Three students from the University of Southampton are using AI technology to document repair work to the Battle of Trafalgar warship HMS Victory.

The university collaborated with the National Museum of the Royal Navy to explore how AI can support the museum in preserving the nation’s maritime heritage.

The project has seen three Master's degree students work alongside archaeologists at the museum to apply AI technologies to projects - including Nelson's flagship.

Since the restoration of HMS Victory began in May 2022, archaeologists have taken more than 3,000 images.

These additional images had to be stored and analysed manually, but aided by AI, this process is now automated.

Dr Rodrigo Pacheco-Ruiz is Visiting Fellow in Maritime Archaeology at the University of Southampton and National Museum of the Royal Navy Archaeological Data Manager for HMS Victory.

Dr Pacheco-Ruiz said: “Archaeologists are obsessed with detail and if records are not accurately stored, vital historical information could be lost forever.

"This is where the help of our University of Southampton students has been invaluable.

"They have developed an AI-based algorithm to match images stored in different locations and add them to our digital 3D model to ensure it’s as accurate as possible.

“This work has pushed the boundaries of what this type of AI is capable of, as our images are extremely high resolution, complex and detailed.

"The project is really at the forefront of how AI is being used in archaeology - we’re demonstrating just what’s possible. It’s certainly a long way from the traditional perception of how an archaeologist spends their time.”

“The project is really at the forefront of how AI is being used in archaeology - we’re demonstrating just what’s possible.”

The students are part of the CORMSIS research group – a collaboration between the Southampton Business School and the School of Mathematical Sciences.

Collections Information and Access Manager for the National Museum of the Royal Navy, Amy Adams, said: “By working with students from the University, we were able to use AI for entity recognition – to retrieve and sort through digital records.

"This has huge time saving potential, but also what was really interesting was AI’s ability to label records from the perspective of a visitor, so records are categorised by what people are likely to be searching for (‘WWII ship’ for example) as well as its official name."