Talking soon on robocars and insurance

Submitted by brad on Tue, 2014-09-23 13:07
Topic: 
Robocars

I've been on the road a lot, talking in places like Singapore, Shenzen and Hong Kong, and visiting Indonesia which is a driving chaos eye-opener. In a bit over 10 hours I will speak at Swiss Re's conference on robocars and insurance in Zurich. While the start will be my standard talk, in the latter section we will have some new discussion of liability and insurance.

A live stream of the event should be available at http://swissre.adobeconnect.com/theautonomouscar/ I talk at 8:45am Central European Summer Time.

A lot of news while I've been on the road -- driving permits in California, new projects and the Singapore effort I was there at the announcement of. And lots of non-news that got people very excited like the "revelation" that Google's car doesn't drive in snow (nobody thought it could) or on all roads (nobody even suggested this) or that it was forced to add a steering wheel for testing (this was always planned, Google participating in the hearings writing those laws.) And lots of car company announcements from the ITS world congress (a conference that 2 years ago barely acknowledged the presence of self-driving cars.)

More to come later.

Share on:
  • Reddit logo
  • Facebook logo
  • Twitter logo
  • LinkedIn logo

Comments

Chris

Tue, 2014-09-23 13:32

I drive in Indonesia (Bali) all the time. It's fun if you like your driving to be like a video game. Roads are very narrow, road is shared by fast and slow vehicles, drivers enter from the sides at any time without looking, no such thing as driver training. It will be a long time before robocars are possible there.

brad

Tue, 2014-09-23 21:50

Generally, I am skeptical of proposals to make a technology work by changing people's behaviour, especially a whole culture's. The rules of Indonesian driving -- that the guy who cuts in front of you has the right of way -- are challenging but not impossible. What is difficult is if the unofficial rules are very different from the legal rules, because cars have a strong incentive to be programmed to follow the law. Individuals can take responsibility for bending and violating the law, it is harder for companies to.

I do wonder, though, if it's possible that robocars will change this. Which is to say that if a nation has such chaotic driving that it's not practical to have robocars, people may eventually change their culture to get them, because they are so useful.

How does this happen? Well, the robocar is making video of what goes on. If somebody plays chicken with you or cuts you off, that might be the norm in the society, but the person being carried in the robocar will be pissed off at the sudden braking and slow progress. So they will pull up that video and send a clip to police. Police might ignore it for a while, but a time may come when just as the companies can't program their cars to do what everybody does, neither can the police ignore it (at least in some places) when there is a formal complaint and a video of the event.

And so it might become the case that people learn, "crap, you can't just cut off those cars." You can't play chicken with them. Oh, you'll win the game of chicken every time. But you'll start getting tickets in the mail.

I'm not sure if this can really happen, and it depends on the culture, but I see it as within the realm of the possible.

Bruce McHenry

Thu, 2014-09-25 11:59

V2V radio for traffic flow control needs to play a role. A V2V device on board would coordinate with often hidden or one mile distant traffic flows that you'll be crossing or with which you'll be merging. Ideally, the V2V helps you to adjust your speed far enough in advance so that you can flow smoothly, minimizing fuel consumption and a lowering the risk that otherwise comes from high peak speed and the jackrabbit style of driving common amongst drivers today.

In order to improve compliance, speeders should automatically get warnings and tickets from the on-board V2V device, which may in fact be part of your phone. (They should also get commendations since we know that positive feedback works better than negative.) These warnings and tickets should be issued immediately so that the driver can better correct his bad behavior. It's also essential to give the warnings instantly so that the correction usually happens before other traffic has to be alerted to the possibility of collision.

brad

Thu, 2014-09-25 22:00

Why would you use v2v? V2V is for applications that requires super-low-latency and can tolerate random operability -- other cars may have it and they may not (and almost certainly will not at the start.) Mobile data networks are what you want for most applications that don't need sub-second latency.

As for your car betraying you and reporting you speeding -- this has been proposed and even tried many times. Politically it gets rejected every time.

Bruce McHenry

Mon, 2014-09-29 04:23

I believe the contrary, that V2V is not for low latency applications at all.

Before we get into why, Brad, please explain how you got that notion.

---------------------------

As for voters, they've changed their minds before.

And automatic warnings and perhaps also citations will not be the only thing they currently refuse but will grudgingly accept when they see how much good it does.

brad

Mon, 2014-09-29 12:12

Well, the big issue with V2V (DSRC) is it's so intermittent. You can only use it to talk to a car that also has the technology (which is no cars for the first users) if that car is within a short range and close to line-of-sight from you. That's a huge limitation, compared to the mobile data networks, which let you talk to any other car anywhere on the road, and which are already deployed in almost every car out there, either in the car's telematics system or more likely the mobile phone of the person in the car. They are not bothered by distance or line of sight. Their only two issues are that there are still a few mobile dead spots out there (though they are vanishing and will have almost entirely vanished within a few years) and they do not do super-low-latency apps. The mobile data networks can actually deliver good latency almost all the time, but don't guarantee it, though if a push were made they could be modified to deliver general low latency and very good "same tower" low latency. Certainly with higher reliability than DSRC, but slightly longer latency.

The mobile data networks have a fee, but we're mostly already paying it.

Only a few apps demand the very low latency of DSRC and can tolerate its highly intermittent (and initially near zero) availability. The main one pushed is blind corner occluded vehicles, particularly vehicles running traffic lights. Traffic signal data and stalled vehicle reports can tolerate multi-second latency and so actual availability is much more important.

DSRC could also be useful for high bandwidth applications (since it is more expensive to send large volumes of data over the mobile data networks) but as yet no high-bandwidth apps have shown up that people want.

Add new comment