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

Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

bloomberg.com

Post reply on HN
Print view View original

171–180 of 340 posts

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#171

Post by a008t »

emodendroket wrote: post #59
fucking_tragedy wrote: post #57

Earlier quoted context omitted.

> It is a bit of a shame that a tech that could save a million lives a year globally won't be deployed because it can't be made perfectly safe This is a bit of a generous assumption. It very well could turn out that self-driving cars are worse, or no better, than human drivers.

Right. So many arguments just start from the premise that self-driving cars must be safer and then ask why we aren't racing to replace human drivers with them. In reality the proposition that they are safer is far from proven.

Not to mention that the risk profile is going to be different. When you are driving your car, there is at least the illusion that you are in control of your risks. A careful driver at least believes that their risks are significantly lower than average. Getting into a self-driving car you have no control and are just playing Russian roulette.

Not to mention that the failure modes of self-driving cars are likely to be very different from failure modes of humans, making it more difficult for other road users to predict what they would do in any given situation.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#172

Post by zaarn »

throwaway76543 wrote: post #150
Retra wrote: post #131

Earlier quoted context omitted.

Yes, it is trivial to calculate how to respond to known hazards. What is hard to calculate is what the hazards actually are in any given situation. Humans don't merely look at the road and compute how much standing water is there, they have prior experiences and a sense of what other drivers are doing on the road and why they are doing those things.

Self driving cars also have prior experiences and a sense of what other drivers are doing and why. That's sort of the point of machine learning. In fact, they will have a lot more prior experience than any human driver in the world.

Machine Learning still confuses Elephants for Cats.

While it's quite fun if you pin up a cat dressed up in an elephant costume and surely amuses a few colleagues, a self-driving car is not something that should confuse an Elephant for a Cat.

ML is IMO not reliable enough for use in self-driving cars, in complicated situations the driver should take over.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#173

Post by jakeogh »

mikejb wrote: post #94
manfredo wrote: post #70

Earlier quoted context omitted.

Stability and traction control is not the same as a self driving car. Keeping wheels from locking, and keeping cars from sliding is largely a solved problem. I think the previous poster was more about referencing recognition of patches of ice ahead and slowing down preemptively. This is a big issue in places with bridges, since ice can form on bridges but not the rest of the roads (since bridges aren't kept warm by t…

I'd argue that Self-driving cars are better capable of recognizing ice patches. They don't just rely on visible light to see the ice, but have a more diverse sensor array. The challenge there of course is having the car learn to interpret the input of these sensors. But that doesn't seem like a unpassable hurdle. On the contrary, it would be pretty high on the list of things to tackle when aiming for driving in snowy…

Ignoring the real problem you mentioned, camera dynamic range (for example) is pathetic compared to the human eye.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#174

Post by umichguy »

bronco21016 wrote: post #119

Waymo’s Michigan testing facility is very close to my home. I’m really curious to see how they do on snowy pothole ridden roads in Michigan. I’ve seen very few of them on the road this summer but if there’s supposed to be heavy testing and validation going on perhaps we’ll see an army of them on the roads here soon. I’m curious how self driving focused groups are solving lane detection and managing surrounding traffi…

"I’m curious how self driving focused groups are solving lane detection and managing surrounding traffic on snow covered roads. It’s not uncommon for a multi lane highway to be completely covered with snow and multiple tracks for different ‘lanes’ that overlap. It’s also not uncommon to see cars just making their own lanes. This occurs even after the snow has stopped flying."

Yep. As a Michigander, I totally agree. You see 6 to 12 lane roads reduced to "random" 2 lane roads when there is a good pileup of snow. And the snow drifts make it even worse!

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#175

Post by Moru »

scj wrote: post #45
grahamburger wrote: post #31

Earlier quoted context omitted.

This sounds like exactly the kind of job for a computer. I live in an area that gets a lot of snow. The only 'safe' things for a human to do in snowy, icy, wet conditions are: slow down (a lot), leave a lot more following distance, stay home. I talk to a lot of people who seem to think that they are able to drive at full speed in the snow and complain about all of the 'out of town' drivers who slow down the roads bec…

> I'd rather share the road with a computer. Not version 1. Factoring in all these variables is going to be a lot of work for a non-universal feature. I have no doubt they'll be included some day, but I doubt it'll be day 1. EDIT: To be clear, I believe that dealing with ice and snow is an order of magnitude more difficult than driving under normal conditions. This is an 80-20 feature that many people won't need. I c…

I'm worried about the south europeans jumping into their self-driving car thinking they can come visit us in northern Sweden early spring (their season, still full winter here).

How do you meassure how slippery the road is? How much wind from the side? Will the wind blow me off the road because it has been polished by the wind just after those trees end, next to the lake? Will the car know to stop before even going because the car doesn't have winter tyres with good enough tracktion left?

