You'd be surprised - different cities have different surcharges for transportation services, you might have origin/destination-based charges e.g. airport fees, etc. Of course geofencing + operating in a small set of cities keeps it simpler but complexity is inevitable.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Autonomous Car Progress
- Thread starter stopcrazypp
- Start date
-
- Tags
- Autonomous Vehicles
Aren't you assuming that Cruise suddenly discovered all this and hasn't been working on it for the past year or more?
powertoold
Active Member
I find it funny that people are still discussing the Cruise and Waymo approaches after seeing Tesla's progress to 10.12
The Cruise and Waymo business models are a clear dead end. Big, diverse, and real data are a prerequisite.
The Cruise and Waymo business models are a clear dead end. Big, diverse, and real data are a prerequisite.
Daniel in SD
(supervised)
I prefer first principles thinking. Most humans learn to drive in one place before expanding to other areas.
diplomat33
Average guy who loves autonomous vehicles
We keep discussing Cruise and Waymo because they have safe and reliable driverless robotaxis that the public can use. When you are deploying safe and reliable driverless robotaxis, there is probably something working with your approach. So it makes sense to talk about them in a thread about AV progress. Regardless of how much progress you think Tesla is making with 10.12, the fact remains that Tesla has not deployed any robotaxis. When Tesla starts testing robotaxis or starts deploying driverless robotaxis, then we can talk. Of course, we should discuss Tesla's progress with 10.12 but we should not ignore the AV progress that Cruise and Waymo are making.
Cruise is starting to charge for their driverless rides and Waymo is both charging for their driverless rides are expanding their robotaxi business quickly. Generally, when you are able to charge money and expand your business, it is not a dead end.
EVNow
Well-Known Member
Thats not first principles. That is just a social observation.
I learnt to drive in India and got a DL there. But first time I drove a car alone was in Seattle
Terminator857
Active Member
Will be interesting when Waymo doesn't lose billions each year. Not familiar with Cruise losses, but I'm sure it is substantial.
diplomat33
Average guy who loves autonomous vehicles
How many billions did Tesla lose when they started? How many years of losing billions before Tesla started making a profit? Tesla almost went bankrupt during the Model 3 ramp up. It is common for businesses to lose money in the beginning while they work to scale up the business. Robotaxis are a completely new business. Autonomous driving is a new technology. It is not finished yet. So yes, I expect Waymo and others to lose billions in the beginning, probably for a few years to come. That is not a sign of failure, it is the price of developing a new technology. But eventually, Waymo will absolutely be profitable, when the tech is more mature and the costs come down, and they have thousands of robotaxis running 24/7 in multiple cities. It will happen. What we are seeing now is just the beginning of the ramp up!
By the way, I am way more bullish on Waymo than I am on Cruise. Waymo has better tech and is showing they can scale faster than Cruise. Waymo testing driverless in SF and downtown Phoenix, including the airport, with plans to connect the downtown area with the Chandler area, not to mention the recent announcement of their partnership with Uber Freight, makes me more confident in Waymo's ability to scale and commercialize their autonomous driving. Cruise only has a very limited driverless service in SF with some autonomous deliveries (with safety drivers) in Scottsdale, AZ. Cruise has not shown an ability to scale yet.
Last edited:
powertoold
Active Member
To those who are paying attention, Tesla has already surpassed Waymo and Cruise in computer-controlled autonomous driving. Which is the whole point of autonomous driving progress, right?
The only "benefit" Waymo and Cruise have is lots of investment in remote *human* monitoring capabilities. They've built lots of software and hardware to have remote human decision makers. Any deviation from the intended path (while on laned roads, not in parking lots) will likely require remote human approval. Waymo and Cruise cars have no real "intelligence" about passing double parked cars (among so many other situations). They simply don't have the breadth of data to deal with these cases, as Tesla does.
I've said this before, but Tesla can already surpass Waymo and Cruise driving performance (minus remote human decision making) if they simply focused on and overfit the training set to an area of their choice (like Chandler for example). That's not Tesla's goal however. They would rather solve the whole problem at once, which is the best business approach.
