Remotely Operated Vehicles (ROVs) are widely used in underwater explorations and constructions. When navigating under the sea, the motions and stabilities of ROVs are influenced by various forces. Operating ROVs to capture images, grasp samples, avoid obstacles, and examine man-made facilities is challenging. This paper proposes an innovative simulator to train users to handle ROVs. Conventional simulators only can mimic undeformable ROVs. Their educational capabilities are limited. The proposed simulator is developed based on advance mathematics models and is able to emulate a ROV equipped with movable robot arms. Therefore, it can perform more realistic and meaningful simulations to enhance the training courses. We deduce a voxel-based method to calculate the mass properties of the ROV. Thus, the hydrodynamic effects caused by its robot arms can be calculated in real time. A numerical algorithm is also invented to coordinate the propellers to produce required forces and moments. Hence, the ROV can attain specified velocities, positions, and orientations under the influences of loads, sea currents, and moving robot arms. Furthermore, an artificial intelligence engine is integrated into the system to perform auto-piloting, auto-balancing, and command interpretation such that the simulator is more user friendly. Besides, the system is augmented with an interactive user interface and a graphics engine to display simulation progressions and information to increase its usability
Ueng, Shyh-Kuang and Chou, Chieh-Shih
"MOTION AND CONTROL SIMULATION FOR UNDERWATER ROBOTS,"
Journal of Marine Science and Technology: Vol. 28
, Article 12.
Available at: https://jmstt.ntou.edu.tw/journal/vol28/iss5/12