Novel design of underwater vehicle with adjustable thruster configuration

Abstract

The uncertainty of underwater environments presents significant challenges to the stable and efficient operation of unmanned underwater vehicles (UUVs). Unpredictable disturbances such as waves, currents, and factors like actuators and manipulators introduce perturbations across six degrees of freedom. What's more, when the thrusters operate near their physical limits (control dead zone), the thrust output becomes unstable, thereby introducing additional noise. To address the aforementioned challenges, this study proposes a novel design for the UUV. Unlike conventional 6-thruster configurations, mainstream 8-thruster layouts, and vector propulsion systems, each thruster in the proposed UUV features two additional degrees of freedom, enabling the vehicle to adjust its configuration to adapt to various scenarios. Additionally, a configuration adjustment strategy is introduced, which consists of offline and online optimization phases. The offline optimization integrates a simulated annealing algorithm combined with sequential quadratic programming (SA-SQP), providing a reference configuration for the online process. Furthermore, an empirically estimated added mass is proposed to account for hydrodynamic effects resulting from changes in thruster orientation, and this model is incorporated into the offline optimization.The online optimization relies on feedback from a disturbance observer based on nonlinear active disturbance rejection control (NLADRC). To validate the proposed method, both offline and online optimization simulations were conducted. The offline optimization results demonstrate that the proposed design achieves, on average, twice the maximum torque in all directions compared to the reference configuration. The online simulation results indicate that the proposed method helps maintain UUV stability under a 150 N disturbance, whereas the system without this strategy loses control. Moreover, when the configuration adjustment strategy is applied under the same disturbance, the thrust decreases by approximately 16%, avoiding saturation and operation in the dead zone. Real-world has also been done to validate the better performance of the reference configuration generated by the offline simulation. The results show an improvement of 1.25 Nm in yaw motion, overcoming the omnidirectional limitations of the reference. Modifying the SA-SQP algorithm's objective function to prioritize specific directions improved surge force by 11 N and pitch torque by 4.5 Nm. Real-world experiments have also been conducted to validate the superior performance of the reference configuration generated through offline simulation. The results demonstrate an improvement of 11.25 Nm in yaw motion, effectively overcoming the omnidirectional limitations of the baseline configuration. Furthermore, by modifying the objective function of the SA-SQP algorithm to prioritize specific directions, the surge force was increased by 11 N and the pitch torque by 4.5 Nm.

Date of Award	15 Jun 2026
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Adam Rushworth (Supervisor), Salman Ijaz (Supervisor) & Ahmed Nasr Abdelwahed (Supervisor)