Construction of a digital twin system for the blended-wing-body underwater glider

Abstract

The complexity of the marine environment adds unpredictable difficulties to researchers’ exploration. To study the application prospects of high-fidelity digital models to the maritime field, this study adopts the blended-wing-body underwater glider (BWBUG) as an example. A novel construction approach of an underwater simulation platform is proposed based on digital twin (DT) technology. The building process is described from the perspectives of geometry, physics, environment and behavior. Based on the Gazebo simulation platform, the proposed approach includes the construction of a high-fidelity model for underwater scenarios and is equipped with a vehicle body containing various sensors. Under the influence of ocean currents, the twin sensor data are integrated in the proportional–integral–derivative (PID) control algorithm, combined with a robot operating system (ROS), to achieve its pitch attitude control. The overshoot was reduced to at most 0.06% and the settling time is reduced to 29.5 s compared with the traditional PID. The results demonstrate that the proposed framework not only validates the implementability of control algorithms for unmanned systems in actual marine environment, but also has the potential for performance prediction, design improvement, and safety assessment in the next generation vehicle design process.