Driverless cars and cycling – how would they mix?

In a rather lengthy twitter conversation on the viability of the driverless car, there was some skepticism expressed about the idea that driverless cars would interact with pedestrians and cyclists.

There’s already been a good deal of fairly impassioned discussion on the topic of how driverless cars will interact with pedestrians.

I agree with the conclusion in David Alpert’s post:<

Driverless cars are sure to lead to big fights. Will they shift the balance farther toward pedestrians, as Kevin Drum believes, or away? I hope the former, but the technology won’t magically solve this problem. Instead, we’ll have to fight it out through the democratic process, as we do most other issues affecting the public sphere.

However, I don’t think there’s much informed doubt expressed that public policy decisions about the interaction of pedestrians with vehicles, including the status quo, could be implemented in robocars.  Even at the status quo, it is highly likely that pedestrians will be much safer than with conventional vehicles, as robocars won’t speed, will have faster reaction times, are unaffected by glare or other vision problems.

As best I can tell, nobody, at least in the public domain, has explicitly considered this question with respect to cyclists.  

Google’s driverless car has undoubtedly passed a few by now, so they must have thought about it, but they’re not saying anything yet.

In short, I do not see any reason why cyclists and driverless cars would not integrate at least as well, and likely much better, than cyclists and conventional vehicles.  

Identifying the presence, speed and direction of a cyclist on the road is not a fundamentally difficult problem, given the sophistication of the sensors available to a driverless car.  It’s worth noting, again, that a driverless car can have 360 degree vision, unaffected by the blind spots that afflict human drivers.  Similarly, identifying and respecting bike lanes – and leaving an overtaking gap on roads without them – is a trivial extension of the myriad other problems that a viable driverless car will have to solve.

One of the more difficult challenges that a driverless car faces when dealing with pedestrians and cyclists (and unleashed pets for that matter) is determining not only where they are and their current trajectory, but what they might do next.   In principle, it shouldn’t be hard to build a system to identify cyclist’s hand signals, but in any case cyclists do not universally follow such signals.  There are other cues that the system could potentially pick up on, including positioning on the road, sudden decelerations, even the direction in which a cyclist is looking.  This kind of thing may sound complex – and it is – but it doesn’t have to work perfectly.  If there’s uncertainty, the car can simply slow sufficiently that if the cyclist does end up on a collision course it can stop in time.

A related question is how a cyclist divines the intention of a driverless car and determines whether the car has “seen” them; making eye contact with a driver is something we all do (whether driving or riding). In my view, that is unlikely to be much of a problem – the driverless car will see you. Furthermore, the most dangerous situations for cyclists are situations where either the cyclist or the driver has no opportunity to make eye contact anyway – getting hit from behind, or hitting a car, for instance, in a “dooring” situation or when a car pulls out from the curb without seeing the cyclist. I’d finally note that driverless cars are more likely to behave predictably, with fewer sudden bursts of hard acceleration, changes of mind in the middle of intersections, and so on.

There are a couple of ways in which driverless cars could improve cyclist safety further.  One dangerous situation for cyclists is when “undertaking” in a cycle lane in heavy traffic, heading into an intersection where cars cross the cycle lane to turn left.  While a driverless car with rear-facing sensors is better placed to detect the cyclist than a human driver, they still may be obscured by other traffic.  However, if driverless cars were equipped with vehicle to vehicle data links (V2V is the favoured acronym, apparently) cars detecting the cyclists could pass that information on to other cars in the vicinity.  

A second piece of infrastructure which may help is equipping bicycles with a V2V transponder.  At a minimum, it would avoid the problem of non-detection of bikes.  However, if the transponder could produce visible or audible warnings, it could warn the rider of cars on a collision course.  Such transponders would be very cheap by the time driverless cars became commonplace (less than $50 is my guess) and could easily be integrated into cycle computers (or Google Goggles, which may be quite common by the time this becomes a live issue) which would reduce the cost even further.  For those worried about mandating such systems, it’s worth pointing out that we mandate that cyclists have to have front and rear lights if they wish to ride at night; a half-decent set of lights easily runs to that, or more.

None of this goes to whether cyclists will be given higher or lower priority on the roads than they currently are; that, as noted earlier, is a policy question. But at a minimum, I don’t see any technical impediment whatsoever to implementing the status quo, with more safety and much less harassment for cyclists than is presently the case.

