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

Waymo recalls software after two self-driving cars hit the same truck

cnn.com

Post reply on HN
Print view View original

71–80 of 82 posts

Re: Waymo recalls software after two self-driving cars hit the same truck

#71

Post by PH95VuimJjqBqy »

piombisallow wrote: post #65
PH95VuimJjqBqy wrote: post #60

Earlier quoted context omitted.

I don't find that acceptable in any way, no human driver is going to do that and by that I mean no human driver is going to drive into something just because it moved in a way they didn't expect. they're going to slam on the brakes and the only way that's going to happen is if momentum is too high. I understand we have to have explanations or we can't fix them, but it's just as important to understand this should nev…

"no human driver"? Really? Ever? Are you willing to bet on that assertion? Even if the human driver downs a bottle of vodka before driving?

well if we're going to play that game, what if the human suffers a medical emergency!

or what if they're driving a bus and they have to keep above 60 mph!

---

I guess it's my fault, I didn't list every single contingency possible...

Re: Waymo recalls software after two self-driving cars hit the same truck

#72

Post by PH95VuimJjqBqy »

dmd wrote: post #67
PH95VuimJjqBqy wrote: post #60

Earlier quoted context omitted.

I don't find that acceptable in any way, no human driver is going to do that and by that I mean no human driver is going to drive into something just because it moved in a way they didn't expect. they're going to slam on the brakes and the only way that's going to happen is if momentum is too high. I understand we have to have explanations or we can't fix them, but it's just as important to understand this should nev…

> no human driver is going to drive into something just because it moved in a way they didn't expect. This is literally the cause of almost every human accident. Imagine you're driving. There's a car in front of you, also driving, at the same speed as you. Do you immediately slam on the brakes? No, because you EXPECT them to keep driving. That is how driving works. If, suddenly, they do something unexpected - like sl…

here, let me complete that quote for you

> they're going to slam on the brakes and the only way that's [hitting the other vehicle] going to happen is if momentum is too high.

there's a difference between hitting something and driving into something.

Re: Waymo recalls software after two self-driving cars hit the same truck

#73

Post by BHSPitMonkey »

PH95VuimJjqBqy wrote: post #60
Mageek wrote: post #6

Earlier quoted context omitted.

The article says it: “We determined that due to the persistent orientation mismatch of the towed pickup truck and tow truck combination, the Waymo AV incorrectly predicted the future motion of the towed vehicle.” It was detected, but it predicted the truck would move in a way that it didn’t end up moving.

I don't find that acceptable in any way, no human driver is going to do that and by that I mean no human driver is going to drive into something just because it moved in a way they didn't expect. they're going to slam on the brakes and the only way that's going to happen is if momentum is too high. I understand we have to have explanations or we can't fix them, but it's just as important to understand this should nev…

> I don't find that acceptable in any way

Well obviously even Waymo agrees, given that they're recalling vehicles to mitigate the issue.

Re: Waymo recalls software after two self-driving cars hit the same truck

#74

Post by bsaul »

what_ever wrote: post #64
bsaul wrote: post #49

Earlier quoted context omitted.

how do we know for sure that it was fixed ? What are the conditions that are fixed ?

Since the comment thread is comparing humans, how do you know the human has fixed their bug?

Humans don't have bugs. they may have disease, or mental troubles, but we're pretty good at assessing them. Thanks in part to their ability to communicate.

Re: Waymo recalls software after two self-driving cars hit the same truck

#75

Post by piombisallow »

PH95VuimJjqBqy wrote: post #71
piombisallow wrote: post #65

Earlier quoted context omitted.

"no human driver"? Really? Ever? Are you willing to bet on that assertion? Even if the human driver downs a bottle of vodka before driving?

well if we're going to play that game, what if the human suffers a medical emergency! or what if they're driving a bus and they have to keep above 60 mph! --- I guess it's my fault, I didn't list every single contingency possible...

Well yeah, it is. This is an edge case in self-driving cars, same as it could be an edge case for humans.

Re: Waymo recalls software after two self-driving cars hit the same truck

#76

Post by antisthenes »

lotsoweiners wrote: post #57
antisthenes wrote: post #29

Earlier quoted context omitted.

This is why autonomous vehicles need a hard cap on the speed limit they are driving at. If the max speed is 35mph, that allows a good braking system to respond by safely stopping from LIDAR info 99% of the time.

Yeah but then people aren’t going to want to take them because they are slow.

The average speed of a commuting car is around ~23mph, when you account for stops and red lights. 35mph is only a hindrance if you commute by freeway every day, which many people don't.

People will take them if they're priced right and they can do things in the car without having to pay attention to the road.

Re: Waymo recalls software after two self-driving cars hit the same truck

#77

Post by RunningDroid »

bsaul wrote: post #74
what_ever wrote: post #64

Earlier quoted context omitted.

Since the comment thread is comparing humans, how do you know the human has fixed their bug?

Humans don't have bugs. they may have disease, or mental troubles, but we're pretty good at assessing them. Thanks in part to their ability to communicate.

I'd argue that humans don't have bugs but their mental models of things do, mostly as a consequence of the fact a model's complexity increases with it's accuracy.

Re: Waymo recalls software after two self-driving cars hit the same truck

#78

Post by tacitusarc »

piombisallow wrote: post #75
PH95VuimJjqBqy wrote: post #71

Earlier quoted context omitted.

well if we're going to play that game, what if the human suffers a medical emergency! or what if they're driving a bus and they have to keep above 60 mph! --- I guess it's my fault, I didn't list every single contingency possible...

Well yeah, it is. This is an edge case in self-driving cars, same as it could be an edge case for humans.

Probably the main point that the software should function at least as well as a non-impaired capable human still stands…

Re: Waymo recalls software after two self-driving cars hit the same truck

#79

Post by roxgib »

antisthenes wrote: post #76
lotsoweiners wrote: post #57

Earlier quoted context omitted.

Yeah but then people aren’t going to want to take them because they are slow.

The average speed of a commuting car is around ~23mph, when you account for stops and red lights. 35mph is only a hindrance if you commute by freeway every day, which many people don't. People will take them if they're priced right and they can do things in the car without having to pay attention to the road.

But autonomous vehicles tend to perform best on freeways, that's where they are most likely to be allowed

Re: Waymo recalls software after two self-driving cars hit the same truck

#80

Post by roxgib »

im3w1l wrote: post #23
digitallis42 wrote: post #20

Earlier quoted context omitted.

It's unlikely to be neutral network based. This sounds like a model prediction failure. You take a mathematical model of car motion: the rear wheels generally don't steer. The front steered wheels can cause the car to drive along an arc. If you want to predict the arc that will be driven, you take the initial starting heading of the vehicle and project forward in time based on your understanding of the vehicles steer…

I'm thinking it might make sense to have a sort of hierarchy of models. The stupidest model predicts that everything will be stationary. The second model predicts that everything will travel in a straight line. The third model tries to predict a circular arc based on same fusion of path history and observed steering input. Fourth model uses a notion of action to predict whats going to happen, like "the car is changin…

These cars can and do slow down or even stop and wait for human assistance in response to unexpected situations. I'm actually quite surprised this didn't trigger here, although we don't really know much about the specifics of the situation.
Post reply on HN