Live data from Hacker News

New Study: Waymo is reducing serious crashes and making streets safer

waymo.com

61–70 of 413 posts

Re: New Study: Waymo is reducing serious crashes and making streets safer

#61
post #43

How do AI cars fare with those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years over the course of a driver's history?? It's happened to me about 2-3 times and I've always made the "(assumedly) correct" split second swerve decision. Wondering if that is a "human/instinctual" skill and if AI cars do just as well or better, or perhaps not as good? I don't ha…

Probably better? 'Who' can process more data faster is likely impossible to answer, but (e.g.) Waymo can train on those scenarios and have way more 'experience' than any individual driver who's seen it once every 10-20 years of their driving career.

Yeah it's interesting, these scenarios are generally at high speeds on an interstate highway, and generally involve some level of predicting what the villain driver is going to do (split second prediction though mind you, it all happens in the blink of an eye), and sometimes the right thing to do is to swerve (if no one is adjacent in surrounding lanes) whilst either not using the brake at all or almost no braking. Basically avoid the brake and do your quick swerve after you've confirmed there is no one in the left/right adjacent lane or close enough behind you. Once again this all happens in the blink of an eye. Perhaps AI would be better? I'm not complete sold though..

Re: New Study: Waymo is reducing serious crashes and making streets safer

#62

Anecdotally, as someone who bikes a lot in SF, Waymo's are a lot safer than human drivers simply because they follow the letter of traffic laws. Stopping at stop signs, waiting for pedestrians to clear the box, following the posted speed limits, etc.

I was just on a business trip to San Francisco for a few days, and I observed the near opposite of this from the Waymo fleet in SoMa: * Waymo vehicle creeping into the pedestrian crosswalk (while the pedestrians had right of way to cross), which caused someone to have to walk around the car into the intersection ahead of the Waymo. * Waymo vehicle entering a dedicated bike lane and practically tailgating the bicyclis…

> I wouldn't want one these anywhere near a high-pedestrian traffic area ever, and I feel the same about human-driven cars, too.

Much of San Francisco is a "a high-pedestrian traffic area" and Waymos operate in those areas constantly and more or less flawlessly. As someone who lived carless in SF for nearly 15 years, I see nothing but upside from more Waymos and less human drivers on those busy streets.

Re: New Study: Waymo is reducing serious crashes and making streets safer

#63

They operate in Los Angeles on the city streets in a square between Marina Del Ray and West Hollywood. They can definitely do better when taking left turns. I've seen situations where Waymo depends on the oncoming drivers to slow down.

It's a bit further than that, the service area now extends to DTLA / Arts District and south toward Inglewood (likely to support LAX soon).

Re: New Study: Waymo is reducing serious crashes and making streets safer

#64

In Holland we have this saying (based on an old commercial): "We from WC Duck recommend... WC Duck". (Which is a toilet cleaning product) It was a funny ad at the time. Unfortunately based in reality more and more these days.

This is almost all research though. (Medical) doctors aren't in the habit of doing longitudinal studies on procedures they don't do because they don't believe in them. (I mean, for one thing, where would they get the data!)

But that's why you have peer review, further studies from different authors perhaps on competing methods that point out some flaws in your approach, etc.

Re: New Study: Waymo is reducing serious crashes and making streets safer

#65
post #56
post #46

Earlier quoted context omitted.

I'm not super familiar with this space, or laws where Waymo is active, what do you mean Google/Waymo don't buy insurance for their vehicles? Is having insurance not legally required? Do they just pay out when there's an accident where they injure someone?

Large companies with fleets of vehicles often self-insure.

Interesting!

When I was much younger, I worked for a couple companies that had (what I would consider) large fleets of vehicles, and they all were insured through an insurance company. I guess I just assumed that's how it was. I wasn't aware self-insuring was a possibility. Thanks.

Re: New Study: Waymo is reducing serious crashes and making streets safer

#66

How do AI cars fare with those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years over the course of a driver's history?? It's happened to me about 2-3 times and I've always made the "(assumedly) correct" split second swerve decision. Wondering if that is a "human/instinctual" skill and if AI cars do just as well or better, or perhaps not as good? I don't ha…

> those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years The correct answer is almost always to hit the brakes. Not to swerve. And Waymo will hit the brakes earlier than you or me.

I would be interested to see waymo adapt to the snow and ice, where “hit the breaks” Is the wrong answer, and the correct is to drop the car into neutral. I believe waymo will figure this out in time, nut SF and Phoenix are idealized driving conditions.

Re: New Study: Waymo is reducing serious crashes and making streets safer

#68
post #37

As someone who is often on SF city streets without a car - I bike and run a lot - I absolutely love Waymo. I am continuously seeing human drivers cut me off, perform illegal maneuvers (i.e. run red lights when I'm going through a crosswalk), and break various other traffic laws. All these things genuinely put people in danger. Just the other day, a guy started running a "no right turn on red" lane in SF, and when I p…

Best part is that they probably have data to show that all that patience costs the typical passenger mere seconds to a minute on 99% of rides. This has always bothered me about aggressive or impatient human drivers: they are probably shaving like 30 seconds off of their daily commute while greatly increasing the odds of an incident.

[flagged]

Re: New Study: Waymo is reducing serious crashes and making streets safer

#69

How do AI cars fare with those instinctual "decide which way to swerve in a split-second" scenarios that come up maybe once every 10-20 years over the course of a driver's history?? It's happened to me about 2-3 times and I've always made the "(assumedly) correct" split second swerve decision. Wondering if that is a "human/instinctual" skill and if AI cars do just as well or better, or perhaps not as good? I don't ha…

I'm ok if the AI car makes an incorrect decision once every 10 years.

Re: New Study: Waymo is reducing serious crashes and making streets safer

#70
post #37

As someone who is often on SF city streets without a car - I bike and run a lot - I absolutely love Waymo. I am continuously seeing human drivers cut me off, perform illegal maneuvers (i.e. run red lights when I'm going through a crosswalk), and break various other traffic laws. All these things genuinely put people in danger. Just the other day, a guy started running a "no right turn on red" lane in SF, and when I p…

Fellow SF cyclist:

Even setting aside the malicious SF stuff, Waymo's have enormous advantages over humans relying on mirrors and accounting for blindspots. I never have to be concerned a Waymo hasn't seen me.

I can't wait until the technology is just standard on cars, and they won't let drivers side-swipe or door cyclists.

Post reply on HN