Some news from Volvo prompts some DTW crystal ball gazing.
Volvo will be getting together with Chalmers University to research driverless cars and sustainable mobility. The ‘Drive Me’ project – a unique public pilot with ordinary drivers behind the steering wheels of 100 self-driving Volvos – has been joined by a new prominent Swedish partner: Chalmers University of Technology,” writes Volvo at their PR portal.
I happen to have a bit of an interest in this topic as I have recenly been involved in a sustainable transport-related project. The interesting possibility with a driverless car is that you can ask it to pootle around while you go off to a meeting. This saves looking for car parking. Driverless cars offer the possibility that you can have more cars driving than there are drivers driving. That means you can have one owner and as many driverless cars as he or she can maintain driving about.
Car rental firms might be in this category. Imagine if Sixt or Europcar have their forty cars at a given office on the move or re-parking themselves instead of taking up valuable space in parking lots.
Driven To Write 
As much as I embrace the idea of driverless cars for stupid commuting jobs or boring long-distance travel (and much less for fun driving on Alpine passes), the idea of cars just driving around and using roads as parking space substitute scares me. Don’t we have too many cars on our roads already? Shouldn’t a good concept of driverless car include the possibility that I can get on and off such a device and when I don’t use it, it’s ready for new passengers? That means, fewer cars than drivers, not more.
I would hope driverless cars were more efficient. The advantages have not been made clear to me yet. It seems to be taken on faith that we need automated cars. As I said before, it surely kills cars´USP, namely the impression of autonomy,
I can see two scenarios where “driverless” makes sense for me.
One is, as I mentioned, a sort of driverless taxi I can use on demand. It should be cheaper than a standard taxi and much more flexible than public transport. This would be a sort of vehicle I don’t own.
The second one is a private vehicle that has an “auto” option. I’ll gladly leave the tedious task of driving 200 km on an overcrowded highway to my virtual driver who is able to communicate with other vehicles on the road as well as weather and traffic services and thus contribute to more safety and efficiency on the road. I’ll hope that most other drivers do so as well, so I can relax and read on DTW meanwhile. However, once the car reminds me with a gentle sound that the highway part of the trip is over, I’m happy to take over again (remember, the passes…).
This is an interesting proposition. Currently it would be easier (in terms of finding space) and cheaper (in fees versus cost of petrol) for me to be able to send even my profligate Audi S6 driving up and down Park Lane than to find a parking space. Apart from the congestion charge, all other costs for driving in London penalise parking. In a polluted city that is pretty daft. Maybe you should charge high to let cars in, but once they are there you should make parking cheap and easy so that they can stop their engines as soon as possible.
Just recently dim, but probably well-meaning, Islington council hiked a £96 levy on parking permits for diesel cars. As a long-term diesel loather I admit to a bit of schadenfreude at finding that those people who embraced the obviously filthy fuel in the idea that they were saving the planet as well as their wallets are getting their bums bitten but, otherwise, must point to the obvious fact that a parked diesel is no less polluting than an EV. Were I an Islington resident, I’d probably feel aggrieved enough to start my engine every day just for the hell of it.
This morning The Guardian ran a story about driverless cars:
http://www.theguardian.com/technology/2015/jun/26/google-delphi-two-self-driving-cars-near-miss
Ignoring the sensationalist tone of the headline, one thing I gleaned from the incident was that the near-miss occurred between two competing driverless software platforms, those being that of Google and automotive supplier Delphi. Is this the first example of incompatibility between driverless systems?
I also found the following passage interesting:
“The news of the close call follows a report from Google in May which revealed that its own fleet of cars has had 11 minor accidents since the programme began in 2009. The director of the programme, Chris Urmson, wrote that not one self-driving car was the “cause” of an accident, and many of the incidents involved the car being rear-ended by another vehicle.”
I am no expert in US driving laws, but I imagine the broad strokes are the same as those in the UK. Here, liability lies with the party hitting the car, not that causing the obstruction. This I believe is what allows Chris Urmson to categorically state that “not one self-driving car was the “cause” of an accident” because in the eyes of the law, they were not.
On the road, however, the law offers a gross simplification of a wide range of circumstances. A driver following a car displaying erratic behaviour can find those actions hard to read and their future actions difficult to anticipate. Thus the circumstances by which an accident can occur is incrementally increased. Liability should at the very least be shared.
Urmson stated that, “many of the incidents involved the car being rear-ended by another vehicle.” That these sorts of incident are recurrent suggests to me that the Google automated car is partly at fault, displaying on road behaviour that is in some way difficult for following (human) drivers to read. Again, liability is shared. Except, of course, in the eyes of the law.
That brings up interesting points. Despite the law, the attitude I have when driving, which I think is wise for my own survival’s sake, is that if I have an accident, some percentage of the responsibility will be mine. By which I mean that, if you keep aware, you can see most accidents coming. Following the arguing couple, the guy who seems to have earache (or is it a phone), the drive with a satnav glued to the windscreen in front of his face and practically any driver from a well-known London private taxi firm, I know that they might suddenly brake or turn without indicating. As far as the police and insurers are concerned I might be blameless, but I’d still kick myself for not seeing it coming. However it’s harder to read the mind of a machine since they have no body (work) language.
From the other side, I fully agree that a nascent technology can’t be condemned for not being faultless and there’s a fair bit about self-driving cars I’d welcome, unless legislation demands I sit hands on wheel, eyes on road, waiting to take over at any time. For most of us here though, we might consider ourselves good drivers, so justifiably less statistically likely to be involved in an accident. There is apparently a 1 in 240 chance of dying from a motor accident in the UK (which seems pretty high but 3 times better than in the wide open spaces of the USA). But I could argue that, as observant drivers, that figure is considerably better for us. So let’s say 1 in 1000. So if Google offers to improve the 1 in 240 figure to, say, 1 in 750 and legislation forces us all into self-driving cars, some of us might actually have increased our risk.