For air traffic controllers of the past, the new world of digital towers, augmented reality and AI will look like science fiction. Controllers of the future will wonder how people managed without all this technological assistance. In late May 2024, I was able to see the present and future at London Heathrow’s control tower as the guest of Andy Taylor, Chief Solutions Officer at NATS.
At the top of the tower is a panoramic view of one of the world’s busiest airports. At the base of the tower is a living demonstration of the future of ATC. This article explores the history of Heathrow’s tower, what it’s like being a controller there and the exciting future that NATS is showcasing in a demo area on the ground floor.
Visiting the tower
Visiting the tower involves a ‘behind the scenes’ trip through airport security. The process is similar to going through security as a passenger in Terminal 3 but perhaps a bit faster and in a rather more industrial environment. After a long walk to the very end of the pier at Terminal 3, we arrived at the tower, which is situated in the middle of the airport with good views out over both runways.
First, Andy took us to the cab, from which we had a panoramic view of the airport and, in the distance, central London.
Then we visited NATS’s innovation lab at the base of the tower where we had a glimpse of the future.
Augmented reality control tower
Not content with building a modern control tower, NATS is already looking to the future. Arranged around the gallery, there are clusters of high-resolution video cameras. They provide 360-degree coverage of the airport with additional PTZ (pan-tilt-zoom) cameras to give remote controllers virtual binoculars. These cameras are an essential part of an ongoing R&D programme into remote and digital control towers. They provide the reality feed for an augmented reality system.
Digital towers
These cameras feed into a prototype of a digital tower located at the base of the control tower and to a Virtual Contingency Facility (VCF) based at an undisclosed location off of the airport grounds, which opened in 2009. The current backup facility can operate at around 70% of the airport’s full capacity but a new facility, due in 2025, should increase that to 100%.
The prototype we saw was developed using technology from Searidge Technologies. It lets controllers visit and give feedback without having to leave their workplace. This is important because involving controllers as early as possible in a project like this is a key success factor, as we discussed in an earlier article about agile development in ATC.
It features three controller stations and one cluster of screens—a quarter of the eventual 360-degree view needed for the VCF.
The experience is quite remarkable. We looked at live video from the cameras we saw at the top of the tower just a few minutes earlier and looked out at the same airport panorama. The displays were unpixellated, sharp and vivid. The refresh rate was fast enough that there was no noticeable lag or rippling.
The system overlays information, displaying call signs and aircraft types that follow the aircraft as they move. This tagging provides an extra level of situational awareness and ‘heads-up’ information for controllers.
Each controller workstation had two large touchscreens and two large monitors. All four screens can be configured in different ways to help controllers manage their workspace. One benefit of the multi-screen approach is that controllers can connect to cameras that cover dead ground that is not visible from the tower itself, such as taxiways and stands between the B and C gates at Terminal 5. They can also connect to the PTZ cameras and use them as virtual binoculars. The use of commercially-available off-the-shelf hardware means that the system can be easily upgraded.
AI assistance
Going beyond remote presence and augmented reality, NATS is also working on AI technology to assist controllers. We saw two demos that show how powerful this assistive technology could be.
The first showed how AI could track aircraft turnarounds on a stand, recognise the movement of vehicles and services, and use that information to predict the likely TOBT (Target Off Block Time), i.e., when the aircraft will be ready for pushback. Watching the demo, it was remarkable to see the software select and highlight different service vehicles and turn that into easily visualised data for the controller.
The second AI tool tracked the position of a landing aircraft’s tail as it turned off runway 09L at Heathrow. Learning to recognise, by sight alone, when the runway has been vacated is a critical controller skill, and it can take months to learn. My colleague Julija has written extensively about how this kind of decision-support AI can improve safety without diminishing human agency and accountability.
I would like to thank Andy Taylor for arranging the visit and Matthew Stibbe, from Articulate Marketing, for the photos and for helping with the article.
