Recent Comments

Follow us on Twitter

Are driverless cars the future?

I wonder if you can guess who wrote the quote below?

Cars are noisy, they’re dangerous, they emit fumes, they take up so much space that our cities seem to be built for cars first and people second, and they are not much fun to sit in on a jammed Auckland motorway.

The price of petrol seems to go higher and higher, and it is true that some of the roads being built which seem like a great idea (e.g. the Puhoi to Wellsford motorway extension) don’t actually pass muster with some cost-benefit analyses.

If you read the quoted text above you could be forgiven for thinking it is something that we had written but no, this text came from a blog post yesterday on the Act party website. Now after picking yourself up off the floor and thinking that there may be a glimmer of hope the party was finally be starting to understand transport and urban issues, you can think again. The rest of the post is dedicated to enthusing how driverless cars and other technologies mean we should keep investing in roads and only roads. So lets have a look at some of their claims in a bit more detail.

The first claim is that driverless cars will give people some of the benefits of PT by allowing them to do things like read books or do work instead of driving while cars will also be driven better without humans at the wheel. Both claims are probably true but to really see the benefit of these cars we will likely have to wait until a substantial proportion of the existing vehicle fleet is replaced. Last year there were ~145,000  new cars registered and the total car fleet rose to just over 2.8 million vehicles. Even if every single car that came into the country tomorrow was driverless, it would take at least 20 years to replace the current fleet. In reality we are probably looking at more like 30-40 years, perhaps longer as it will take a while for the technology to start being embedded in all new cars made, that’s a long time to wait.

Of course what happens after your driverless car drops you at work. Well Act suggest that the car will take itself away somewhere and park itself. That all sounds nice but when every other person is using a car at the same time it means we either need to spend millions on additional parking buildings to house them all. Alternatively they could just go home and return when you need to be picked up but that brings about its own problems. What happens if you don’t have a set time to leave a certain place, does that mean you need to request your car and wait for perhaps 30 minutes or more for it to arrive?  Alternatively perhaps they are suggesting some kind of shared car scheme in which case it would likely require the government/council to purchase huge numbers of these driverless cars, how much would that cost taxpayers.

Act and many driverless car advocates also suggest that these computer controlled cars will be able to run closely together allowing for much more capacity to be achieved out of the existing road network. If that is the case then isn’t it a perfect reason NOT to invest in new roads as we would be better off to wait and see exactly what impact these cars have before spending millions on roads we may not need. I can’t help but think the party still ignores the reality of real live and still has this video on repeat:

Of course one thing all of the driverless car advocates don’t realise is there is a pretty big elephant in the room. As these cars will be so much smarter and also programmed to avoid crashes they will automatically stop (and tell the other vehicles around them to stop) if someone walks out on to the road. It shouldn’t take long for people to realise that they can easily take back the streets but just using them how they want and the cars will be forced let it happen. So perhaps we should welcome driverless cars, we can get shared spaces without it needing to cost ratepayers a thing (bet they didn’t think about this issue).

85 comments to Are driverless cars the future?

  • Well done for opening the topic.

    Driverless cars will begin as a TAXI service. You won’t own it. It will drive you to your destination, then drop you off and move on to the nest customer. No parking issues. Incredibly productive. Maybe 1 driverless car can take 10 or 20 traditional cars off the road?

    Also, the huge advantage is that you can scale the cars directly to demand. Just you commuting? Order up an enclosed single-seater motorcycle. A big group? Order up a taxi-van. Imagine the efficiency gains from this alone.

    It will of course take time to saturate the market, but not many decades. You’re looking at explosive growth because the advantages and economics are too good.

    You have a credible point about capacity. With full-automation mode the closer the vehicles operate together, the safer it is (because large inter-vehicle speeds can’t develop if the worst happens).

    And what about using tiny cars for freight? Why do an errand when a “micro-car” can do it for you?

    http://andrewatkin.blogspot.co.nz/2009/06/automated-transportation-network.html

    …oh. And if people start screwing with the system by forcing the cars to stop in front of them, the car can ultimately take a picture and forward it to the police…if it is obviously disruptive behavior.

    Driverless cars are the future. Best we all think carefully about what this means!

    • Even in the best case situation we are still looking at these being 10-20 years away before we start seeing the technology in production cars and it will be a lot longer before there were even sufficient numbers to make any kind of impact. In the mean time the rest of us still want to be able to get around so other options are still needed. As for capacity, even if you you ran small cars close together you still would have considerably less capacity than other methods as that capacity is limited by intersections. At the moment arterial roads average only about 800 cars per hour in each direction at peak times, even if you double that it is still a pretty small number.

      Of course driverless cars have been promised to be just around the corner for decades and in reality they are no closer than they were 20 years ago. If they ever do come on stream I personally don’t seem them doing anything other than filling a niche role.

    • Oh and driverless cars aren’t going to do anything about one of the points raised that they take up a lot of space in the city, in fact many roads would probably become harder to cross due to a wall of cars with no gaps what so ever in them.

    • Steve D

      Yeah, if you send that picture I’m sure the police will get right on that. They’ll have tracker dogs and everything.

      But sarcasm aside, I genuinely think Andrew is right – driverless cars will come. I just doubt they’ll have nearly as big an effect on our cities as trams or cars did.

      “Driverless” cars are already here, and they’re called taxis. Not having a driver will make them cheaper by, oh, $15 per hour? To guesstimate a bit, if they’re going straight from job to job with 50% dead running at an average of 30km/h, that’d make fares perhaps 50c/km less? Given they’re $2.50/km or so at the moment, that’s cheaper, but not a revolution.

      Meanwhile, we’ve already got driverless cars that can follow each other without gaps – those are called trains. And the tiny cars that can do errands and fetch stuff? From couriers to truckies to delivery boys on scooters we’re already most of the way there.

