Adaptive Event-Triggered Predictive Control for Agile Motion of Underwater Vehicles

Abstract

As the demand for underwater robots in complex environments continues to grow, research on their agile motion capabilities becomes increasingly crucial. This paper focuses on the design and agile motion control of autonomous underwater vehicles (AUVs) operating in subsea environments, addressing key issues such as structural design, system modeling, and control algorithm development. An optimization model for the arrangement of propellers is formulated and solved using a Sequential Quadratic Programming (SQP) algorithm. Computational Fluid Dynamics (CFD) software is employed for hydrodynamic analysis and shape optimization. A novel adaptive event-triggered nonlinear model predictive control (AET-NMPC) algorithm is proposed and compared with traditional Cascaded Proportional–Integral–Derivative (PID) control and event-triggered cascaded PID control algorithms. Simulation and experimental results demonstrate that the AET-NMPC algorithm significantly enhances the response capability and control accuracy of underwater robots in complex tasks, with the trajectory tracking error being reduced to 4.89%. This study provides valuable insights into the design and control strategies for AUVs, paving the way for more sophisticated underwater operations in challenging environments.