Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Waymo

This site may earn commission on affiliate links.
#981
They used to drive around Mountain View for years. I would see many per day, sometimes 3 Pacificas lined up right behind each other at lights.

They didn't transport a single paying customer in Mountain View after all those years. I've come to see Waymo driving around some area like a type of advertising and marketing ("Hey, we're still here and relevant!").

Google/Waymo has a huge presence in Mountain View are with many thousands of employees that are potential customers. Most Silicon Valley companies pay for coaches and shuttles, reimburse for public transport like buses and trains, pay for Ubers/Lyfts...

powertoold said:
Waymo's been ramping up their gen-5 i-Pace cars in San Francisco big time. I saw more than 5 of them near the Dolores Park area within the span of 15 minutes yesterday. They all seemed to be going the same route though.
Click to expand...
 
  • Like
Reactions: mikes_fsd
#983
ThomasD said:
How many cars would Waymo need to replace the UBER drivers in the San Francisco area.
Click to expand...

I think that's hard to say because it would depend on lot of variables. Uber currently has 50,000 Ubers in SF. But Waymo could probably be able to do the same area with less cars since driverless cars don't need to rest or sleep. Driverless cars can do more rides, with less downtime. So a smaller fleet can do more.

ThomasD said:
Could they make the ride cheaper and more convenient than an UBER
Click to expand...

Short answer: Yes

Long answer: They already have made it cheaper than Uber. According to JJ Ricks who rides in Waymo and has tracked all his rides, the current cost of Waymo in Chandler is cheaper than an Uber. Waymo's cost is roughly $0.80 per mile with $4.99 minimum. In SF, Waymo would use the 5th Gen I-Pace which Waymo says is cheaper than the Pacificas. So a ride-hailing service in SF, could be even cheaper than in Chandler. So Waymo could definitely make it cheaper than Uber.

Convenience is a matter of personal preference, I think. Some might argue that having the entire Waymo car all to yourself is a convenience. The wait time is shorter for an Uber, so that might be a convenience point in favor of Uber. However, Waymo is working to reduce that. With more cars, Waymo could probably reduce the wait time. Most Waymo rides are very smooth so that it is a point of convenience in Waymo's favor. The rare instance when Waymo gets stuck would certainly be inconvenient. But again, Waymo is working to reduce those instances.

Overall, I would say that it should be very possible for Waymo to make their rides cheaper and more convenient than Uber.
 
#984
Waymo presentation on their multi-path prediction model:


"Long-term human motion prediction is a critical component of scalable autonomous driving systems. There has been a multitude of core modeling improvements for this task in recent years, in large part fueled by popular public benchmarks. However, existing datasets, metrics, and output representations leave much to be desired in their ability to capture interactions between agents. In this talk, we go over recent advances in modeling, datasets, evaluation and output representations at Waymo, with a particular emphasis on modeling interactions."
 
  • Like
  • Funny
  • Informative
Reactions: Bladerskb, mikes_fsd, Microterf and 1 other person
#985
Would love to know Waymo’s thoughts on this whole shift at Tesla to Vision only. That must be an interesting topic of discussion lately. Waymo previously stated how Tesla had a slight advantage, primarily due to their huge data collection and data sampling they have to use and improve their systems with similar technologies. Now part of that is out the window.
 
#986
techlogik said:
Waymo previously stated how Tesla had a slight advantage, primarily due to their huge data collection and data sampling they have to use and improve their systems with similar technologies. Now part of that is out the window.
Click to expand...

Which part is out the window?

I don't think Tesla was maintaining a large neural network training set of radar returns. The valuable training data is real-world 360 video.
 
  • Like
Reactions: rxlawdude and mikes_fsd
#987
techlogik said:
Would love to know Waymo’s thoughts on this whole shift at Tesla to Vision only. That must be an interesting topic of discussion lately.
Click to expand...

I think the Waymo blog on the 5th Gen hardware basically answers that question:

"In order to navigate the complexities of driving - from the biker weaving and speeding through traffic on a foggy San Francisco morning, to the family pet rushing into the street to greet you at night - the Waymo Driver uses a comprehensive view of its surroundings and a detailed understanding of its environment to accurately reason about its next move. No one type of sensor on its own can provide that level of detail in all operating conditions; therefore, we've spent over a decade developing a single integrated system comprised of complementary sensors to give our Driver this comprehensive view of the world so that it can safely navigate complex environments."

blog.waymo.com

Introducing the 5th-generation Waymo Driver: Informed by experience, designed for scale, engineered to tackle more environments

At Waymo, we’re building the World's Most Experienced Driver - our combination of hardware, software, and compute that powers vehicles to safely get people and things where they’re going.
blog.waymo.com blog.waymo.com

So, based on that quote, I think Waymo might say something like "We think Tesla's vision-only can work great for an advanced driver assist but we don't think Tesla's vision-only approach will be good enough to reliably and safely handle all the complexities of autonomous driving, in all operating conditions."

techlogik said:
Waymo previously stated how Tesla had a slight advantage, primarily due to their huge data collection and data sampling they have to use and improve their systems with similar technologies.
Click to expand...

