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

Uber wants to turn its drivers into a sensor grid for self-driving companies

techcrunch.com

Post reply on HN
Print view View original

71–80 of 152 posts

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#71

Post by gcheong »

ra7 wrote: post #32

> The insight driving the program, Naga said, is that the limiting factor for AV development is no longer the underlying technology. “The bottleneck is data,” he said. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all the…

Didn't they need the data from the 200 million miles or so from actual driving before they could get to the generative model though? Data isn't everything, as you point out with Telsa (mainly because they decided to forego using lidar it would seem), but it is pretty fundamental.

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#72

Post by ra7 »

suddenexample wrote: post #59
ra7 wrote: post #32

> The insight driving the program, Naga said, is that the limiting factor for AV development is no longer the underlying technology. “The bottleneck is data,” he said. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all the…

Yeah I'm not so sure this CTO is on the mark here, but to be fair, I do think some of this IRL long tail/edge case data is important for Waymo. The simulation software is super interesting to me - the real world can be so chaotic, and even if they could generate every possible real life case, there needs to be validation on whether the Waymo driver is responding in the optimal way. They certainly haven't solved this…

I agree real world data is important for Waymo. I didn't mean to say it wasn't, so I've edited my comment to reflect that. It's just that data is not some magic bullet to achieve self driving like Tesla and others suggest.

Of course, Waymo still has much more room for improvement. But it's much more efficient to supplement less but higher quality IRL data with large amounts of synthetic data, than to run a million data collection vehicles 24x7 because most IRL data is boring and useless.

Waymo said 6 years ago they simulate 20 million miles every single day [1]. Clearly, it's working for them given their scale of deployment right now.

[1] https://waymo.com/blog/2020/04/off-road-but-not-offline--sim...

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#73

Post by ra7 »

gcheong wrote: post #71
ra7 wrote: post #32

> The insight driving the program, Naga said, is that the limiting factor for AV development is no longer the underlying technology. “The bottleneck is data,” he said. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all the…

Didn't they need the data from the 200 million miles or so from actual driving before they could get to the generative model though? Data isn't everything, as you point out with Telsa (mainly because they decided to forego using lidar it would seem), but it is pretty fundamental.

IIRC, they had clocked 20 million real world miles before starting to scale their deployment. But they were also driving 20 million miles in the simulator every day: https://waymo.com/blog/2020/04/off-road-but-not-offline--sim...

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#74

Post by reaperducer »

nerdsniper wrote: post #5

I feel like they should have done this 6 years ago. Most AV companies already have tons of their own data today. But how would it work to install expensive LIDAR sensors on privately-owned vehicles?

Most AV companies already have tons of their own data today.

Real-world data spoils faster than a gas station banana.

If your AV company is relying on data from six years ago, you're going to kill someone.

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#75

Post by ra7 »

whiplash451 wrote: post #65
ra7 wrote: post #32

> The insight driving the program, Naga said, is that the limiting factor for AV development is no longer the underlying technology. “The bottleneck is data,” he said. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all the…

Waymo might very well be missing specific kinds of data (e.g more incidents/accidents, near-collisions etc) Also, Uber’s data might be useful for eval, not training (e.g « here is how Waymo would behave vs human drivers therefore it is safer »)

> Waymo might very well be missing specific kinds of data (e.g more incidents/accidents, near-collisions etc)

Accidents and near-collisions are exactly the kind of scenarios perfect for simulation. You don't test them out in the real world and risk injuries/deaths. You need to have confidence they're handled before you deploy.

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#76

Post by ra7 »

simmonmt wrote: post #58
ra7 wrote: post #32

> The insight driving the program, Naga said, is that the limiting factor for AV development is no longer the underlying technology. “The bottleneck is data,” he said. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all the…

> When they want data about a school intersection in SF at a certain time of day, they just... synthetically generate it and simulate I think it's more about detecting changes to the world. You need boots on the ground, so to speak, to see that new speed limit sign or the new lane paint. The Waymo vehicle can no doubt react to changes in the world when it encounters them, relaying them back to the mothership, but it'…

Most AVs, definitely Waymo vehicles, are self mapping. They can detect environment changes and relay it to the entire fleet. That's because they map using the same vehicles as the fleet.

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#77

Post by skybrian »

ra7 wrote: post #72
suddenexample wrote: post #59

Earlier quoted context omitted.

Yeah I'm not so sure this CTO is on the mark here, but to be fair, I do think some of this IRL long tail/edge case data is important for Waymo. The simulation software is super interesting to me - the real world can be so chaotic, and even if they could generate every possible real life case, there needs to be validation on whether the Waymo driver is responding in the optimal way. They certainly haven't solved this…

I agree real world data is important for Waymo. I didn't mean to say it wasn't, so I've edited my comment to reflect that. It's just that data is not some magic bullet to achieve self driving like Tesla and others suggest. Of course, Waymo still has much more room for improvement. But it's much more efficient to supplement less but higher quality IRL data with large amounts of synthetic data, than to run a million da…

Although most of the real-world data is probably boring, collecting more of it likely makes discovering rare edge cases more likely. But since they happen rarely, I imagine that after discovering them, they would then need to figure out how to simulate them.

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#79

Post by ninjagoo »

AndrewKemendo wrote: post #35
JumpCrisscross wrote: post #30

Earlier quoted context omitted.

> for a moment of pleasure The pleasure of being an Uber driver? Wouldn’t the better analogy be survival for most gig workers?

The majority of people have absolutely no foundational belief for their actions it is simply how do I get more money sex property attention etc… There are no monks door dashing

> The majority of people have absolutely no foundational belief for their actions

The majority of people have absolutely no choice for their actions

FTFY

Especially given that for an increasing number of folks the alternative to doing what you're told is starvation (food insecurity), homelessness, and death (from lack of healthcare).

Re: Uber wants to turn its drivers into a sensor grid for self-driving companies

#80

Post by JumpCrisscross »

jcgrillo wrote: post #64
JumpCrisscross wrote: post #34

Earlier quoted context omitted.

> it doesn't even snow there My Subaru can lane keep in a Wyoming blizzard. There isn’t some unsolved technical problem with snow for any system with radar, i.e. anyone who isn’t Tesla. Keep in mind that like a fifth of Americans and half of humans live somewhere is rarely or never snows.

I guess once they demonstrate it working smoothly and profitably in cities like Boston, NY, Detroit, etc. I'll be more concerned? Given that it doesn't really even work too well in places with good signage, lane markings, etc--not to mention no weather--I'm not worried yet. And frankly I'm done with "the future". If it doesn't deliver results now, it's not real. Until it's real, it's nothing to get alarmed about.

> cities like Boston

Couldn't come soon enough [1].

> Given that it doesn't really even work too well in places with good signage, lane markings, etc

Works fine in Phoenix, Miami and Los Angeles. Plenty of neighbourhoods there have non-existent, defaced and degraded signage and markings.

> Until it's real, it's nothing to get alarmed about

I don't think there is anything to be alarmed about, period. Driving is a silly job when you think about it. We made these machines to do our bidding, not enslave us behind their wheels.

My belief in a smooth roll-out is reinforced by those who would probably oppose AVs also not believing it's real. Once the first factory mass manufacturing AVs breaks ground, any limited local opposition can be preëmpted.

[1] https://waymo.com/blog/shorts/back-to-boston/

Post reply on HN