    • Ash

      Driverless cars are one of those ideas, like flying cars, that the world just isn’t likely to be ready for for a long while yet. I’m reminded of the autogyros in Brave New World. I think time and time again futurists tend to underestimate the cultural attachment to driving in its current form that people have.

      Driving has not substantially changed in more than 100 years. People expect to be in control of the vehicles they own. And for many males, the car can function as an extension of their self-concept, even virility, and so to be an inert passenger could be felt emasculating.

      I wonder how many years of testing will be required before driverless cars can safely make U-turns in traffic, safely avoid debris or unexpected objects on the road? Co-existing with unpredictable cyclists and pedestrians? Driving at high speed in close formations under software and GPS control. What could possibly go wrong?

      What happens when you are trying to join an intersection against the dominant traffic flow, where a normal driver might after a time be required to bite the bullet and pull out knowing they’ll cause somebody to have to brake. Would the driverless car proceed so boldly, or would a 20 minute wait be in order? Traffic lights at every intersection? What happens when you see a friend or something of interest as you’re passing it and want to turn around and go back?

      I think there’s a great deal of fuzzy logic required for which the human brain is superbly equipped, and which computers will continue to struggle with for a long time yet.

      Or perhaps they’d be part-time driverless cars, just handling the stop-start crawl along an Auckland motorway then becoming normal cars when you exit? Having the steering wheel, etc in the way would be too conducive to writing that report or catching up on your Facebook either.

  • Greg N

    An atomic powered Tunnel Boring Machine (TBM), now thats the killer technology the CRL needs to really fire the imagination of the public, and finally sort out the traffic issues in Auckland. How did we not see this before! Thanks Disney.

    Also seems ironic, yet darkly fitting, that a “driver-less political party (ACT) should be going on about “driverless” cars as the way of the PT future. Perhaps they see that as the way of the future for politics as well?

  • Matt,

    Always only 20 years away? According to who? Obviously people who don’t know what they’re talking about. Driverless cars have only just been actualised today – and proven. The technology is rapidly refining itself.

    http://youtu.be/bp9KBrH8H04

    -I would like to see tunneling for electric car only bypass lines, for capacity relief.

    • I have heard auto industry execs say they could have build them decades ago but that the demand for it isn’t there as people have an inherent distrust in the tech so they didn’t bother pursuing it, can’t find that online right now though. That is why it is the likes of Google who are doing it as they are doing it for themselves.

      And if there is such a great capacity increase delivered then why would you need bypass tunnels?

      • Scott

        They are sure showing interest now. Especially Volvo: http://www.stuff.co.nz/motoring/7867230/Traffic-jam-help-targets-daily-commute

        Did you go down and have a look at the Volvo ocean race yachts when they were in auckland? Volvo was running a demonstration of one of its cars crash avoidance systems (automatic breaking) at that event. Very impressive.

        I think the best capacity gains can only be realized if every vehicle is automated. That way things like intersections can be optimized without restrictions like marked lanes etc that make it easier for human drivers.

      • Felix Alexander

        Because if you didn’t tunnel a bypass I would do exactly what you suggest. Picnic in the middle of the freeway? Why not! And when it’s too noisy, just throw a ball across a few lanes and have the dog stop the traffic for a few minutes — do it a couple times and the algorithm will designate this a 30 k zone for a while, and the braking/reacceleration will be pushed further away.

        (And as for taking a photo for “obviously disruptive behavior”, I don’t know what’s so obviously disruptive about just randomly crossing the road here, instead of walking three minutes to the lights, waiting two or threee for a change of cycle (or two, because you have to push the button and missed it by milliseconds), and walking three minutes back just to get the shops on the other side. In fact, all those cars are obviously disruptive from my perspective. But would the photo on the car know that? Of course not.)

        You have, Matt, made be the foremost advocate of driverless cars as of five minutes ago. Thanks.

  • Matt,

    Platooning does the opposite of your concern. It “clumps” the cars together creating nice big gaps between them (to walk through).

    • No because there would be more vehicles on the roads to fill those gaps, especially if you have taxi type services always running to and from places (your substituting car parking for more travel)

  • OrangeKiwi

    Driverless cars don’t buy stuff.

  • Matt,

    Yes it may well amplify demand for cars and travel (note: a well co-ordinated network should have minimal empty-vehicle operation, as a ratio to directly productive operation). But this is a good sign – high demand proves success. Externalities, and how they are managed and valued, is another concern. In sensitive areas tolling can always be employed where needed. In fact it can be done so seamlessly, because it just goes on the users account as a higher fare.

    tunnels:

    You are right that there are limits to capacity relief with platooning, depending on the bottlenecks. Bypass tunnels for small electric cars (only) would be desirable. The tunnel can be cheap, 2-lane, and have ample capacity as a ratio to cost IF it specialises to small electric automated cars.

  • jonno1

    I’m delighted to see this topic raised on this blog, and trust the regular posters will keep us up to date with progress on this future mode. Sure, I have no idea how long it will take to be commercially viable, but it will come; as Andrew’s link demonstrates the technology already exists. Just a thought – could existing busways be an initial channel while driverless cars are limited in number?

    One downside: I’ve just returned from a road trip, and part of the enjoyment was the driving – maybe driverless cars should be limited to the city!

  • Electric driver-less buses, trucks and trains. Buses would come first. Forget the driver-less cars as they’re way too inefficient.

  • Well this technology is currently being developed, and marketed hard, and has of course been used for decades on railways, but does this change anything about how wise it is to be aggressively building uneconomic motorways. No of course not. In fact the appearance of this ploy by the road lobby through their politically irrelevant stooges is a very pleasing sign of desperation.

    We will, however, be getting extremely high capacity electric powered automated vehicles operating on their own purpose built congestion free subterranean right of way before this decade is out: The ‘cars’ have been ordered and are being built for this right now: The CRL, I’m told that the new route will be fully automated, so therefore driverless, even though we will probably have someone in the cab.