This entry was posted in Economics, politics, Science and tagged , , . Bookmark the permalink.

15 Responses to Driverless cars and cycling – how would they mix?

  1. Iain Hall says:

    Sorry Robert but I can’t see why anyone would want to own or ride in a diriverless car.
    Further I just can’t believe that they can be fail-safe.
    Soft ware failure or short comings are one thing but so to are hardware failures and knowing what I do about the mechanism of a car I can’t help but think that the sort of devices you envisage will just de-skill the population and make us all slaves to the technology. On top of that there are the legal liability issues to consider if there is a fatal accident with your driverless car are you responsible or is it the builder/programmer liable?

    Human beings are fallible when it comes to operating their cars and its wise to take that on board when we use the public roads There is just no viable alternative to humans being responsible fro driving the cars on our roads as far as I can see.

    • Tony says:

      I would trust driver-less cars much more than I would cars driven by humans. Although primitive, the Google Car is showing much promise. New scanning technologies, such as femtocameras paired with computers that will begin to be indistinguishable from human intelligence (however, not AI quite yet), the car will be much more aware of its environment and others on the road than any human could be – and would even be capable of ‘talking’ with the infrastructure as well, informing maintenance crews where roads are in need of repair.

      With the issue of pedestrians, bicyclists, and legacy drivers, the roads would most likely only be safer. Self driving cars would give you more space, be more courteous, and actually STOP for pedestrians. I would only assume that manually operated vehicles would be given a higher priority, since the computers won’t care about much other than safety and efficiency. Also, how many people are killed by drunk drivers each year? Yeah, I don’t need to go into that.

      And lastly, over the issue you brought up about mechanical or hardware problems, mind you that self driving cars would most definitely be electric so mechanical failures would be a rarity. And hardware failures, although extremely uncommon most likely, would most definitely have redundancies – which humans don’t have when they have errors like falling asleep at the wheel or just a failure in judgement.

  2. Moz in Oz says:

    I’m also pretty sure driverless cars will be safer for cyclists. Less worried about failure modes, as it happens, because mostly those will involve the car abruptly stopping rather than ploughing into things the way human-operated cars tend to do. More worried about cycles having common behaviours that are outside the expectations of the programmers. In a way glad it’s silicon valley based so they probably do see recumbents and faster cyclists. More so than say, Monash, anyway.

    The dooring question is an excellent one, because I expect that driverless cars (like many modern cars) will have computer mediated door locking and unlocking. It should be easy enough in engineering terms to have an override on the door opening that locks the door if a cyclist is too close. But I can imagine some very unhappy motorists if the news of that feature got out.

    • Tony says:

      It should be common place technology I would say to have sensors on cars warning drivers if a bicycle or other vehicle is coming, and would hold the door shut.

  3. Moz in Oz says:

    If you want more obsession with them, Brad Templeton covers a lot of the issues on his blog:

  4. Tony Alvarado says:

    Maybe adding coexisting technologies such as the BiModal Glideway can ease some of the questions and concerns that face the driverless car programs of today.

    • Interesting idea – not directly on point, but interesting nonetheless 🙂

      Interesting enough that I’m going to have a careful read and critique the proposal.

    • Moz in Oz says:

      Except that it requires huge infrastructure investment, isn’t compatible with existing vehicles or roads, and can’t mix with either…. nope,

      It’s just another dumb idea from an engineer with little to no experience of how roads get built. See also: monorails, aerial anything, on-road bicycle paths, busways etc etc. In a way, bike lanes and busways are the counter to things like the glideway (of which there have been many). We can’t even get funding to put paint on the road, there’s no chance of getting major structural changes. Especially since it takes away car space even more emphatically than bike or bus paint does. And another set of parallel transport infrastructure… ain’t gonna happen.

      For a similar but easier proposal, look at the narrow vehicle proposals. The idea is that we have lanes dedicated to vehicles less than 1.2m wide, so the lane itself will be 1.5m or so. Half width, in other words. But my gosh, the howls of outrage from just about everyone at this proposal…

      • I think the point is that it’s supposed to be compatible with existing roads by being elevated above them.

        That said, driverless road vehicles take away a key advantage of this technology – by allowing cars and trucks to “platoon”, the potential throughput advantage of this system is much reduced.

