Bachelor and Master Theses

Currently, there is no universally accepted standard for developing and verifying autonomous vessel systems, with jurisdictions imposing varied requirements that hinder the creation of a unified verification protocol. This makes determining liability and accountability in the event of a collision, complex, as verifying whether collision avoidance protocols were properly followed or whether a failure was due to technical glitches, software bugs, or external factors is challenging. Autonomous vessels must also navigate international and territorial waters governed by different regulatory frameworks, leading to operational difficulties and potential legal conflicts. Additionally, maritime law is continuously evolving to address technological advancements, and vessels registered in one country may face additional compliance requirements when entering the waters of another.

To address these challenges, Sreekant Sreedharan and collaborators at NTNU have developed Legata within the context of the SFI Autoship project (https://www.ntnu.edu/sfi-autoship). Legata is a domain-specific language designed to ensure regulatory compliance in autonomous maritime vessels. Using large-scale simulations, Legata translates legal regulations into computable terms, enabling precise risk assessment based on regulatory violations across diverse scenarios.

In this thesis, we will study how we can use large-language models to transform Legata clauses into other formats to extend the analysis possibilities of the language.

We are looking for a motivated student to take on this exciting research opportunity and to work together with Sreekant Sreedharan and collaborators at NTNU in Norway. The student should be excited to experiment with new software engineering methods and tools. Knowlege of model-driven engineering is helpful but not mandatory.