    I look forward to the ACT Party [all three members] and Mr Atkin above hailing this and look forward to them lobbying hard for the earliest possible commitment from government to redirect sufficient funding from their wasteful motorway plans to complete this exciting programme.

  • Tom

    This is fantasy. I mean, sure, we’ll have autonomous cars on the road but they will be very expensive (just the lidar on Google’s car is US$70000). I’m guessing that they will be used mostly by businesses to replace professional drivers (yay efficiency!). They are already doing this on mining sites:
    http://www.riotinto.com/media/18435_media_releases_20677.asp

    How do politicians get away with these things? To argue we need to build highways today so that we can accommodate a hypothetical fleet of high-tech personal transports is just absurd. This looks like a standard attention diverting trick that gives a good excuse for continuing the “business as usual” theme. Maybe it’s also aimed at people who like cars but feel uncomfortable with the current trends. But then again, people who like cars usually enjoy driving them.

  • steve

    Indeed, however using that logic, in the future due to flying cars, no land based motorways will be required. Wonder when Warner Bros will release the Jetson car designs they have been hoarding for all these years…..:) Memories….http://www.youtube.com/watch?v=FyinD6ZDqeg

  • Icebird

    I love the fact that the blog post jumps straight to driverless cars as the solution to all traffic problems. I’ve advocated Personal Rapid Transit here in the past, to a healthy degree of skepticism. I won’t encourage a re-ignition of the “will PRT work or not debate?” (since the arguments are all theoretical till a system is deployed in an existing urban environment rather than a brand new city (Madar) or relatively controlled environment (Heathrow)). But I will look at the “PRT versus driverless cars” debate:

    - Where do the unused cars go? In the driverless model, they need to be parked. With PRT they can be recycled for other users.
    - PRT operates in a separate completely controlled system. It doesn’t have to account for autonomous elements like the vagaries of human drivers.
    - Infrastructure is cheaper than roads.
    - PRT offers non-stop journeys (driverless cars haven’t solves the stop-light yet).
    - The transit user doesn’t have to purchase a vehicle to take advantage of PRT.

    That’s really the rub of the ACT argument – the state shouldn’t invest in transit, transit users should… no matter how much those driverless vehicles cost.

    • But Icebird the ACT argument is that the the state [and city] must continue to subsidise roads [and parking and all the other driving externalities]…. it’s just another attempt to corner taxation as a sector subsidy under the guise of technological fantasy. Rates and fuels tax and RUCs are taxes, they are not a fund for one sector to own for their benefit alone as the RUF believe.

      This is just another clumsy press release from the road lobby.

      You however would seem to be obsessed with an obscure although current technology with few benefits and not insignificant costs. Who knows why?

    • swan

      - “Where do the unused cars go? In the driverless model, they need to be parked. With PRT they can be recycled for other users.”

      Driverless vehicles as a taxi or demand responsive PT service will be able to do exactly that.

      - “PRT operates in a separate completely controlled system. It doesn’t have to account for autonomous elements like the vagaries of human drivers.”

      It operates on a system that does not exist and would have to be built.

      - “Infrastructure is cheaper than roads.”

      Regardless we already have the roads.

      - “PRT offers non-stop journeys (driverless cars haven’t solves the stop-light yet).”

      Hard to see how a full driverless vehicle system wouldnt be able to do this, as long as peds were grade separated. But is it really desirable?

      - “The transit user doesn’t have to purchase a vehicle to take advantage of PRT.”

      Ditto for driverless vehicles. This is one of the potential benefits of driverless vehicles. Frankly, if we were ever to want or get a city scale PRT system, it would be in the form of driverless vehicles on conventional roads.

    • I love the fact that the blog post jumps straight to driverless cars as the solution to all traffic problems.

      Not at all, the point of the post is that driverless cars won’t be that revolutionary. I do think that Driverless cars would win out over PRT though as the infrastructure to run them on is already in place.
      Also PRT will have to have intersections of some kind unless you are planning on a PRT line from everywhere to everywhere else. Even if you had flying junctions (like motorway to motorway ramps) you would still have issues on heavily used corridors where there would be simply to many pods so some would have to slow down to accommodate a new one coming in.

  • Sanctuary

    Will driverless cars appeal? Ask yourself this. What would Jeremy Clarkson say? Can you imagine him sitting, cross armed and looking bored, as his 2021 Ford Mondeo II responsibly navigates it’s way around the racetrack? I am not being facetious here. People have a relationship with driving. Except for the drudgery of rush hour at Mt. Wellington, most people LIKE driving their cars (I know I do – I have driven my own car since the day I turned fifteen, and I still enjoy simply driving my car about) for all sorts of reasons – time to be alone, the ability to take a different route for the scenery or sights or just get away and go shopping in an out of the way place, the sense of freedom & independence, the freedom to just up and go, the fact that you might own a 1968 Mustang notchback and enjoy driving it about to show off or you might just enjoy cruising on the motorway to Raglan in your Overlander, or even something as simple as doing a 30 second spot of enjoying the risky sensation of recklessly accelerating onto the motorway and secretly pretending you are a formula One motor car driver.

    The point is, the driverless car removes one of the prime arguments in favour of cars at all – their recreational value to their owner and the pleasure driving still mostly gives to people. Driverless cars are simply an incredibly inefficient form of PT, a decaffeinated, trim soy flat white form of driving.

    Cars are fun. Cars are flexible. Cars are quite useless at moving large numbers of people about in the rush hour. Driverless cars will be attacked by everyone in the motoring press as draining all the fun from one of the few pleasures still accessable to most people and yet another example of the nanny state plot to turn us all into babies and they are an epic fail as a viable form of PT. Driverless cars are an expensive invention looking for an application.

    • swan

      Sanctuary,

      I hear what you are saying but would have to disagree. Motor vehicle accidents and the associated carnage is a weeping sore of our society. We only put up with it because we have to. As soon as there is technology available to save hundreds of lives and thousands of serious injuries a year, it will be jumped at. I work with the construction industry – another part of our society with an unacceptable safety record. In the construction industry, as soon as a practice or technology is readily available that reduces risk (even slightly) it is picked up.

