Robots That Park Your Car: Twelve Years On, and Now Bookable
When you spend hours researching computer hardware and AI chips, as I did recently, the social media algorithms eventually figure you out. My feed shifted from random clips to heavy industrial robotics, and I’m not complaining, because sometimes you find a solution to a problem you didn’t realize was that urgent.
The clip that stopped my scroll showed a low, flat robot sliding under a large Mercedes SUV, lifting it, and carrying it off to park it. It looks like a gadget for the very rich. Then I went looking into it, and the interesting part turned out not to be the machine at all. This technology has been parking cars at a French airport since around 2018, it went on sale to passengers at Gatwick this August, and it already failed once in Germany for reasons that had nothing to do with the engineering.
Why a robot beats a self-driving car here
The obvious objection is that cars will drive themselves soon enough, so why build a separate machine for this. That misses what a parking garage is short of, which isn’t intelligence. It’s space.
A self-driving car still needs room for a person to open the door and get out, and a lane wide enough to turn into a bay. A garage built for humans is full of dead air: door swing, walking routes, turning circles, and the margin everyone leaves so they don’t scrape the car beside them.
A lifting robot removes all of that. You leave the car in a wide bay and walk away, and the machine can then set vehicles down centimeters apart, in rows nobody has to walk through. Gatwick sums up the customer side in one line on its booking page: drop the vehicle in an extra-wide garage, let the robots park it, keep your keys.
How much does that gain you? Stanley Robotics, which builds the Gatwick system, advertises “+50% capacity in your car park” and describes its high-density storage as “increasing the number of cars in a given area by up to 50%”. I could find no audit of that figure by anyone who isn’t selling the system. Worth noting alongside it: Stanley’s own Gatwick announcement contains no capacity claim at all, and no figure for how many robots or spaces are involved.
Germany went first, and it didn’t take
A robot called Ray started handling cars at Düsseldorf Airport in June 2014, built by Serva Transport Systems in Grabenstätt, Bavaria. The German press called it the world’s first parking robot, and that needs one qualification: automated parking garages are much older, going back to fixed mechanical systems in the 1920s. What was new about Ray was the direction of travel. The machine came to the car instead of the car being driven into a machine.
Two years in, an airport spokesperson gave the German press agency dpa some numbers. At peak the robot ran at about 60 percent of capacity, and in normal operation closer to 30 to 40 percent. The airport’s own explanation was that people had reservations about handing their car to a machine. Serva’s managing director pushed back in the same report, saying no damage to a customer vehicle had ever been reported to them and that the risk in a conventional garage was higher.
I wanted to know whether Ray still runs. Düsseldorf’s current parking overview lists twelve car parks, P1 through P25, and none of them is a robot. No press coverage I could find goes past 2016. That isn’t proof it was switched off, and no source I saw announced a shutdown, so I’ll leave it there. What stayed with me is the shape of it: the engineering worked, and the machine was not what stopped it.
Where it is now, and how long that took
Stanley Robotics, founded in France in 2015, put robots into the outdoor car parks at Lyon Saint-Exupéry after a partnership signed around 2017, which the company describes as the first time robots parked vehicles outdoors. London Gatwick and Basel Mulhouse signed up by 2020.
Then not much visible happened for years. In July 2026 Stanley announced that Gatwick would become the first UK airport to offer robotic parking as a product, at the South Terminal, with the first passengers using it in August. Gatwick now lists Robotic Parking on its own booking page next to Long Stay and Valet.
Six years between an airport signing up and an airport selling it is the number I keep coming back to. It says more about how this technology arrives than any launch video does, and a global pandemic that emptied airports explains part of it but not all of it.
Stanley’s Gatwick announcement lists what the robots can take, in its own words: “a maximum wheelbase of 3.3 metres, height of 2.3 metres, wheel diameter of 21 inches and weight of 2.6 tonnes”.
This is the least glamorous part of the story and the only part that decides whether it works for you. The technology is usually filmed carrying something large, so it’s worth putting your own car against those four figures before you book.
The case I find most convincing
It’s easy to look at AI and robotics as things that arrive to replace people. Here I see the opposite, and it has more to do with demographics than with technology.
Picture an older driver in a dark, tight, multi-level underground garage. Reversing into a narrow bay asks for neck rotation, spatial judgment and quick correction. I have no data on this and I’m reasoning from people I know rather than from a study, but my strong suspicion is that parking, not driving, is what pushes a lot of people to hand back their keys sooner than they had to.
A wide transfer bay removes that specific obstacle. You stop, you get out with room to open the door, you leave. It’s a narrow benefit and I like it for being narrow.
What it doesn’t fix
It doesn’t reduce the number of cars. It packs them better, which is worth real money in a city where a garage can’t grow sideways, and it’s a different thing from needing fewer parking spaces. A denser garage can just as easily mean more cars arriving downtown.
And it adds a failure mode a normal garage doesn’t have: your car sits in a rack you can’t walk to. I assumed this was a hypothetical until I looked it up, and it isn’t.
In August 2006, the Garden Street Garage in Hoboken, New Jersey stopped working with hundreds of cars inside it. The cause was not a breakdown. The city’s contract with the vendor, Robotic Parking, had ended, the software license expired with it, and the garage simply stopped. Owners got their cars back after a settlement in which the city paid for a new three-year license.
That is the version of the risk I hadn’t considered, and it isn’t really about robots. It’s about a car park that only runs while a commercial relationship holds. Stanley Robotics itself has been a subsidiary of the Korean group HL Robotics since 2024, which is not a criticism of anyone, just the answer to the question of whose machine your car depends on. It’s a fair thing to ask at the desk.
In a nutshell
Robotic parking wins over a self-driving car for this job because it removes the space people need, not the space cars need. It isn’t new: Düsseldorf ran one from 2014 and the obstacle was drivers, not the machine, and it no longer appears among the airport’s car parks. It’s now bookable at Gatwick, six years after the airport signed up, with four vehicle limits worth checking first. And the documented worst case, in Hoboken in 2006, was a contract expiring rather than anything breaking.
Stanley Robotics, press release of 23 July 2026, for the Gatwick launch, the South Terminal, the August start and the four vehicle limits; the company’s own history page for the Lyon partnership, Gatwick and Basel Mulhouse signing up by 2020, and the 2024 acquisition by HL Robotics; and its valet parking page for the capacity claim. London Gatwick’s parking page for the live listing and its wording. Heise online, Parkroboter Ray hat nicht genügend Freunde, 10 July 2016, reporting dpa, for the Düsseldorf figures and both quotations. Düsseldorf Airport’s car park overview for what is on offer there today. Autoblog, 8 August 2006, for the Hoboken shutdown and the license settlement. Deployment status changes quickly, and the Düsseldorf figures describe 2016.
