Autonomous Car Progress

[Tikib2020]

Bloomberg - Are you a robot?

I am telling you Tesla are further ahead than you guys think behind closed doors.

I also don't think Lidar is necessary for L5.

Some of you are going to say Elon has been wrong before and it's true but nobody's perfect.

How many rockets have you sent to the ISS by the way?

 

[diplomat33]

Bloomberg - Are you a robot?

I am telling you Tesla are further ahead than you guys think behind closed doors.

I also don't think Lidar is necessary for L5.

Then why don't we see more features on our cars? Or just more demos? Tesla could show off this amazing L5 that they are secretly testing.

What about the following?
- Roundabouts
- Unprotected left turns
- Right turns
- pulling over for emergency vehicles
- Anticipating pedestrians crossing the adjacent road that you are coming to after you make a turn
- Responding to stopped cars blocking your lane
- Intersections with no traffic controls
- Reverse Summon
- Not hitting curbs in Smart Summon
- Automatic lane changing into turning lane
- Tracking paths for all objects at a busy intersection during rush hour

Those are just a few things that I could come up with off the top of my head. Some of these are quite complicated. I am sure you could come up with more things that L5 requires.

So has Tesla solved all of these?

"Musk reiterated that the electric vehicle maker has solved most of the essential challenges toward achieving Level 5 autonomy, or a fully self-driven automobile that needs no human behind the wheel."

“I’m confident that we will have the basic functionality of L5 autonomous driving this year,” Musk said. “There are no fundamental challenges.”

I think Elon is sincere in these statements but I think this is just more of Elon underestimating the challenges of autonomous driving.

Elon thinks autonomous driving is just a matter of perception. And yes, Tesla is probably close to "feature complete" vision that can see depth and see different objects like other vehicles, lanes, traffic controls, curbs, road markings etc... Hence, Elon thinks that they are close to solving the main challenges of L5.

But note also that Elon is just saying they are close to solving the basics. There is a big difference between having the basics of autonomous driving done and actually getting driverless L5 done. Going from "feature complete" to actually removing the driver completely is a huge gap. Even Waymo has not completely removed the driver in all cases yet.

And L5 is more than perception. There is driving policy too. The car has to drive "smart" and respond to other vehicles. There are a lot of driving cases like responding to double parked cars or emergency vehicles or pedestrians or cyclists that have to be solved.

So yes, Tesla probably has pretty good perception but there is lot more to do.

But if I am wrong and Tesla releases Waymo level autonomous driving in 5 months, I will be super happy. But I doubt it.

 

[Tikib2020]

I don't think Tesla will have L5 autonomous driving by the end of the year.

I don't think that's what Musk's actually saying anyway, I think he's alluding to having the route to (theory behind) L5 autonomous driving down by the end of the year.

As for all the features you mentioned I would imagine they are working on all of them. I think the idea that they work on one at a time in serial is completely wrong. So it's entirely possible that you could see several features released at a similar time.

 

[diplomat33]

I don't think that's what Musk's actually saying anyway, I think he's alluding to having the route to (theory behind) L5 autonomous driving down by the end of the year.

I agree that is what Elon is saying. But that's my point. Having the theory done and actually getting L5 done are two different things. Waymo, Cruise have solved the theory of autonomous driving many years ago. They are still working on solving difficult driving cases.

So even if we assume that Tesla is close to solving the theory of autonomous driving, that is still a long way from having driverless L5.

Elon acts like solving the basic challenges is just two skips away from having L5 and that is just false.

if anything, it shows how far behind Tesla is. Elon is boasting about almost solving the basics. Everybody else has solved the basics a long time ago and are still working on complex driving policy to actually get a driverless L5.

As for all the features you mentioned I would imagine they are working on all of them. I think the idea that they work on one at a time in serial is completely wrong. So it's entirely possible that you could see several features released at a similar time.

That would great. I would love to see more features soon.

 

[Tikib2020]

As I said before I think within a couple of years we will see Tesla robotaxis on the road.

And in saying that I am not saying Tesla are ahead. Cruise and Waymo are both further ahead.

But probably neither will have the guts to put on their big boy panties and actually release a product before Tesla.

Musk knows that autonomous vehicles don't have to be 100 percent perfect before being released they just have to be as safe as humans and then you use what you learn from that to make them much better.

Perfect is the enemy of good.

