[persistent]

Waymo is so far ahead in this game. Tesla has the biggest online cheerleading section and an "autopilot" that behaves as a low-budget adaptive cruise control with lane keeping that sometimes works, and a pretty big body count. Uber has a dead pedestrian and a major lawsuit. Cruise has a lot of people on staff but virtually nothing to show for it. Lyft exists. Waymo is in production revenue service.

[grecy]

> Tesla has the biggest online cheerleading section and an "autopilot" that behaves as a low-budget adaptive cruise control with lane keeping that sometimes works, and a pretty big body count

I don't think you're being very accurate there. There are lots of videos of people commuting 50+ miles to and from work every day without touching anything while on the interstates, including interchanges. It's doing a lot more than just lane keeping.

[mdorazio]

I disagree. It’s completely fair. Handling basic tasks on nice freeways is exactly what adaptive cruise control + lane keeping is for. Last I checked, autopilot couldn’t even stop for a stop sign unless you count “shadow mode” claims.

[grecy]

What other vehicle will change lanes by itself, and take on/off ramps through interchanges to take you where your navigation is routed, without you touching a single thing?

[new_realist]

You have to keep touching the wheel or it will disengage. It sounds like you’ve never used it.

[grecy]

OK, sure. You have to touch the wheel or it turns off.

What I mean is, you are not providing input to the wheel, or pedals, or turn signal or driving the car in any way. You're just letting it know you have not fallen asleep.

And, yes, thanks for asking, I have used it.

[mdorazio]

None, because every other manufacturer is more concerned about not killing people than they are about releasing beta features for cars. The most surprising thing to me about autopilot is that Tesla hasn't been sued into oblivion for it.

[alasdair_]

>Last I checked, autopilot couldn’t even stop for a stop sign unless you count “shadow mode” claims.

Perhaps you should check more frequently? Here is a video of a Tesla on Autopilot stopping for a stop sign:

https://electrek.co/2019/05/29/tesla-autopilot-stop-sign-mak...

[danso]

Based on the description in that article, it doesn’t seem that the Tesla autonomously reacted to the stop sign, so much as it slowed to a stop before making a right turn based on its navigation system. The driver says he had to manually start AutoPilot after the stop for it to initiate the right turn:

> “The car always stops at the end of a highway exit since Navigate on Autopilot. At the stop, you just need to touch the accelerator to start Autopilot again without Navigate on Autopilot. I didn’t accelerate, you just need to touch the accelerator for less than a second and the car will do the rest at the intersection.”

[mdorazio]

Perhaps you should confirm what it's actually doing in the video? Tesla itself says that stop sign and traffic light handling is in shadow mode only [1].

[1] https://www.teslarati.com/tesla-autopilot-fsd-traffic-light-...

[xvector]

I have never seen another TACC implementation that doesn’t ping from side to side.

Which cars have you drive with a superior cruise control implementation?

[persistent]

Tesla will "ping" you straight into a concrete wall at full speed.

https://www.youtube.com/watch?v=5z8v9he74po

[shiftpgdn]

It's trying to center in an extremely large lane. An easy fix would be to not have a huge concrete wall jutting into the middle of the freeway without lane markers.

I feel like people forget human drivers hit those same barriers with alarming frequency.

[gundmc]

The lane markings looked pretty clear in this instance.

That being said, Tesla's cruise control capabilities are best in class, but it's misleading marketing-speak to call it true self-driving.

[chris11]

Comma ai's OpenPilot makes freeway driving really nice. I've gone miles without touching pedals or the wheel, it's definitely better than stock. Though I haven't tried AP on Tesla, and OpenPilot is basically just a really good LKA and ACC.

[mdorazio]

Cadillac super cruise worked pretty damn well when I used it in a rental last month.

[paxys]

Freeways are a very ideal environment - wide, clearly marked lanes, no traffic lights, no pedestrians/bikes, no random stops, no navigation ambiguity. Unless Tesla has self driving cars on city roads (which all of the others so), they can't really be part of the same conversation.

[dehrmann]

I recently drove on the other side of the road for the first time. Staying in the same lane on a freeway was the easiest part of it.

[persistent]

Sure but specifically the staying in its lane feature doesn't work.

[xvector]

While Tesla is far behind Waymo and probably never will catch up, especially because of CAPTCHA, I think your description is unduly harsh. Autopilot works much better than every other TACC out there and is the best thing on the market you can get.

[carlivar]

Since it lets you fall asleep as long as your hand (or something) is on the wheel, I'm not sure it's the best thing you can get. Cadillac SuperCruise seems better from a driver awareness standpoint, given that this is "predictable abuse" otherwise.

[mlindner]

It doesn't let you fall asleep. It needs force on the wheel to prevent disabling itself, and just a weight attached isn't enough. It needs a change in force.

Honestly, if your only critiques are based on hearsay then maybe you shouldn't critique.

[xvector]

I am not sure what you are talking about. Tesla autopilot regularly makes you either tug at the steering wheel or press one of the buttons on the steering wheel. Having your hand on it isn’t enough.

