In the Cooperative Research Ships (CRS) Data Driven working group, participants worked on several problem cases in ship hydrodynamics using data driven methods. The problem cases included significant event detection, performance prediction, whipping assessment and rudder control.
In this webinar on Thursday April 28, 14.00-16.30 CET, we will summarize these projects and highlight the main results and conclusions. The results of the Data Driven working group motivated us to continue our research in the CRS Fusion working group, where we aim to fuse domain knowledge and data driven approaches to tackle various challenges in ship hydrodynamics.
Scientific Machine Learning for Ship Hydromechanics | Gabriel D. Weymouth, University of Southampton
Can machine learning identify the conditions leading to significant green water impact events? | Tormod Landet, DNV
Ship performance prediction using machine learning | Thomas Scholcz, MARIN & Nico van den Heuvel, DAMEN & Eivind Ruth, DNV
Machine learning for fast assessment of whipping nonlinearities | Alexandru Andoniu, Bureau Veritas
Rudder control with Reinforcement Learning | Bülent Düz, MARIN