The Environmental Impact of Driverless Cars

Will Truman

Will Truman is the Editor-in-Chief of Ordinary Times. He is also on Twitter.

Related Post Roulette

95 Responses

  1. Kim says:

    One of the main problems of public transportation is that it needs to go “everywhere” in a city. And going everywhere means that you don’t get to any one place terribly often or efficiently.

    I can see robocars as awesome feeders into a train/subway infrastructure, which can be way more efficient because it doesn’t need to cover Every.Single.Block.

    (And I don’t see why folks who live near Zipcars don’t buy smaller cars. The zipcar membership (and cost per use) is cheap enough for when you need a pickup or van).Report

  2. Vikram Bath says:

    I do find #1 intriguing, and it’s an idea I hadn’t thought of before. My response to it would be

    1. Vehicles have not actually become smaller over time if you measure by usable space within the vehicle. I think instead of “reverse the trend toward smaller vehicles” it should be “continue the trend toward larger vehicles.”

    2. How much space is actually needed for a productive workspace? I’m productive on airplanes, and they offer much less space than my car. Driverless cars are not momentum-less cars, so writing legibly is probably out of the question, so a large desk wouldn’t be that helpful anyway.

    3. Gas taxes and CAFE standards can present a barrier to cars getting much bigger even if consumers would like them to be.Report

  3. Kazzy says:

    I have different concerns. When 22-year-old Kazzy would get hammered and think, “I should go over to my ex’s and give her a piece of my mind,” there were multiple natural impediments that overwhelmed my poot decision making.

    Similarly, when contempary Kazzy and his friends get drunk and think, “To the casino,” again we are thwarted.

    Driverless cars will mean we need to make better decisions when drunk! Noooooooooo!Report

  4. NewDealer says:

    Argh my comment gotten eaten.

    I think it is too early to tell whether driverless cars will be good or bad for the environment. There are only so many hours in the day and we have probably already reached maximum super commuter though. 2-3 hours one way is probably the most realistic that anyone can do. A 4 hour one way commute seems insane to me. Only teleportation will increase the commuting differences or much faster train travel.

    I don’t see how rideshare would work with suburbs which you seem to think is still going to exist. Rideshare works best when you only need a car every now and then, not when you need one multiple times a day. People will still own their driverless cars if they live in the burbs.

    The whole debate smacks of futurisitc speculation which tends to produce wild support and wild dissent but nothing inbetween.

    My big concern about driverless cars is that they will be hacked by marketers and advertisers. It is one thing to hear adverts on the radio or pandora. It is another thing to have a car analyze the environment (temperature and stuff) and lay of the land (Google Maps) and say “It is hot out there? Why don’t we make a little detour for some Ben and Jerry’s ice cream?” Would you want to be in a car with your daughter when it did that?Report

    • Will Truman in reply to NewDealer says:

      Lots of people will still own cars, but the dynamics are likely to change a great deal. The driverless cars they own will often be smaller because most of the time you don’t need the extra space and you can rent more easily whenever you do. A lot of two-car households would become one-car households. Driverless car takes Mom to work and then returns home for Dad’s use, until lunch when it goes back to Mom, and then Dad’s for the afternoon. And some people who need a car now won’t need one, and so won’t have one because they won’t break the bank for one.

      Advertising will probably be an option, but I wouldn’t expect it to be an unavoidable one.Report

    • Kim in reply to NewDealer says:

      Ha. You’re still assuming there will be people in the vehicles.
      Driverless TRUCKS are already in prototype, and working pretty damn well.
      (Ten years before truck drivers are obsolete)
      keep the trucks off the roads during rush hour, no?Report

    • Kim in reply to NewDealer says:

      Google’s car was doing doughnuts in a parking lot around here.
      I’m pretty sure they’ve fixed their wireless so it isn’t hackable.
      (this is not to say that advertisers can’t get your data. legally)Report

      • Will Truman in reply to Kim says:

        I assumed he meant legal hacking. In the case of illegal hacking, advertisers are the least of one’s concerns, I think.

        To pull a couple of your ideas together, think driverless fuel trucks!Report

    • Vikram Bath in reply to NewDealer says:

      “It is hot out there? Why don’t we make a little detour for some Ben and Jerry’s ice cream?”

