One thing Tesla and Comma.ai overlooked in self-driving • Hacker News Board

Re: One thing Tesla and Comma.ai overlooked in self-driving

#41

Post by fragmede » Thu, Jun 05, 2025, 3:37 AM UTC

skywhopper wrote: post #29

Animats wrote: post #10

Earlier quoted context omitted.

Right. In the US, once the top 12 taxi markets have been served, Waymo will have most of the industry.[1] That's all the cities with more than 1000 taxis. Serve the next 10 cities, and it's down to Saginaw, MI, with 41 taxis. In San Francisco, Waymo has passed Lyft and is gaining on Uber. [1] https://en.wikipedia.org/wiki/Taxis_of_the_United_States

Uh, that list is wildly incomplete. There are 240 metro areas bigger than Saginaw, Michigan in the United States.

City vs metropolitan area vs CSA. Where cities are a legal jurisdiction, and self driving cars need a license to operate fully autonomously, legal city limits are the thing to look at, not the wider area that gets considered when eg LA is colloquially described.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#42

Post by pests » Thu, Jun 05, 2025, 5:38 AM UTC

Animats wrote: post #39

DennisP wrote: post #37

Earlier quoted context omitted.

Your link says it's the same thing Waymo is doing: > Waymo uses what it calls a “fleet response agent,” a human assistant the vehicle can ping when it gets confused by a complex traffic scenario. These agents can view real-time exterior camera feeds, examine a 3D map of what the vehicle sees and even rewind the footage like a DVR to get better context. “As with the rest of our operations, a helpful human is no more t…

Waymo's people don't remotely drive. Waymo wasn't willing to trust cell phone network latency. They just give hints to the car's control system, like "turn around and take another route", when a car gets stuck.

Interesting. I believe you, but do known of any announcements or sources? Just curious to learn more.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#43

Post by Animats » Thu, Jun 05, 2025, 6:27 AM UTC

pests wrote: post #42

Animats wrote: post #39

Earlier quoted context omitted.

Waymo's people don't remotely drive. Waymo wasn't willing to trust cell phone network latency. They just give hints to the car's control system, like "turn around and take another route", when a car gets stuck.

Interesting. I believe you, but do known of any announcements or sources? Just curious to learn more.

"Fleet response can influence the Waymo Driver's path, whether indirectly through indicating lane closures, explicitly requesting the AV use a particular lane, or, in the most complex scenarios, explicitly proposing a path for the vehicle to consider. The Waymo Driver evaluates the input from fleet response and independently remains in control of driving. This collaboration enhances the rider experience by efficiently guiding them to their destinations."[1]

It looks like the way this works is that the vehicle first has to come to a stop autonomously, and the control center then gives it instructions on what to do next. It's for getting out of stuck situations.

California's CPUC permit for Waymo operation does not allow teleoperation.[2]

Reviews of the job on Glassdoor indicate that it's more or less OK. No opportunity for advancement, but snacks are included.

[1] https://waymo.com/blog/2024/05/fleet-response

[2] https://www.cpuc.ca.gov/-/media/cpuc-website/divisions/consu...

Re: One thing Tesla and Comma.ai overlooked in self-driving

#44

Post by actinium226 » Thu, Jun 05, 2025, 5:57 PM UTC

> Waymo was frequently mocked that they are heavily geofenced, and “it cannot drive anywhere”-mentality. I have to say that I was also a subscriber to this camp and it seemed like universal “no map”-approach was to go. I was cheering for Tesla Autopilot team.

I never thought this argument made much sense because that's not really how humans work. If you take someone who's an excellent driver in NYC and plop them in SF they'll need some time to learn the streets before they can be a really good driver in SF.

For sure they'll be able to navigate safely, but perhaps not well, and I don't want my robotaxi making a wrong turn onto a 1-way street or missing a turn.

Can a neural net be trained to drive well in NYC, SF, and elsewhere? Maybe, but why wait until it's perfect everywhere before starting somewhere?

Re: One thing Tesla and Comma.ai overlooked in self-driving

#45

Post by DennisP » Fri, Jun 06, 2025, 12:55 AM UTC

Animats wrote: post #39

DennisP wrote: post #37

Earlier quoted context omitted.

Your link says it's the same thing Waymo is doing: > Waymo uses what it calls a “fleet response agent,” a human assistant the vehicle can ping when it gets confused by a complex traffic scenario. These agents can view real-time exterior camera feeds, examine a 3D map of what the vehicle sees and even rewind the footage like a DVR to get better context. “As with the rest of our operations, a helpful human is no more t…

Waymo's people don't remotely drive. Waymo wasn't willing to trust cell phone network latency. They just give hints to the car's control system, like "turn around and take another route", when a car gets stuck.

Yes, and your link says that "Tesla's setup appears to be similar." I haven't seen any real evidence that Tesla's remote operators will actually drive the cars.

Re: One thing Tesla and Comma.ai overlooked in self-driving

#46

Post by Animats » Fri, Jun 06, 2025, 2:04 AM UTC

DennisP wrote: post #45

Animats wrote: post #39

Earlier quoted context omitted.

Waymo's people don't remotely drive. Waymo wasn't willing to trust cell phone network latency. They just give hints to the car's control system, like "turn around and take another route", when a car gets stuck.

Yes, and your link says that "Tesla's setup appears to be similar." I haven't seen any real evidence that Tesla's remote operators will actually drive the cars.

Here's one of several stories.[1] Job ads for the remote drivers have been seen.

[1] https://electrek.co/2025/05/29/tesla-testing-robotaxi-servic...

Re: One thing Tesla and Comma.ai overlooked in self-driving

#47

Post by actinium226 » Fri, Jun 06, 2025, 4:47 AM UTC

Animats wrote: post #10

wagwang wrote: post #4

Wait... the big insight here is that people live in cities? Brother, this has been talked to death in the "is level 4 good enough" conversation.

Right. In the US, once the top 12 taxi markets have been served, Waymo will have most of the industry.[1] That's all the cities with more than 1000 taxis. Serve the next 10 cities, and it's down to Saginaw, MI, with 41 taxis. In San Francisco, Waymo has passed Lyft and is gaining on Uber. [1] https://en.wikipedia.org/wiki/Taxis_of_the_United_States

Other have already pointed out that some of the most populous cities in the US aren't on that list, but also if the autonomous can be cheaper and/or more reliable we might see people using them more.