Paul Burton, MASS Technical Engagement Manager at the UK Hydrographic Office (UKHO), addressed the 11th MASRWG Conference on the critical role of navigation data for Maritime Autonomous Surface Ships (MASS).
The forthcoming IMO MASS Code, a high-level, goal-based regulation, mandates that data for autonomous systems must be in a “machine-readable format.” However, interpreting this requirement is complex. The extent of data needed depends on the specific navigation sub-functions (voyage planning, situational awareness, collision avoidance, etc.) the system is designed to perform. While basic functions might rely on today’s Electronic Navigational Charts (ENCs), more advanced autonomy requires a much broader dataset.
Burton highlighted a fundamental challenge: for over 200 years, nautical information has been optimized for human interpretation. Critical data like Radio Navigational Warnings, Sailing Directions, Local Port Regulations, and numerous textual notes on ENCs are often descriptive and unstructured. “Machine-readable” does not automatically mean “machine-understandable.”
The solution involves two parallel paths:
- Data Standardization: The International Hydrographic Organization’s S-100 framework (including S-101 for ENCs, S-124 for warnings) is developing structured, machine-friendly formats.
- Technology Advancement: Systems must evolve to interpret unstructured data. Promising work with Large Language Models (LLMs) is underway to enable AI to parse textual instructions and define appropriate courses of action. ( ref. MARINEAI )
Significant limitations remain: incomplete, outdated, or non-existent source data in many parts of the world. An autonomous system must recognize areas of low confidence (e.g., based on lead-line surveys from the 1800s) and adjust its planning and reactions accordingly, just as a trained human navigator would.
Furthermore, global coverage of new S-100-based products will take years due to varying national capabilities, potentially constraining where advanced autonomy can be safely deployed.
Note on Physical vs. Virtual Navigation AND Human vs Machine Perception : In response to a question during the Q&A session, Burton noted that autonomous ships, like crewed vessels, will continue to rely on physical conspicuous points and real aids to navigation (buoys, beacons) for position fixing. He expressed skepticism about the widespread adoption of “Virtual Aids to Navigation,” emphasizing that it is not reassuring for safety to depend on a virtual signal (e.g., AIS) if the corresponding physical object is not visible or does not exist. The physical navigation environment will therefore remain fundamental: “machines” will do what many OOWs are not doing any more – look outside of the bridge windows. IALA
Source: Paul Burton, MASS Technical Engagement Manager at the UK Hydrographic Office (UKHO). Presentation given at the 11th Maritime Autonomous Systems Regulatory Working Group (MASRWG) Conference, January 22, 2026.