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

Autonomous Car Progress

This site may earn commission on affiliate links.
Tesla already did, rendering the term useless. Self-driving now means you drive it yourself, lol. "Full Self Driving" means Level 2 ADAS. Since you can't go beyond "Full", a car that drives with nobody in the driver's seat needs a different term. The industry uses Autonomous. And this thread is about Autonomous Cars.

Hugely useful, since 95%* of all driving days stay within 40 miles.

True autonomy restricted to congested US metro highways would free up hundreds of man-hours per year per commuter. Billions of man-hours in total.

"Full Self Driving" is the gimmick. It frees up no man hours. It's a parlor trick.

"Be ready to take over" is Mercedes Drive Pilot, if it ever ships. Answer e-mails, read a book, etc. as long as you can respond to a takeover request within 10-15 seconds.

FSD is not "be ready to take over". FSD is "watch like a hawk or suffer decapitation". FSD frees up zero man hours. I taught four teenagers how to drive, I'm done watching like a hawk.

All else equal, multiple sensor types reduce computational load.

Way, way off. Human vision near the focal point is equivalent to a 576 megapixel camera. Tesla cameras are 1 megapixel.

*Completely made up statistic, but true :)
you kind of missed my point and just used my words to attack FSD Beta lol, i was talking about a theoritical complete full self driving package, not the FSD Beta currently deployed.

even in vision you miss the point, resolution doesn't matter so long as it can resolve the horizon. if it can resolve the horizon, it doesn't need more detail than that since there is nothing beyond that. futhermore, even if the resolution at max distance is less than the human eye, again it doesn't entirely matter, in theory a computer would be able to use context to theorize about that object and react accordingly (kind of like a human does when you see something but can't exactly tell what it is cause its so far away).... again talking about a complete FSD package, NOT FSD BETA lol.

why people insist on talking about FSD beta like its a competed feature is beyond me and frankly exahusting lol.
 
In Chandler, Waymo charges $4.99 + $0.80 per mile. We don't know the pricing in SF yet. I guess we will find out soon.
Here is what Lyft charges in Chandler, apparently. So, Lyft might be cheaper than Waymo.

I feel the only way Robotaxis make big inroads in the taxi market would be by undercutting Uber/Lyft. Not everyone is a nerd who wants to ride in driverless cars.


1646158209577.png
 
  • Informative
Reactions: scottf200
I feel the only way Robotaxis make big inroads in the taxi market would be by undercutting Uber/Lyft. Not everyone is a nerd who wants to ride in driverless cars.

There are a couple ways they can make big inroads.

Consistency of service, and pricing - With Uber/Lyft its extremely variable.
Vehicle features - They could differentiate themselves with a consistent feature set like bike racks.
Monthly Plans - Working with businesses to offer Robotaxi service to employees as a perk
Vehicles Designed to carry more people like shuttle type services where it reduces the cost per person.

Long term I do think they'll be able to compete on price because of the high cost of labor. It's a lot more expensive in the PNW now days on average than it was before the pandemic.
 
There are a couple ways they can make big inroads.

Consistency of service, and pricing - With Uber/Lyft its extremely variable.
True. But Robotaxi will also have variability of pricing. Only way to handle highly variable demand.

Vehicle features - They could differentiate themselves with a consistent feature set like bike racks.
Uber does some of that - including wheel chair accessible cars.

Monthly Plans - Working with businesses to offer Robotaxi service to employees as a perk
Uber does that. (BTW, that’s a taxable perk, making it expensive for companies and employees)

Vehicles Designed to carry more people like shuttle type services where it reduces the cost per person.
Like shared Uber ?

Long term I do think they'll be able to compete on price because of the high cost of labor. It's a lot more expensive in the PNW now days on average than it was before the pandemic.
Yes.
 
True. But Robotaxi will also have variability of pricing. Only way to handle highly variable demand.


Uber does some of that - including wheel chair accessible cars.


Uber does that. (BTW, that’s a taxable perk, making it expensive for companies and employees)


Like shared Uber ?


Yes.

Uber doesn't do any of those things well.

Dabbling in them isn't doing them well. They don't have much control over the type of vehicle so they can't have any consistency that depends on vehicle type. Plus they simply can't have a vehicle like a Zoox.

Some variability of pricing is expected, but Uber in particular is so variable it would be pointless to look at pricing tables. A recent 4.25 mile trip that took 8 minutes was $12.51 where a trip that was .70 miles that lasted 3 min was $18.89. Neither of those prices include a tip where ride share people feel pressure to do so due to low driver pay.

I wouldn't have taken the second one if I wasn't with someone that didn't want to walk in the cold.

I believe Waymo/cruise/zoox/etc will be better suited towards providing more consistent service, and pricing overall.
 
Waymo is expensive.

How so? Waymo charges $4.99 + $0.80 per mile and Lyft charges $3.50 + $0.81 per mile. So Lyft is only $1.49 per ride cheaper. That seems pretty comparable. And Lyft has additional fees life a safe ride fee of $2.65 which Waymo does not have. You also don't have to tip the Waymo. So, unless I am misreading the Lyft pricing, I don't see Waymo being super expensive compared to Lyft.
 