      Also, while people enjoy driving, I am not sure they enjoy paying for it. Driverless cars will not only be safer, and faster, they will also be much cheaper (goodbye the high fixed costs of vehicle ownership). They will also reform our cities with respect to physical parking.

      These are far too big opportunities to pass up.

    • Ash

      Yes, driverless cars will possibly never appear in large enough numbers to make any difference. Even once they’re a fully mature technology, If the consumer doesn’t buy them, then they’ll stay a novelty/fringe item like so many other fully-developed techs that never found a market.

      I also think they serve as useful “pie in the sky” for those who have ideological opposition to public transport (particularly rail), because this vapourware technology proposes to solve many of the downsides of automobility as we currently know it. That is why the likes of ACT like to bring it up. Electric cars may be in the same category to some extent too, I think – at least the expectation that everyone will want them.

  • Ari

    Overall though, they are just another feature of the transport network. They will not solve congestion problems like ACT is suggesting.

    @ Draco – I totally Agree. Driverless buses and taxis will be the very first use of this technology. A taxi company with no drivers will start, where people will use a smart phone ap and their GPS to allow the nearest driverless taxi to find them. As part of profit seeking measures by bus companies, they will replace all those pesky bus drivers and all buses will become driverless. Next will be all the trucking companies which would allow them to drive their trucks 24/7 instead of being restricted by pesky driver time restrictions. All these three applications have very powerful implications from a profit driven perspective. And none have anything to do with personal driverless cars.

    @Tom – I recall a certain IBM CEO declaring there were only a market for 5 computers due to their great cost. (This is a misquote as there is no record of such a statement being made, but my point remains the same.) You can say this is fantasy because they are too expensive now. PC’s costing $5,000 in 1984. The first cellphone cost $4000 in 1984 (thats $8000 in todays money). Technology almost always goes down in price as market demand increases because of economies of scale. No different with cars. Though cars will remain at the prices they are currently, just the features will improve. I think it is 10 years away before these vehicles become mainstream, but luxury cars will have it much sooner. Audi/Mercedes/Cadillac are all planning semi autonomous vehicles within the next 2-3 years.

    @Icebird – I like PRT, but I suspect a driverless taxi network where you pay a subscription is likely to supercede any PRT network, simply because the infrasturcture is all there already. I would see a cooperative of some form where you pay $50,000 investment to a cooperative taxi company and you get paid 10% return every year or something like that. It would cost less than rolling out a PRT system which would have to go up against too many nimbys. In a taxi based system you achieve most the benefits of PRT. The only issue with a taxi system would be that the safety issues remain. While driverless cars will be safer, there will still be significant conflicts within the existing road network. PRT and rail retain the advantage in this respect.

  • PBY

    There seems to be at least 2 separate discussions here…
    1. Will driverless technology replace current technology?
    2. Will driverless cars replace Public Transport?
    I think driverless cars are inevitable in some form. But that technology won’t be limited to cars it will continue to be rolled out across all forms of transport, including Buses, trains, trucks, planes, emergency vehicles etc….
    But despite the high likelihood that driverless technology will be a huge part of our transport future, I can’t see it replacing Public Transport. For the simple reason that a driverless bus carrying 50 people in a congested urban environment will still be more efficient and effective than 50 driverless cars travelling in the same environment.
    This is how Jarrett Walker put it on his Human Transit blog….
    “If you define a “car” as “a separate enclosed vehicle for every passenger or party”, then the geometric fact about all cars, self-driving or not, miniaturized or not, is that they take vastly more space per passenger than effective public transit. This will not be a problem in low-density suburbs, but cities, by definition, are places with relatively little space per person. Self-driving cars will certainly improve the efficiency with which cars use space, so they will shift the calculus somewhat. But the bottom line will still be that if you want two crash-safe metal walls between every two strangers going down the same street, you will need a lot more space than if those two people can sit next to each other on civilized public transit.
    You will also need vastly more metal and equipment, which means that the self-driving-car-replaces-transit fantasy involves massive industrial production with severe consequences for energy security and greenhouse-gas emissions.”

  • Don’t forget that driverless cars will not be a new system under and old roof – the will be a new system under a new roof. Most people won’t own one, nor will they need to (they will get far superior performance and cost reduction from not having to park their car and own it). It will be the age of the “robo-taxi”. We won’t need the parking space because they won’t be parked up (mostly) – they will be working for the next passenger.

    I agree with some of the comments on here that shuttle vans and buses will be a first-application for this technology. This is where they can deliver the biggest immediate bang for the buck.

    http://andrewatkin.blogspot.co.nz/2011/09/end-of-buses.html

    However, even if it cost, say, $50,000 more for full-automation per-car, that would pay for itself quickly with a car that serves 10x the number of people. Moore’s law is predicted to hold out for another 10 years at least, with other technologies coming down the pike, and with refined software the costs of full-automation should predictably plummet.

    [youtube http://www.youtube.com/watch?v=VdXnuVIT4rY&w=560&h=315

    • Andrew but you would still need a huge amount available to cater for peak demand that will be little used during the rest of the day and those unused vehicles will need to be stored somewhere. If that storage happens to be on cheaper land outside of the city centre then you have a lot more travelling to meet the demand when it is needed.

      • swan

        Matt,

        When we no longer have individual vehicle ownership, people will be able to choose the vehicle that most suits them for each and every trip. So for peak time commuting, a lot more people will car pool, take minibuses, buses etc. as the price will be responsive to demand (eg. it might be $5 for a car for half an hour off peak, but $15 during the peak – numbers picked randomly).

