Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cyclist hit by driverless Waymo car in San Francisco, police say

sfgate.com

Post reply on HN
Print view View original

51–60 of 750 posts

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#51

Post by itake »

km3r wrote: post #11

So the scenario (as described by waymo, surely with some spin): "Driving through a 4 way intersection, while a large truck is also going through the intersection. As you go through, a biker quickly passes behind the truck with minimal time to react once in sightlines." A human driver would take significantly longer to react, and could be deadly. Waymo managed to hit the breaks before the collision, potentially leadin…

> When they became fully visible, our vehicle applied heavy braking but was not able to avoid the collision.

The "fully visible" part made me thinks humans can identify a bicycle and rider with just partial visibility, which means they would stop faster than the car.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#53

Post by readthenotes1 »

TulliusCicero wrote: post #32
mvdtnz wrote: post #27

Earlier quoted context omitted.

Are you implying that we shouldn't be concerned about automated vehicles hitting cyclists if the cyclist wasn't following the lawn to the letter?

According to the Waymo description, the cyclist was behind a truck, and not visible to the Waymo car. If that's true, then a human driver likely would've done the same thing: take its turn at the stop sign.

In most of my defensive driving classes, they tell me that if I can't see it, that means there's probably a car there.

So a safe human driver would not have entered the intersection while the truck was obstructing vision.

That may have pissed off the people behind me, but at least I wouldn't be turning while blind.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#54

Post by readthenotes1 »

tobyjsullivan wrote: post #26
kevingadd wrote: post #10

Earlier quoted context omitted.

It is pretty common to see people go from the opposite side of a stop sign since it's "safe", even if it happens to be against the rules. I can understand how the Waymo AI would end up being configured to do this. Hopefully they amend that based on this incident.

I don’t know about California but that’s legal where I live. Vehicles on opposite sides of a 4-way stop can enter simultaneously if both are going straight.

The way Mo was turning

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#55

Post by nicbou »

ugh123 wrote: post #20
km3r wrote: post #11

So the scenario (as described by waymo, surely with some spin): "Driving through a 4 way intersection, while a large truck is also going through the intersection. As you go through, a biker quickly passes behind the truck with minimal time to react once in sightlines." A human driver would take significantly longer to react, and could be deadly. Waymo managed to hit the breaks before the collision, potentially leadin…

>What I hope Waymo learbx for this, as well applies as similar scenarios: never cross through these zones of high bike traffic when the obstruction from another vehicle prevents a wide enough view to brake before this zone. I hope it's as simple as that, and that regulators will understand just how little fine-tuning could yield substantial safety increases not possible with human drivers.

Not only that, but fine-tuning is cumulative. Test situations can be created and tested against forever. Lessons can be learned and applied across the whole fleet.

This is radically different from sharing the roads with a bunch of Sunday drivers, some of which might be distracted or impaired.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#56

Post by readthenotes1 »

tzs wrote: post #35
kevingadd wrote: post #10

Earlier quoted context omitted.

It is pretty common to see people go from the opposite side of a stop sign since it's "safe", even if it happens to be against the rules. I can understand how the Waymo AI would end up being configured to do this. Hopefully they amend that based on this incident.

> It is pretty common to see people go from the opposite side of a stop sign since it's "safe", even if it happens to be against the rules. Where is it against the rules?

Is it not against the rules to drive into an area that you do not have proper visibility?

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#57

Post by hx8 »

xianshou wrote: post #7

If Waymo is truly as much better than Cruise as the statistics claim, this is a perfect opportunity to release the full video of the incident. Most collisions are avoidable, but a few are the result of a perfect storm of conditions that would be difficult to navigate for even the most skilled human driver. To know whether the truck's occlusion of the cyclist is one of those cases requires the actual footage.

If the footage is not released to the public I'm going to assume it makes Waymo look bad.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#58

Post by babl-yc »

xianshou wrote: post #7

If Waymo is truly as much better than Cruise as the statistics claim, this is a perfect opportunity to release the full video of the incident. Most collisions are avoidable, but a few are the result of a perfect storm of conditions that would be difficult to navigate for even the most skilled human driver. To know whether the truck's occlusion of the cyclist is one of those cases requires the actual footage.

I agree it would be good to share video if possible, but Waymo cannot realistically expect that there will never be an accident where they look bad.

There are 43,000 US road deaths per year. If all driving was autonomous, even if autonomous driving was 100x safer than humans, that would still be 430 deaths per year.

And some of those accidents will be failures that a human would not make (sensor failure, bad software update, edge cases).

What you _can't_ do is what Cruise did and hide evidence from the public and regulators fearing that the truth would impact your business.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#59

Post by readthenotes1 »

mewpmewp2 wrote: post #18
lcnPylGDnU4H9OF wrote: post #14

Earlier quoted context omitted.

I can’t believe those things are allowed on the road!

I can't believe we have roads in the first place. It is the invention of the romans. And what good have they ever done for us?

Possibly, perhapsily, they left someone behind in England who became the seed of the King Arthur stories, and those are pretty good.

Re: Cyclist hit by driverless Waymo car in San Francisco, police say

#60

Post by roenxi »

itake wrote: post #51
km3r wrote: post #11

So the scenario (as described by waymo, surely with some spin): "Driving through a 4 way intersection, while a large truck is also going through the intersection. As you go through, a biker quickly passes behind the truck with minimal time to react once in sightlines." A human driver would take significantly longer to react, and could be deadly. Waymo managed to hit the breaks before the collision, potentially leadin…

> When they became fully visible, our vehicle applied heavy braking but was not able to avoid the collision. The "fully visible" part made me thinks humans can identify a bicycle and rider with just partial visibility, which means they would stop faster than the car.

Based on what? Humans run in to cyclists too and although I have know idea about Waymo's tech, in a research setting computers are better than humans at figuring out what objects are based on visual data.
Post reply on HN