      Ha! Of course, the argument advertisers make are that they are really just providing information for you to make an informed decision.Report

    • Troublesome Frog in reply to NewDealer says:

      I think you’re right about the upper end of commute time. I peaked out at a little over an hour each way, which was burning a huge amount of fuel (even in a compact car). I made it more “productive” by starting up on audio books (2 hours a day, roughly 2-3 extra books per month, 24-36 books per year I otherwise wouldn’t have made it through), but the amount of time it cuts out of your day gets prohibitive after that, even if fuel is free and you can sleep on the way.

      Of course, I don’t think we’re worried about the upper tail of the distribution going from 1 hour to 2 or from 2 to 3 hours. It’s more like the middle of the distribution going from 20 minutes to 40 minutes, which would have much larger total effect.

      As for the ride share, my suburban life doesn’t have too many ‘several times a day’ driving days. I may have to go to several places, but that’s more a matter of the car going from one higher-density location to another. Trips home in the middle of a set of errands aren’t all that common. I can definitely see an on-demand “fabric” of self-delivering cars for rent as a net win.Report

      • You’re right that it’s more about the 20 minutes to 40 minutes effect. And that would be an area of vulnerability. Some people might prefer longer commutes, within specific parameters, if it gives them time to relax (instead of time to honk at bad drivers).Report

      • NewDealer in reply to Troublesome Frog says:

        There are people who commute from Napa/Sonoma to San Francisco every day because they want property and you need to go that far to get a good deal. I don’t really need a few acres of property so no commuting that distance for me (yet).

        They are building an Sonoma-Marin train that goes to the Ferry Station at Larkspur. Maybe it will extend to SF one day. Though there is enough demand for commuting between Sonoma and Marin it seems.

        http://www.sonomamarintrain.org/

        There are people who commute to and from Poughkeepsie to NYC everyday and this is about two hours on the train plus getting to work or home from the train stations.

        A few years ago, The New Yorker had an article on people who did super-commutes into NYC. They mainly seemed to live in Pennsylvania. One thing I noticed is that people will sometimes have a small studio in NYC and a large property in Massachusetts or Pennsylvania. They will work Monday-Thursday in NYC and then have a three day weekend at home. My guess is that they get up really early on Monday morning to be at work on time.Report

    • dhex in reply to NewDealer says:

      it’s google, so of course advertising will be involved somehow. nobody works for free.Report

  5. Mo says:

    I wonder how pricing will work for these. Automated Zipcars will work better for people that live in places where public transportation is crap, but there’s a lot of shared routes or routes are within a small radius (e.g. college towns) than far flung spokes. It would be ironic that this may work better in South Bend, IN than Pawling, NYReport

    • Will Truman in reply to Mo says:

      I would expect it to work particularly well in college towns. I think @kim makes a great point about how this will be able to help public transportation. So people in far-flung suburbs and exburbs could maybe take an autocar to the train station (or whatever).Report

      • veronica dire in reply to Will Truman says:

        Right, and with smart routing, and with discounts for ride-share, ’cause why not?, this could really change everything. Everyone in the neighborhood logs in, sets their schedule, desired pickup times, desired train schedule, and then goes downstairs to meet the car.

        The car will probably be late sometimes. People will complain.

        Myself, I live 2 blocks from the subway on one side and work three blocks from it on the other. So for me walking is totally fine.

        But then, the other day the ground was covered with snow, I was in heels, and I was heading to the doctor’s. His office, it turns out, is about eight blocks from the subway station. So I decided to take a cab from the train to his office. Except it can be hard to catch a cab from that train station. So I changed trains downtown and went to another station attached to a mall with a hotel with a cab stand. That way I could catch a cab and avoid the snow.

        How much easier to just schedule the car to meet the train.Report

  6. Jim Heffman says:

    You make a good point about the ability to have driverless small cars (coupes, really) for when you don’t need the full capability of a big car…

    …buuuuuuut it’s more efficient for manufacturers to make all the same model of car, to squeeze the largest benefits from mass-production. And it’s more efficient for rental-fleet operators to have the same model of car, to simplify their maintenance procedures and spare-parts needs. And if you’re going to be a public service then you need to have cars that can handle ADA requirements (wheelchair ramps) which means they *have* to be big cars, minivans or SUVs.

