David Alpert on driverless cars

I’ve been meaning to write a “how this urbanist stopped worrying and learned to love the driverless car” post for a while, but I’ve finally been spurred into action by this piece in the Atlantic Cities by Greater Greater Washington founder David Alpert.  Right up front I want to say I still have a lot of concerns about how we plan and incorporate robot cars, but on this issue of competing road users, I take a different view.

Alpert’s contention is that in our society’s haste to adopt driverless cars, we will “intensify current tensions” between drivers (more accurately called passengers in a robot car-filled world of the future) and non-auto users, such as pedestrians and cyclists, who are trying to use the same right of way.  I think this case is overstated for a number of reasons.

The author’s main evidence for the idea that tensions will be increased is reference to an animation done by some computer scientists that showed how to optimize an intersection when most of the cars are driverless, thus increasing flow.  According to the article,

[H]uman-driven cars would have to wait for a signal that would be optimized based on what everyone else is doing. And the same would be true of pedestrians and bike riders.

And to that Alpert reacts:

That certainly sounds like all other users of the road will have to act at the convenience of the driverless cars, under constraints designed to maximize vehicle movement instead of balancing the needs of various users…

The video even depicts an intersection with a whopping 12 lanes for each roadway, at a time when most transportation professionals have come to believe that grids of smaller roads, not mega-arterials, are the best approach to mobility in metropolitan areas.

Driverless cars, therefore, are poised to trigger a whole new round of pressure to further redesign intersections for the throughput of vehicles above all else.

I’m not sure how this one animation demonstrates why driverless cars would trigger a gush of road-building or elimination of non-auto facilities.  Setting aside the fact that I’m sure this animation was developed as a proof-of-concept (I can hear the research team now: “If we use 12 lanes in each direction, it will look even more impressive!”), this leads me to my first objection to the premise that driverless cars will increase tensions.

Driverless cars don’t make bad roads, people make bad roads

As Alpert himself states, “Already, cities host ongoing and raucous debates over the role of cars versus people on their streets. For over 50 years, traffic engineers with the same dreams about optimizing whizzing cars have designed and redesigned intersections to move more and more vehicles.”  Yes, and we’ll continue to have this debate into the future whether robot cars are adopted or not.  Given that gradual adoption of this technology is the most likely scenario (more on that later), I don’t see auto users getting more vocal (than they already are) about road capacity because there car has a few more widgets.

Building a balanced transportation system that looks at the full picture of quality of life rather than just mobility and speed will continue to be a challenge, although we seem to be making some progress in that direction.  Issues of public health, environmental impact and land use impacts will probably always take some extra effort to incorporate into transportation decision-making, an effort organizations like Greater Greater Washington should continue to make.  I view this as an institutional problem, failing to bring full information about transportation systems impacts to the design table, and it should be addressed in our decision-making processes.

12-lane at-grade intersections would make any cityscape pretty awful, but that leads me to my second objection:

Driverless cars can do more with less

Maybe the computer scientists at UT Austin should have showed a 2-lane 4-way intersection with driverless cars instead of a 12-lane intersection.  They also should have showed a comparison with a present day intersection.  One of the potential benefits of driverless cars is squeezing more flow or capacity out of the road systems we already have.  Cars can drive closer together, and yes, maybe intersections can look more India-like.  Potentially, we’ll get more from our existing concrete without having to widen or reduce non-auto infrastructure.

There is also this nagging funding issue.  In Minnesota for example, we already can’t pay for all the roads we want.  So 1) a huge explosion of more road-building probably isn’t likely and 2) driverless cars give us kind of another way out: if we’re intent on adding more capacity, maybe we can make our vehicles smarter rather than our roads wider.

And finally:

Driverless cars are safer

The first forays into “driverless cars” are about collision detection and avoidance (see a long list of existing implementation here).  Google’s driverless car has driven 200,000 miles and been involved in two accidents (both while being driven by a human).  Before any cars are driving themselves around, their computer brains will just be allowed to stop us from having accidents.  This is good for auto users and others alike.   And their adoption will happen gradually (they’ll be pretty expensive at first).

It seems obvious that driverless cars will be programmed to not hit pedestrians and cyclists.  Driverless cars will never (or very rarely) drive in a bike lane or right-hook a cyclist.  And for the next fifty years, they’ll probably be operating on roadways that look very similar to what we have today, pedestrian cross-walks and all.  The dys/utopian future where we have streets with tightly-spaced driverless cars traveling 200 mph is quite a ways off, and when that happens, why shouldn’t they be limited access and/or grade separated?  Wouldn’t we require the same of high-speed rail?

Again, there are lots of other potential negative impacts we need to be aware of as driverless cars become common (see my summary here), but I think these can be addressed by human policy decisions.  We also need to take some drastic action on emissions from transportation that contribute to climate change, and robot cars will likely not have a measurable impact there for some time (it’s also possible our action, if we take any, may actually delay their deployment).

Robot cars could offer urbanists a myriad of benefits that Alpert doesn’t address (but which others have covered in detail, but that should probably wait for another post.

3 thoughts on “David Alpert on driverless cars

  1. Great post. I’m glad that you called out UT Austin for their portrayal of that insane intersection. They should apply their models and analysis to some real world examples!

  2. Pingback: Why urbanists and others should love the coming of the robot car (Part 2) | streets.mn

  3. Pingback: The greenhouse gas benefits of autonomous vehicles | Net Density

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s