More
I^3 Project: Predicting Ship Hydrodynamics to Enable Autonomous Shipping: Nonlinear Physics and Machine Learning
Project partners:
Institute for Ship Structural Design and Analysis (TU Hamburg), Dynamics Group (TU Hamburg)
Project members:
Sören Ehlers, Franz von Bock und Polach, Simon Haberl
Funding party:
TUHH i3 Programm
Duration:
01.2022 – 12.2023
Summary:
This I3-project focuses on enhancing the hydrodynamics of ships in waves. The challenge is to create a cyber-physical system, driven by computer algorithms and environmental sensors, ensuring collision avoidance, safety, efficiency, and environmental impact reduction in real-time. The project aims to develop a simulator-based test environment, creating a Digital Twin (DT) for ships and structures in waves. Two DTs will be built: one based on physical and numerical models, and the other on Machine Learning (ML) algorithms. The physical DT will serve as the foundation for an accurate simulator, addressing non-linear effects efficiently for rapid ML training data generation. Efforts include overcoming challenges in coupling seakeeping simulations with Finite Element Method (FEM) tools and using ML to perform complex tasks in real time. Results from the physical DT will train the ML-DT. The project concludes with verifying and demonstrating the applicability of both DTs.
