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.
I’m not saying that they hit a bottleneck. I’m saying that their bottleneck has always been there. Waymo has a huge challenge when it comes to validating their software updates in real world conditions. For example, a minor change in the driving policy may work well in Chandler Arizona, but it presents a new edge case in San Francisco. Without a fleet that is constantly encountering edge cases, it’s difficult to know how your changes affect one driving locality vs another.

That is helpful. But just to be clear. Waymo has a fleet. It is smaller than Tesla's fleet but they have a fleet. The question is whether the fleet is big enough or not. We can debate how big the fleet should be but they do have a fleet of autonomous cars constantly encountering edge cases and helping to train the software. Waymo has test cars in California, Arizona, Washington State, Texas, Florida, Michigan etc... And Waymo has greatly expanded their testing in SF.
 
Last edited:
Of course, we don't know what Tesla's disengagement rate is since they won't release it.

I guess it's technically 0 since Tesla only has a level 2 system.

Honestly seems like Waymo is at least 5 years ahead of Tesla. The major difference being that Tesla is releasing what they have as a beta to customers whereas Waymo will keep building until they figure everything out.
 
  • Like
Reactions: diplomat33
I guess it's technically 0 since Tesla only has a level 2 system.

Honestly seems like Waymo is at least 5 years ahead of Tesla. The major difference being that Tesla is releasing what they have as a beta to customers whereas Waymo will keep building until they figure everything out.

Tesla reported 0 autonomous miles. So Tesla's disengagement rate would be 0/0 which is mathematically undefined. So I think "N/A" would be more accurate than "0". "0" is an actual mathematical number that does not correspond to what Tesla reported.
 
Yesterday, Waymo released an report on how their system would have handled past fatal crashes by human drivers. Waymo looked at every fatal car crash between 2008 and 2017 that occurred in the Chandler area where Waymo has robotaxis. They reconstructed those accidents in their simulations to see how the Waymo Driver would have handled them. The results are quite encouraging IMO and show that Waymo Driver would significantly reduce accidents and fatalities if every car was replaced with a Waymo.

Key Summary:

"In total, the simulated Waymo Driver completely avoided or mitigated 100% of crashes aside from the crashes in which it was struck from behind, including every instance that involved a pedestrian or cyclist."

"We started by simulating 72 fatal crashes as they occurred on public roads in our operating domain, which covers thousands of miles of road in southeast Phoenix. Since many of these crashes involved two vehicles, we ran separate experiments simulating the Waymo Driver in the role of each vehicle—first replacing the vehicle that initiated the crash (the “initiator”) with the Waymo Driver, and then replacing the vehicle that responded to the other vehicle’s actions (the “responder”) with the Waymo Driver. That left us with 91 simulations in total.

"When we swapped in the Waymo Driver as the simulated initiator (52 simulations), it avoided every crash by consistent, competent driving, and obeying the rules of the road—yielding appropriately to traffic, executing proper gap selection, and observing traffic signals."

"When the Waymo Driver was placed in the responder role, it completely avoided 82% of simulated crashes. In fact, in the vast majority of events, it did so with smooth, consistent driving—without the need to brake hard or make an urgent evasive response."

"In another 10% of scenarios when the simulated Waymo Driver was the responder—all at an intersection when another vehicle turned across its path— it took action that mitigated the severity of the crash."

"Only 8% of responder crash simulations were unchanged. In all of these, the human-driven vehicle struck the rear of the simulated Waymo Driver when it was either stopped or traveling at a constant speed, giving the Waymo Driver little opportunity to respond."

"In other words, even when a human driver did something to initiate a crash, such as running a red light, the simulated Waymo Driver avoided or mitigated the vast majority of these fatal crashes."

It's not very long. You can the entire blog that discusses the key results here: Waypoint - The official Waymo blog: Replaying real life: how the Waymo Driver avoids fatal human crashes
 
Last edited:
Waymo has made 2 improvements to their ride-hailing to improve the customer experience:

They added the ability to do multiple stops in the same trip:

"The Waymo app now has an "add a stop" option. Riders using the Waymo One autonomous ride service can pick up a friend or some groceries without having to reorder a car at each stop.

You can make a short stop, like dropping off your sister, or a long stop, like picking up a package at FedEx. For a longer stop, Waymo cars — which are all modified Chrysler Pacifica minivans, but will soon include electric Jaguar I-Pace SUVs —will wait nearby and pick you back up once you're ready to head to the next location.

Waymo product manager Sam Kansara said in a recent call "the meter is not running while on your long stop." In a traditional taxi cab, you're charged for the time the driver spends idling while you get your coffee. Waymo, however, shows you the cost of the trip with all your stops before you get into the car."

Waymo is also ditching Google Play Music and let's users play music from their phones through the car's speaker system:

"In another attempt to improve the rider experience, Waymo is officially ditching its Google Play Music playlists for the ability to connect riders' phones to the computer screens in the self-driving cars.

The new music system works best for Android users. On Android devices, you can connect with Google Cast and play any audio apps like Spotify, Google Podcasts, or YouTube Music. For iPhone and other devices, you need to have the Google Assistant app, which connects to the in-car audio system. Then you are plugged into your own playlists or audiobooks you were listening to before the car arrived."

 
That is awesome, but instead of actually expanding the service area (the #1 limiting factor of their system) they keep highlighting mouse farts...

And it seems like they keep shrinking the usable service area as they still follow the "don't fix it, restrict it" method. Or as JJRicks says: "Don't fix the problem apply duct tape to it." Avoiding that turns takes another 5-10 minutes. (Which it started doing after there was a disengagement at that turn before.)


It seems that every time he gets the Waymo driver to have a problem they remove that turn or drop off/pickup spot.

