Renowned British marine geospatial data, seabed mapping, and chart-producing agency, the UK Hydrographic Office (UKHO), and British maritime autonomy software company Marine AI, have jointly launched the world’s first programme to give Maritime Autonomous Surface Ships (MASS) the ability to read, interpret and act on official navigational information that currently could only be used by humans. This research project would run for eight months and would involve Marine AI refining its baseline large language model (LLM) to process ADMIRALTY Sailing Directions and Radio Navigation Warnings, which were currently written in natural language for interpretation by people. (ADMIRALTY was the brand under which the UKHO supplied its products to its customers.)

Once Marine AI’s LLM had processed this information, it would be fed, in structured form, into the company’s GuardianAI software suite, allowing for the autonomous control of an uncrewed vessel. Achieving this would allow MASS to make real-time safe decisions using the same data as professional mariners. This, in turn, could transform the global shipping industry.

“This is the first time anyone has attempted to process ADMIRALTY Sailing Directions and Radio Navigation Warnings in a way that an autonomous control system can act upon,” highlighted Marine AI technical director Oliver Thompson. “By proving this capability on the water, we are closing one of the biggest gaps in MASS autonomy and taking a major step toward safe, fully automated operations.”

The project required that a bespoke LLM be retrained, and supporting AI agents developed, to restructure data contained in unstructured text and legacy broadcast formats and which was not machine-readable. This structured data would then be fed into the GuardianAI tactical engine and human-machine interface.

The climax of the project would be live on-water trials using an Oceanus12 vessel, from ZeroUSV. This would be fitted with the GuardianAI suite and would be demonstrated on the waterways of the English naval port city of Plymouth, in parallel with simulations. These tests would take place in the northern spring, next year.

The project would also support the development of the International Hydrographic Office’s (IHO’s) S-100 data framework. This would be the new universal standard for future navigation technologies.

“Working with Marine AI allows us to push the boundaries of how autonomous systems can use official hydrographic information,” affirmed UKHO Research, Design and Innovation head Mark Casey. “The outcomes will not only support the safety of lives at sea but also feed directly into the development of the IHO’s S-100 framework, ensuring that UKHO data continues to set the global benchmark for safe navigation in both crewed and uncrewed vessels.”