Yup, it should be boring other than the first wow factor of no one being in the driver's seat.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Waymo
- Thread starter Daniel in SD
- Start date
Elon went for a different kind of wow. Totally different strategies; can’t compare.
diplomat33
Average guy who loves autonomous vehicles
Funny but no. I don't like the term "warlord". It has a negative connotation. Maybe "TechnoKing of Waymo"? Would Elon sue me? LOL.
Frankly, I am a big fan of autonomous driving tech and driverless cars. I like Waymo because they have really good autonomous driving tech and really good driverless cars. I just wish they would expand faster. But I don't have any ties to the company.
Man. Ok ok Waymo Wizard or or Waymo Wanker hmm maybe Waymo Wanderer…
EVNow
Well-Known Member
Terminator857
Active Member
Waymo pokes a bit a fun at Tesla mistaking a moon for a traffic light.
The problem with this is that was last July. Anyone notice trouble with moon lately? Tesla seems to move (make updates) much faster than Waymo. Another reason why I feel Tesla is ahead of Waymo in the long run, 5 years.
EVNow
Well-Known Member
It takes a long time for Waymo to produce a blog - after it has gone through 500 levels of wetting
rxlawdude
Active Member
Sounds like they'll need Bounty. The quicker picker upper.... after it has gone through 500 levels of wetting
EVNow
Well-Known Member
Yes, thats what they call the new CEO. Though, "less slow" is more apt than "quicker".
Terminator857
Active Member
Tesla wins by 2 minutes because Waymo avoids freeways.
Could this be because Waymo is optimizing for something other than speed?
Terminator857
Active Member
haha good one. But we do know freeways are safer than streets.
Last edited:
diplomat33
Average guy who loves autonomous vehicles
Yes. Waymo driverless rides with passengers are optimized for safety and following traffic laws, not speed.
Like you, I am a fan of autonomous driving technology. I also agree that Waymo has made some impressive strides, especially given that they are limited by human software-writing capacity, which is probably close to topping out. Tesla has chosen an entirely different strategy, as you know. By building a system on deep learning neural nets and inexpensive sensor suites, they are enabling scale and flexibility, without the need for HD maps, geofencing, multiple sensors and so forth. In Tesla's case, the limitation is no longer human software, but rather the curated training data. Kaparthy argues that the performance ceiling of the neural net approach is much much higher, even with existing hardware. Time will tell, but if I were betting, I would bet on Tesla to get to the more general solution first.
I don't agree with folks who deprecate Waymo and other non-Tesla attempts at driverless vehicles. On the contrary, the competition is good for everyone. And we need companies pursuing multiple strategies to increase the likelihood of eventual success. So, as much as I love my Tesla and as much as I'm enjoying testing the FSD beta, I also say: Go Waymo! Go everyone else trying to do this complex and difficult thing!
It's probably because they are trying to limit the scope of the problem as much as possible. Their goal is to get to driverless operation as fast as possible because a Waymo with a safety driver is useless.
The real question is how many times could the average human driver complete that route without a collision? How many times could the Waymo? That's what they're optimizing for.
EVNow
Well-Known Member
Ofcourse, otherwise they wouldn't just serve a tiny
Here in the issue - to work on FSD takes a long time. So it is ideal for someone with a LOT of money - like Google or Apple. BUT - those companies are very risk averse given their cash position and every lawyer worth his degree (or not) waiting to sue them.
Hmmm ... wait. Were you talking about some other kind of speed ?
diplomat33
Average guy who loves autonomous vehicles
I am not sure what you mean by "human software writing capacity". Waymo is using deep NN and ML for every part of their stack, including perception, prediction and planning. Most of Waymo's stack is NN. In fact, Waymo uses more ML than Tesla actually. The HD maps are of course specialized to a geofenced area but the rest of perception (camera, radar, lidar), prediction and planning stack is all generalized. It would be more accurate to say that Waymo is using both HD maps and generalized NN.
Terminator857
Active Member
The inside scoop says it is because traffic laws aren't friendly to cars trying to merge onto a freeway. If a Waymo is trying to merge then it can be baited by someone seemingly allowing it onto the freeway, then accelerate onto merging Waymo and it would be Waymo's fault. As a datapoint consider that Waymo said freeways are easy and that is why they don't bother testing there. If traffic law was changed to allow easier merging, Waymo would be using the freeways. In Germany the law is better. Germany says every car has to let one merging car in or something like that.
Last edited:
diplomat33
Average guy who loves autonomous vehicles
Someone spotted a driverless Waymo truck on I-10 outside Tucson, AZ.
Similar threads
- Article