        • People don’t carpool now and I don’t see that changing much in the future, demand responsive pricing is good and will probably work against these systems, driving people to to use PT

          • swan

            People dont carpool because it is a hassle. (requires organisation and coordination) People do take shuttles to the airport. I don’t see it as being carpooling in the traditional sense. Instead, you order a “seat” on your smartphone. You might have different people in the car/van/minibus each day. Different levels of service will cost different amounts. i.e. How direct the service is between your origin and destination.

          • Steve D

            Swan – a van that constantly dives down residential side streets to pick people up is going to be way slower than a bus that stays on a fixed route on the main road the whole way, whether or not it has a driver.

          • swan

            Not necessarily. If it has 10 seats, it may have an algorithm to pick up 10 people in a suburb and take them to a destination (e.g. the CBD). So it might dart down a few side streets at the start of the trip, but then join the main streets. I would expect this to overall be more convenient and faster (including waiting/walking at start and end of trips), than a conventional bus service. But the larger point is we dont need to pick winners. The market can determine the optimum vehicle size.

          • Steve D

            Size has nothing to do with it. Driverless tech will probably mean we run more smaller busses more frequently, and on more routes, but it’s not going to make tootling around suburban streets any quicker. It’s still going to be quicker to get off a bus and get onto another one, rather than taking detours to pick up and set down other people. It’s also cheaper, since you’re taking a more direct route.

            Peak time bus services to the CBD already run pretty much as you describe: they run around the suburbs picking people up then run express to the CBD. At other times and on other routes, there’s just not enough people travelling to and from similar places at once to make it feasible to pick them up and drop them off individually.

            I again recommend Jarrett Walker. Have a read of the links in http://www.humantransit.org/2012/10/los-angeles-a-nightmarish-fantasy-that-refuses-to-die.html

          • We already have picked winners by only investing in one mode for 60 years, the market has already been skewed by this activity.

          • Swan

            “there’s just not enough people travelling to and from similar places at once to make it feasible to pick them up and drop them off individually.”

            Remember we are talking about largely replacing cars. There are a lot of people traveling to similar places in a city like Auckland. Any evidence from PT to date is not that relevant.

          • Steve D

            First, the people who are currently taking the trips most likely to be effectively served by these shuttles are most likely to be taking public transport already. Second, how are these shuttles going to attract people out of cars? Car drivers already don’t have to pay for a driver, since they’re doing the driving themselves. And they can go direct, without picking anyone up along the way.

          • swan

            The reason is – people will no longer need to own cars. In my opinion one of the reasons why so many people use cars for commuting is because they are faced only with the marginal cost. People really want to be able to use cars, at least some of the time so they end up owning cars. People who own cars are then more likely to use the car all the time, as they face only the marginal cost.

            However once people are faced with the full cost each and every trip (and at peak time the price will be highest), they will be able to economise, and use the service that best suits them at the time.

    • Andrew – I have just been looking at the Newgeography.com blog and I was wondering if you seriously stand by this comment you made on there (at least I assume it is the same Andrew Atkins): “For example, I think the NZ Green party is, functionally, the propaganda arm of the UN and maybe nothing more.”

      I certainly hope you do as that will allow me to file your view of the world and how it works under “C” for “crazy”.

  • Matt,

    Peak demand will drive higher prices for usage during the peaks, which will lead to shuttle van (type) demand to offset it, etc, at least to a given degree (note: You can ultimately compartmentalise seating for privacy in collective transport, which could be popular). I understand that modern ‘peak’ is more like 4 hours than 1, so we should see a huge net reduction in the volume of cars (and enclosed motorcycles) needed in the city, regardless. But yes – the remaining cars need to be stored. That’s hardly a problem though.

    • As mentioned above, demand responsive pricing is a good thing and will get even more people using PT options instead, perhaps we should invest in that to ensure it can handle the demand.

  • Matt,

    I have no problem with collective versus individual automated transport finding its balance with an economically rationalised (road priced) context.

  • As far as taxis and small buses go, this sounds like a good idea. I have just come from living in a city of 3m people where the taxis were incredibly cheap. I could travel anywhere in the city for no more than NZ$15. It was a great alternative but the problem of course was the if people can afford to buy a taxi they can afford to buy a car, so the roads were clogged just the same. Even worse than Auckland.

    Fundamentally though the technology misses a big point of public transport. Reclaiming lost roads for pedestrians and more human scaled transport that allows city centres to be destinations. Not just somewhere you try and get in and out of as fast as possible and avoid.

    I suspect that most of the people who want something like this are not really city people, they are rural people living in cities who want the best of both worlds’ The opportunities and vibrancy of the city but the ease of getting around and independence of the rural community. This sems to me where the National Party sits.

    I can see the good sides of both those worlds (I grew up in a small city) but I think if we try and have both in a big city (and Auckland will be 2.5 million in 30 years time) it will be a disaster and by the time we go to fix it when we have no choice (petrol prices, congestion), the cost of doing so will make it prohibitive.

    If this is the revolution of transport, why is it not being embraced and talked about more openly? Are all the leaders of all the cities in the EU, Hong Kong, Singapore, Australia and the East Coast of the US deluded in investing in rail, walking and cycling? If so, that seems like a massive oversight by a large group of intelligent, well informed people who have no vested interest in pushing that agenda. They dont own rail companies or cycle manufacturers and there is far less profit in PT and cycling than traffic infrastructure.

    On the other hand car manufacturers have a massive vested interest in making sure cars remain vital to transport along with the road construction industry and oil companies.

    I am not against this technology, I just see it as a nice cherry on the top of a well integrated, rail based PT system.

  • Tomislav

    I would just like to advise caution to people who are applying Moore’s law (and the implications of the law) to anything other than IC electronics. Different tech advances at different paces and encounters different roadblocks. Just because the transistor count on a CPU die is doubling every two years doesn’t mean that there will be cheap lidars on the market in 5-10 years. I mean, they have existed as long as the microcomputer and they are still pricey. Who knows if it will ever be cheap enough? Not to mention that we are actually pushing the limits on Moore’s law which is showing diminishing returns.