Waymo got their permit to charge for rides in SF with a safety driver:


For those who don't know, CA requires AV companies to get permits from the DMV to test AVs and separate permits from the Public Utilities Commission to charge money for AV rides.
Cruise also got approved at the same time with a safety driver. It is notable however, driverless commercial operations is still not approved for Cruise by CPUC yet, so we still don't have resolution on that end (plus even if CPUC approves, SFMTA may throw roadblocks given its opposition previously).
 
  • Like
Reactions: diplomat33
How so? Waymo charges $4.99 + $0.80 per mile and Lyft charges $3.50 + $0.81 per mile. So Lyft is only $1.49 per ride cheaper. That seems pretty comparable.


Except we've had numerous reports that since Waymo can't handle a lot of situations it avoids hard things like left turns, and takes significantly longer routes.

Meaning more miles you get charged for. Plus wasted time with longer rides too.

And longer pickup times if it's a "hard" route to get to you I suppose.
 
  • Helpful
Reactions: scottf200
Except we've had numerous reports that since Waymo can't handle a lot of situations it avoids hard things like left turns, and takes significantly longer routes.

I would not say it happens "a lot". Waymo avoids some unprotected left turns but it does not avoid all left turns. Also, the Chandler robotaxis are the inferior 4th Gen FSD. The 5th Gen FSD will likely not do this.

Meaning more miles you get charged for. Plus wasted time with longer rides too.

And longer pickup times if it's a "hard" route to get to you I suppose.

These issues will be fixed as the autonomous driving gets better.
 
I would not say it happens "a lot".

Do you know of any actual numbers?

I mean, I'm sure Waymo knows how many "miles out of the way" each ride takes compared to "optimal route if a human was driving" pathing, but I doubt they publish that number..... I've just seen it often mentioned in stories from folks who've actually ridden in one, enough to think it's not especially uncommon- but then I don't have #s either.

(if I did I imagine Waymo would sue to redact my posts like they did with all their accident data :p)





These issues will be fixed as the autonomous driving gets better.

Ok Elon :)
 
  • Funny
Reactions: EVNow
Except we've had numerous reports that since Waymo can't handle a lot of situations it avoids hard things like left turns, and takes significantly longer routes.

Meaning more miles you get charged for. Plus wasted time with longer rides too.

And longer pickup times if it's a "hard" route to get to you I suppose.
I would assume any robotaxi would optimize a route based on time, and safety. Things like no uncontrolled lefts that could take forever, and avoiding dangerous intersections.

There will also be a fair amount of city regulation telling the fleet company what they can, and can't do. Like I could easily see Waze style "avoid the traffic" trick not being allowed as residents want to keep traffic low.

Hopefully cities will help optimize fleet robo shuttles as we want people to travel in groups, and not alone.
 
I would assume any robotaxi would optimize a route based on time, and safety. Things like no uncontrolled lefts that could take forever, and avoiding dangerous intersections.

You would assume that, but that's not what Waymo does today based on numerous reports from riders and reporters.

For example:


Trips are generally longer and less direct than an Uber or Lyft trip. The taxis still aren't using shared turn lanes, the company said. That forces them to sometimes take roundabout routes, like three right turns instead of one left turn.

Waymo's vehicles avoid making a left turn from a busy street into his neighborhood. Mitkowski said that instead they take a slight detour into a neighborhood across the busy street so that they can cross the busy street at a traffic signal.

Waymo rider Sophia Lovasz told CNN Business that a recent trip that would usually be a 10-minute drive took her 27 minutes with Waymo. "If you had somewhere to be at a specific time, you may not want to call a Waymo," Lovasz said.
 
I believe Waymo/cruise/zoox/etc will be better suited towards providing more consistent service, and pricing overall.
I don’t see why robotaxis would want to lose out on revenue. They will charge the max they can.

BTW, there are definitely cases where some people might feel safer in a robotaxi at night than a Uber. But overall price will dictate the market share.
 
  • Helpful
Reactions: scottf200
How so? Waymo charges $4.99 + $0.80 per mile and Lyft charges $3.50 + $0.81 per mile. So Lyft is only $1.49 per ride cheaper. That seems pretty comparable. And Lyft has additional fees life a safe ride fee of $2.65 which Waymo does not have. You also don't have to tip the Waymo. So, unless I am misreading the Lyft pricing, I don't see Waymo being super expensive compared to Lyft.
They are similar … but I’d have thought robotaxis would be cheaper …. Given they don’t have to pay the driver.

Or May be the idea is for big companies to gobble up the money that supports a lot of working families now.
 
They are similar … but I’d have thought robotaxis would be cheaper …. Given they don’t have to pay the driver.

Well remember that the Waymo robotaxis in Chandler now are still "in development". Technology that is in development tends to be more expensive. The costs will come down. As the autonomous driving gets better, Waymo will need less remote assistance. And the mass scale custom robotaxis like the driverless vehicle being designed by Zeekr, those vehicles will be cheaper. Eventually, I think robotaxis will be way cheaper when they are electric, driverless all the time and doing routes efficiently with low wait times. The fact that Waymo's prices are already comparable to Lyft when Waymo is at their most expensive stage, is a good sign that Waymo will get cheaper than Lyft in the future.