Learning Six Degree-of-Freedom Ship Motion in Random Seas with Multi-Fidelity Data
Please login to view abstract download link
The design of ships and autonomous marine vehicles requires the understanding of the six-degree-of- freedom (6DOF) motion in the random ocean environment. Computational Fluid Dynamics can predict motion in breaking waves and extreme maneuvers, but its cost severely limits it use to assess many designs or environmental conditions. Current work is aimed towards learning 6DOF motion of a ship in random waves using combined CFD and lower-fidelity potential flow. The results will assess the ability to predict the RMS motion, as well as the distribution of motion with particular focus on the tails that are used for extreme values. If successful, the model will leverage data to generate the ability to evaluate additional wave environments at negligible additional cost.