    But let’s say this happens and the price goes down 10x, comparative to a 1990 and a 2012 PC price difference. Using the Google car as an example, let’s say that half of the equipment is not required on a production car. That would still mean a $7500 premium (in current money) on any vehicle that is driverless.

    Let’s just agree to disagree, delay the road building for 5 years and revisit the issue. It doesn’t take that long to upgrade roads but once it’s done you’re stuck with them and with the debt incurred by building them.

  • Tomislav,

    Please don’t forget that full-automation means network-based transport. 1 cars serves many families – not just one. The cost of the computer and supporting electronics is not too concerning in this format. Indeed, it means you can go hard with other goodies like electronic suspension for comfort and energy-efficiency investment, etc.

    The movement will actually represent an ultimate morph of PT, PRT, dual-mode (PRT’s cousin), and private transport in that you hire your car all to yourself while you use it.

    -Getting cars off the roads for the aesthetic sake of it? I agree cars suck in that they destroy your ability to go for a nice walk among nice ambiance. But they’re going nowhere in themselves because they are a functional necessity in our modern world, can be hugely less intrusive with a network format (because they can be programmed for electric-operation only in sensitive areas), and it makes it easier to create pedestrian-only zones in parts of the city center (and everywhere else, for that matter) as they can be better coordinated as a network.

    Also, they can revolutionise “sprawl” property development. You can easily build new townships based on electric operation only mode, where the cars/roads are segregated into the back streets in a “hyper-green” township.

    http://andrewatkin.blogspot.co.nz/2011/12/green-sprawl-why-not.html

  • Steve D

    But why are they going to switch to shuttles, rather than individual driverless cars, which will give them a quicker trip? At the moment, people usually take taxis over shuttles, since shuttles are slower than a direct taxi trip, and usually not much cheaper. Driverless tech will improve the lower cost of taxis MORE than shuttles, since you’re saving the cost of a whole driver, not just a share of one.

    Meanwhile, even if everyone did switch to your shuttles, there just isn’t enough demand for total travel anywhere to expect there will often be someone else, by chance, living near you who want to go somewhere along your route within a few minutes of you. Even for peak travel to the CBD, your neighbours don’t leave home at exactly the same time: the time it takes the shuttle to come detour down your street and wait for 10 people to be ready to leave is going to be longer than the few minutes it would take to walk to the bus stop.

    So you either end up with people riding alone, or taking a fixed-route service (because it’s faster).

    Some more good Jarrett:

    http://www.humantransit.org/2012/08/bus-stigma-and-driverless-cars-email-of-the-month.html

  • Liz

    The driverless car utopia also seems to conveniently overlook a few points:

    - Not everyone has a smart phone, or can afford one. Even if they can afford one, they may not want one. Driverless cars booked through smart-phones would disadvantage the elderly, children, visitors, and poorer people.
    - Trains are nicer than cars. There is more room, less stop-starting, less turning. It’s much easier to work on a train than in a car.
    - There is no guarantee that driverless cars will be safe from crime or vandalism, whereas at least on a train or a bus there are other people and usually a driver. Mini-van type vehicles have been suggested, but I for one would refuse to get into a driverless car with people I didn’t know. It wouldn’t feel public enough.
    - On PT you can ask a driver, or other passengers, for help if you need directions or assistance.
    - It’s not a bad thing to be surrounded by other people. Sure, sometimes someone will sneeze on you or be rude, but other times you can help a fellow passenger, or bump into an old friend, etc. Let’s stop locking ourselves away in little metal boxes.

    If we want to reduce car ownership then we need to improve PT, and perhaps have alternatives such as the zipcar-style model – i.e. cars for short-term hire. These could well be driverless in the long run, although that wouldn’t be a necessity, but for now we can start with the cars that are already available.

  • Liz:

    -You don’t need a smart phone as such. Soon they will be very cheap anyway (for a basic or secondhand model). Smart phones are cheaper than private cars.
    -The vast majority of people prefer cars, other things being equal. A self-tilting gyro-stabilised enclosed motorcycle, with a long wheel-base, will have no side-forces. Long distance you can sleep on it. Sleeping-friendly models will be around the corner – maybe competing with domestic flight, even?
    -Video-intercom reduces the security problem. In good time you can always take a private car if you don’t feel safe. Biometrically linked access would make it extremely safe, as identities (and therefore accountability) cannot be faked.
    -You can ask anyone around, or in or out of the van if you need help.
    -Virtually everyone socialises better in a car than a public bus. This is a value for the market to decide, regardless. Most just want their privacy.

    Steve D:

    I think mass-demand will be overwhelmingly for private transport. Collective transport will be a niche demand (almost entirely in CBD’s), resulting from big fares as a result of the need for aggressive demand control in any given area. So if it costs, say, $20 per-vehicle to access a given point, then a lot people will want to get on a shuttle to dilute that cost – driverless or not.

    • Liz

      Try living in a city where you can turn up to a bus stop or train station with a frequency of bus/train every few minutes, and then see if you like that or want to book a ride every single time you go out. I really think that the ease of just turning up is better and fairer. No matter how cheap smart phones are, there will always be people who don’t have them and they shouldn’t be penalised for that. What about visitors to our country? Do you think that tourists will want to pay roaming charges on their phone each time they need to book a car?

      Perhaps a driverless car would be good for long distances… I’m sure a train can go faster more safely, but we have a shortage of fast long-distance trains in NZ. But I’d still feel safer and more comfortable in a train when travelling at speeds, especially out on our roads with all the logging trucks and landslides. Also, trains and planes can incorporate amenities such as toilets and food, as well as allowing space for people to walk around.

      So, now we are not only going to have cars that drive themselves, but are gyro-stabilised and have biometrically linked access? Sounds like we are needing more and more extras to make this a viable option. The scanners at airports fail often, so why would the cars be any different? Would it refuse me entry because the retinal scanner isn’t working? Why don’t we just go with the transit technology we already have that works? And the security issue isn’t gone either, but just reduced to ‘take a private car’ – in which case, we might as well just stick with the cars we have, or take a taxi. With a bus or train, if I feel threatened I can get off at the next stop, which hopefully has other people around and perhaps staff or security guards. With a driverless car taking whatever route it has decided, what happens if I want to get out half-way? Will I know where I am?

