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.
There’s no way this will work; lidar can’t handle rain. This is the end for Waymo and they will revert to vision only.

Are you being facetious or serious?

Lidar does work in the rain, although the range is reduced:


And cameras have less range in rain, just like lidar. So vision-only would not solve the problem.

That is why Waymo uses sensor fusion of cameras, lidar and lidar. Radar is not degraded at all in rain or fog. So the fusion of radar will compensate for any loss by the cameras and lidar. Waymo is obviously testing their sensor fusion to make sure the perception can maintain high reliability even in adverse weather conditions.

And "revert" implies they once did vision-only. I don't think Waymo ever did vision-only. AFAIK, Waymo always did some form of sensor fusion.
 
  • Funny
Reactions: AlanSubie4Life
Waymo has spent 15 years with their lidars and sensors, and now, they're still trying to figure out how everything works when it's raining? Is this bizarro world?
Just spitballing here, but they probably realize hundreds of thousands of collision-free miles in the rain with their sensor suite is not enough for proper validation.
 
Waymo has spent 15 years with their lidars and sensors, and now, they're still trying to figure out how everything works when it's raining? Is this bizarro world?

No, they are not trying to "figure out how everything works when it rains". They are validating their 5th Gen sensors. When you have a new sensors and new software, you need to test to see how it performs and make sure it is 99.9999% reliable in ALL conditions. Are you trolling or just that clueless?
 
  • Like
Reactions: AlanSubie4Life
Waymo has spent 15 years with their lidars and sensors, and now, they're still trying to figure out how everything works when it's raining? Is this bizarro world?

Also the 5th gen sensors waymo uses now are completely different from the sensors they used 15 years ago. When you use completely new sensors, you need to revalidate. And the software is radically new from 15 years ago. That requires testing too.
 
Also the 5th gen sensors waymo uses now are completely different from the sensors they used 15 years ago. When you use completely new sensors, you need to revalidate. And the software is radically new from 15 years ago. That requires testing too.

Aren't those the waymos in SF?

Don't they have water in labs? Don't they have engineering studies? 6 months to drive around in the rain to figure out what water does? For real?
 
What is there to "understand" about weather or water? And why does it take 6 months? Waymo is comedy sometimes.
Aren't those the waymos in SF?

Don't they have water in labs? Don't they have engineering studies? 6 months to drive around in the rain to figure out what water does? For real?
FSD Beta 10.69.1.1 in light to moderate rain.

is this the comedy you speak of? what happened to trillions of miles of rain data?

 
Aren't those the waymos in SF?

Don't they have water in labs? Don't they have engineering studies? 6 months to drive around in the rain to figure out what water does? For real?
Being such a huge proponent of Tesla's approach I'm surprised that you think they can verify performance in a lab.
Tesla brags about having 100,000 testers. Does it really require 100,000 testers to make FSD work? For real?
 
What? We're talking about how water affects a sensor.

No, that is not what we are talking about at all. We are talking about how rain affects the overall reliability of the autonomous driving. That could be the rain reducing visibility, rain drops on the camera lens, water on the road that makes the road slippery, water puddles that could confuse the cameras, rain might affect the driving of other vehicles on the road which affects your behavior prediction etc... That is why you need lots of real world testing because the best way to make sure your autonomous driving can handle rain or other adverse weather conditions, is to drive a lot and see if it can still drive safely. When Waymo completes millions of miles in the rain and they see the Waymo Driver can do it safely without a safety driver, then they will have the data that they can deploy driverless in rain and it will be safe enough.
 
Last edited: