IMEC - AI for Smart Shipping

Outline

1. Intersection of AI and Smart Shipping
2. Solutions related to situational awareness
3. Solutions related to autonomous navigation
4. Existing Challenges
   1. Lack of open datasets: Issues with the existing datasets
   2. Lack of transparency and safety guarantees: How to incorporate varying rules in the trained models
5. Conclusions

• AI systems are maturing gradually in MASS
• Early applications are already witnessed in Industry ref. Seafar, MAHI
• Further research should be focused on responsible design of AI with the focus on safety guarantees and generalized behavior
• Open sourcing of data and simulators will push the boundaries of research in this domain

Published papers

• (Trinh et all. 2024) A comprehensive review of datasets and deep learning techniques for vision in Unmanned Surface Vehicles
• (Billast et all. 2022): Physical Ergonomics Anticipation with Human Motion Prediction