[revscat]

> Waymo is in production revenue service.

I’m going to ignore the fact that you’re flatly incorrect about the rest of your posts regarding Tesla, since others have more ably corrected you than I could, and simply ask what the hell this even means.

A “production revenue service”? This sounds like meaningless babble. Waymo does not currently have anything available that you can buy, lease, subscribe to, or otherwise exchange money for. That is, in fact, the entire point of the article: “coming soon,” they say, with an additional clause on top of even that: “coming soon you may...”

So: what do you mean when you say “Waymo is in production revenue service”? Because that sounds like nonsense, to be honest.

[losvedir]

Nah man. They have a ride hailing app and seevice in Arizona. The email was to customers of this "production revenue service" that some rides soon will not have a safety driver in the car.

[eloff]

I used to think that, but now I agree with George Hotz. Tesla uses machine learning to map inputs to control outputs. Google is using it just to build a model of the world and then use program logic to decide the control action to take.

The problem with the latter approach is the long tail of real life variety doesn't lend itself to a fixed set of programmed rules. You'll always have more edge cases.

I think Waymos approach gets you to 90% faster, but then plateaus.

[jdashg]

I think it's the other way around: Direct mapping of inputs to outputs will work great for a subset of conditions, but the last N% is infeasible without explicit logical programming.

[eloff]

You'll never run out of edge cases if you have to explicitly program the logic. At some point you have to let the machine learning decide it. Maybe there is room for a hybrid approach, but I think anything relying on programmers having thought of every situation ahead of time is doomed to fail.

[jdashg]

I believe relying on machine learning is doomed to fail because I don't think it can feasibly offer the robust safety guarantees required.

I honestly don't expect go-anywhere L5 in the next decade at least, and I definitely expect an engineered approach to hit L4 first.

[eloff]

You're wrong about the requirements. It doesn't have to be perfect, just better than us, and that's easily doable for computers that don't get tired, distracted, or drunk.

I'm not sure how long it will take, but I'm sure we'll get there.

[persistent]

Not sure there is enough evidence in support of those claims about Tesla. Tesla doesn't collect anywhere near enough data from its customer fleet to support the kind of massive training being claimed. In contrast Waymo and Cruise vehicles are regularly in the depot where each car offloads terabytes of data. Who has the bigger training set?

[revscat]

Tesla routinely gets data from the fleet and analyzes it. Here is Andre Karpathy speaking on this subject:

https://youtu.be/Ucp0TTmvqOE

1:49:30. I’m on mobile otherwise I would link to the exact spot.

Tesla has tens (hundreds?) of thousands of vehicles on the road right now that are continuously feeding training data back to Tesla. What evidence is there that this is insufficient, or even deficient? From what I can gather, Waymo has limited testing geofenced to Phoenix, Arizona. Tesla has cars all over the world with autopilot, all providing training data.

I don’t see a reason for your conviction that Waymo has an advantage here.

[losvedir]

Unless Teslas are uploading gigabytes every night over your home WiFi (do they?) they're limited by a 4G connection.

So you have a lot more Teslas sending a lot less data each. I'm not sure which company has the richer training set.

[jodrellblank]

they're limited by a 4G connection

Are they?

- they could store data to be collected in bulk when the car is in the garage for maintenance or servicing or repair.

- Superchargers could include fast data transfer in their cable for cars using them.

- Superchargers could include a wireless data collection from cars sopped at or near them.

- Cars could connect to public wifi to upload data at any car park, any place, any time.

- Tesla could have vehicles driving around, like Google mapping cars, collecting data wirelessly from nearby Teslas.

I don't know if they do any of them, but none are unreasonable things for companies of 2019 to think about doing.

[tigershark]

Hundreds of thousands. The 200k mark only in US was achieved more than 1 year ago, and it triggered the phase out of the incentives. For the last couple of quarters Tesla has been delivering almost 100k vehicles so it won’t be that long to reach the 1 million mark globally.

[xvector]

On top of this, Waymo has the full force of Google’s CAPTCHA behind it.

It is already full self driving in some cities, whereas my Tesla cannot even detect stop signs.

I love my Tesla but I harbor no illusions about the eventual victor of this self-driving race.

[macspoofing]

>Tesla uses machine learning to map inputs to control outputs. Google is using it just to build a model of the world and then use program logic to decide the control action to take.

I suspect that isn't true in so far that engineers will build what works. Building a complicated system like an autonomous car will necessarily involve a messy combination of a bunch of different strategies and with the result converging on a common solution.

[eloff]

I don't know if it's still true, but that was the state of things as of a year ago.

[danso]

That said, seems like Tesla will still benefit from Waymo being the first out of the gate. While Tesla plays catchup in technology, Waymo will take all the early criticism, while ostensibly making driverless cars more palatable to mainstream society. Similar to how Lyft benefitted from Uber being the first to bully local governments into legalizing non-taxi app-driven car service.