Self-Driving Cars: How Ford Is Taking on The Parking Problem
A few months ago, I wrote a piece highlighting a few important issues that stand in the way of autonomous cars working well on California’s highways and actually being capable of driving themselves entirely without human input in all situations. One of those problems I highlighted was how difficult it would be to integrate autonomous cars into the current parking infrastructure.
Recently though, it came to my attention that Ford is working on a system that would address the parking problem. I wanted to know more, so I scheduled a call with Mike Tinskey, engineer at Ford and fellow Georgia Tech alum to talk about what he and the rest of his team are up to.
As we chatted about the big picture, Tinskey said that he doesn’t necessarily see privately-owned autonomous cars dropping people off at the door and then parking themselves. That’s, of course, something that autonomous taxis will probably be able to do, but what Ford is working towards is a future where your car takes you from your door to an open parking spot near your destination.
Theoretically, this system would also allow you to set a limit on how much you are willing to pay for parking and how far you are willing to walk. It should come as no surprise, but in one of their earlier studies, Ford’s engineers found that affluence was the most important factor in how far people were willing to walk for free parking. The more money someone has, the more likely they are to be willing to pay for the privilege of not having to walk.
If a car is going to be able to direct itself to an open parking space, it needs to be able to tell in advance which spaces are open and which spaces are full. Tinskey said it’s theoretically possible to put a parking sensor in every single space, but those cost around $600 each, and in order to put one on every parking space in the U.S. would be astronomically expensive. It’s also possible to use security cameras with built-in analytics to monitor available spaces, but the cost of a system like that works out to around $200 per space.
After looking at a number of solutions, Tinskey said it became clear to the team that the sheer number of parking spaces makes changing and updating the infrastructure prohibitively expensive. When they started looking at non-infrastructure solutions, though, they realized they could use cars themselves to track open parking spaces.
The brilliance of the solution Ford’s engineers came up with is that it only requires a software update to enable a car to start gathering parking data. That’s because what they’re using are the parking sensors that are already installed on a lot of cars.
Theoretically, with all the parking spaces geo-fenced, any time a car got close to a parking area, its sensors would turn on and start probing for open spaces. That information would then be recorded in a database for other other drivers to have access to. Their cars would know which spaces were most likely to still be open and would be able to drive directly to an open space.
The idea has already been successful in single vehicle demonstrations, and Tinskey said the next step is for them to try it out at their campus in Dearborn, Mich. Employees aren’t already riding to work in autonomous Fords, but even if they’re driving their own cars, the goal is to see how well the system works for helping them find parking spaces more quickly.
Before the end of 2015, Tinskey said Ford would also be launching a city-wide program in a location that has yet to be determined. Customer vehicles equipped with the necessary sensors will be asked to participate in a volunteer program to test out the software on a much larger scale. They will also have access to an app that will crowdsource information that’s hard for computers and radar sensors to capture, like which spaces are handicap spaces, which spaces are reserved for employees, etc.
The future of this parking system largely depends on finding a way to get more than just Ford vehicles to contribute data. No official steps have been taken yet to bring on other automakers, but the ability to offer a parking-finder feature to their customers has the potential to motivate several other companies to license its use from Ford. Tinskey said there is the chance the company may decide to open up its patents for other companies to use license-free as a gesture of goodwill, but it’s too early in the program to know for certain.
Assuming the city-wide test goes well, it looks like Ford is looking to implement this system sooner rather than later. If enough automakers see the value and license the technology for their cars as well, in 25 years, the idea of spending 20 minutes hunting for a parking space could be as antiquated as an in-car cassette player.