Very interesting concept.
Obviously inductive charging relies on pushing a current through a coiled wire to generate a magnetic field, which then induces a current in the coiled wire in the device to be charged. This would use a static magnetic field instead on the road, but ostensibly the movement of the car over that magnetic field could induce current.
I doubt that this would allow cars to run indefinitely. Because you require the motion of the car through the field to induce electric current, it will only work when the car is moving, and thus it relies on the car's own power. I suspect that by definition there will be a less than 100% efficiency, i.e. it will be impossible at say 70 mph for the rate of charge to exceed the rate of depletion (you know, no perpetual motion machines and all).
But if it works, it could be hell of a range extension feature. Let's say that at 70 mph, it is able to replace charge at 50% of the rate of depletion. All of a sudden a car with 300 mile range might be able to go 450. That's a major improvement, and while the road surface would be more costly, it doesn't require constant injection of electricity like most of the other inductive charging ideas.
The only thing that jumps out at me is the nagging feeling that I'm missing something--that there is a reactionary force applied when trying to move a coil through a magnetic field. I recall physics courses from long ago that if you tried to push a magnet into the middle of a coil, there would actually be force generated trying to repel the magnet's motion by the induced magnetic field being opposite to that you were introducing... I don't know if this would generate a mechanical resistance on the car opposite the motion of travel, which would defeat the purpose because then the motors would have to work extra hard to keep the car moving. You wouldn't be only encountering air resistance and the rolling resistance of the tires, you'd be encountering EMF opposite the direction of travel too...