At Heathrow, a Wheelchair That Drives Itself to the Gate

On June 3, 2026, at Heathrow's Terminal 3, a traveler with reduced mobility settled into a wheelchair, tapped a screen to pick a gate, and let it carry them. No one pushed. Bristling with sensors, the chair found its way through the crowd, steered around suitcases and children, slowed at junctions, then stopped outside the boarding area. Once the passenger stepped off, it drove itself back to its charging dock.

The chair is called WHILL. It already runs in Tokyo, Los Angeles, Miami and Copenhagen, and its maker claims twenty-six airports equipped and more than a million autonomous rides. Behind the demo sits a precise idea: giving someone who can no longer walk the ability to cross a vast space without waiting for a person to come and fetch them.

Waiting is the real handicap of travel

For anyone who moves by wheelchair, an airport is not first a walking problem, it is a waiting problem. The assistance service exists, and in Europe it is even mandatory, but it runs on its own schedule: you book, you wait, you depend on an agent pushing several passengers in a row. A flight missed because assistance ran late is nothing unusual.

The autonomous chair shifts the balance. The traveler selects a destination on a screen, leaves when they want, stops at the restroom or in front of a shop if they decide to. At four kilometers an hour, the pace of a brisk walker, with a range of about ten kilometers, it covers the distances of a large terminal with ease. What it gives back is not walking, it is control over one's own route.

The gain also shows on the ground. Airports struggle to recruit assistance agents, a hard and poorly paid job, just as demand swells with an aging traveling public. A chair that drives itself back to its dock frees staff for the gestures that truly matter: boarding, the transfer, the luggage.

The need is anything but marginal. At Europe's big hubs, requests for assistance to passengers with reduced mobility run into the millions each year and climb faster than traffic itself. Against that curve, adding human arms no longer suffices, and automating part of the journeys becomes less a gadget than a logistics answer.

A LIDAR on your lap

The machinery borrows from the self-driving car, only slower and more cautious. The chair carries a LIDAR, cameras and ultrasonic sensors that sweep the few meters ahead of it without pause. An obstacle appears, a traveler cuts across its path: the machine slows or stops dead, exactly like a car in an emergency halt.

That caution has a price: slowness. Where an agent quickens the pace to catch a boarding call, the chair keeps its regulated speed and halts at the slightest doubt. In a packed terminal, safety imposes a timidity the hurried traveler will have to learn to accept.

But the comparison ends fast. Unlike a car turned loose in an unpredictable city, this chair travels only on routes mapped in advance, inside a closed building, air-conditioned, lit, with no rain, no ice, no cyclist bursting from a side street. The airport is the ideal ground for autonomous driving, a marked-out world where almost everything is foreseeable.

That is where the real feat lies, and its limit. The chair does not find its way in the world, it finds its way in a map. Take it out of the terminal and it becomes an ordinary wheelchair again. The autonomy it offers is real, but it hugs exactly the contours of the territory the operator was willing to digitize.

A freedom that stops at the gate

The chair does not do everything. It drives to the boarding area, not to the aircraft seat. Airports say it plainly, the machine complements assistance, it does not replace it:

• boarding the aircraft and stepping off it stay manual;
• the transfer into the narrow cabin wheelchair remains an agent's job;
• so does handling bags and helping in case of a fall or faintness.

Those who need to be carried will still be carried.

Then there is the matter of the gaze. A device that films and scans its surroundings constantly in order to move produces, by design, data about the places and people it passes. Nothing points to misuse, but freedom of movement is paid for with a presence of sensors, in a space, the airport, already saturated with surveillance.

Finally there is dependence displaced. Yesterday the traveler depended on an agent; tomorrow they will depend on a fleet, on software, on a map kept up to date and on a free dock. The day the system breaks down, the route is stale, or every machine is taken, someone will still have to come, the old way. The independence regained hangs on an infrastructure the user does not control.

The terminal as a prototype of the rest

The Heathrow chair is not an endpoint, it is a test bench. The same machines are starting to roll through hospitals, shopping malls, campuses, anywhere a space vast and regular enough can be mapped once and for all. With each newly digitized place, the territory of autonomy widens by a notch.

The real question is not whether the machine can drive itself, it already can. It is how far we will agree to map the world so that it keeps doing so. Giving back freedom of movement to those who had lost it is a promise kept, within a perimeter. The whole game now turns on the size of that perimeter, and on what we will consent to in order to enlarge it.