Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
  • The final cut of the 9th episode of the Tesla Motors Club Podcast, featuring Chad Schwitters, the former president of Plug In America, is now available. You can watch it now on YouTube or listen to it on all major podcast networks.

Is Tesla set for a massive expense with autonomous failures?

diplomat33

Average guy who loves autonomous vehicles
Aug 3, 2017
9,296
13,245
Terre Haute, IN USA
Following on from the Model 3 crash & considering autopilot fails with fixed objects which is a pretty rudimentary ‘obstacle’, what if after 1 million+ Model S3XY cars are on the road Tesla say “sorry, but we can’t offer full self driving”?
What would be the cost of retrofitting 1 million+ cars for FSD?

Depending on what the retrofit is, it could be potentially be very expensive of course.

But I don't think it is a big problem.

1) Tesla has already committed to upgrading the CPU's should regulators require it for FSD. So, I am sure Tesla has already looked at what that would cost.

2) Not braking for fixed objects can be fixed with a software patch and won't require any hardware upgrades at all. Tesla just needs the software to get the cameras to recognize fixed objects and brake instead of relying on the radar. This is something that FSD will do of course. So, this is a problem that should be fixable without needing any hardware upgrades at all.
 
  • Informative
Reactions: MelaniainLA

ewoodrick

Well-Known Member
Apr 13, 2018
5,285
4,289
Buford, GA
What you see today isn't full autopilot. There's really little auto in it, hence the requirement that drivers are required to stay vigilant when driving the cars. And the statement "when driving the cars" is important, because drivers do drive the cars today.

Full autopilot is not in the car today and will require upgrades, supposedly software and the answer to the question is easy. It won't cost anything.

Of course your question is pretty dubious, slight reworded, "How much will it cost to provide full AutoPilot when it can't be implemented"
 
  • Like
Reactions: MTL_HABS1909
1) Tesla has already committed to upgrading the CPU's should regulators require it for FSD. So, I am sure Tesla has already looked at what that would cost.
That is assuming the CPU is the only upgrade required.

What about if another camera is required? What about if they decide LIDAR is required?
Ridiculously expensive, borderline full refund.
 

TT97

Active Member
Aug 6, 2017
2,178
3,003
Los Angeles
Mobileye (which I'm sure many Tesla owners will remember) has been testing FSD in Jerusalem. They seem to be well ahead of Tesla, but more importantly, they are doing it all with just cameras (no LIDAR). They probably have the same equipment that is currently on the 3.

A lap of Jerusalem in Intel's surprisingly aggressive self-driving car

I think the hardware on the 3 is sufficient and in likelihood would just upgrade the CPU to process all the information needed (although I wouldn't mind if they can also add V2V and V2I).
 
  • Informative
Reactions: bhzmark

diplomat33

Average guy who loves autonomous vehicles
Aug 3, 2017
9,296
13,245
Terre Haute, IN USA
That is assuming the CPU is the only upgrade required.

What about if another camera is required? What about if they decide LIDAR is required?
Ridiculously expensive, borderline full refund.

I maintain that another camera or LIDAR are not needed. The car already has full coverage with 8 cameras, why would it need another one? And in terms of LIDAR, I know it is controversial, but I don't think it is needed either, at least not from a technical point of view. MobilEye has achieved FSD without LIDAR so it is doable. LIDAR is just another form of radar. It's not some magical "FSD tech". You can use radar and ultrasonics with really good camera vision instead of LIDAR.

Now, it is possible that government regulators might require LIDAR even though it is not technically necessary but that has not happened yet. Let's wait and see. There is no point in assuming that they will demand it and therefore writing off Tesla's FSD before we've even seen the product.
 

malcolm

Active Member
Nov 12, 2006
3,072
1,758
I maintain that another camera or LIDAR are not needed. The car already has full coverage with 8 cameras, why would it need another one?

Yeah, but Mobileye are going to put them in anyway:

Mind you, it's a lot of cameras, 12 in total, with 3 dedicated to the task of looking straight ahead, each at a different focal length. Another camera looks back, 2 each on the left and right look diagonally forward and back, while a further 4 are embedded to survey the immediate surroundings of the car. Like I said, that's a lot of imaging sensors but at a cost of $10 or $20 each, you could outfit a whole fleet of cars for what Velodyne will charge you for a single lidar scanner.
They're cheap because they're basic, just 1.3 megapixels each, but the company is upgrading to 8 megapixels soon. That dozen cameras, plus of course GPS and a couple decades' worth of learning from the company's various imaging-based driver assistance systems, is enough to make the car drive itself.
Except that it isn't. At least, it won't be when it comes to delivering the kind of on-road redundancy that Mobileye's engineers demand. And that's why Mobileye is now working on a lidar and radar-based solution. However, unlike most autonomous car developers, which unify their sensor data into a single input for its AI driver, Mobileye is actually keeping things wholly separate. The goal? Full redundancy.

