Life Auto We have lift off: Toyota studying hover cars

We have lift off: Toyota studying hover cars
Twitter Facebook Reddit Pinterest Email

The future of cars may involve some lessons learned from the aerospace industry, if a comment from the head of Toyota’s global R&D effort is any guide.

Hiroyoshi Yoshiki was responding to a flippant question put to him by interviewer Brad Stone at the end of a discussion concerning driverless cars at Bloomberg’s Next Big Thing 2014 summit recently.

Acknowledging that the question was not to be taken too seriously, Stone asked Yoshiki and his fellow panellist, David Friedman of the National Highway Traffic and Safety Administration, to offer their best guesses as to when we might see something like the Terrafugia flying above the nation’s highways.

• Self-driving cars are ‘coming soon’, says Google
 This is not just a car. It’s a revolution
 Five ways to drastically cut your petrol bill

Friedman responded first, and his views could be summed up by his conclusion: “Let’s get the ones on the ground working really well first.”

Yoshiki’s reply was also ‘down to earth’, but in an aerodynamic context.

“We have been studying flying cars – in our most advanced R&D area,” he stated. But clarifying his point, Yoshiki explained that he wasn’t thinking of flying cars in the classical Jetsons style.


“A flying car means [one] just a little bit away from the road, so that it doesn’t have any kind of friction, resistance from the road.”

What Yoshiki was describing suggests Toyota is investigating ground effects to reduce aerodynamic drag under the car. It’s the same principle, but employed for a different purpose than in motorsport, in which the effect helps keep the car closer to the road and builds up downforce on the wheels for improved traction, braking and cornering.

Ground effects in other transport applications are used to break up vortices that can occur around low-wing aircraft wing tips at speed. Typically this manifests itself as increased lift and reduced drag as the aircraft approaches or leaves the ground.

“A flying car means [one] just a little bit away from the road, so that it doesn’t have any kind of friction, resistance from the road.”

In the area of automotive aerodynamics, car manufacturers have made progress in leaps and bounds since the early 1980s in reducing drag and eking out more speed or greater range for less energy expenditure, but most of the major gains are the result of careful styling of the car’s underside. Ground effects may be the next major step forward in that direction.

But it will likely be a long way off, as could be inferred from Yoshiki’s remark that Toyota was studying the effect in its most advanced R&D area. It therefore appears to remain very much at the theoretical end of research into this concept.


Driverless cars actually appear to be a much closer prospect. Yoshiki noted that the technology for cars to be driverless already existed, but there are a few gateways along the road ahead.

Toyota would not want to swap human error behind the wheel for human error coding software, as Yoshiki pointed out when asked about a timeframe for broad-based implementation of driverless car technology.

“It’s a tough question … our ultimate goal is to eliminate traffic casualties … Maybe Google shares [that] same value.”

The Toyota executive suggested that driverless vehicles might be introduced progressively, with larger, slower moving vehicles first (and networked).

“Maybe it’s a commercial vehicle… and [those] vehicles have the ability of communicating with each other – and communicating with the infrastructure – we may be able to let them go.

“But as far as regular, normal, city-driving type of situation probably it’s years away… really.”


Given that Google is already well advanced developing a vehicle that can drive without human intervention, and the software company had developed its system with the Toyota Prius first, Toyota’s reluctance to commit to a driverless future in the short-term seems conservative, not least of all when a number of European and American manufacturers are working hard at developing C2X systems.

As to what drew Google to the Prius as a test bed in the first place, Yoshiki ventured that it was the ‘drive-by-wire’ systems in the hybrid hatch.

“I don’t know why they picked the Prius,” he said. “But … the key technology is ‘by-wire’. Its brake pedal [and] throttle pedal are controlled by wire. What that means is you may think you’re physically kicking the brake down, but you are just sending the electronic signals to the computer – and the computer makes a judgement: ‘OK, now we need this much braking [effort]…”

Of course it’s just as likely Google wanted a car that would be an ideal showcase for new technology, and a poster child for current environmental thinking within the automotive industry.

This story first appeared in