 

[diplomat33]

Here is a full quote from Elon from WAIC in more context:

“I think there are no fundamental challenges remaining for Level 5 autonomy. There are many small problems. And then there’s the challenge of solving all those small problems and then putting the whole system together and just keep addressing the long tail of problems. So you’ll find that you’re able to handle the vast majority of situations. But then there will be something very odd. You have to have a system figure out a train to deal with these odd situations. This is why you need real world situations. Nothing is more complex and weird than the real world,”
Tesla's Elon Musk talks Autopilot and Level 5 Autonomy at China AI conference

I am sorry to sound "anti-Tesla" but that first sentence sounds incredibly naive or arrogant to me. There are really no fundamental challenges remaining for L5? It's all just small problems from here on out? Who knew!

 

[mikes_fsd]

Perfect is the enemy of good.

...
Don't worry @Tikib2020 ... @diplomat33 will take anything Waymo or what any other lidar user says as gospel, but anything that Musk or Karpathy says must not be taken seriously.

Because you know Elon time...
And waymo delivers everything on schedule...

 

[mikes_fsd]

to sound "anti-Tesla" but that first sentence sounds incredibly naive or arrogant to me.

Oh you don't have to worry about that we know you're anti Tesla.
And you've proven time and again that the one naive and arrogant is you, not Musk.

 

[diplomat33]

...
Don't worry @Tikib2020 ... @diplomat33 will take anything Waymo or what any other lidar user says as gospel, but anything that Musk or Karpathy says must not be taken seriously.

I take Waymo seriously because they are the leader in autonomous driving.

 

[diplomat33]

Oh you don't have to worry about that we know you're anti Tesla.
And you've proven time and again that the one naive and arrogant is you, not Musk.

I am not "anti-Tesla". If you look at my record, it is clear.

I am "anti-BS" when it comes to FSD. And sadly, Elon has said some B.S when it comes to FSD.

So I am naive and arrogant for listening to the leaders in autonomous driving and for understanding that FSD is hard? WOW.

Elon says that FSD is super easy so it must be true.

You accuse me of taking Waymo as gospel but you do the same with Elon. For you, anything that Elon says about FSD is absolute gospel truth without question and you reject the expertise of Waymo and others.

 

[john5520]

Here is a full quote from Elon from WAIC in more context:

“I think there are no fundamental challenges remaining for Level 5 autonomy. There are many small problems. And then there’s the challenge of solving all those small problems and then putting the whole system together and just keep addressing the long tail of problems. So you’ll find that you’re able to handle the vast majority of situations. But then there will be something very odd. You have to have a system figure out a train to deal with these odd situations. This is why you need real world situations. Nothing is more complex and weird than the real world,”
Tesla's Elon Musk talks Autopilot and Level 5 Autonomy at China AI conference

I am sorry to sound "anti-Tesla" but that first sentence sounds incredibly naive or arrogant to me. There are really no fundamental challenges remaining for L5? It's all just small problems from here on out? Who knew!

I agree. You're not being 'anti-tesla', you're just echoing the reality of the situation. True L5 is a really tall order to fulfill.

What I admire about Musk is his drive. He does not give up on a vision. However, his vision can be overly ambitious at times and this is one of them, IMO.
What he should be concentrating on is reliable (as possible) level 2-3 automation. And create a separate division with different hardware for just the robotaxi project.

Do people really believe their Model 3s in the current form, are capable of being a Robotaxi? That you can just let it do its thing and make money, lol? Sorry, that ain't happening with
the current hardware. And if it can, not anytime soon. I would love to be wrong about that though.

 

[diplomat33]

I agree. You're not being 'anti-tesla', you're just echoing the reality of the situation. True L5 is a really tall order to fulfill.

What I admire about Musk is his drive. He does not give up on a vision. However, his vision can be overly ambitious at times and this is one of them, IMO.
What he should be concentrating on is reliable (as possible) level 2-3 automation. And create a separate division with different hardware for just the robotaxi project.

Thank you.

The fact is that I use AP 99% of my drives. For what it does, AP is fantastic. Yes, I think Tesla would do well to focus on reliable L2 and L3. That would be amazing. Heck, if Tesla delivered reliable L3 on the highway, I would be the first to applaud Tesla and Elon.

But saying that Tesla has no fundamental challenges for L5 and it's just small problems now? Sorry. I think Elon is getting way ahead of himself there.

Do people really believe their Model 3s in the current form, are capable of being a Robotaxi? That you can just let it do its thing and make money, lol? Sorry, that ain't happening with the current hardware. And if it can, not anytime soon. I would love to be wrong about that though.

I think most Tesla owners probably realize that the current hardware will fall short of true driverless L5. And anybody who understands autonomous driving knows that the current hardware will fall short of true L5.

 

[mikes_fsd]

Do people really believe their Model 3s in the current form, are capable of being a Robotaxi?

Yes. I'm a person and an owner...

