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

Beginning fully autonomous operations with the 6th-generation Waymo driver

waymo.com

Post reply on HN
Print view View original

291–300 of 425 posts

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#291

Post by rhubarbtree »

gf000 wrote: post #263
rhubarbtree wrote: post #204

Earlier quoted context omitted.

Unsure if you’re trolling, but you haven’t listened to what Tesla are actually saying. Having more sensors is complicating the matter, but yes sure you can do that if you want to. But just using vision simplifies training a huge amount. The more you think about it, the stronger this argument is. Synthesising data is a lot easier if you’re dealing with one fairly homogenous input. But the real point is that cameras ar…

That assumes that hardware was/is/will be more expensive than.. simply scaling up data collection and training? Which seems like a very bad assumption, I'm not even sure it was ever true and is getting less and less true.

The data is key. You need a lot of homogenous data collected at vast scale over places and time, and you need to be able to synthesise data accurately.

Waymo gets limited data from very limited locations, and will have a harder time synthesising data than others.

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#293

Post by rhubarbtree »

jacquesm wrote: post #222
rhubarbtree wrote: post #201

Earlier quoted context omitted.

Last time I checked I relied entirely on vision to drive autonomously.

That's a very naive way of looking at this. You have an extremely detailed world model including a mental model of the drivers and other road users around you. You rely on sight, sound, experience and lots of knowledge. You are aware of the social contracts at work when dealing with shared resources and your brain is many orders of magnitude more powerful than any box full of electronics. What you can do with 'just v…

I don’t disagree - what you’re saying is major improvements in AI would be needed to make this work. You are correct.

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#294

Post by WarmWash »

small_model wrote: post #238
WarmWash wrote: post #156

Earlier quoted context omitted.

It seems they run without a safety driver or follow car (mostly?). However the area it operates is extremely small, and they are still only allowing Tesla bros to try it.

Nope is open to the public and covers bigger area then Waymo. EDS is limiting a lot of people here's ability to critically evaluate the current autonomous auto rollout.

The unsupervised area is a tiny subset of the supervised area.

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#295

Post by testing22321 »

UltraSane wrote: post #183
testing22321 wrote: post #180

Earlier quoted context omitted.

> and there won't be, for years, if ever. That is a lot of confidence. Do you work in the autonomous vehicle space? What makes you so certain?

Because camera only simply won't be reliable enough with current technology.

Technology is moving fast.

When do you think it will be reliable enough?

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#296

Post by schmidtleonard »

zoenolan wrote: post #207
coffeemug wrote: post #179

Earlier quoted context omitted.

That's fascinating, I didn't know that! What are some example cities/towns where this is common?

Tom Scott visits Peachtree City, Georgia https://youtu.be/pcVGqtmd2wM

Is there anything in Georgia not named Peachtree?

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#297

Post by rcxdude »

YeGoblynQueenne wrote: post #277
rcxdude wrote: post #142

Earlier quoted context omitted.

A safety driver is a driver that needs to stay focused on the vehicle's movements 100% of the time, to intervene if the autonomous system fails and starts to become unsafe. This is not what Waymo uses. The autonomous system is trusted with safety, the remote agents are there to resolve cases where the autonomous system cannot proceed (or, more precisely, is not allowed to proceed out of an abundance of caution) witho…

This is just quibbling about semantics, and with arbitrary definitions of "safety driver" on top. Waymo's remote operators are clearly safety drivers: they are there to improve the safety of their safe-driving cars. Otherwise, why are they needed?

They're there to improve the availability of their self- driving cars. If they were to all just walk away, there would be no danger of crashes, but a lot more of their cars stopped in inconvenient places.

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#298

Post by corford »

rhubarbtree wrote: post #199
qwerpy wrote: post #114

Earlier quoted context omitted.

I hope this prods Tesla to up their game. I love my Teslas but if Waymo’s approach is shown to be truly better then I’d happily switch to a car that used their tech. For now I have no choice but to stick with the self-driving that’s available for personal cars. Hopefully Waymo works on licensing their tech for other manufacturers and expanding their geographical coverage.

Tesla’s approach is smarter than Waymo. Maybe Tesla won’t win, perhaps a Chinese firm, but the winner will follow a vision based system and not a micro-mapping LIDAR system. “Vision is all you need.”

With costs approaching $200 (if Hesai Group is to be believed - and there's no major reason I know of to doubt them), why won't LIDAR win?

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#299

Post by circuit10 »

chpatrick wrote: post #250
lysace wrote: post #241

Earlier quoted context omitted.

Because trains are famously awesome for the last mile(s)?

I think the idea is that they could go full speed in sync, then afterwards they could still take you to your doorstep.

The commenter was being sarcastic about trains I think, but while we should be using public transport where possible there is sometimes a need for door-to-door transport so I don't see why we can't explore new ideas that may combine some of the advantages of both for certain use cases

Re: Beginning fully autonomous operations with the 6th-generation Waymo driver

#300

Post by badgersnake »

Rudybega wrote: post #38
dweinus wrote: post #31

[flagged]

There aren't remote "drivers" in the Philippines, that's not how Fleet Response works. You can see how it works here if you're curious: https://waymo.com/blog/2024/05/fleet-response but the TLDR is that they give the Waymo driver options in confusing situations (things like, you can go use this driveway on the side to pass this blocked traffic).

Directing the movements of a car is now “fleet response”, not “driving”. Give them a fucking innovation award or something.
Post reply on HN