    And besides, the reasons for not having big cars in the city are that they’re hard to park. But if you don’t *have* to park the car–because it is, effectively, a taxi–then who cares if it’s a big car? Just stop in the street, everybody get out, and let the car drive itself back to the garage (which can be in a location that’s inconvenient for travellers to get to, but just fine for cars to drive to.)Report

    • it’s more efficient for manufacturers to make all the same model of car,

      That’s a reason that the number of models won’t explode, but despite the efficiency you mention carmakers release lots of kind of cars. I’m talking about the addition of one line (a two-seater) which may well come at the expense of another (economy, for example, or maybe full-size).

      Rental agencies also rent out cars that come in multiple sizes.

      And if you’re going to be a public service then you need to have cars that can handle ADA requirements (wheelchair ramps) which means they *have* to be big cars, minivans or SUVs.

      They might have to have larger vehicles in their fleets, but there’s not much reason to expect that it will be limited to them. Just as rental agencies are not limited to them or taxi companies.

      And besides, the reasons for not having big cars in the city are that they’re hard to park.

      That’s one of them. For me, it’s one of the biggest. However, fuel consumption and basic cost are also reasons.Report

      • Jim Heffman in reply to Will Truman says:

        It is not axiomatic that smaller cars are better. The example of the Prius versus the Prius C is instructive; the smaller car has worse fuel economy due to aerodynamics.Report

      • Rod in reply to Will Truman says:

        Has anyone made that precise claim? All else being equal , a smaller (lighter) car will be more economical. Not sure what your point is here.Report

      • Jim Heffman in reply to Will Truman says:

        “All else being equal , a smaller (lighter) car will be more economical.”

        The example of the Prius versus the Prius C is instructive; the smaller car has worse fuel economy due to aerodynamics.

        The assertion was that if cars moved to a rent-as-needed self-driving/self-delivering paradigm, then manufacturers would obviously start making more things like the Smart ForTwo and the Scion iQ, because those are obviously more efficient and preferable to big stupid wasteful four-door cars that people only buy for that one time a year when they have more than one passenger. I reply that A: small cars aren’t necessarily vastly more efficient than the larger ones (Will might thinking about SUVs, but I’m thinking about Toyota Corollas), and B: economies of scale apply to fleet maintenance–and business activity. If the only platform in my fleet is the Toyota Corolla, then I don’t have to worry about anything beyond making sure the staging lots have enough cars to handle demand. If I now have some Corollas and some Scions, then I now have twice as much inventory-management work to do, because if a customer needs a Corolla and can’t get one because the nearest lot is full of Scions, that customer will not be happy.Report

      • As things stand, Smarts are some of the most fuel-efficient cars on the road. Not the most fuel-efficient, though. Maybe they can be made the most fuel-efficient if more effort were put into that end or maybe not. Maybe they will achieve aerodynamism by expecting the person to lean back in such a way we wouldn’t want them to if they were driving. Maybe it will lead to a one-wide segment. I suspect that there will be something below the economy size, but’s if I’m wrong that’s not actually central to my point.

        But even if that is the case, it can lead to a lot of gravitation from compact to subcompact, midsize to compact, full-size to midsize, and so on. One of the present arguments against getting a Smart is that you won’t have the space when you need it. The same goes for economy over midsize. When gauging fuel efficiency, size for people and cargo space that you rarely need will be less of a factor. That’s significant whether it leads to two-seaters or not. Even if there are exceptions, most of the time (within a single maker particularly) the smaller the vehicle the better the mileage. True for SUV’s, true for sedans.

        And beyond all that, someone mentioned electric cars. Could be good gains there, because most of the time we won’t expand beyond the range. Right now, a lot of the hesitation is because of special use rather than typical use, and if you can call on a special use car for special uses, then you can rely on something with a smaller range most of the time.

        The ultimate point is that whether we’re talking about cargo space, passenger space, or range, we buy what we want to be able to use in special cases. If those things can be handled easily and separately, then we can go smaller, electric, and more efficient.Report

      • Mad Rocket Scientist in reply to Will Truman says:

        Assuming the future of autocars that are available on demand happens, what we are more likely to see if manufacturers adopting universal chassis systems. This is almost the way things are now, as most makers tend to have standardized their cars around given frames & engines, and then they differentiate with body & features. It can go another step further.

        A car service buys a fleet made of two chassis, coupe & van. The coupe chassis is good for up to 4 passengers plus limited cargo. The van can be an SUV, a Van, or a truck. Same chassis, same powerplant, just different body and features. The body is (relatively) easily swappable. So if a service gets a call for a truck, but only has vans or SUVs on the lot because yesterday was Friday & today is Saturday, they could, inside of 30 minutes or so, pull a Van in & have it leave as a truck.

