Driverless cars + Urbanism

I would love to start a discussion about the impact of autonomous vehicles on urbanism and real estate development patterns.

I think most people assume that it is too early to see an impact of this technology on cities, but with people sleeping in their Teslas on the highway becoming a regular occurrence (https://www.nytimes.com/2019/11/27/magazine/tesla-autopilot-sleeping.html), and Waymo offering for hire AV in parts of Phoenix, my impression is that this technology will have visible impacts in the near future.

Curious to hear any thoughts on this topic.

There is a great report from the American Planning Association on the topic.
See https://www.planning.org/publications/report/9157605/ .

PS: the summary is free. The full report requires a subscription. If you are in an academic environment, check with the urban planning or architecture department.

The report looks interesting. I think there has been a lot of discussion from an Urban Planning perspective which focus on road design and perhaps reduced parking. But I feel like this is somehow too narrow of a focus. It seems to me that the shift from cars to AVs is most similar to the shift from horses to cars. It is even kind of funny that some of the same rhetoric is being applied, for example one of the earlier names for cars was automobiles (which makes an AV an autonomous automobile I suppose). And while cars certainly had a huge impact on urban planning, there was also an arguably much larger and more pervasive impact on society in general. It is this larger impact which I don’t really see being fleshed out all to often.


You may enjoy this then

I happen to be one of the weirdos, unlike Musk and his acolytes, who likes public transportation and believes it serves much more than just transporting us in hyper-efficient manners from point A to B.

A diffusion of AVs in cities at the scales that rival the model T on urban roadways will require an unprecedented amount of control of the maintenance of street markings and street signs (currently the realm of public works agencies).

Highways are another story and I’m more than happy for all them Teslas cruising away and cutting the yearly 40k deaths in America by 90%

As a wildcard hot take, I think AVs would be great to modernize NYC Subways.
The MTA is on track to spend $7 Billion for Communication based Train Control which effectively means installed 40,000 sensors every 100 feet across 700 miles of track. For what purpose? Collision avoidance.

Instead, I would borrow all the cool Computer Vision, LiDar / Radar tech from the AV industry and make subways cheaper and more reliable.

Thanks for those, links that is some very interesting content that I hadn’t seen before. I’m excited to read Taming the AV.

I am also a big fan of public transportation. I recently got married and we took some wedding photos on the subway. That is a very interesting suggestion for retrofitting the NY system, they could even go a step further and rip out the tracks and put the vehicles on rubber wheels, in a similar vein to the Montreal metro. The tracks are primarily there to prevent the trains from hitting the walls, but with the LiDar setup that would also be redundant. I think single occupancy AV will never be able to match the same kind of capacity as a subway/light rail, and therefore these systems will always have an application in dense urban areas. However, I also feel like there is sometimes a false dichotomy that is constructed between AVs and public transportation. There are vast swathes of Brooklyn and Queens that have no subway access, and the bus system is not very reliable. AVs owned and operated as mass transit probably have some part to play in providing better service in situations like these.

I just spent some time in LA which as far as I understand one of the first major cities to massively adopt cars at the beginning of the 20th century. It is such a decentralized, variegated place, that I feel like it is a pretty good example of what the AV city will be like, although without the need to find parking, and probably roadways will increasingly dissolve into other types of shared public space.

Adrian,

Congrats on getting married and celebrating the moment on public transit! That’s super duper dope!

they could even go a step further and rip out the tracks and put the vehicles on rubber wheels, in a similar vein to the Montreal metro.

Emily Badger had some good thoughts

However, I also feel like there is sometimes a false dichotomy that is constructed between AVs and public transportation. There are vast swathes of Brooklyn and Queens that have no subway access, and the bus system is not very reliable. AVs owned and operated as mass transit probably have some part to play in providing better service in situations like these.

To me, AVs represents basic violations of accessibility and equity. AVs promise autonomy without addressing key equity issues. A basic proof of equity would be to demonstrate that an Autonomous vehicle can drive “autonomously” on “all” streets in a city - not just the few that have been paved and ploughed and marked.

Uber tried to “navigate” away from these issues with Mayor Peduto when asking that the city prioritize the ploughing of certain streets over others.

Alphabet via Sidewalk Labs effectively implying that they need to own the entire infrastructure supply chain to ensure autonomy.

Waymo testing in places like Tempe, AZ where residential roadways are effectively highways and not representative of median urban traffic conditions.

China’s Pony AI doing interesting things but knowing China, they’ll probably make it a federal offense punishable by death if anyone even accidentally vandalizes a street sign that ends up disrupting an autonomous trip

All for fixed route AVs aka Buses.

I just spent some time in LA which as far as I understand one of the first major cities to massively adopt cars at the beginning of the 20th century. It is such a decentralized, variegated place, that I feel like it is a pretty good example of what the AV city will be like, although without the need to find parking, and probably roadways will increasingly dissolve into other types of shared public space.

You will find an audience with Robin Chase and her R.A.T. thesis.

Yeah makes sense. I think in general we have a pretty similar perspective.

Here are a list of questions I feel like still remain for myself:
+What level of urban density will ultimately be best supported by AV
+Will total hours driven per capita increase or decrease
+Will the massive and autonomous application of electric vehicles help stabilize or destabilize an electrical grid based on renewables

Re: Density - there’s some good work done by Columbia some time ago that models the “optimal” number of AV taxis to serve NYC demand. It results in far fewer vehicles than current state.

Re: vehicle miles travelled (VMT) - I do believe that we will find new ways of monetizing mobility and that means “restaurant on wheels” which is more of a reaction to real estate prices than anything else. So I expect VMT to increase

Re: GRID reliability / electric utilities are already building capacity to deal with this but I think this is also a function of battery storage. Check usv’s climate investment thesis on this

Copyright Cornell Tech — 2020.