As early as 1926, Henry Ford had conceived of a flying car. It was called the “Sky Flivver’’ but was abandoned when an early crash killed the pilot. “Mark my word,’’ Ford said in 1940, “a combination airplane and motorcar is coming. You may smile, but it will come.’’
If you Google “flying car,’’ you will find dozens of examples of this future vision, including the Moller M400 Skycar.
The Skycar looks the part, with an aerodynamic fuselage, a large, “Jetsons’’-like bubble window, and four massive ducted fans that rotate. Paul Moller has been working on the M400 car for more than 50 years, but the only demonstration of one came in 2003, and the product has largely appeared to be a $100 million bust.
Why haven’t these vehicles “taken off?’’
Simply put, it is hard enough to guarantee the safety of your daily commute on the ground. It is difficult, if not frightening, to imagine hundreds, or even thousands, of personal flying vehicles traversing the skies without accidents occurring.
For all the press a single plane crash receives, there are many more highway deaths each year: 32,675, according to recent data from the National Highway Traffic Safety Administration. It’s anybody’s guess what would happen if cars went airborne.
There is one thing that makes the development of a flying car more than just wishful thinking, however; it’s the autonomous car. As companies like Google work to create a car that drives itself, the possibility that such a car can achieve liftoff is far more likely than if such a vehicle were driven by mere mortals.
Cost and fuel efficiency are still among the largest hurdles to overcome to achieve a high-volume flying car, but if the public can accept autonomous cars, they might also accept autonomous flying commuter vehicles.
One technological challenge for autonomous cars is processing power. Modern cars may boast multiple sensors and cameras, but autonomous cars have a much more concentrated array of sensors, including radar and lidar, and multiple cameras. The simultaneous processing of all of their data is pushing current computational power to its limits.
In January of this year, at the Consumer Electronics Show, technology firm NVIDIA announced creation of the PX 2, an artificial intelligence supercar computer designed specifically for application in autonomous vehicles. The AI supercomputer is capable of 24 trillion computations a second. That kind of processing power can operate all the cameras on a car that will be looking for road lines, other cars, bicyclists, pedestrians, and anything else the average commute might throw in your path.
It also means the car can respond as fast, or faster, than if a human were driving it. In fact, Volvo has announced that some of the autonomous XC90s being made for the public will feature the PX 2 supercomputer.
Another key component in autonomous vehicle safety is vehicle-to-vehicle, or V2V, communications. This ability for vehicles to “talk’’ to one another means that numerous autonomous vehicles can be on the road at the same time. This same ability would be crucial for flying cars.
Woburn-based Terrafugia, a small, privately held corporation that is developing a flying car called the TF-X, understands these challenges completely. Founded by graduates from MIT, the firm’s initial product, the Transition, was a folding-wing aircraft that was more of a drivable plane. One would have to drive the Transition to a local airport, convert it to flight mode, and take off from there.
But after several years of showing the vehicle, the company went back to the drawing board and announced the design of the TF-X.
Instead of wings, the TF-X features a pair of tilting rotors, much like the US Marines’ V-22 Osprey. The TF-X, which is still years away from production, will be capable of autonomous flight, including auto takeoff and landing. Terrafugia claims the fully electric TF-X also will be capable of “automatically avoiding other air traffic,’’ suggesting the presence of this V2V capability. The company appears to have added many redundancies, specifically with safety in mind.
Ultimately, road-worthy autonomous cars will have to pre-date the flying car if “winging it’’ is to gain any sort of real traction. The only way regulators—and the public—would be willing to accept flying cars, is if a reliable track record can be established among autonomous cars. At the rate we are going, that day isn’t too far off.