A self-driving truck, does it know enough not to slow down in the uphill because it will start to glide sideways or backwards where the road leans too much to the side?

I get gray hair when I start thinking of all the problems that needs to be solved before we have self-driving cars.

One of the developer voes, am I wise enough to be humble about my limits of knowledge or do I know enough to be dangerous?

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#176

Post by rjsw »

graycat wrote: post #167
Hamuko wrote: post #110

Earlier quoted context omitted.

This is so incorrect it hurts.

My point is clear if just consider a little of how snow tires work: The main point is the big bumps in the tread. These big bumps permit spinning drive wheels to dig down in the snow to find some traction, maybe the hard surface below. This digging down has the bumps grab a lot of snow and throw it back and out of the way and let the tires get down, through the snow into some traction. Ordinary tires just spin and th…

Modern snow tyres don't have big bumps in the tread.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#177

Post by agurk »

realandreskytt wrote: post #145
mdhen wrote: post #54

Earlier quoted context omitted.

Those conditions sound like you wouldn't be driving. So, solved. Pull over and stop

I’d recommend visiting Finland. In these parts, there’s basically 3 months of optimal visibility. The rest is fog, rain, snow, sleet and road conditions that would shut down most of US. Yet people drive and do so safely. “What people can do” is a bit different from “what an average us driver can be expected to do”

And as a taxi driver once told me in Helsinki there are four seasons in Finland - Autumn, Winter, Spring and Construction. Even when the visibility is optimal they'll likely be tearing up the road anyway.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#178

Post by Scarblac »

edejong wrote: post #138
hef19898 wrote: post #122

Earlier quoted context omitted.

Can that be that in US cities there is not trmendous amount of left turns? At least that was the rwason I heard why the Amazon Logistocs routes in Europe were planned without any left turns driving everybody crazy and adding a lot of wasted time.

The reason Amazon plans like that is because it is faster. Left turns with traffic lights have a higher average waiting time. Also, the route can have left turns, they just have a much higher penalty.

That sounds like an assumption that holds in the US much more than elsewhere, because in the US you don't have to wait for the traffic light if you turn right, but elsewhere you do.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#179

Post by Tor3 »

jakeogh wrote: post #170
mortehu wrote: post #64

Earlier quoted context omitted.

Yet electronic stability control[1] works far better than human control on snow, wet roads, or ice. If you don't believe it, try an obstacle course on an ice track with and without it sometime (a required part of driver's ed in Norway). Example: https://youtu.be/hd5JMQr4qe0?t=56s https://en.wikipedia.org/wiki/Electronic_stability_control

Unscientific test. Give the human individual brake control (I can think of simple analog systems to do it). I learned to drive on ice covered dirt roads, many times I would have liked to have the ability to asymmetrically brake. A simple 2nd order gimble would be awesome. Having a single symmetric pedal dumbs it down way too much to make a pre-programmed car vs human comparison. Same thing goes for torque distributio…

> I would have liked to have the ability to asymmetrically brake

Yes, I have been in that situation too.

As for brakes.. ABS is very bad at handling hard snow with a drizzle of fresh snow on top. It's nearly impossible to stop. I've had to actually turn off the engine in order to stop the car when driving (slowly) down a hill from a hotel in a certain snowy area, in order not to end up on the main road in front. That was with various rental cars (yes, I go to that place regularly in my job). The first time this happened I thought here was a technical problem with the car. So I asked around. Same experience from others. The cars vary, but they're all much worse than fully manual brakes, in certain conditions.

And then there was the time there was a reindeer (yes, a reindeer) running along the road, zig-zagging in front of the car. Snow conditions as described above. ABS refused to brake. Didn't want to switch off the engine (busy steering), so after about fifteen seconds of this I drove the car off the road instead, safely into a lot of snow.

And people want computers to drive cars, when computers can't even be used safely for brakes?

In any case, I've yet to hear about a self-driving car which can predict that another driver will get into problems in a few seconds unless I slow down or increase the speed, in order to allow a third car to adjust. This is something I see or do nearly daily, on my commute.

Re: Self-Driving Cars Can Handle Neither Rain nor Sleet nor Snow

#180

Post by yogrish »

Not to be Cynical,Somehow the article sounded like a sales pitch for "Wavesense". They too want to have pie out in the self driving race. As mentioned in the article, each sensor(Radar,Lidar,Camera) has a limitation on its own. But, Self driving cars depend not just on one sensor but on Fused output of sensor. Also, they use V2V, HD maps and GPS/DR for localisation. Wavesense is another sensor that can solve localisation problem. They too might have difficulty in cases where there are fallen metal objects that reflect different signatures, causing trouble. More the sensors, more the confidence for Self driving.

Edit: Grammar & some terminology

Post reply on HN