Ed and Eddie / Flickr
A few of the students in my UBC Urban Transportation Planning class are doing a major project on autonomous vehicles (driverless cars), so it has sparked some thinking on the topic. That’s why I found this post on Planetizen by William Riggs and Michael Boswell intriguing. In the transportation planning world, autonomous vehicles (AVs) are a wild card – we simply don’t know how they will impact the future of urban mobility. At a high level, there are two basic models in thinking about how AVs will change mobility and the cities: 1. radically increased efficiency will allow cities to accommodate increased auto-mobility without infrastructure expansion; and 2. the convenience of AVs (with electric/battery technology) will expand motordom through the constraints of congestion and resource limits, further feeling sprawl and auto-centric urban development. The Planetizen post takes the first model as its starting post and the argument is compelling.

“In this environment of uncertainty, we argue that the only certainty in how autonomous vehicles (AVs) will manifest in cities is uncertainty. While this unclear future might imply no need for a policy response, we believe there is a pragmatic approach to planning for the future of AVs: a temporary moratorium on roadway expansion. Under this moratorium, safety enhancements and regular maintenance would continue, but projects aimed primarily at capacity expansion would stop. This moratorium would include new freeways, interchanges, and major arterials as well as lane additions and intersection widening”

This conservative approach to estimating the future is probably the wisest considering the uncertainty (and the huge costs of expanding auto-centric infrastructure) and calls into question multi-billion dollar boondoggles like the proposed Massey Bridge (as if we needed another reason). But it also highlights the need for some serious policy thinking on the issue. Closer to home, UBC SALA’s AnnaLisa Meyboom (someone I respect) has called for Vancouver to play an agenda-setting role on planning for AVs. Meyboom, the director of UBC’s Transportation Infrastructure and Public Space Lab (what a fantastic initiative), identifies the uncertainty and the need for proactive work to determine whether these technologies will have negative or consequences for social equity and the environment. Notably, Vancouver City Councillor Geoff Meggs has called for City staff to look into this issue and I am looking forward to their report (of course, I think this is a great area for collaboration between the City and UBC – I know that my students would make some amazing contributions).


  1. Just hit my inbox and may be of interest to people reading this article who are in the UBC area this afternoon:
    “It’s a beautiful day out, so why not take Veemo for a spin?
    Surprise! We’re heading out to UBC to offer the first test rides of our beta prototype velomobile!
    You can find us at the flagpole at the end of main mall, near the Rose Garden, between noon and 3PM today”

    1. That looks like a very interesting product. A bike with enhanced cargo capacity, hill climbing ability and weather protection is certainly a welcome addition.
      I found an article suggesting a cost of 28 cents/minute which sounds quite reasonable for a vehicle with such flexibility. As an occasional use vehicle I can see value.
      For regular trips one is far more likely to choose a traditional bicycle supplemented by transit at just $4.20/day.

      1. Does the veemo have larger cargo capacity than a bike? Other than the shell isn’t it just an electric assist bike, only larger and heavier?

  2. Thank you for addressing the costs of road space.
    It is, of course, a public cost designed to support mostly private transport. As it is currently configured, there are very damaging external capital and social costs usually not accounted for when attaching figures to projects. There is also no attempt to evaluate the permanent operational costs of road infrastructure. Perhaps the cumulative figures are so high over time that they are politically unpalatable to publish.
    As many people are killed on the roads every six days on average in Canada as those who died last week in the terrorist attacks in Brussels. One data set is ignored; the other with equally tragic results gets extraordinary media attention. These are the priorities we have chosen in our society.
    It doesn’t matter what label one wishes to attach to cars, or what new technology will power and drive them, they are still motorized personal transport vehicles depending on public infrastructure, and in that form they will always impose a steep cost on public budgets and taxes.
    The laws of physics and a call to account for life cycle costs will force a levelling of these costs as sustainable urbanism moves closer to the forefront. Building walkable communities first, followed by transit-oriented urban design in all its forms will always be more cost efficient and safe than supporting personal transport with public roads and public energy on a massive scale.