 

[ItsNotAboutTheMoney]

I agree. You're not being 'anti-tesla', you're just echoing the reality of the situation. True L5 is a really tall order to fulfill.

What I admire about Musk is his drive. He does not give up on a vision. However, his vision can be overly ambitious at times and this is one of them, IMO.
What he should be concentrating on is reliable (as possible) level 2-3 automation. And create a separate division with different hardware for just the robotaxi project.

Do people really believe their Model 3s in the current form, are capable of being a Robotaxi? That you can just let it do its thing and make money, lol? Sorry, that ain't happening with
the current hardware. And if it can, not anytime soon. I would love to be wrong about that though.

Level 5 is just perfect Level 2.

 

[mikes_fsd]

you reject the expertise of Waymo and others.

I've followed Waymo/Google self driving way longer than Tesla's Autopilot/FSD work.

And I reject them because they've delivered nothing. Their business case for commercializing it is terrible, not scalable and prone to the whims of each local municipality wherever they try to expand into.

In the 5 years of watching Tesla approach this problem, I've seen way more innovation and progress. It's like they took a blank sheet approach to the problem, what a concept!

I guess in a way I can compare it to Boeing versus SpaceX especially with the way the crewed capsules worked for both companies and their demo missions.

If it wasn't clear Waymo is Boeing and the example above.

 

[diplomat33]

Elon is just making the same promises that he's made before about how Tesla is close to L5. Of course, the optimists (I used to be one) will get super excited. The skeptics will be doubtful. Based on the current features we have in our cars, I think we have every reason to be skeptical about claims that Tesla is close to L5.

 

[diplomat33]

I've followed Waymo/Google self driving way longer than Tesla's Autopilot/FSD work.

And I reject them because they've delivered nothing. Their business case for commercializing it is terrible, not scalable and prone to the whims of each local municipality wherever they try to expand into.

In the 5 years of watching Tesla approach this problem, I've seen way more innovation and progress. It's like they took a blank sheet approach to the problem, what a concept!

I guess in a way I can compare it to Boeing versus SpaceX especially with the way the crewed capsules worked for both companies and their demo missions.

If it wasn't clear Waymo is Boeing and the example above.

Waymo only did the first ever driverless ride back in 2015 and has L4 robotaxis. They also did over 20M autonomous miles. Other than that, they have not delivered anything.

Sure, Tesla has made quite a bit of progress with AP but that is only L2. I wish Tesla the best and I am confident we will see a lot more AP progress. But that is a far cry from L5.

 

[john5520]

Level 5 is just perfect Level 2.

Picture your Model 3 without a steering wheel and pedals, or you sitting in the back seat. Then picture yourself relaxing or sleeping on the way to work and not worrying about anything (people running lights, cutting you off, merging into you, sudden stops etc. Even in bad weather). That's what L5 is supposed to accomplish. Do you think your Model 3 would be capable of that level of automation by the end of the year?

 

[Soda Popinski]

How many rockets have you sent to the ISS by the way?

Since you asked, I've docked Dragon 2 twice with the ISS in the SpaceX web simulator, at berthed the original Dragon with ISS at least twice in VR in Mission:ISS, and sent countless rockets to Low Kerbin Orbit with a bunch of boosters doing return to launch site. So clearly, I'm an expert in autonomous cars...

In all seriousness, we don't have any evidence of Tesla being way ahead in FSD behind the scenes. That said, I don't think we can extrapolate FSD progress from what we're seeing with the current software build, which seems to be primarily based on AP2 software. I think we can safely assume whatever Tesla is working on, it's way ahead of what's currently available in Teslas today.

 

[Dutchie]

I am not "anti-Tesla". If you look at my record, it is clear.

I am "anti-BS" when it comes to FSD. And sadly, Elon has said some B.S when it comes to FSD.

So I am naive and arrogant for listening to the leaders in autonomous driving and for understanding that FSD is hard? WOW.

Elon says that FSD is super easy so it must be true.

You accuse me of taking Waymo as gospel but you do the same with Elon. For you, anything that Elon says about FSD is absolute gospel truth without question and you reject the expertise of Waymo and others.

I have been saying and will say it agin. Here is guy who lands boosters on barges in the middle of the the ocean and send people to the ISS for pennies to the dollar for price it used to be. Talk about hard things. On a conference call he said most R&D goes to FSD, Not only Elon but all people who work on FSD within Tesla are positive they can crack the code soon. Elon also mentioned that eventually FSD is not a gradual process but will be implemented with a "flip of a switch" to hundreds of thousands of cars.
Yes, he was of in his time but Elon alway comes through eventually. His timing has become better recently, especially short time prognoses.
Don't underestimate him.