The Autonomous Vehicle - Savior or Killer of Cities?

Few doubt that the autonomous vehicle (AV) will come and that the self driving car will be a certain reality. The only question is when. Meanwhile there are many questions about what those vehicles will mean for cities. I was reminded of this when the $2.9 billion Baltimore Red Line rail transit project was stopped in the last minute before construction and someone commented that the AV would have arrived before the Red Line would have been completed in 2021 anyway and would have made it obsolete before it was operational.  Would AVs substitute for transit? Kill cities or enhance them? How can we know?

AV operations are inherently different from human-driven vehicles. They may be programmed to not break traffic laws. They do not drink and drive. Their reaction times are quicker and they can be optimized to smooth traffic flows, improve fuel economy, and reduce emissions. They can deliver freight and unlicensed travelers to their destinations. Daniel J. Fagnant, Kara Kockelman, The University of Texas at Austin 2013

Advertisement from 1957 for “America’s Independent Electric Light and Power Companies” (art by H. Miller).  Text with original: “ELECTRICITY MAY BETHE DRIVER. One day your car may speed along an electric super-highway,  its speed andsteering automatically controlled by electronic  devices embedded in the road. Highways will be made safe—by electricity! 

The experimental Google AV   

Will these attributes help to foster the renaissance of American cities that we observed in recent years or will they reverse decades of urban progress and open the floodgates to even more driving and sprawl?
As this Rand Corporation report notes:

While AV technology offers the potential of substantial benefits, there are also important costs. Ironically, many of the costs of AV technology stem in part from its benefits. For example, since AV technology is likely to decrease the cost of congestion and increase fuel economy, it will also likely decrease the private cost of driving that a particular user incurs. Because of this decline (and because of the increase in mobility that AVs offer to the elderly or disabled), AV technology may increase total VMT, which in turn may lead to increases in the negative externalities of driving, including congestion and an increase in overall fuel consumption. (Rand Corporation Report, James M. Anderson, Nidhi Kalra, Karlyn D. Stanley, Paul Sorensen, Constantine Samaras, Oluwatobi A. Oluwatola)

The two scenarios below are intended to demonstrate that even opposing futures can be envisioned, both seemingly plausible and possible. This means that to achieve desirable outcomes more is needed than allowing a new technology to run its course. It will take the right policies to actually shape the future. The time to think about what these may be in light of the AV is now!

Scenario 1, the AV is a city killer

With road capacity magically increased through AV, congestion has become a thing of the past and a commute has become a breeze. Who would mind moving a bit further out? Without the long stressful commutes which had effectively put the brakes on sprawl, the dispersal model of development has revved up again and jobs, housing, retail and entertainment have spread further and further out from the old town centers.

AVs stimulate additional road construction and sprawl

It’s simple, says Ken Laberteaux, a senior scientist at Toyota. If you make transportation faster, easier and perhaps cheaper, then people won’t mind commuting. “What a consumer is expected to do is see what they can gain by moving a little further from the job centers or the cultural centers,” he says. That’s bad news: Urban sprawl is linked to economic, environmental, and health hardships. (Max Ufberg, Wired Magazine)

In city centers, finding a parking space had once been a major barrier to driving. The hassle of parking had fueled interest in transit and all major US cities had begun building out their transit systems. No longer. Now AVs find the parking space by themselves, they park way more accurately than human drivers and they can move aside on demand. These abilities have reduced space demand drastically and increased parking supply without building any new garages. Similarly, the chaos of taxis, bicyclists, buses and pedestrians in city streets had once deterred many less confident drivers from even dreaming about entering big cities. All this is no longer a headache. The self driving car smoothly finds its way without any intervention while occupants do useful things. As a result people want to go places at all times of the day and night, including those who before were never found in a car because they were too old, too frail, too young or otherwise averse to drive. Transit investments have come to a standstill, buses have become a thing of the past and only huge metros still run subways. The new form of transit are large platoons of AVs moving at breathtaking speeds down the once congested freeways.

The new hyper-mobility has instantly eaten up whatever additional capacity the added efficiency had provided, a phenomenon long know by road builders as "induced demand." But where before centrifugal and centripetal forces found some kind of equilibrium that consisted in a ring around job centers equivalent to the area one can reach in about one hour commute (an isochrome), the higher speeds, convenience and safety of AVs have increased this "isochrome" line considerably. The last remaining green spaces have been gobbled up quickly in the same manner as the initial motorization made cities explode in area while often shrinking population. All external costs of driving such as land and energy consumption, pollution and infrastructure inefficiencies continue to rise while core areas are as congested as ever if not more so. The key difference is that fewer people care because being delayed in back-ups is now productive time and never means complete stand still thanks to the efficient management of overall flow of AVs.


Scenario 2, the AV is a savior of cities

Back when congestion and being fed up with long commutes had resulted in a renaissance and building boom in cities such as Denver, Austin or San Diego, there was still blight over large swaths of their downtowns from all the space that had be devoted to parking and storing automobiles. The arteries leading into the cities were an endless string of gas stations, drive through drugstores and fast food joints and city centers were pock marked way too frequently with surface parking lots, gigantic dull garages into which all the rolling sheet-metal piled in the morning, dozed all day long until departing in another big rush in the evening.

