On Wednesday, October 18, from 11 a.m. to noon, the DigiShape AI working group is hosting an online technical session on Ship as a wave buoy: Directional sea state estimation from ship motions. The speaker is Bart Mak from Marin.
Content and goal
Reliable sea state estimation is important for maritime operations as it improves safety and efficiency. Direct measurement of the sea state is complex and expensive, especially on moving ships. However, under the assumption that a specific wave train, the loading conditions of the ship and the forward speed of the ship uniquely determine the resulting ship motions, a long enough measurement of the ship’s motions can be used to infer the sea state. Well-known methods based on this idea use the motion spectrum and a mathematical model of the ship with Bayesian inference to get to the wave spectrum.
In our research we focused on deep learning on time series measurement data and known sea states in order to overcome some of the limitations of existing methods. We will explore how and why we implemented deep learning for sea state estimation and some of the benefits and pitfalls of this approach as well as some practical aspects.
Practical information
- What: Online technical session
- When: Wednesday, October 20, from 11 to 12 a.m.
- Location: Microsoft Teams
- Register here
