Advocates of self-driving cars have proposed that this emerging technology can provide the solution to the problems of urban mobility. Debates often center on whether the technology will be ready, and if regulations can catch up to the technological advances.
However, even if all of the technological and regulatory issues of self-driving cars are solved, they do not overcome one of the basic issues of urban transportation: capacity.
Cities are the key. They are engines of culture, economy, technological innovation, and ecological sustainability. For a city to be effective, it must move large numbers of people in, out, and around its core and region on a daily basis.
Cities function because they put so many people, ideas, cultures, and enterprises in such close proximity. If there were no value to agglomeration, population density on this planet would be spread roughly evenly across its habitable climate zones, but instead people cluster despite congestion and all the other headaches of urban life.
The very thing that makes cities tick makes them too crowded, too expensive, too dangerous, or too ugly. Jane Jacobs in The Economy of Cities called them “valuable inefficiencies.”
As it stands, cars take up too much space. Self-driving cars may prove to be a useful tool, but a successful future for urban transportation means reviving and modernizing mature technologies rather than replacing them with new ones.
“Passenger transportation is ultimately a simple problem: how do you get everyone to where they’re going?”
Does this mean that the future of cities has no place for self-driving cars? Not necessarily. They will be a useful tool if kept in perspective. Self-driving cars could be effective as feeders for a regional transit system in outlying areas. They could make car sharing services a more attractive alternative to private car ownership. They could increase the number of taxis in urban areas.
Despite its complexity in practice, passenger transportation is ultimately a simple problem: how do you get everyone to where they’re going?
How Transportation Shapes Cities
Urban historians like Kenneth T. Jackson in Crabgrass Frontier have traced how transportation advances have affected the patterns of urban space.
For the great majority of urban history, the footprint of a city was limited by the distance a person could walk in a day.
In the 19th century, the development of streetcars and railways allowed cities and suburbs to extend outward, creating gridded industrial cities and suburban town centers that clustered around stations.
Finally, the development of the automobile and the construction of highways led to the kind of suburban sprawl we see today: strip malls, subdivisions, and the concentrations of malls and office parks Joel Garreu calls Edge City in his book of the same name.
“The problem with this approach to self-driving cars is that it repeats the same mistake of the 20th century: seeing the problem of passenger transportation and the problem of car traffic as the same thing, and failing to recognize that older modes can do things that newer modes can’t.”
Each of these transportation revolutions radically altered the possibilities of moving through urban space, but each carried a tradeoff that made one aspect of the new mode worse than the others.
Trade-offs of Transportation
Transit – buses, rail, and ferries – is much faster than walking, but is less flexible. Urban transit generally follows fixed lines, and it requires people to depart and arrive from fixed points along those lines.
Automobiles combined the flexibility of walking with the speed of transit, and at the time they seemed to make transit – and even walking itself – obsolete.
Accordingly, we redesigned our cities to accommodate the great metal machines. We ripped out streetcar tracks, let passenger railroads wither, and tore out entire neighborhoods to build urban freeways and parking lots.
The great tradeoffs we failed to foresee were that automobiles burn a great deal of fossil fuels, they kill or maim a great deal of people who get in their way, and they take up a great deal of space. The people alive today have been living with the unintended consequences of these decisions ever since.
The promise of better cars is that they will supposedly solve these issues. The story goes that hybrids and electric vehicles, if combined with cleaner electricity sources, will cut the pollution of our atmosphere. Self-driving cars will be safer than human drivers. The problems of space will be overcome by the smarter, more efficient driving algorithms. Some, like YouTuber CGPGrey, have even suggested that self-driving cars will even eliminate the need for stops at intersections, but it is not clear when pedestrians are expected to cross the street.
The problem with this approach to self-driving cars is that it repeats the same mistake of the 20th century: seeing the problem of passenger transportation and the problem of car traffic as the same thing, and failing to recognize that older modes can do things that newer modes can’t.
Passenger transportation is not about moving the vehicles, it’s about moving the people in them. Even if self-driving cars solve all the problems of energy usage and safety, they cannot overcome the fact that cars take up so much space.
Solving the Capacity Issue: Cars vs.Trains
It may be possible for self-driving cars to increase their capacity relative to driven vehicles through the methods CGPGrey proposes.
However, these gains will be marginal compared to what cities can gain by devoting the same amount of space to transit.
If we assume that a typical car is 5 meters long and can carry 5 people (though in practice they usually only have 1 or 2), then this means that each vehicle carries up to 1 person per meter.
If cars in heavy traffic drive about 2 seconds apart, that means that in any given minute, a lane can put through 30 cars, for a capacity of 150 people, in theory.
If self-driving cars, thanks to their fast reaction times, allow vehicles to follow as close as half a second behind one another, then this increases the capacity of the lane quite a bit – up to 120. If each car has 5 people inside, that’s an impressive (if optimistic) 600 people flowing past a single point each second.
“A person in a car takes up the footprint of a car, while a person riding transit takes up the footprint of a person.”
How Does that Compare to a Train?
According to the Transit Capacity and Quality of Service Manual, a typical light rail car can hold up to about 8 people per meter, while a rapid transit car can hold about 11.5 per meter.
If each car is about 23 meters long, then a light rail train with 5 cars can hold 920 people while a typical rapid transit train with 10 cars can hold more than 2,500 people.
Under traditional signaling systems, it is possible to safely run trains just under one minute apart, but more effective signaling systems can cut that in half.
The constraint is often the capacity of the stations to load and unload people fast enough to make room for the next car or train. But, for the same reasons, stations have more person capacity than onramps and intersections. A person in a car takes up the footprint of a car, while a person riding transit takes up the footprint of a person.
We Can’t Self-Drive Our Cities out of Gridlock
Whether self-driving cars deliver on their promise is not just a question of technology but a question of how we choose to plan our cities and share our society. If we use self-driving cars to complement our fixed-route regional transit systems, we gain an effective tool to reduce our greenhouse emissions, instead of fitting more vehicles on the road.
If we address automation through stronger social insurance and a universal basic income, we share the benefits of self-driving cars fairly, instead of pushing the costs onto people who make their living driving cabs, buses, and trucks.
If we commit to modernizing our higher-capacity transit systems in cities regardless of whether self-driving cars overcome their technical design challenges, we can make room for even more of the cultural, economic, and technological exchange that makes cities so powerful.
On the other hand, if we convince ourselves that self-driving cars will make transit obsolete, we will repeat the mistakes of the 20th century, and spend the 21st stuck in traffic all over again.