      I would rather be able to ask a driver who is employed by a bus/train company, as I’d expect them to generally be more knowledgeable and trustworthy. You know, I expected a response along the lines of ‘use your smartphone or a computer interface in the cab’, but even that doesn’t always give the answers that are needed.

      Re socialising – if the cars are small, you have a much smaller chance of seeing anyone you know. Personally, when I’m travelling through the city I don’t expect a huge amount of privacy. I’m in a public space and therefore i have to deal with other people. You could try this, it’s really not that bad. Yes, some people may want or need privacy, but maybe having a fast, frequent, and affordable PT system is more important. Your note to Steve also underlines the affordability issue – I’m not going to pay $20 per trip, or even $10 per trip. Your answer to this problem is to share a vehicle – I suggest that we already have such shared vehicles (buses and trains), and therefore we don’t need new ones. With regards to those people who want to travel privately and not drive, there are taxis.

      • Liz you clearly are a well adjusted person without an irrational fear of sharing space with other people while moving around a city… behind all PRT and driverless pod obsessives are misanthropic fears of others, irrational control freakery, and/or plain old fashion snobbery… Luckily there is no need to design our cities to pander to their issues, in fact making them ‘get out more’ would probably therapeutic for the poor dears….

        • Liz

          Haha, thanks! That’s the thing though, I actually would feel SO much safer on a crowded bus or train than in a driverless car with a couple of other people. Sure, the pickpocketing or bag-snatching potential may not be as high, but what’s to stop someone assaulting me? Or stop a driver-full car from smashing into me? No thanks, I’ll take buses and trains any day.

          I really think that the people who want these technologies should go live somewhere like London, where you can get pretty much anywhere on PT (and where people who do drive will stop for pedestrians – we don’t need driverless cars for this to happen, just some manners). Or if they want a smaller city, try somewhere in the Midlands where the bus-driver says ‘Ey up duck’ when you get on. Social interaction is nice. A PT network is nice. Being part of humanity instead of shut in a box.. so nice, even if crowded at times. I have made several friends on buses and at bus-stops. I’ve helped tourists and homestay students find the right stop. I have enjoyed a viewing carriage and amusing commentary on a long-distance train, I’ve seen Mt Fuji from the bullet train. I have had strangers convince me that I could fit into a sardine-packed tube carriage (just!). All these things that I’d miss out on if I was in a little car, in a line of little cars, on a boring grey motorway.

          • swan

            Liz,

            Even in London the dominant mode of transport is the private car. Driverless vehicles are about usurping the private car. This is something that PT has never done in the developed world.

          • Swan – that is a complete straw man argument. I have never heard or read any PT advocate say that PT will usurp or replace the automobile completely. Even in Copenhagen, maybe the least auto dependent city in the developed world, has about a 30%/70% modal share between automobiles (the 30%) and PT/cycling. Auckland is so far from that it may as well be on another planet, let alone continent.

            If we were to reduce the overall kms travelled by car by 10% tp 20%, with the massive motorways Auckland already has, it would basically solve our congestion problems. That would still mean that about 60-70% of people would be drivinjg to work as it is currently 85%.

            Even the “nutty” Greens are only advocating a 50/50 transport spend in NZ.

          • Steve D

            If by dominant, you mean overbearing and threatening to everything else, I agree completely. If you mean how the majority of trips are taken, you are mistaken:

            http://www.tfl.gov.uk/assets/downloads/Travel-in-London-report-1.pdf (search for mode share).

          • Actually swan that conclusion is only possible by expanding both the definition of London wildly and by assuming every type of journey is of the same value. But also it is the PT journeys that make all the car journeys possible. Urban London can function without cars, but not without Transit. If all the train, bus, tube, and bike users tried to drive- total dysfunction. So your metric totally misstates the value of Transit. The private car is an entirely contingent and subservient mode in London is want you really should say.

          • swan

            I think you misunderstood me. Liz is comparing driverless vehicles to current PT. It should be private cars (primarily) that driverless cars be compared to. It is this market where the impact will be greatest.

          • Liz

            Swan, perhaps you were only comparing driverless cars to current private cars, but you seem to have overlooked the fact that others are not. Many proponents of driverless cars seem to be advocating for them to replace buses. Andrew said that he expects most people will want private journeys, and that driverless cars are better than buses/trains. Taking those people out of buses and trains and putting them in cars, even if driverless, will result in more cars on the road – which we definitely do NOT want.

            Even if we are only looking at driverless cars to replace private cars, I think that there are other ways to reduce car ownership. Having a frequent and extensive PT network will help, as will having a hubbing/connection based PT model (then the local buses can have slightly slower windier routes, as they are only a short part of a longer journey). So will things like zipcar, where you can hire a car for an hour or so as needed. In cities like London, some apartment buildings own a few cars that the residents can use for a fee… car access without car ownership, and many fewer cars than if everyone in the building had one. Perhaps we do also need an improved and cheaper taxi service (not as cheap as PT, otherwise congestion will just get worse) for when people need a driving-free but private or (more likely) door-to-door service. But this mindset can be worked towards without driverless cars – then if they become cheap enough, they can be integrated as well.

          • swan

            I would expect driverless buses to replace buses.

          • Liz

            But until that happens, I suggest that you focus on other ways to reduce car use/ownership… and I’ll continue to enjoy saying “hello” and “thank you” to my friendly bus driver.

          • swan

            Agreed – the best way to reduce car use is to price roads appropriately. That would, hands down, be the biggest boon for buses. And needn’t cost a thing!

