What question did this study set out to answer?

The research aims to develop a maritime routing system that utilizes AIS data to create an optimal routing graph for vessels.

April 1, 2026Open Access

AIS-Derived Maritime Routing Graph with Collaborative Fleet Learning

Key Points

The research aims to develop a maritime routing system that utilizes AIS data to create an optimal routing graph for vessels.
Constructed a waypoint graph from AIS vessel track data using Douglas-Peucker compression.
Implemented Fleet Collaborative Route Consensus for real-time route sharing among vessels.
Applied zone-stratified AIS compression to achieve significant storage reduction.
Enabled vessel-type-aware graph traversal considering draft, length, and beam constraints.
Achieved a median error of 1.2%, compared to 30-50% for traditional straight-line methods.
Demonstrated a 99% reduction in storage requirements while maintaining over 95% accuracy.
Utilized over 52 million AIS positions across 36,000 vessels and 12,714 ports.

Abstract

We disclose a novel maritime routing system (Mari8X OSRM) that constructs an ocean routing graph from real AIS vessel track data. Key novelties: (1) AIS-derived waypoint graph using Douglas-Peucker compression; (2) Fleet Collaborative Route Consensus (FCRC) — vessels improve shared route representations in real time; (3) Zone-stratified AIS compression (99% storage reduction, 95%+ accuracy); (4) Vessel-type-aware graph traversal (draft/LOA/beam constraints). Achieves 1.2% median error vs 30-50% for straight-line methods. 52M+ AIS positions, 36K vessels, 12,714 ports.

AIS-Derived Maritime Routing Graph with Collaborative Fleet Learning

Key Points

Abstract

Cite This Study