Exploring LLM Capabilities for Situational Understanding and COLREG compliance on real-world maritime navigation scenarios

2026-08-08

AI Summary

This study presents the first application of large language models (LLMs) to real-world maritime collision avoidance, aiming to enable them to interpret complex navigational scenarios and adhere to the International Regulations for Preventing Collisions at Sea (COLREGs) alongside principles of good seamanship. To this end, the authors construct a dataset comprising 50 real AIS-based encounter scenarios, annotated with applicable COLREGs provisions, recommended maneuvers, and underlying reasoning. They further propose an evaluation framework grounded in scenario modeling, rule alignment, and chain-of-thought analysis. Experimental results demonstrate that current state-of-the-art LLMs, without domain-specific fine-tuning, struggle to reliably support maritime decision-making, underscoring the necessity of domain adaptation. This work establishes the first specialized benchmark and methodological foundation for intelligent decision support in maritime navigation.

Abstract

Recently, Large Language Models (LLMs) have shown considerable capability for situational understanding, reasoning, and decision making in different domains, most notable in the automotive sector. Therefore, we explore current state-of-the-art LLMs as a tool for maritime navigation, which includes both codified rules in the Collision Regulations (COLREGs) and uncodified best practices summarized in the concept of ``Good Seamanship''. We construct a dataset consisting of 50 diverse, real-world navigation scenarios from AIS data, label scenarios with applicable COLREG rules, recommended actions, and the reasoning for the action. We explore a variety of different LLM architectures and sizes to determine their understanding of maritime navigation tasks as well as evaluate their reasoning capabilities in this domain. The results obtained indicate that the maritime navigation task remains difficult to solve without fine-tuning, even for larger online models.

Problem

• Large Language Models
• Situational Understanding
• COLREGs
• Maritime Navigation
• Good Seamanship

Innovation

• Large Language Models
• Maritime Navigation
• COLREGs Compliance
• Situational Understanding
• AIS Dataset