  3. If we’re using the two models of thinking of AV’s outlined by Planetizen…
    Under the “radically increased efficiency model”, the business model for (1) vehicle ownership, and (2) the taxation approach to road use are likely to impact this result significantly. You can think of this a 2×2 matrix with “own the vehicle” and “pay for vehicle time” along one axis, and “distance charges” and “no distance charges” along the other axis. The resulting efficiency gains are likely quite different in these four scenarios. And additional factors will also play into this (changes in parking availability/pricing, price of competing modes, etc.)
    Under the “convenience expands motordom model”, I’m often curious as to why discussion of AV’s ignores the potential of AV Transit. Presumably if we can design a 4-seat vehicle to stop at lights/stop signs, turn corners, avoid cyclists than we can do that with 45-seat vehicles as well. AV Transit fundamentally changes the cost profile of transit operating costs (likely lower), potentially changing service delivery levels. So while AV vehicles might increase convenience, so too might AV Transit. In this case (1) corridor demand/capacity, (2) corridor price by mode, and (3) travel time by mode, are likely to be the major determinants of which vehicle capacity is deployed and how popular it becomes. Remembering Marchetti’s Constant, motordom and sprawl are generally only more convenient/attractive if travel speeds increase to open up that land. By providing priority for different vehicle capacities and/or pricing them differently is likely to influence mode choice….not much different than today.

  4. Autonomous transit vehicles are similar to driverless SkyTrain configurations with the exception that trains run on fixed and 100% separated tracks. If autonomous buses work, then bravo. They would be a lot more efficient and take up far less public road space than autonomous cars, even when they are joined in a “train.” But without designated, separate road space, there will always be delays with unpredictable congestion.

  5. Planetizen tells us that, “Elon Musk is marketing improved cameras, radars, redundant electronics, and software as a part of the Tesla Autopilot. This feature will allow drivers to “summon” a car from across the city…”. Sounds great. I will be able to head in to work, then set my AV free to zip back home across the bridge for my spouse to go off looking for the best deal on spot prawns. When I’m ready to go home I’ll summon my AV to pick me up downtown. If I am delayed, or if my AV is early it can just toddle around the block at 3km, with all the others, waiting in slo-mo without getting a parking ticket. Cool.
    When I get home I can just send it out as a nighttime Uber earner. There will have to be an app for that.
    I have a question. If my AV vehicle somehow breaks a traffic law, or, heaven forbid, squashes a darting pedestrian that’s careened into a vehicle lane, will police need to stop it if there’s nobody to issue a violation ticket to? If they can stop it, then how? How will police be able to stop a specific, empty, AV and why bother because any incident would obviously be a software issue?

    1. Do you also worry about how you are going to flag down autonomous trains?
      Can’t understand why you see the need to flag down an autonomous vehicle. Who do you think the police office is going to talk to? Take a picture of it. Done. Send it to the owner along with the citation. Want to stop it? Park in front of it.
      Doesn’t seem to be too much of an issue. Google reports 1.2 million miles in autonomous mode without a ticket for their test vehicles.

      1. The real world is far more complex. Bicyclists cutting in front of you or weaving into your lane. Lines unpainted on dual roads. Roads with snow. People not stopping at stop signs. Pedestrian waiting at crosswalk, then not crossing. Even pulling out of your garage necessitates someone opening the garage door. Who gets the ticket if software, camera or sensor is erroneous and AV hits a pedestrian. Who is at fault? The owner ? The AV manufacturer ? The sensor maker ? The software firm ? The camera firm ? The pedestrian ? The city that didn’t send a proper signal to sensor ?
        Parking allowed in neighborhoods, like E-Van, close to downtown ? Idling ? Looping around the same block 28 times ok to kill time ? Or only 12 times ? Who pays for the meter at a traditional parking meter ?
        With now more potential AV car users, namely kids under 18, the handicapped and seniors we will likely get far more cars on the road not less.
        We can’t even get Uber organized in Vancouver, and we somehow think that AV’s on a large scale are just around the corner ? Maybe 2030 in trial areas of MetroVan for limited pre-approved AVs ? Likely far later than that due to all the human-machine-legal-union-bureaucracy interface complexities .. let alone on a mass scale. Will TransLink abandon all small buses, especially in off hours, or will TransLink now operate the biggest AV fleet in MetroVan ?
        It shall be interesting.

      2. Jeff; We can now be pretty sure that these AVs will park themselves using lateral movements. I know that the ones that can get themselves out of a tight parking spot will be the main sellers. Once an AV has a destination set there won’t be much even a pistol packing cop will be able to do. Maybe a baby is asleep in the back.
        Rest assured, there won’t be any police parking in front to try and block one.
        What the cops might have will be master zappers that will have to be aimed precisely. Of course, none of these will ever be misused, hacked or make it to the black market.
        Much more likely is a dedicated ID on each vehicle, like that now used by TREO. We will all be monitored, all the time, by Transportation Central. It will be through them that rogue vehicles will be controlled.

        1. Surely your AV will have a transponder. How else to charge for all that road use you are planning? And if the vehicle is empty, when that type of use is eventually permitted, charge an unoccupied vehicle premium.

