Why Tesla will win the Autonomous Driving race

[Alketi]

I think if you go based on pure projections of what we (think we) know --

Model 3 will ship with hardware capable of L4. That means S/X will get that hardware sooner. The fleet will move quickly from L2 to L2.5 to L3 autopilot, acquiring mileage and fleet learning as it goes.

A likely middle-of-the-road bet for L4 autopilot with regulatory approval from Tesla is 2019 +/- 1 year.

Ford and others are throwing around a 2021 date for their own efforts. Though it's not entirely clear how they get there without the intermediate steps and billions of miles of safety data.

I would conclude, with a lot of hand-waving, that Tesla has perhaps a 2 year lead on the industry. This may sound like a lot or a little depending on your perspective, but to me it's another indication that the Model 3 and Y will sell in whatever quantities Tesla can manufacture.

 

[EnzoXYZ]

I think if you go based on pure projections of what we (think we) know --

I would conclude, with a lot of hand-waving, that Tesla has perhaps a 2 year lead on the industry. This may sound like a lot or a little depending on your perspective, but to me it's another indication that the Model 3 and Y will sell in whatever quantities Tesla can manufacture.

Couldn't agree more with this! I don't think it's just the autonomous capability but everything else a TSLA is.

 

[cluster]

Any comment on the Carnegie Mellon University's self-driving car? Looks very impressive. It does a lot of things Tesla AP cannot.

 

[GoTslaGo]

Any comment on the Carnegie Mellon University's self-driving car? Looks very impressive. It does a lot of things Tesla AP cannot.

Very impressive!

Sept 2013 as well. Really impressed by how the car handles local traffic early on (stops signs, etc.. did he run a stop sign early on after the previous car turned? Not really a full stop... ). Only one or two times where the driver grabs the wheel. Not sure if he's actually doing anything, since the entire time he has his right hand on the "button" in the center which I assume is some kind of take-over switch.

For anyone who wants to watch, I would suggest running it at 2x speed. Pretty funny at around 15:45 or so, car in the middle lane is busy taking pics/videos of the car.

This would be Level 4-5 I assume. Also I assume there are loads of cameras/sensors up top, which is why it must have been obvious to passing drivers that this is a test car (that they are taking videos).

 

[CarlK]

Any comment on the Carnegie Mellon University's self-driving car? Looks very impressive. It does a lot of things Tesla AP cannot.

It is not close to Google's self driving car which does much more in various traffic conditions without even a steering wheel or pedal. I'm pretty sure Tesla has at least the same technology too just that it's not ready for a production car. Remember a university science project can work 99% of time and fails 1% of time but a real production car can't fail even 0.0001% of times. If you have not noticed it yet that big red button in the middle should give you a clue of state of technology in that car.

 

[schonelucht]

I'm pretty sure Tesla has at least the same technology too just that it's not ready for a production car.

Why are you pretty sure of this?

 

[bonaire]

There is no real winner in autonomous driving.

Puts millions of taxi drivers out of work. Puts over the road truckers out of work as well. Already putting fork lift workers out inside of automated warehouses. But good for profit margins of companies that possibly some who chat here may own and are looking to cut those pesky and costly workers.

 

[Spiderz]

I agree with @bonaire 100%. Uber initially won the battles by letting government know it is doing it for shared economy, everyone will have jobs using Uber. Now they want all autonomous vehicles, putting millions out of job, for what? profits, increasing poverty. I think we all should raise our voices to our senators so only 5+/- autonomous cars can be owned by a business so small business owners or a taxi driver can have income as well.

 

[Spiderz]

Coming back to the topic of this discussion, check out Volvo XC90 drive me, already on roads in sweden and will beat Tesla without a doubt. They are already using NVIDIA PX 2 , deep learning and have 6 radars, 4 cameras, 1 lidar and 12 sensors... I don't think you need more hardware than that. Volvo is beating Tesla in technology....

 

[techmaven]

Coming back to the topic of this discussion, check out Volvo XC90 drive me, already on roads in sweden and will beat Tesla without a doubt. They are already using NVIDIA PX 2 , deep learning and have 6 radars, 4 cameras, 1 lidar and 12 sensors... I don't think you need more hardware than that. Volvo is beating Tesla in technology....

Since you are talking about research and development hardware and not what is shipping, how do you know what R&D hardware Tesla is using these days?

 

[Garlan Garner]

I believe Tesla will win because of the amount of experience its allowing itself to acquire on a daily/hourly basis.

Tesla is collecting a million miles of data every 10 minutes.