        • Liz: A network of driverless cars would be evenly dispersed throughout the city. You would have many of them parked-up at common access points, like bus stops. If one breaks down you get on another. Like buses.

          Patrick Reynolds
          @ October 29, 2012 at 6:22 pm:

          Grow up.

      • Liz – I agree 100% with all you are saying.

        Andrew – I have asked a number of times but you didnt answer. Have you ever lived in a city with a population similar to Auckland with really good integrated, frequent public transport? I have to believe no, or otherwise you wouldnt be so against it.

        You are wrong that people prefer driving 100% of the time. They would prefer to use a really good, FREQUENT (the major thing Auckland lacks) public transport system at least some of the time and some people want to use it all the time (me for example).

        You obviously really hate public transport and having to share space with other people. Fine, maybe you will part of the 50% or so who choose never to use PT. Those people exist even in cities like Copenhagen.

        I like to paraphrase Abe Lincoln: “You can have all the people driving some of the time and some of the people driving all the time, but you cant have all the people driving all the time”. Rail based public transport is the only real option for a city of 2.5 million people (where Auckland will be in 30 years time) as is reflected all over the developed world, other than a small corner in South Western United States and Auckland. We have the tracks, 2013/2014 we will have the electric trains now all we need is the CRL. Buses are a great compliment to that as the Northern Busway has demonstrated.

        You might believe that the leaders and planners in all those European and Asian cities are suffering some kind of mass delusion but I choose to believe (and have seen a plethora of evidence to support this) that they have studied all options and that is the best one.

        • I have lived in Auckland and Melbourne, and I currently live in Wellington.

          I have nothing emotional against PT. I occasionally use it. But what I know is that if you can get in a small car (1 or 2 seater) and have it take you from origin-to-destination, for cheap, and fast, then the demand for collective PT will collapse. The latter is what full-automation will give us.

      • No 12 could be quite nice. Note that you have to “see” the developments from the position of a human, not a bird, if you want to get a feel for what they would actually be like to live in.

        • Luke C

          all these developments make a mockery of your shuttle idea, as they are all terribly disconnected from anywhere else. Transport disastrously inefficient. Some sort of shuttle system may work, however only where you have good connected subdivision design. Won’t work in suburbs of the 80′s, 90′s and early 2000′s for sure.

  • Steve D

    Andrew:

    Yes, all else being the same, private transport would still be the main mode if driverless cars take over. But most of the remaining people will still prefer direct fixed-route PT services, rather than shuttles taking slow detours down back streets, especially since driverless tech will mean more frequent services in smaller vehicles.

    But if you think driverless cars will have such an effect, I’d expect you agree with Matt that we should hold off building roads that we won’t be needing, stop requiring new developments to have carparks they won’t need, and get to work redeveloping all our old useless carparks and car dealerships with new affordable housing. And we can start this today!

    • Filde

      Exactly; who would sit around for a shuttle which may pick you up and drive around for the next 10 minutes more than the journey will take? This brings in another point, which may have been mentioned already (skim read)… Is time certainty. People don’t want a mode which may drop you off in 15 or 45 minutes. They just want to know when they have to leave and when they will get there. Which is why I ride a bicycle around Central City rather than drive a car.

      Driverless vehicles has a future with PT.

    • I agree. If automated shuttles come first they will be mostly fixed-route. They will be the new ‘bus’, just in broken-down high-frequency form.

      More highway building? It depends when we’ll get this technology, exactly. I would hope we don’t need them.

  • To all you CBD lovers:

    Driverless, fast, shuttle-buses operating on congestion-charged roads will give you your CBD recreational center better than any other transport mode can, because it means less cars (and noisy buses) and more people. You should be on-board with this.

    It was a loose prediction of mine from a while back that with automated transport technology we may see an evolution of two extremes – high-density people-centric centers (people like to visit their “Disneyland’s” – including me), and a movement to very low density exurban residential development, that I personally call “green sprawl”.

    Full automation can give us the best of both worlds, each highly accessible.

  • ben

    What a depressing thread. I see irrational opposition to the motor car is boundless, as is irrational support for rail. Let driverless cars in, and let them compete on their merits knowing that if nothing else there will be fewer crashes and safer driving to the extent this technology is picked up. Let the users of the technology pay its costs, including environmental. If Liz et al is right then the driverless technology won’t get very far anyway, so what is lost. Let the technology be tried in the marketplace is all.

    • I don’t think anyone is saying that we shouldn’t have driverless cars, at least I’m not, I just don’t think they are something that are going to magically solve our problems and mean we can or should stop investing in alternatives PT

    • Bryce P

      Ben, I’m a mechanic by trade (25 years) and love cars, driving and motor racing. In saying that, driverless cars will not be the solution to Auckland’s problems any time soon and may never be. Building a more compact city based around all types of transport (motor vehicles, PT, cycling and walking) will be what makes the difference. Do some reading and see what many other enlightened cities around the world are doing. I suggest starting with a book called Walkable City by Jeff Speck.

  • Jamie Walton

    Cars still take up a lot of space, whether they’re “driverless” or not, and despite “auto-queue” spacing technologies. Do we really want every main road to be a continuos stream of traffic, making it hazardous for other road users (pedestrians, cyclists, horses (and non-”riderless” motorcycles)), and blocking the vista ahead for every car occupant? – No (obviously a rhetorical question to anyone who isn’t Gerry & Co). And I am pro-”driverless” cars (and light delivery vans, but perhaps not heavy trucks without a driver on active duty, like a train driver).

  • There are a lot of good points here for and against but the basic point is that driverless cars will not be like our current vehicle fleet – most urban dwellers will not own they’re own vehicle, they’ll use one on demand, most of the vehicles will be smaller (2 seat) unless a larger vehicle is required. They’ll probably be electric and able to form road-trains or platoon to use the roads much more efficiently. The economic benefits to drivers and society will force rapid change once a tipping point is reached. Lots of people used to own horses, don’t see many in Auckland CBD nowadays!

Leave a Reply