        2. Eric, as you continue to be concerned about emergency stop controls for AVs, presuming only unoccupied AVs which are further off, let’s put your mind to rest.
          Commercially available remote stop for autonomous vehicles, available for many years now, and with an OEM option for the radio to be incorporated into the vehicle hardware. The technology is straightforward. So regulate it. Next problem?

          This is all coming faster than some believe it to be coming.

        3. As we see with Uber regulations will be years behind the available technology. Cell phones existed in the 1970s, but only in the 2000s where they widespread and only in the 2010s as smart phones. 40+ years ! Ditto with AVs but they will revolutionize transportation choices and create a much broader continuum between private vehicles and public transit.
          If I can indeed get a car within minutes, from cheap crappo to high-end luxury limo depending on my budget, life style choice or occasion many people will indeed do away with a car and opt for shared cars instead, much like airplanes today as only the top 1% of the top 1% own a plane and even wealthy folks do not own one but rent a seat when needed.
          Not unusual in the 2030s I’d say and in the 2040s widespread.

        4. Oh Jeff, Yes, we all still love those big yellow Tonka toys, eh? But this is just like model aircraft, you need a line-of-sight. The AVs we are really talking about are those that you just summon to your door when wanted. Nobody is spending billions just so you can climb aboard your chariot after having a bottle of wine and sit back and have snooze. You will, of course, be able to do that. The laws regarding incapacitation will naturally become redundant once drivers are no longer needed. Nobody is going to be required to hover their hands over any steering wheel incase there’s need to quickly swerve. Neither will they have to be resting a foot on a brake pedal. The whole idea is living-room travel. The idea is that these puppies will roam on-demand, to and fro. Any airline operator will tell you that aircraft have to be flying all the time to be most cost effective. Parking all day or overnight is a big drain, both financially and from a space point of view.
          BMWs AV tests are using an $120,000 system for guidance. The price will eventually come down, after a couple of decades. The first models will probably start around $150,000.
          Thomas is probably right. The first major impact will be in the 2040s. Still plenty of time to enjoy the Nurburgring.

    2. There’s a lot of deadheading there, Eric, basically wasted return trips burning up resources and wearing out the vehicle, not to mention the public infrastructure, and racking up higher operational costs, like maintenance.

      1. EVs recharge for free. Bumper-to-bumper maintenance and warranty comes cheap too. Solo HOV cruising too. We will have to figure out how many of these cuties we need, methinks one going back and forward is probably less wasteful than two only cruising occasionally. Numbers will be crunched.

  6. I look forward to the driverless car. I look forward to buying it, and enrolling it in Uber, and having it work 24/7 earning me money. With enough cars driving people around, and checking themselves in for servicing, I will make a living.

    1. Report back here in 2035 to tell us how that is going, as unions, cities, provinces, lawyers, federal regulators, car companies, taxi companies, public transit authorities, bike associations and developers will all weigh in here with commentary, opposition, requirements and rules that will take a decade to sort out even if the technology exists.

    2. I’ve had pretty good service from cabs, and know they are covered by licencing and minimum insurance requirements. I used to drive one way back when and know it isn’t the best way to earn a living.
      If I ever call up Uber, my first question to the driver will be, So, what’s your liability insurance coverage? Followed by, Are you earning enough to support a family, or even above minimum wage levels on your 12-hour shifts? Next would be, Do you have anything more than a Class 5 licence? Then, How much effort do you put in to cleaning up passenger’s harf? The last item happens at least once every couple of months to average cabbies and requires at least a donated, pay-free hour to clean up. Good luck getting the sick ride to pay extra.
      So many comments here by people who have never driven a cab or Uber.

      1. Uber beat cabs BY A MILE: better cars, cheaper, one knows where the car is, no phon calls but online booking at the touch of a button, no hassle with payment at all as all is automated, customer feedback so dirty, weird or smelly drivers get weeded out fast. In summary: a vastly superior service at a lower price. That always always wins.
        That is why ALL cities where Uber is running cab use has been in a deep DEEP decline. The customer is never wrong. Even in Vancouver cab licenses have dropped from almost $1M to almost nil.
        Here is what is coming, using Edmonton as an example:
        – upgraded insurance required
        – annual car check
        – criminal record check for driver
        – some $s per ride paid to city
        – minimum price per ride
        That all makes sense to me.
        More here at
        or here on plummeting cab licenses in Vancouver:

  7. The thing that’s a real drag about these is that since they require someone with a drivers license to be present in the car, these won’t help all the people who do not have a drivers license. They will still be dependent on someone else and will need other mobility options.

Leave a Reply

Your email address will not be published. Required fields are marked *