AI-Sail
Machine learning applied to the maritime industry
Running
In MARIN Strategy Plan 2022-2025 ‘Beyond the horizon’, we have adopted challenging perspectives for the future of the maritime sector. One of these perspectives is taking the lead in artificial intelligence (AI), with the sea effectively deployed as a ‘digital lab’. This is not because we intend to create a maritime future full of emotionless robots but because we recognise how powerfully AI can support human decisions, from design to operations.
BACKGROUND
In the AI-Sail project, we will let an algorithm learn to sail an Optimist in our simulation and test environments with the overall objective of exploring the possibilities (and limitations) of Reinforcement Learning (RL) to control dynamic systems. This pioneering project will bring RL from simulation exercises into the physical world.PARTNERS
Each partner can choose between three levels of involvement:
- Listener: the participant will follow-up the project and will receive an overview of the progress done.
- Applier: The participant will have hands-on experience with machine learning using the Optimist environment. Additionally, an overview of RL in the maritime industry will be given.
- Designer: The participant will have an understanding of what is needed to implement machine learning to a specific application (this will be done using MARIN’s numerical environment). If applicable, a follow-up project can be proposed to tack a chosen application.
Deliverables
It is foreseen that this project will have the following results:- An experimental Optimist set-up prototype.
- Optimist and Offshore basin numerical environment to train Reinforcement Learning agents.
- Trained Reinforcement Learning agents with different learning approaches.
- Knowledge transfer to the industry through seminars, workshops and follow-up project proposals.
Contact
Julio Polo
Project Manager
Hannes Bogaert
President
Bart Mak
Senior Researcher
Work On AI-Sail Progress
In late April we had our first basin entry with the fully instrumented MARIN Optimist. The goal of the experiments was to assess the connectivity of the instrumented Optimist, our Offshore basin, the numerical environment and the Reinforcement Learning (RL) agent. We also gave a seminar describing the AI-Sail project and background. At the end of the seminar, we gave a sneak peak of the project in the basin to the attendees.
The next step will be to train the RL agent in the numerical environment. Progress on this has already been done. Together with the insights obtained in April, we are closer to train/test a RL agent in the basin.
The next step will be to train the RL agent in the numerical environment. Progress on this has already been done. Together with the insights obtained in April, we are closer to train/test a RL agent in the basin.
download
AI Sail leaflet (PDF)
Tags
data science