Network Rail to deploy AI-powered ‘crowd monitoring’ at Waterloo station

Automated security technology has been trialled at various transport hubs in recent years, having been developed by a Cumbrian SME backed by funding support from a Ministry of Defence agency

Network Rail is to deploy artificial intelligence technology to monitor crowds at London Waterloo station and provide staff with “digital ‘eyes on the ground’ that identify safety and security concerns”.

The Station Guardian platform, provided by Cumbria-based tech firm Createc, is based on light detection and ranging (LiDAR) technology – a type of remote sensor that uses lasers to measure distances. The development of the system was backed by monetary and programme support from the Defence and Security Accelerator (DASA) – a unit that sits within the Ministry of Defence with the remit of funding new defence and national security technologies.

DASA originally supported work on developing the tool as it had been identified as a potential means to help protect against the use of explosives in public spaces. Following several funding and development stages – and year-long trials at London Euston train station and Luton Airport – the tech has become geared towards more everyday concerns of managing the flow of people in busy transport hubs, according to an online post from DASA.

The MoD unit added that the platform now provides “machine learning to offer users a day-to-day system that notifies them of events requiring attention to help improve crowd management and safety, [which] could include notifications of suspicious activity, as well as incidents like a person falling, a malfunctioning escalator or elevator, or predications of how public transport delays can affect crowd flow”.

Related content

The sensor system “creates an anonymous track for each individual in the observation area, with the data being sent to cloud servers for analysis and alert generation for staff”, DASA said. It also provides users with “real-time comparison of crowd density is conducted with historical data from the same time of day, week, and year”.

As well as the longer-term exercises at Euston and Luton, the platform was also piloted for a month at Waterloo – and it is here that Network Rail has “acquired the autonomous crowd monitoring technology for deployment” later this year.

“The Euston and Waterloo trials proved the system can operate in a rail station environment, producing useful alerts when crowd density surged, and helping to predict and avoid overcrowding in the station during delays,” DASA said.

Waterloo is the UK’s largest train station, and data from regulator the Office of Rail and Road shows that it was the third busiest by footfall in the 2023 fiscal year, with almost 58 million people using the station. Close to three million trains arrived or departed from the station – which also connects to a London Underground station served by four different lines.

Sam Trendall

Learn More →

Leave a Reply

Your email address will not be published. Required fields are marked *

Thank you! Your subscription has been confirmed. You'll hear from us soon.
Subscribe to our newsletter