The internet seemed to resound with almost unmitigated delight when Google announced their progress on driverless cars last week. German scientists see a “golden future” for their driverless vehicles. There are, however, some key implications that are being missed about what it means if our cars are driven by robots. I’ll preface the rest of this post by saying that I think the benefits of robot cars probably outweigh the drawbacks. However, robot cars are not a panacea, and we shouldn’t overlook unintended consequences.
David Levinson at The Transportationist does an excellent job summarizing why robot cars matter, but in my opinion doesn’t go far enough explaining the potential downsides. Here are some of my thoughts on why we should adopt robot cars carefully, even with their myriad advantages.
(Another really detailed look at the potential benefits and drawbacks of robot cars, or “robocars”, comes from Brad Templeton, and predates all of the recent discussion by quite a few years.)
Our Waistlines and Our Public Places
As Levinson notes in his post, the adoption of robot cars will likely increase the number of trips by car, and increase their length (because the cost of these trips will be lower, both in monetary cost and time cost). We already know that heavy reliance on automobiles for transportation effects our health, and making auto transportation even cheaper will only increase this reliance.
Designing primarily around the automobile also turns our cities into “places not worth caring about“, in the words of James Howard Kunstler, as more and more land area is turned over to parking lots and right of way.
Sprawl (or Mega-Sprawl)
Robot cars hold the promise of reducing many of the negative effects of a long commute: travel times may actually go down for long trips, drivers won’t get stressed out about traffic, and time spent commuting can actually be used productively. With increasing mobile internet speeds, the commute may become just another part of their work day. Removing the negative aspects of commuting means that people can live farther and farther away from their jobs, shopping and other people. Levinson notes that the same phenomenon is evident with commuter rail, but robot cars would have a much greater impact. While there probably isn’t an unlimited appetite for road expansion (and certainly not unlimited funding), it is easy to imagine much more land being developed on the fringes if super-commutes become normal and tolerable thanks to robot cars.
Low-density land use has a greater impact on animal habitat, the water quality of our lakes and rivers, our physical health and climate change. The broad adoption of the robot car will make it even easier to continue developing in a low-density pattern.
What central authority will “operate” these robot cars? While they can be mostly autonomous, the Google robot cars uses detailed map information that was collected in advance and stored on Google servers. Google has raised hackles before over privacy, and if we can’t even adopt a mileage tax due to privacy concerns, how will people feel about a car that will probably be continuously communicating location, speed and many other pieces of “personal” data to server somewhere? I can only imagine the howls if the “government” were somehow put in charge of such a program. Brad Templeton has much more detail on all the operational hurdles robot cars would need to overcome.
While robot cars hold the promise (although it may be many years or decades off) of dramatically increasing the capacity of existing roads, they don’t solve the problem that has always plagued transportation planners, induced demand. This concept is also known by the phrase “you can’t build your way out of congestion”. If robot cars are successful at decreasing trip cost (time and money), then naturally more people will want to take more and longer trips. Capacity issues on roads will continue to be a problem, along with the attendant issues of construction, funding and land use impacts.
Are Cars The Best Way to Improve Accessibility?
In the end, the purpose of robot cars, like any transportation innovation is to give people access to more locations more quickly (and perhaps to improve safety). However, accessibility is a two-sided coin, on one side is transportation, on the other is land use. Could we achieve similar gains in accessibility by rethinking how we construct our communities? More density and more connected street networks can improve accessibility, and have the added benefit of reducing transportation energy use, keeping us healthy and reducing negative impacts to the environment.
I’m not aware of any efforts to compare the full costs of an accessibility improvement based only on transportation infrastructure to a similar improvement using land use changes, but perhaps we should work on that. We can’t really say for sure that the best way to improve accessibility is only on the transportation side without weighing all these potential costs and benefits.