        Also, one area may be served by multiple services, each specializing in a given vehicle type. So if you need a minicar, you choose from services A, B, or C; but if you need a Van or truck, you choose from D or E.Report

      • With regard to what the rental agencies will offer, that will vary from place to place. In a smaller place like where I live, they will probably default larger because they are dealing with a relatively small pool. The larger the pool, though, the more they can (and I think will) differentiate. They will be able to assume that if you need an economy car, one will be available somewhere in the vacinity of where you need.

        Regular rental agencies don’t default to a particular size, even though the same dynamics are in play. The bigger the rental lot, in my experience, the more likely they are to have different sizes. I don’t see why that would change when you’re dealing with what I believe would be larger fleets generally. If you don’t have a midsize here, you’re likely to have one there. They might have to wait ten minutes instead of five. This is all much easier to deal with than the rental car agencies now that carry the different sizes. More mobility and larger fleets.

        The other thing is that a lot of what I’m talking about when I am talking about people taking smaller cars when they need smaller cars is that I am also looking at ownership. Which is to say that childless Will and Clancy can get a small car because they know that a larger car is available for whenever they need one.Report

      • I hadn’t thought of the swappability, but that actually solves one of the dilemmas: Peak hours versus non-peak. You can have it be a van during rush hour for carpooling, and then something lighter and more fuel-efficient the rest of the time.Report

    • Troublesome Frog in reply to Jim Heffman says:

      I think that “big cars in the city” is potentially one of the biggest efficiency wins. If you call the car using a smartphone with your location and destination, it should be possible for dispatch to figure out how to efficiently get more than one person with more than one destination into a cab and accurately split the fare. Sort of a hybrid between a bus and a solo cab.Report

  7. Dan Miller says:

    I think you’re right that they have the potential to change suburbs much more than downtowns. In a dense urban environment, it’s purely a question of space–a bus or train can carry people more space-efficiently than a car, no matter how good the driverless car software (especially when said software has to deal with jaywalking pedestrians, people getting into and out of cabs, etc). In the suburbs, the challenge of fitting everybody is replaced by the challenge of transporting people to their relatively far-flung and lower-density destinations, and that’s where driverless cars could prove really useful.Report

    • Will Truman in reply to Dan Miller says:

      I think we have to make distinctions between dense places with good public transportation and dense places without good public transportation. The inner core of cities probably wouldn’t change much – except for the increased density that fewer parking lots would bring – but in a lot of places you don’t have to go out very far before things become very car-centric.Report

      • Kim in reply to Will Truman says:

        Even if they aren’t car centric, the capability of having ‘single routes’ for large parts of the city, would make a big difference.

        Pittsburgh is full of hills, our topography makes buses sometimes a bit odd.Report

  8. Damon says:

    There’s only SO much size you can reduce. Is anyone suggesting that space will take a priorty to safety? Doubtfull, so you’ll still have cars with 6+ airbags, collision avoidance, etc. All that takes up space, because, in the unlikely event of an accident in a driverless car, the manuf is getting sued if they aren’t there.Report

    • Will Truman in reply to Damon says:

      I assume that driverless cars will be of roughly the same range as regular cars are now. So the smaller cars I am thinking of are Smart sized, roughly. Though I would expect there to be more action on that side of the market since the purchase market is tilted against such cars since we are inclined towards buying the capacity we will only sometimes need.Report

      • Damon in reply to Will Truman says:

        @will-truman
        Have you driven/ridden in a Smart car?

        I’ve seen them on highways and ridden in one on local roads. No way I’m getting into it again. The damn thing feels jittery, has a high center of gravity and is barely noticeable on the highway. Strikes me more of a deathtrap than a car.Report

      • Kim in reply to Will Truman says:

        Damon,
        if you don’t need to see, and the other cars are finding you wirelessly, you can lie down in your car and sleep the whole way. a low center of gravity car would be kinda cool. Bonus if you can tilt it up so that it parks pointing upwards (taking up more vertical space and nearly no horizontal space)Report

    • Kim in reply to Damon says:

      you might be able to cap speed, and thus shed some of the nuisance regs.Report

    • Michael Cain in reply to Damon says:

      The rules for designing safe very small cars are apparently different than those for designing safe large cars. The goals are fundamentally the same — crush-proof passenger compartment and as much “ride-down” room for decelerating the passengers as possible — but the methods aren’t. In a big car, the designer provides as much as possible within the “standard” package: a rigid passenger compartment, crumple zones, collapsing steering columns, air bags, etc.