        I can see a host of other problems with this thing. For one, the weight and bulk of the wheels, axle, drive system, and connecting it to the monocoque (most existing road-rail vehicles are chassis-on-body utes, but no modern passenger vehicles are built this way).

        Another issue is the power engineering. Just how much power would those rails have to supply?

        The only place I could see this making sense is fast, fuel efficient long-distance freight, and in those circumstances I wonder whether it would make more financial sense to not electrify the rails. The big advantage of rails is lower friction and more efficiency, so any conceivable battery drivetrain would have a much longer range.

        In any case, like you, I just can’t see this being built, anywhere, ever.

      • Moz in Oz says:

        Robert, the technical problems are pretty much irrelevant. There’s a huge history of these sorts of “revolutionary” transport infrastructure ideas. Some of them have even been built, at least in the “2km of track” sense.

        What kills them is not technical, it’s budget. Whatever is to be spent has to come from somewhere and the obvious place is the transport budget. So the new toy is competing with everything from fixing potholes to tunnels under the CBD (normally the latter). And it’s doing so in a system with multiple points of control (local council through to federal highway people, multiple emergency services, cable TV companies, you name it) and a lot of interested parties.

        In this environment changing the colour of paint is a radical step that takes a long time and costs a lot of money (I kid you not. Green surfaced bike lanes were a big fight and needed legislation).

        And elevated = expensive. Very, very expensive. Look at the Sydney monorail construction costs if you want head-explodingly awful. Just buying the airspace to run that thing would be tedious and costly. If it is to go above the roadway it needs to be 8m or more up, or removable. Otherwise you need to spend a lot of time making sure absolutely everywhere you want to put it remains accessible to overheight vehicles. Then the fire brigades will demand access, both for fire engines and firefighters dealing with buildings next to the path that are on fire. And so on.

        Flat, paved surfaces that people can drive trucks on at high speed are much easier to get permission for than any obstruction.

      • With regards to your latter points, if the technical advantages were sufficiently compelling (say it allowed safe, cheap travel at 500 km/h, or gave you capacity at 1/20th the cost of freeways), you might see the political/organizational hassles being overcome.

        But absent those order-of-magnitude type gains I agree completely that budgetary/political/organizational factors will kill this kind of thing stone dead.

  5. BilB says:

    A new (not actually that new this has been in the making ofr quite some years) dimension to personal transport that is gett government recognition in Europe

    One of the reasons for the acceptance is that these will form a new wing for emergency, policing and military services, and will work in and around the community.

  6. Brett Bordelon says:

    Rear endings are actually one of the *least* dangerous collisions for cyclists, if only for their rarity. Most car/bike collisions happen during turns.

    Not to mention that in a head-on collision, the velocities are added, whereas in a rear-ending, they are subtracted.

  7. As a regular cyclist on our roads, I’d rather every single car on the road replaced right now with whatever prototype software exists for driverless cars. There’s no way it can be worse than human drivers.

  8. ianbrettcooper says:

    “identifying and respecting bike lanes – and leaving an overtaking gap on roads without them”

    Erm… you still have to leave a safe overtaking gap on roads WITH bike lanes, and the fact that you’re not alone in thinking that you don’t is part of the reason why tests of bike lanes show that they’re less safe for cyclists than is a standard road. Hopefully those who program the software of driverless cars will understand this and other road rules as they apply to driving around cyclists, otherwise the driverless car will be almost as hazardous to cyclists as regular drivers are.

    “making eye contact with a driver is something we all do”

    Not me. Eye contact is a very bad indicator for what a driver is going to do, as I found out 20-odd years ago when a driver smiled at me right before pulling straight into my path. Motorists tend to ignore things that are not threatening them. They can do this even when they visibly acknowledge a person, because the parts of the brain that do these things are separate and unconnected.

    In my opinion, driverless cars will increase risk on the road simply because they will be risk-averse and because not every car will be driverless. This will lead to drivers of non-driverless cars taking even greater risks to get ahead of a line of traffic because they know that a driverless car will be risk-averse. Some drivers do this now, relying on other humans’ tendency to be risk-averse. But when they KNOW a certain car type will definitely be risk averse, well, I think we can imagine the level of risk that some drivers might apply in such a situation.

Comments are closed.