Hopefully the restrictions are just temporary while they make improvements, but I haven't seen evidence of that yet.
 
  • Funny
Reactions: mikes_fsd
That is awesome, but instead of actually expanding the service area (the #1 limiting factor of their system) they keep highlighting mouse farts...

Waymo is working on expanding the service area and they are actively testing their 5th Gen i-Pace robotaxis. You know they can work on both things at the same time, right? And these are not "mouse farts", they are quality of life improvements that will enhance the customer experience.

And I don't see you complaining when Tesla releases an actual fart update.
 
  • Disagree
Reactions: mikes_fsd
And it seems like they keep shrinking the usable service area as they still follow the "don't fix it, restrict it" method. Or as JJRicks says: "Don't fix the problem apply duct tape to it." Avoiding that turns takes another 5-10 minutes. (Which it started doing after there was a disengagement at that turn before.)


It seems that every time he gets the Waymo driver to have a problem they remove that turn or drop off/pickup spot.

Hopefully the restrictions are just temporary while they make improvements, but I haven't seen evidence of that yet.

Waymo is not shrinking the overall service area. But yes, they do sometimes reroute. Since the cars have no driver to intervene in a problem, Waymo prefers to take the safest route rather than the quickest route. And yes it is temporary while Waymo improves the software.

Also, those are the older 4th gen cars. You will see much better improvements when Waymo rolls out the 5th Gen i-Pace to the public.
 
Also, those are the older 4th gen cars. You will see much better improvements when Waymo rolls out the 5th Gen i-Pace to the public.

What is going to make the 5th gen cars so much better? Did they have to add sensors? (Surely it isn't just that they can't update the software on the 4th gen, unless the new software requires more compute power than is available.)
 
What is going to make the 5th gen cars so much better? Did they have to add sensors? (Surely it isn't just that they can't update the software on the 4th gen, unless the new software requires more compute power than is available.)

My understanding is that it's a new sensor set altogether for 5th gen. The 4th gen sensors were expensive. The 5th gen has newer sensors that are able to see further and wider while being about half the cost of the 4th gen, and is made in a way that makes it easy to manufacture the car.
 
  • Like
Reactions: diplomat33
What is going to make the 5th gen cars so much better? Did they have to add sensors? (Surely it isn't just that they can't update the software on the 4th gen, unless the new software requires more compute power than is available.)

The 5th gen has better hardware. The cameras, lidar and radar are better, higher resolution, longer range etc... Also, the computer is more powerful. If you have not already done so, you can read more about the 5th gen hardware here:


Also, the 5th gen hardware is easier, simpler and cheaper to manufacture. So Waymo believes the 5th gen hardware will be more suitable for scaling up.
 
  • Informative
Reactions: Dan D.
Also, the 5th gen hardware is easier, simpler and cheaper to manufacture. So Waymo believes the 5th gen hardware will be more suitable for scaling up.
Each new generation h/w will be more suitable than the last, obviously. That said, 4th gen is not what's stopping them from scaling. If the software and the business model both work, then the hardware on the first 25-50k cars doesn't really matter.

I still say their business model doesn't work. But it's possible they're just incompetent.
 
Each new generation h/w will be more suitable than the last, obviously. That said, 4th gen is not what's stopping them from scaling. If the software and the business model both work, then the hardware on the first 25-50k cars doesn't really matter.

I still say their business model doesn't work. But it's possible they're just incompetent.

Waymo has shown it has safe and reliable L4 autonomous driving. Their disengagement rate is the best of all the companies testing in CA. They even have FSD with no driver! They are clearly not incompetent when it comes to FSD. Their software clearly works. And as for their business model, Waymo has raised billions so clearly investors believe in their business model. And Waymo is the only company with commercial driverless ride-hailing robotaxis. And the service is receiving rave reviews. Waymo has proven that their ride-hailing model works. Now they can scale it up to more cities with the 5th gen FSD. So yeah, I think their business model is working just fine.
 
  • Funny
Reactions: mikes_fsd
Just an observation...
I remember seeing a few Waymo Pacificas every single day. I've seen three of them lined up right behind each other at lights. It's been probably 6+ months seen I've seen a single one in Mountain View.

They moved them to other areas. They still have the autonomous semi trucks that they're testing around the area though.
 
Just an observation...
I remember seeing a few Waymo Pacificas every single day. I've seen three of them lined up right behind each other at lights. It's been probably 6+ months seen I've seen a single one in Mountain View.

Keep your eyes open for the Waymo i-Pace. As Waymo shifts to the 5th Gen, you should see fewer Pacificas and more i-Pace's.
 
  • Funny
Reactions: mikes_fsd
How do they plan on keeping passengers safe from each other. How do these cars handle an intersection that has a non functioning light. Can the car make command decisions. Such as there is a car accident and I need to drive over the sidewalk and across the grass to go around it. Can it take directions from emergency officials. Can the cars sense danger and avoid it? A person standing in the street shooting a gun what would the car do?
 
  • Funny
Reactions: powertoold
How do they plan on keeping passengers safe from each other. How do these cars handle an intersection that has a non functioning light. Can the car make command decisions. Such as there is a car accident and I need to drive over the sidewalk and across the grass to go around it. Can it take directions from emergency officials. Can the cars sense danger and avoid it? A person standing in the street shooting a gun what would the car do?

Some of these questions make no sense. No autonomous car is expected to protect passengers from each other or sense a person with a gun and avoid it. That is not a realistic expectation. Autonomous cars are only expected to be able to drive safely on public roads and obey traffic laws.

But Waymo cars can handle intersections with non functioning lights. We see that in the videos. I think Waymo cars can take direction from emergency officials too. I am not sure about driving on the sidewalk or grass. Waymo cars can detect an accident and they would follow the rules, like pull over, change lanes, follow cones to get around an accident.