      In a very small car, you start with making the entire car rigid so that in a collision with a large car, the small car makes use of the big car’s crumple zones. You increase ride-down space in the interior by getting rid of the steering wheel and most of the dash and using a side-stick (a la contemporary fighter jets). Better restraints, like four- or five-point belts instead of three. At least with respect to this post about self-driving cars, you make the car smarter about avoiding the most dangerous accidents (eg, striking a concrete barrier after drifting out of the lane) even when the driver is nominally in charge.

      I seem to recall something from a few years back by an academic who had been doing simulations for years criticizing one of the common crash tests, the one where the front of the car strikes a massive barrier off-center. He said something like, “Of course the very small car does badly on that when the authorities mandate the use of the same crumple-zone strategy that works on big cars. The proper approach is to make the vehicle rigid enough that it spins off and comes to a halt farther down the road.”Report

  9. LeeEsq says:

    The biggest problem with driverless cars from an environmental standpoint is that its goign to lead to a lot more sprawl. Every innovation in transportation has increased the range between home and work. When you had to walk everywhere, cities were limited in size.Report

    • Will Truman in reply to LeeEsq says:

      I think your brother is right that we’ve probably hit the upper bounds of how long people are willing to spend commuting. Even if they don’t have the stress of driving, they’re just not going to want to lose more than 2-3 hours a day commuting.

      I think that the cutting down on parking spaces could actually increase density, making this an exception to the rule.Report

    • LeeEsq in reply to LeeEsq says:

      Continuing, every mode of transportation allowed the distance between home and work to increase, especially with cars. Driverless cars are going to decrease a lot of the human error and stress of driving. With really good GPS systems, they might also allow for a better trafic flow and reduce many traffic jams, making commutes less frustrating and faster. This will lead to more sprawl, which is bad for the environment.Report

      • Will Truman in reply to LeeEsq says:

        @leeesq You’re right that the increased efficiency alone could lead to more sprawl. On the other hand, the increased efficiency would probably make more room for lowering speed limits and increasing fuel efficiency that way. It also, as @troublesome-frog points out, makes carpooling easier.Report

  10. Shazbot3 says:

    We’ve had driverless cars since the early 1980’s. One of them was called Christine and it didn’t work out so good.Report

  11. LeeEsq says:

    I do not think that driverless cars are going to be a big threat to public transportation. Lots of people are still going to prefer urban living and some cities and metropolitan areas are still going to be desirable places to live than others. As density increases, so will the need for public transportation even with driverless cars. LA spent decades as the epitome of the auto-centric suburb but increased density, LA county as ten million people in around 4000 square miles, and demand has led to the recreation of a public transportation. If the original car did not kill cities in their entirety than I don’t see why driverless cars would.

    What driverless cars might hurt is short to long distance plane travel. By making driving a more relaxing experience, people that would have used air travel to cover a half-day or day’s drive previously might decide to go driverless instead if they thought they had the time.Report

    • Will Truman in reply to LeeEsq says:

      I can’t imagine that driverless cars would kill cities. In some ways they would help by reducing the amount of parking spaces needed. I do think that they will hurt – though not kill – demand for rail and whatnot because what a lot of the boosters want, driverless cars will give them.Report

      • Kim in reply to Will Truman says:

        If you can “lock” the driverless cars together, so that you’re reducing drag, what’s the harm in having separate compartments? (some, yes, but not as much as you think…)Report

  12. DavidTC says:

    This article is ignoring a lot of things that would make driverless cars a lot more efficient.

    People have already mentioned ‘Not owning a car’, but having rentable cars that show up also means those cars can *easily* be electric. Because no one cares how long they take to recharge, as long as the system is smart enough to not run out of charge in the middle of a trip. You go the mall, get out of one car, get back in a different one an hour later while the first one is still recharging.

    Heck, you could even use them for longer-than-max-battery trips, as long as you were willing to spend five minutes switching to another car mid-trip. It’s not *ideal*, but at some point it will be cheaper than gas.

    What is more likely is that there will be different range cars, and you usually rent a really short range one that make it barely five miles, you use to get groceries and whatnot. (Remember, you get a different car when you leave.) This allow the battery to be tiny, thus reducing the cost of lugging it around. When you need longer trips, you simply order a different car.

    Or you even *own* the short-range one. Worse comes to worse, you run low on charge, you rent a car, and your car wanders off to charge and make it home by itself.

    And here’s a fun question: How much of driving is dropping people off somewhere? Especially children. What if you could just put them in a car and send them off, and didn’t need to get the car back? (Perhaps with video monitoring or something.)

    @will-truman mentioned a hypothetical above where people could take the car somewhere, and it could then go elsewhere and pick up other people…and I say, why? Why not just have a *different*, closer, car pick them up?

    Hell, I keep talking ‘rentals’…what if this was more like a *time share*? A half dozen people own the same car, and whoever needs it, uses it. (If two people need a car at the same time, they both pay half the rental cost of an extra car, or whatever.) Or fifty people in the same subdivision own five cars and a truck.

    Turn car drives into a *commodity*, where people just *buy them*, instead of maintaining their own vehicle. We could probably do with a *tenth* of the cars in existence, and half the total miles traveled, if people could just *get* a car, use it, and *put it back*.Report

    • Will Truman in reply to DavidTC says:

      mentioned a hypothetical above where people could take the car somewhere, and it could then go elsewhere and pick up other people…and I say, why? Why not just have a *different*, closer, car pick them up?

      Depending on which thing I said that you’re referring to, I was assuming that the people they go pick up would be nearby or that it’s an owned car. In the former case, the newest rider’s “closer” car. In the latter, car ownership would still have its perks and if it was your car you’d want to limit who would be riding in it.

      I think we could do with far, far less than a tenth. Though I still think a lot of people would want to own their own cars. The difference is that it would be a luxury item instead the necessity it often currently is.Report

      • North in reply to Will Truman says:

        Yeah I suspect that driverless cars would lead to a massive plummet in car ownership in most urban areas and most dense suburban areas as well. Why on earth would you wish to own a car (short of as a highly expensive luxury item) when whatever car you’d need is a few minutes and a telephone call away?

        Frankly I wonder if it might be a near singularity level transformation in the way we structure communities and lay out roads.Report

      • Troublesome Frog in reply to Will Truman says:

        One of the perks of car ownership may be the ability to join a car co-op and rent it out during your down time to defray some of the costs of ownership. If a company can do a good job of building the infrastructure for something like that (monitoring, dispatch, accountability for damage), they may not even have to supply much of their own fleet.

        I can imagine people buying easily cleaned rental-ready models as an investment much the same way people buy rental property.Report

      • DavidTC in reply to Will Truman says:

        I think you might be right about ‘less than a tenth’ at *most* times. OTOH, we still have some pretty stupid rush hours in this part of the world.

        On the third hand, car pooling works a lot better if it can simply be spontaneous and no one has to plan it all, because you’re really just splitting a cab. (As opposed to ‘I might need to work late, so can’t carpool *into* work either’.) They’d probably be a phone app with a list of a few dozen people who live near you, and you just sorta check yourself off as you leave, and group up in random sized vehicles.

        This discussion spent a lot of through on the tiny cars were might have…but,there’s probably just as big a place for twelve person vans operating between cities and subdivisions also.

        You could even have multi-stage trips…one vehicle goes north, and when it gets far enough, half the people get out to go east-ish, and the remaining vehicle turn west. Or both groups get in new vehicles, and the big one turns around to shuttle another group. Whatever.

        Changing vehicles sounds like a hassle, but people just think it’s a hassle because of the wait. If the other vehicle is sitting there ready to leave as soon as you get in, if people could just step out of one and sit down in another, people would put up with it. (Especially if we started putting areas on the side of the road designed for that.)Report

      • @davidtc According to this article we’d need about a third of the cars we presently have.Report

    • Kim in reply to DavidTC says:

      Further AWESOME implications:
      Why have a human at all?
      Send mini-car out to deposit your laundry.
      Send mini-car out to get your takeout.
      Send mini-car out to get your groceries.

      (Minicar, by the way, doesn’t need half the safety regs we have for current cars, because minicar can be designed to be a total loss in case of accident. No human to preserve. Lightweight, tiny — and such a timesaver!!)Report

      • DavidTC in reply to Kim says:

        I suspect the minicars will work the other way, and tend to be owned by companies that want to deliver things.

        Aka, I’m imaging a pizza minicar, with a built in warmer, that operates almost like a vending machine. It pulls up in your driveway and sends you a text, you wave your phone at it, it opens a door and slides out your pizza.

