Self-driving cars will be a part of our future transportation systems, but I don't think they will transform those systems as drastically or in the specific ways most think they will. While autonomous vehicles will increase roadway utilization efficiencies (assuming we get to a point where they’re no longer sharing roadways with human operators), there's still the problem with urban geometry and the physical capacity of our roadways to accommodate traffic of any type.
I see autonomous vehicles being utilized more for commercial distribution and public transit systems. While personal automobiles will also benefit, I think they will be more useful as part of a feeder system in lower-density communities that provide convenient access to higher capacity transit systems nearby that bring them into the higher-density urban centers.
Not to mention most cities are starting to reconfigure how they manage traffic, no longer using level of service calculations and instead focusing on Vehicle Miles Traveled (VMT) as the metric for measuring significant impacts. This means that new developments will be required to provide less parking and other amenities that cater to and encourage any personal automobile usage.
While autonomous cars could potentially provide more efficient taxi or ridesharing services, this will be a relatively small percentage of the general mode share in most major cities. This would increase the efficiency of each vehicle and would hopefully reduce the number of cars circulating the streets looking to pick up passengers.
The real issue is that there is still a physical limit to how many single occupancy vehicles you can fit on a roadway, no matter how closely together we can pack them. With the continued increase in population, this is not going to be a solution to traffic congestion. Autonomous buses, trains, and other high-capacity transit modes would help to increase efficiencies, reducing wait times, and increasing capacity.
Many people seem to think that autonomous and electric cars will save the suburban way of life, but that just glazes over all of the other negative externalities associated with low-density sprawl. These systems won’t make infrastructure maintenance cheaper, won’t make traffic congestion go away, they won’t help to create a sense of community, won’t solve our environmental issues, won’t lower energy demand, won’t lower tax burdens, won’t lower the costs of transportation significantly if at all, and won’t fix the tangled web of streets that don’t connect or go anywhere.
Even if we shift to 100% electric and autonomous vehicles, our energy demand would far exceed the current rate of energy growth. Our supply of energy would therefore limit how many people can utilize a personal automobile, as energy costs increase due to demand. So while the technology seems exciting, I just don’t see it changing the fact that we need to continue to invest in high-capacity public transit systems.
People also seem to miss the fact that personal automobiles come with an extremely high hidden cost that we all pay without realizing it. The cost of parking in a suburban retail strip mall isn’t actually free. Consumers are paying for that parking through increased product/service prices. Parking adds significant cost to housing and office developments that are typically rolled into the price of the structure, whether the buyer or leasee wants that parking or not (although this is slowly changing in some cities that now require any parking be unbundled from the housing or office use). There’s also the added costs that go to pay transportation planners and engineers (like myself) to analyze all new developments and determine their impacts on traffic, parking, etc. in the surrounding environment.
When it comes down to it, we would save a lot of time, money, and frustration if we weren’t so depended on personal automobiles of any kind.