This certainly would not solve traffic congestion issues. Other modes of traffic address traffic much better than more cars/same number of cars does. If a car can't make it across town in an hour at rush time, then "everywhere cars an go" isn't that much of a selling point since cars can't really go that far then in a good amount of time.
> where every vehicle on the road is self-driving
Once everything is self-driving, they can coordinate much better than humans and fill roads to their capacity (as well as routing intelligently since we'll be able to have completely accurate traffic predictions, although that demands a level of widespread coordination that won't necessarily come at the same time as 100% self-driving vehicles). You can also throw vans/busses into the mix if the roads still can't keep up, or automated carpooling services.
100% self-driving vehicles will be able to fill roads to capacity for 2 reasons:
1. It's all automated, so most of the common causes for traffic jams simply won't happen (most traffic jams happen spontaneously once the road reaches a high percentage of capacity, rather than for any concrete reason that demands a jam, and self-driving cars will be able to avoid the turbulence that causes the jams). 2. Because it's automated, cars can also drive faster while remaining safe, and drive closer together as well. By the time we're at 100% self-driving cars the driving software will be pretty well-tested, so the only significant risk* will be a mechanical failure in the car, and even if that happens, because everything's automated the surrounding cars will be able to adapt much better.
*I'm talking here about accidental risk. Any kind of intentional issues (hacking the car, throwing heavy stuff at the car from another vehicle, etc) are not particularly relevant here.
Still, this does ignore adoption rates. In a more real world, I think you'd argue that adoption of self-driving cars would be quicker and easier than adoption of PRT, that might hamper my model. So, perhaps the conclusion of the article is right after all.
But in a city with people, cars will need to go slower and stop for people (or some combination). Even with instantaneous reaction, a car can only stop so fast.
By contrast, a road lane (which uses the same space) currently typically carries around 1000-2000 cars per hour. In ideal circumstances it can be almost 5000 cars per hour. Even assuming quadruple capacity from self driving cars (not realistic), 20,000 cars per hour in ideal circumstances would compete with trains at a far higher environmental cost.
But in more realistic scenarios with quadruple capacity, 4000-8000 cars per hour would fall way short of a train. Given that we have to account for pedestrians, cyclists, etc., I can easily see traffic lights continuing to exist, making self driving cars not achieve the miracles people expect.
