This technology coupled with solar panel covered highways could mean that your vehicule would just need enough to get to or leave the highway.
Putting solar panels flat on the ground loses 30% of their potential power.
Even ignoring all of that, covering solar panels with cars is a terrible idea.
If you don't mount flat you can then do tracking too.
Like you say. Mounting flat is probably the worst way to go from a utility and return standpoint.
To have solar win against fossils you need an area that is sunny and very dry. You also need very very cheap land. Watt/square meter of land fossils will win in most places within the US.
I'm not for or against, just that coke bottle glass is not the only way
The two problems I see with this idea are people complaining about the "view" (which is why it shouldn't be done on some really nice scenic part of a road, but in the middle of an ugly part of the city it should), and the possible problems with wind-loading, especially during storms, though surely this has been dealt with before, as there's a lot of places where they cover parking lots with elevated solar panels like this. For the view problem, they don't even have to cover the whole road with panels, they could just cover some lanes, put them in the median and extending over the inner lanes some, etc.
As a side note, you have to consider how in parking lots it is rare to have cars and trucks passing under the panels at 70 Mph, and more generally accidents are usually very low speed. Besides the "normal" resistance to wind, if you want to build a sort of half-pipe covering a highway you need to consider the pressure induced from the inside by high-speed moving objects and the possibility to resist accidents (preventing an 18 wheeler from completely crash a - say - 100 m section of the half-pipe in case of an accident needs a rather sturdy structure).
So, more or less it would be needed to build an artificial tunnel (concrete, rather thick) similar to a "snow shed":
https://en.wikipedia.org/wiki/Snow_shed
http://wsdotblog.blogspot.it/2014/03/the-i-90-snowshed-retir...
The "lighter" metal/lattice ones are only used usually for railways.
I mean, aren't these issues already basically solved? I frequently drive through miles-long tunnels under Boston at 70mph.
However, you bring up a great point with the "snow shed": covering all or part of a road with solar panels like this would also have the side effect of keeping bad weather off the road. How much money would that save in show-shoveling costs, and the costs to society of having roads blocked by snow (and snowplows), and having weather-induced accidents (both rain and snow)? Now, it's probably not feasible to cover every little road in a city with these things, but at least covering major arteries, while probably somewhat expensive due to the engineering challenges you raise, would have some major benefits: keeping the main roads clear even during horrible weather, and also providing a lot of energy from the solar panels.
And you have to consider how - in places where there is snow - you have to choose to either forfeit the (possibly little) energy the panels can produce in winter or anyway clean them from the snow anyway.
The good ol' rule of the thumb is that (of course it depends on the country) the cost is 10,000,000 Euro's per km of highway (plain, mainly earthworks), 20,000,000 per km on bridges/viaducts, 30,000,000 per km in tunnels. (the "base" price may of course vary, but the relative proportions remain roughly valid)
An artifical tunnel or sbow shed, besides the solar panels, would be probably 1.5 to 1.7 times the cost of the "simple" highway.
On the whole idea powering cars wirelessly while they drive just doesn't work on mass scales. It's going to be massively expensive to build it out.