Tesla has 780 million miles of driving data, and adds another million every 10 hours

 

[Turing]

Since you are talking about research and development hardware and not what is shipping, how do you know what R&D hardware Tesla is using these days?

Was about to post the same thing. The poster is comparing Volvo's R&D testing to Tesla's _____?

 

[Saghost]

There is no real winner in autonomous driving.

Puts millions of taxi drivers out of work. Puts over the road truckers out of work as well. Already putting fork lift workers out inside of automated warehouses. But good for profit margins of companies that possibly some who chat here may own and are looking to cut those pesky and costly workers.

Yup. Takes millions of people who spend their lives doing annoying dangerous tasks and frees them up to do something more useful, while reducing the delays and costs and risks for everyone. Sounds horrible.

(There are specialized cases where it seems unfortunate, like folks who spent literally a million dollars on a NYC taxi medallion that will presumably be worth a lot less soon.)

 

[Turing]

There is no real winner in autonomous driving.

Puts millions of taxi drivers out of work. Puts over the road truckers out of work as well. Already putting fork lift workers out inside of automated warehouses. But good for profit margins of companies that possibly some who chat here may own and are looking to cut those pesky and costly workers.

No real winners, are you kidding? How about the 10,000's of people who needlessly die every year in human error caused auto accidents? How about the millions of hours wasted annually in mind numbing traffic related mostly to human driver error? Not to mention there is quite a lot of research showing how much better off lower income people are with good and cheap transportion, something that is severely lacking in the US but which could be mostly solved with the application of self driving technology bringing down the cost of transportation, making it not a significant financial barrier for those stuck in poverty in many places wrt to owning and maintaining a car.

 

[CliffG]

Uber/Lyft again will have to license Tesla's superior system or go under.

Can't say about Lyft, but Uber's already there:
Uber self-driving cars in Pittsburgh this month.

And, to respond to other posts in this thread, note that Uber got there by absorbing many from the Carnegie-Mellon AI and Robotics faculty (cf. post #23). Not sure about the relationship to the NREC, which is right down the street from Uber's ATC. And as for post #29, it looks like these will be those XC90s. Not sure if they'll have the superstructure or not (the cars in the Bloomberg video look different from those on the Uber site).

 

[Turing]

Can't say about Lyft, but Uber's already there:
Uber self-driving cars in Pittsburgh this month.

And, to respond to other posts in this thread, note that Uber got there by absorbing many from the Carnegie-Mellon AI and Robotics faculty. Not sure about the relationship to the NREC, which is right down the street from Uber's ATC.

Already there with what? a driver assist system they are aiming to be self driving by 2021 per your article?

 

[CliffG]

Already there with what? a driver assist system they are aiming to be self driving by 2021 per your article?

From what I've read and been told, it's a self-driving vehicle, with a fallback driver and observer. These are admittedly still working out the kinks, but then so is Tesla. These are aiming to be as self-driving as one of our cars (no, mine is a Classic) now. We'll see.

There are more and more participants in this effort, and more emphasis (and funding!) is coming. Much as I'm cheering on Tesla and stocking up on the koolaid, I think this is a case of the more the merrier.

I suppose my initial impulse was to point out that Uber, for one, wasn't waiting for Tesla's technology. I think this is progressing further and faster that most realize. Interesting times.

 

[EnzoXYZ]

The people saying automomous technology is detrimental to society is the same people championing for car stealerships.

 

[Turing]

From what I've read and been told, it's a self-driving vehicle, with a fallback driver and observer. These are admittedly still working out the kinks, but then so is Tesla. These are aiming to be as self-driving as one of our cars (no, mine is a Classic) now. We'll see.

If it needs a driver to oversee the car, as it does, then it is by definition a driver assist system and not a self-driving car. Google's car with no steering wheel and no driver, are self driving cars.

There are more and more participants in this effort, and more emphasis (and funding!) is coming. Much as I'm cheering on Tesla and stocking up on the koolaid, I think this is a case of the more the better.

Sure, the quicker we get to self driving cars the better imo. I'm very much looking forward to my baby daughter growing up and never being driven by other teenage drivers.

 

[Chopr147]

I believe Tesla will win because of the amount of experience its allowing itself to acquire on a daily/hourly basis.

Tesla is collecting a million miles of data every 10 minutes.

Tesla has 780 million miles of driving data, and adds another million every 10 hours

I really don't see how any other car company can compete with a billion miles of AP driving. (Elon's beta target) Volvo is doing well but they test in house 1 car at a time. And as some other posters mentioned , some of the big guys do not even seem to be trying. Tesla is small enough right now, and tech savvy enough, and free from "old ways" to do it their own way. There is no leash on Tesla like the big guys.