Source: linked above by @TT97
 
  • Informative
Reactions: diplomat33
I think the real issue is that Tesla has done a terrible job of explaining what "autopilot" is and is not. They have also oversold (over promised) what the system can do and they have created a lot of confusion about the system. I have found that most Tesla employees don't understand the system (and its limitations) so how can they hope to explain it correctly to new owners???

I believe this is not a "hardware" issue as much as a "training" issue.

This is based on almost 80,000 miles of AP 1.0 use between the 3 MSs I have owned.

ps. I really hope that someday they actually deliver on all of their promises.
 
Following on from the Model 3 crash & considering autopilot fails with fixed objects which is a pretty rudimentary ‘obstacle’, what if after 1 million+ Model S3XY cars are on the road Tesla say “sorry, but we can’t offer full self driving”?
What would be the cost of retrofitting 1 million+ cars for FSD?

They just won't say that they can't offer full self-driving. Even if in five years all new Tesla's, new Chevy's and most new cars offer full self-driving; Tesla can just say that full self-driving is still possible on ap2 cars and they are working on the software to make it happen. There will be some lawsuits, but Tesla has never said that FSD will be available in any amount of time. Eventually, the amount of ap2 cars will dwindle due to age or breakthroughs will make it cost effective enough, and Tesla will do it.
 

gregd

Active Member
Dec 31, 2014
2,621
1,849
CM98
Wouldn't LIDAR solve the problem of running into barriers (concrete walls, fire trucks, etc) being reported? I thought the base issue was that RADAR is prevented from noticing anything stationary because it can't tell where exactly the things are (e.g. elevation off the roadway). Too much stationary clutter that prevents it from determining what's there. LIDAR creates a real 3-D image of the surrounding area, and presumably should be able to notice that the road doesn't pass through that solid thing. Can RADAR do that? If so, why isn't it?
 

diplomat33

Average guy who loves autonomous vehicles
Aug 3, 2017
9,296
13,245
Terre Haute, IN USA
Wouldn't LIDAR solve the problem of running into barriers (concrete walls, fire trucks, etc) being reported? I thought the base issue was that RADAR is prevented from noticing anything stationary because it can't tell where exactly the things are (e.g. elevation off the roadway). Too much stationary clutter that prevents it from determining what's there. LIDAR creates a real 3-D image of the surrounding area, and presumably should be able to notice that the road doesn't pass through that solid thing. Can RADAR do that? If so, why isn't it?

Yes, LIDAR would solve the problem. But good camera vision algorithms would also solve the problem of running into barriers too. And camera vision is a heck of a lot cheaper.
 
  • Like
Reactions: D3xDt3Reaction

kengchang

Active Member
Jul 17, 2017
2,468
15,067
California
Mobileye (which I'm sure many Tesla owners will remember) has been testing FSD in Jerusalem. They seem to be well ahead of Tesla, but more importantly, they are doing it all with just cameras (no LIDAR). They probably have the same equipment that is currently on the 3.

A lap of Jerusalem in Intel's surprisingly aggressive self-driving car

I think the hardware on the 3 is sufficient and in likelihood would just upgrade the CPU to process all the information needed (although I wouldn't mind if they can also add V2V and V2I).

Way ahead in running red light Mobileye autonomous vehicle runs red light in Jerusalem
 

Runt8

Active Member
May 19, 2017
1,986
2,448
Colorado
Mobileye (which I'm sure many Tesla owners will remember) has been testing FSD in Jerusalem. They seem to be well ahead of Tesla, but more importantly, they are doing it all with just cameras (no LIDAR). They probably have the same equipment that is currently on the 3.

A lap of Jerusalem in Intel's surprisingly aggressive self-driving car

I think the hardware on the 3 is sufficient and in likelihood would just upgrade the CPU to process all the information needed (although I wouldn't mind if they can also add V2V and V2I).
Intel’s Mobileye wants to dominate driverless cars—but there’s a problem

Mobileye’s approach is pretty sketchy...
 

Products we're discussing on TMC...

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top