        As for grocery shopping, the real innovation is going to be when you can show up at a grocery store, turn in a list, and get back a cart full of that stuff you can walk out with. Until that can happen, there’s no point in trying to get things ‘delivered’. I rather suspect grocery stores will be resistant to this form of shopping, but they’ll have to adapt or die.Report

      • Will Truman in reply to Kim says:

        As an aside, and different from what you’re saying but I was reminded of it nonethelss, I’ve speculated previously that higher fuel prices could ironically result in even more stuff being delivered instead of people going out and shopping for stuff, as it probably does save fuel to have one person delivering 10 things to Simon Street than people in Simon Street all going to the store.Report

      • Kim in reply to Kim says:

        David,
        you mean how dry goods stores used to work?
        Yeah. We do have programs around the city to pass a grocery list and get it delivered, this will just make that a lot easier.Report

    • Fnord in reply to DavidTC says:

      Recharging times are also less of a problem if your car can drive to a recharging station on its own while you’re at work (and/or while you’re sleeping) and drive back when it’s done. And the same ideas about renting solving the occasional demand for high capacity apply to occasional demand for high endurance.Report

    • Will Truman in reply to DavidTC says:

      @davidtc I was going off “cars in existence” rather than “cars on the road (at any given time)”

      I think it’s easy to underestimate the appeal of being in a car as “alone time” as a motivator in taking your own car or riding solo. Especially when the alternative to dealing with other people also does not include having to drive. It has the markings of a collective action problem. One that public policy could mitigate, though.

      Would it? That’s a good question. On the one hand, arguments against higher fuel or VMT taxes would be harder to make because reasons why people need to take their own car would be fewer. On the other hand… taxes. and the desire to be alone. My own status as an introvert could be influencing my thinking here.

      It could well be that you and @north are right, though, about buses being replaced by vans. Depending in part on policy mitigations.Report

  13. Mad Rocket Scientist says:

    Imagine if you will:

    Cars are all built with driver-less capability & the ability to wirelessly interface with an urban public transit grid (PTG).

    If you live in a rural area without a PTG and/or with roads that have poor capacity for autodrive, you get manual control. If you enter a city or area with good roads, the car can take over.
    If you are in an area with a PTG, you can have your car interface with the grid & as long as you stay on autodrive, the car can become a car for hire, responding to transport requests that meet certain criteria along your route (like no pickups that would add more than 20 minutes to the overall trip, etc.). The car would then drive itself to people, & take them to their destination. For every public transport the owner does, they get a fuel &/or tax credit or a portion of the public transit fee the rider paid.

    Other benefits of driverless cars – the removal of so much signage and other visual signals. Cars in autodrive would communicate with each other for smooth flow, and manual cars would receive traffic signals & notices in the car (audio tone and center console or heads-up display).Report

    • Mad Rocket Scientist in reply to Mad Rocket Scientist says:

      Toss in a fleet of public cars (or private company rentals), and you could do away with trains & buses.

      One major design change, however, would be cars would need a universal system for quickly & securely installing a child car seat.Report

      • The more I think about it, the more these cars would work in tandem with public transportation. Specifically during peak usage hours. It would make a lot more sense to have the autocar ferrying people two or three miles to a bus stop or train stop than it does ferrying them 20 miles to work. During rush hour, you’re likely to have some serious capacity problems.Report

      • North in reply to Mad Rocket Scientist says:

        I’m not convinced that there wouldn’t be public transportation… in the form of larger cars or small busses.Report

      • Kim in reply to Mad Rocket Scientist says:

        North,
        yeah, a lot of people would take public transit if it didn’t have ‘those people’ on it.Report

      • DavidTC in reply to Mad Rocket Scientist says:

        @north
        I’m not convinced that there wouldn’t be public transportation… in the form of larger cars or small busses.

        Exactly. I’m seeing thousands of cars turn into hundreds of 12 person vans, or whatever the ideal number of people is.Report

      • I don’t know about quickly but aren’t fixing points for child seats fairly standard with most cars these days?Report

      • Mad Rocket Scientist in reply to Mad Rocket Scientist says:

        Newer cars & car seats can use the Latch system, which works pretty easily when the car seat is facing forward (child is 2 or older). The system has 3 points to secure with, two at the seat, and one behind the seat to keep the seat from rocking forward too much. Infants, however, face backward, and the only way to secure the third latch is to tie it to the front seat rail, which is a pain to do.