AVs reduce the space demand for the automobile and free cities to be
safe and pleasant spaces for pedestrians

The AV finally put an end to all this. Because the AV requires so much less space to park, developers rushed to find higher and better uses for those enormous garages adding tremendous new cash flow to city coffers from the intensified land use. Driving and possessing a car has completely fallen out of favor as a favorite past-time or a staus symbol, just like having a horse or a shine carriage had become obsolete a good century earlier. An AV is just not a good status symbol and there is only utility and no thrill in the ride. AVs have become the last step in a steady process towards the sharing economy. AVs took the Uber and the Zipcar principles of mobility, combined both and brought them to their logical conclusion. Instead of wasting large amounts of disposable income on purchasing a shiny car that sat around about 90% of the time doing nothing at all, the AVs are now commanded and accessed only when needed just like a Zipcar had been before. AVs are being ordered like Uber taxis used to be via smart phone and they now drive you where you want to go like a taxi, but without the driver. The few who still own their AVs rent them out during downtimes, a trend that had already begun in 2015 with flightcar.

A recent study by MIT's Aero/Astro department estimated that a fleet of 300,000 autonomous shared vehicles could serve the entire population of Singapore (almost 6 million people) with a maximum 15 minute waiting time during peak hours. (Ryan Chin, MIT)

The AV has led to a drastic decrease of the number of vehicles on the roads and in the garages giving valuable space back in cities, suburbs and anywhere in between. There is now much more space for people on foot, on bicycles or for outdoor activities like sitting in cafes to enjoy the newly gained spare time of the "post work" society where regular 9-5 work has become the exception. Cities have become quieter, cleaner and safer. AVs are really good at avoiding collisions and their efficiency in negotiating available streetspace has drastically decreased congestion, eliminated honking and reduced emissions. Most importantly, traffic fatalities have fallen to near zero. Hardly anybody, though uses the big AVs to get from out areas to the downtown centers. Many people live close to where they work or want to be for their activities. Others use the AVs to take them to transit hubs from where they get whisked into high density centers. Much of the transit of second and third tier cities can now be routed on the surface without much delay since smart programming made AVs and transit vehicles move optimally together.


Which scenario will be our future?

If one finds either one of these scenarios plausible, how could one be blamed? If only one of them seems attractive, action is required. Predicting the future is incredibly difficult. The hardest part is usually not the prediction of the technology itself but how it will be used and how technology will affect social change. A desirable future will hardly just arrive by itself like a self driven car.

automated pod

Given the apparent promise of AVs, policymakers and the public would be wise to seek a 6 smooth and intelligently planned introduction for and transition to this new technology. AV 7 technology seems likely to advance with or without legislative or agency actions. However, the 8 manner in which AV technologies progress and will eventually be implemented depend on these 9 efforts. Intelligent planning, meaningful vision, regulatory action, and reform are required to 10 address the issues. Daniel J. Fagnant, Kara Kockelman, The University of Texas at Austin 2013

To avoid the terrible scenario in which sprawl kicks in again, strict land use policies need to be put in place, controls that would have been useful all along to avoid the depletion of resources, the spoiling of pristine landscapes, the squandering of energy, the waste in infrastructure investment for unsustainable low densities. If those controls and safeguards will not be in place by the time autonomous vehicles will become commonplace, sprawl will indeed accelerate. As a result the US will fall behind further in urban quality of life, social and environmental justice, innovation and transit service. Worse, maybe, could decline because sprawl is not only environmentally unsustainable but also socially and economically. Roads, infrastructure and common cost for constructing and maintaining ever larger metropolitan areas will not become any more affordable or sustainable through AV technology.

Even if AVs can take the stress out of navigating through city traffic, cities can hardly tolerate an even larger number of individual vehicles with low occupancies. Even in the age of AVs transit is still necessary to move high volumes of people along high density corridors and towards concentrated destinations. AVs may be a substitute for transit in outlying areas where the problem is "the last mile" (from transit to the destination). AVs can enlarge the last mile to several miles, but when it comes to urban centers, AVs have to be kept out just like automobiles today. Neither London's congestion pricing nor San Francisco's variable bridge tolls would become obsolete only because vehicles can control themselves. To the contrary, with the potentially larger number of AV users compared to current driving, controls at the perimeter of core areas with intercepts are becoming even more essential. So will be good reliable transit which can pick riders up at those intercept points.

Innovation districts

The deciding difference in the two scenarios may, indeed be, how desirable private ownership of AVs turns out to be. Given the added space efficiency which will very likely make individual cars more attractive, the total number of vehicles must be decoupled from the increased number of trips to manage the outcome. Other consieerations should involve the question how autonomous vehicle technology can change transit itself. For example through self driving buses and an automated on-demand van-pool system that could serve mobility needs in currently unknown ways.

Given that AV technology will not become available on every vehicle all at once, pilots may be advisable to study effects and impacts in a controlled manner. What better place than progressive cities to create AV-only pilot districts? The innovation districts springing up in many US cities right now could be the right place to start. Linking AV technology to open data, sensing, sustainability and resilience, the main topics defining many innovation districts right now, seem to be the perfect guardians to keep the impacts of AV technology in check.


Klaus Philipsen, FAIA
edited by Ben Groff, JD
updated 8/17/15

You may find this interesting as well, also on this blog:

The next big thing in transportation will change everything

Links:
The Impact of Autonmous Vehicles on Cities
Guardian: Driverless cars, the future of transport in cities?
The self driving Tesla might make us love sprawl again (Slate)
How will driverless cars affect our cities?

Preparing anation for autonomous vehicles: 2 opportunities, barriers and policyrecommendations for 3 capitalizing on self-driven vehicles

Rand Corporation Report. Autonomous Vehicle Technology, A Guide for Policymakers