£300,000 award to commercialise iceberg and sea ice detection using AI

There are approximately two to three shipping incidents involving icebergs each year in the Northern Hemisphere © Orlando Florin Rosu
There are approximately two to three shipping incidents involving icebergs each year in the Northern Hemisphere

A technique used to search for stars in the sky is being repurposed to detect icebergs at sea.

The highly successful method of detecting icebergs using AI is to be commercialised by Lancaster University with the help of a £300,000 award from the STFC.

The system uses an AI algorithm combined with satellite imaging to detect and track icebergs and sea ice in shipping routes in any cloud condition with a 94% success rate.

A number of maritime organisations are already interested in potentially developing this technology with the research team, who will use the £300,000 award to recruit a researcher and supply high-end computing equipment for Lancaster University’s supercomputer.

The project aims to collaborate with the maritime industry in developing a robust and precise detection system. This system is designed to enhance safety measures, refine risk modelling, enable real-time route identification, and drive substantial efficiencies in an evolving environment.

The technology is based on astrophysics research at Lancaster University, which used the same AI techniques to search for and characterise clusters of galaxies in large areas of the sky.

Astrophysicist Dr John Stott developed the AI technique with Dr Matthew Chan.

Dr Stott said: “We looked to apply our technique to areas beyond astrophysics and the vision for this project is to further develop our automated iceberg and sea ice detection system, such that it can be used as a commercial product.”

There are approximately two to three shipping incidents involving icebergs each year in the Northern Hemisphere, despite modern radar technology onboard ships. Several recent incidents have even involved tourist cruises having to abandon ship or caused severe damage to fishing vessels. There is also the commercial cost of diversionary routes to avoid icebergs in addition to false alarms detected on radar.

“Broadly, icebergs are pieces of ice in the sea that have broken away from a glacier and sea ice is thick ice that has formed in-situ. Ice that formed during the previous winter is of the greatest concern to shipping.”

The system takes large area synthetic aperture radar (SAR) images of the ocean from satellites and identifies icebergs and sea ice with a high level of accuracy. The aim is to send the locations of these potential hazards to clients within the UK and global maritime industry, including merchant shipping, fishing, tourist vessels and shipping insurers.

Throughout the development phase, the team will actively engage with InnovateUK's ICURe program, seeking both developmental support and industry partnerships. Simultaneously, they will explore the potential for establishing a high-growth spin-out company, specialising in delivering an AI-enabled software solution.

Lancaster University's innovative technology is initially tailored to deliver safety, logistical, and efficiency enhancements for the UK and international shipping, fishing, tourism sectors, and scientific research fleets. However, the scope of remote sensing technologies extends beyond these applications, offering diverse possibilities from agriculture to climate monitoring. The anticipated market growth and societal benefits from these advancements are substantial.

Back to News