Paul Moller has been working on this concept for nearly 40 years (www.moller.com/skycar). They’ve done tethered flight tests and appear close to doing untethered tests.

Wouldn’t geometry obviate concerns about traffic jams? With a maximum ceiling of 32,000 feet, and assuming (for simplicity’s sake) flight corridors every 1,000 feet, we multiply today’s capacity by 32 even if we restrict ourselves to the space over existing roads.

Given the low cost of opening more “lanes” (zero, or the cost of another flight-traffic computer), we gain exponentially more traffic corridors.

If we really can make the cars fully automated, I think we’ll see a change in the market model from individually-owned vehicles to rental or taxi-like set-ups. Without the high overhead of a driver, flying taxis can be sent to pick up a passenger and return empty (no more soccer moms–just stick the kids in the taxi and send them to practice).

There certainly would be a lot of hybrid service models, like a service plan that ensures a flying car at your door at the right time every morning to get you to work. Different pricing levels would help distribute the fleet efficiently (7:45 a.m. costs 5X 7:00, and 10X the cost of 6:00?)

I’m sure there are things a simple layman like me is missing, but this appears promising.