The only "benefit" Waymo and Cruise have is lots of investment in remote *human* monitoring capabilities. They've built lots of software and hardware to have remote human decision makers. Any deviation from the intended path (while on laned roads, not in parking lots) will likely require remote human approval. Waymo and Cruise cars have no real "intelligence" about passing double parked cars (among so many other situations). They simply don't have the breadth of data to deal with these cases, as Tesla does.
I've said this before, but Tesla can already surpass Waymo and Cruise driving performance (minus remote human decision making) if they simply focused on and overfit the training set to an area of their choice (like Chandler for example). That's not Tesla's goal however. They would rather solve the whole problem at once, which is the best business approach.
diplomat33
Average guy who loves autonomous vehicles
This is the most outrageous BS I've ever seen. Thanks for the laugh.
Tesla's FSD beta is years behind Waymo. FSD Beta 10.12 can't even do unprotected left turns! If you really think 10.12 is better than the 5th Gen Waymo Driver, then you are totally clueless. The Waymo 5th Gen software has better NN, better perception, better prediction, better planning than tesla. It is far more reliable than what Tesla has. It is not even close. And Waymo's ML is way ahead of Tesla. Just look at the new ML papers and patents Waymo is putting out every year compared to Tesla.
First, Elon admitted that they overfit to the Bay Area. So tesla is already overfitting and they still can't do geofenced robotaxis. If Tesla could do geofenced robotaxis, then they should do it now. They could prove everybody wrong, corner the robotaxi market and put Waymo out of business. Heck, just do 1 demo of a driverless ride! Tesla does not do geofenced robotaxis because they can't! FSD Beta is simply not good enough. It can do some simple drives with zero interventions and it is getting better but it is nowhere near the reliability to actually remove the driver, not even in a geofenced area like Chandler. Just look at Chuck's unprotected left turn and you will see that Tesla is struggling to solve FSD. "Solving the whole problem at once" is just an excuse to cover up the fact that Tesla is years behind.
diplomat33
Average guy who loves autonomous vehicles
This is total BS. You are just flat out lying. Waymo Driver uses sophisticated AI NN that handle ALL perception, prediction and planning. And Waymo Driver handles double parked cars and more complicated scenarios completely on their own without any remote help. In fact, Waymo Driver has much better AI than Tesla.
Please stop lying!
powertoold
Active Member
Blah blah blah, another year, same old story about Waymo Gen XYZ:
I watch Waymo ride videos from time to time, and they're still driving through the most boring and easy to drive areas (with or without safety drivers), avoiding anything approaching the easiest fsd beta drives.
15 years of development and the best sensors still trip up at any signs of cones.
I watch Waymo ride videos from time to time, and they're still driving through the most boring and easy to drive areas (with or without safety drivers), avoiding anything approaching the easiest fsd beta drives.
15 years of development and the best sensors still trip up at any signs of cones.
diplomat33
Average guy who loves autonomous vehicles
powertoold
Active Member
Yup, Waymo's AI is actually a human who isn't in the car, lol!
diplomat33
Average guy who loves autonomous vehicles
Stop your silly trolling. It is a fact that Waymo has sophisticated AI NN in the car that makes all the driving decisions.
diplomat33
Average guy who loves autonomous vehicles
Again, this is a lie. The 5th Gen does not get tripped by "any sign of cones".
powertoold
Active Member
diplomat33
Average guy who loves autonomous vehicles
Yes you are spreading lies. You are spreading the same old debunked talking points.
I am stating established facts. It is a fact that Waymo has sophisticated AI NN in the car that makes all the driving decisions.
Terminator857
Active Member
Tesla reported its first profitable quarter in 2013. Waymo has never been semi profitable and there is no end in site to losing billions per year. Hopefully they can turn that around, but we've been saying that for several years. Suspect trucking business will be first to be profitable. Truck seems easier because can be profitable just working on limited access freeways.
Daniel in SD
(supervised)
Can you quantify Tesla's performance?
The whole point of autonomous driving is to automate the task of driving. If you still require one attentive licensed driver per vehicle you haven't really accomplished anything useful (yet). Having many AV's assisted by a single human is an improvement in efficiency. Think of it like factory automation.
Similar threads
- Article