I know that is something that Tesla fans say a lot but I am not aware of Waymo making that statement. Do you have a link or source?
 
#993
diplomat33 said:
I think the Waymo blog on the 5th Gen hardware basically answers that question:

"In order to navigate the complexities of driving - from the biker weaving and speeding through traffic on a foggy San Francisco morning, to the family pet rushing into the street to greet you at night - the Waymo Driver uses a comprehensive view of its surroundings and a detailed understanding of its environment to accurately reason about its next move. No one type of sensor on its own can provide that level of detail in all operating conditions; therefore, we've spent over a decade developing a single integrated system comprised of complementary sensors to give our Driver this comprehensive view of the world so that it can safely navigate complex environments."

blog.waymo.com

Introducing the 5th-generation Waymo Driver: Informed by experience, designed for scale, engineered to tackle more environments

At Waymo, we’re building the World's Most Experienced Driver - our combination of hardware, software, and compute that powers vehicles to safely get people and things where they’re going.
blog.waymo.com blog.waymo.com

So, based on that quote, I think Waymo might say something like "We think Tesla's vision-only can work great for an advanced driver assist but we don't think Tesla's vision-only approach will be good enough to reliably and safely handle all the complexities of autonomous driving, in all operating conditions."



I know that is something that Tesla fans say a lot but I am not aware of Waymo making that statement. Do you have a link or source?
Click to expand...
www.google.com

Inside the lab where Waymo is building the brains for its driverless cars

The Google spinoff has a head start in AI, but can they maintain the lead?
www.google.com www.google.com

The article is 3yrs ago and much has changed. They try to walk the massive data collection advantage somewhat in the article, which has only gained more by Tesla due to producing so many cars to collect info. But then say it is diminishing returns having more data. But the latest statement from the Waymo CEO talking about Tesla’s pipe dream and nonsense about magical driving cars all of the sudden I agree with. It doesn’t work like that. Tesla have not shown or proven much and now went backwards again.
 
  • Like
Reactions: diplomat33
#994
techlogik said:
www.google.com

Inside the lab where Waymo is building the brains for its driverless cars

The Google spinoff has a head start in AI, but can they maintain the lead?
www.google.com www.google.com

The article is 3yrs ago and much has changed. They try to walk the massive data collection advantage somewhat in the article, which has only gained more by Tesla due to producing so many cars to collect info. But then say it is diminishing returns having more data. But the latest statement from the Waymo CEO talking about Tesla’s pipe dream and nonsense about magical driving cars all of the sudden I agree with. It doesn’t work like that. Tesla have not shown or proven much and now went backwards again.
Click to expand...

I see you are a new member to this forum. Welcome to TMC!! Thanks for sharing your thoughts.
 
#995
diplomat33 said:
I think the Waymo blog on the 5th Gen hardware basically answers that question:

"No one type of sensor on its own can provide that level of detail in all operating conditions; therefore, we've spent over a decade developing a single integrated system comprised of complementary sensors to give our Driver this comprehensive view of the world so that it can safely navigate complex environments."
Click to expand...
Traffic cones excepted. I can see the diplomat's down vote. Facts matter, dude.
 
  • Love
  • Disagree
Reactions: mikes_fsd and diplomat33
You must log in or register to reply here.

Similar threads

diplomat33
  • Article
Karpathy: Self-driving as a case study for AGI
Replies
20
Views
2K
AI, Autopilot, & Autonomous/FSD
Tony Y
T
diplomat33
  • Article
Glossary for AI, Machine Learning and Autonomous Driving
Replies
4
Views
2K
AI, Autopilot, & Autonomous/FSD
Larry Hutchinson
L
diplomat33
Post your autonomous driving predictions for 2023!
Replies
23
Views
3K
AI, Autopilot, & Autonomous/FSD
ThomasD
T
I
6,000+ miles of continental autopilot use….
Replies
59
Views
4K
AI, Autopilot, & Autonomous/FSD
Supcom
S
T
  • Article
FSD reflections after 3mo of use
Replies
30
Views
6K
AI, Autopilot, & Autonomous/FSD
mgs333
mgs333
Top
Back
Blog
New
Forum list
Marketplace
Menu