from significant challenges due to internal model uncertainties and external disturbances, stemming from unmodeled hydrodynamics, parametric uncertainties, and time-varying environmental forces. Moreover, the presence of inherent velocity constraints may further restrict actuator performance, leading to increased tracking errors or even system instability. Taking into account these issues, this study proposes a tightly coupled nonlinear continuous sliding mode predictive control (NCSMPC) scheme to tackle the trajectory tracking problem for AUV. First, the Lyapunov-based nonlinear model predictive control (NMPC) kinematic controller is developed for the position loop to achieve finite-time tracking, which generates the constrained velocity signals to consistently achieve optimal tracking performance in accordance with the online optimization function. Meanwhile, an adaptive integral event-triggering mechanism (ETM) is introduced to reduce the computational burden by adaptively regulating the frequency of optimization updates. Then, an adaptive continuous sliding mode controller (ACSM) based on the adaptive super-twisting algorithm is proposed for the velocity loop. The designed dynamic controller can actively compensate for unmodeled hydrodynamics and time-varying disturbances without requiring prior knowledge of their bounds, thereby achieving finite-time velocity tracking with low oscillation. Guaranteed by Lyapunov stability, the proposed control framework integrates the optimization of NMPC with the robust performance of ACSM, thereby ensuring high-performance position-velocity closed-loop tracking. Finally, simulations and real-time experiments on a robot-operating-system (ROS)-based AUV model validate the effectiveness of the proposed controller.