Zelim, a U.K.-based developer of maritime safety technology, has developed an artificial intelligence (AI)-powered system capable of automatically detecting and tracking persons overboard.

Currently deployed at sea on an oil rig for the first time, the person in water (PIW) detection system, dubbed Zoe, is designed to assist search and rescue operations at sea.

Source: ValarisSource: Valaris

To accomplish this, Zoe features a suite of cameras and sensors to monitor areas surrounding maritime vessels as well as maritime structures such as oil rigs. In the event someone falls from a vessel or other maritime structure, an alert is raised. As soon as that person falls in the water, Zoe’s software tracks them with a reported 97% rate of accuracy to a distance of roughly 330 m, the company explained.

The system, which has been trained on a dataset of over 4.9 million maritime rescue images over an assortment of ocean conditions, is currently undergoing a trial on the jack-up rig Valaris Stavanger. There, Zelim has installed seven infrared and optical cameras all around the vessel, which enables the entire rig to be monitored, giving operators a 360° view of the rig.

A PIW event will reportedly trigger an automatic alert from Zoe to a radio operator who can view a seconds-long clip of the moment when the alert was triggered while simultaneously viewing live footage. Zoe will continue to track the PIW, logging the position of the vessel, rig or other structure as well as the location of the person overboard for mayday call geo-location, along with a mayday script and actions checklist.

Zelim suggests that the technology promises to reduce rescue response times and improve the number of successful rescue outcomes.

For more on Zoe, watch the accompanying video that appears courtesy of Zelim.

To contact the author of this article, email mdonlon@globalspec.com