        Finally, the amount of effort one has to expend to get a car seat firmly in place in considerable, since you need to put enough weight on it while tightening straps to compress the seat cushion quite a bit. It’s enough of a struggle that most of the parents I know have at least two car seats, one that stays in the car the child normally rides in, and another for the second car, or for travel or overnight visits with relatives (we took our spare when we went on vacation, and this weekend when my son visits his aunts, they’ll have the spare).

        I can just imagine the struggle a single parent would have managing a kid while also going through the gymnastics of installing a car seat, all the while other riders are getting annoyed at the delay. A standardized quick connect system for car seats would be ideal, maybe something where the mount in actually under the seat cushion and can be accessed easily by flipping up the seat.Report

      • Jim Heffman in reply to Mad Rocket Scientist says:

        The LATCH system is not legal for use if the child weighs more than 35 pounds.Report

      • Mad Rocket Scientist in reply to Mad Rocket Scientist says:

        @jim-heffman

        Seems like a clusterfudge of epic proportions:

        http://carseatblog.com/20202/weight-limits-the-death-of-latch/Report

  14. Michael Cain says:

    Once the powers that be concede drive-by-wire — and you don’t get more drive-by-wire than the software running everything — they let you challenge traditional complex designs that haven’t changed all that much in a hundred years. Put an electric motor on/in each of four wheels and everything becomes a simple function of voltage: traction, forward and backwards; braking; steering. Electric motors can be made ridiculously rugged, and you’ve done away with a substantial number of moving parts. Isn’t one of Elon Musk’s arguments for not using dealerships that mechanically, electric cars need a whole lot less service than traditional designs? Tesla has barely started on the simplifications that are possible…Report

  15. Matty says:

    More broadly than myself, the vast majority of pickups would become unnecessary. A large number of SUV’s would as well.

    What proportion are actually necessary at the moment? I know it’s a different country and you have a lot more rural space but my impression is that the popularity of things like SUV’s is largely driven by fashion rather than any need to actually drive on rough terrain and I’m not sure that incentive changes with driverless cars.Report

    • Kim in reply to Matty says:

      no, that incentive changes with gas prices, mainly.
      And loss of income in general.Report

      • Matty in reply to Kim says:

        You’re broadly right, though my subjective impression is that the kind of people who drive a 4×4 from the suburbs into a city are relatively wealthy anyway so they have as it were further to fall before a declining income hits discretionary spending.Report

    • Will Truman in reply to Matty says:

      One of the ways that people who rarely need a pickup or a 4×4 convince themselves that they do is that sometimes they do. We have a crossover SUV that we got primarily due to procreational plans. However, it’s proven quite helpful. Sometimes we need it. Sometimes, actually, we need a pickup. If I were less frugal or if I liked driving large vehicles, it’s definitely something I would consider. Likewise, my father-in-law owns a pickup because he sometimes needs it to move stuff. 95% of the time he doesn’t need it, but the 5% of the time that he does, it’s handy to have around. He can rent, but have you ever tried to rent such vehicles? It gets tough. In good part because of the various safety requirements, maintenance costs, and insurance costs associated with actual drivers.Report

      • Mad Rocket Scientist in reply to Will Truman says:

        Same here, our Subaru Tribeca is big & comfy and a nice safe place for the child to ride. Plus AWD & traction control are nice, especially when it is wet out, or icy. Bigger than a sedan, but smaller than a minivan, with enough room that the child seat doesn’t interfere with the front seats.Report

      • We have a Forester. Our next one is likely to be larger still. In part because of the “special needs” that would be alleviated by relatively simple access to a larger vehicle when we need one.Report

  16. Miss Mary says:

    The robocar future you speak of would be of huge benefit to my employer and those like us. I’m in the people business and people have to get places, so a chunk of my job is logistics. Money is always an issue at our non profit, but if we lived in your robocar future we could better utilize our fleet of vehicles. Instead of the four large vans, three minivans, and three sedans we own, we could have two minivans, one large van, and five two-passenger vehicles for the same reasons you describe having a larger vehicle for your family.

    Public transportation is nearly extinct in our rural community; forget vehicle sharing programs. It’s surprising how close we are to progressive Portland, and yet how behind we are.

    When I attend the board meeting this Tuesday as they discuss their choices for purchasing a new vehicle, I will be dreaming of the robocar future.Report