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

Tesla autopilot HW3

This site may earn commission on affiliate links.
Are all highways in the US covered now? I was under the impression that NoA didn't work everywhere in the beginning and was still rolling out to more and more areas.

Autopilot used ADAS tiles for info about the road thru version 2019.8.x. And if I have understood verygreen correctly they have changed the format on those from 2019.12.x.
My guess would be that NoA uses a combo of navigation data and the ADAS tiles to get into the correct lane.

Here in Norway NoA works only on select highways. And the logic explanation is that some map data is missing for those highways that doesn't work.
 
Are all highways in the US covered now? I was under the impression that NoA didn't work everywhere in the beginning and was still rolling out to more and more areas.

Autopilot used ADAS tiles for info about the road thru version 2019.8.x. And if I have understood verygreen correctly they have changed the format on those from 2019.12.x.
My guess would be that NoA uses a combo of navigation data and the ADAS tiles to get into the correct lane.

Here in Norway NoA works only on select highways. And the logic explanation is that some map data is missing for those highways that doesn't work.

Works for all North American highways as far as I know. There is a navigation map data requirement (navigation maps are different) but the system isn't reliant on the ADAS tiles being current.
 
  • Informative
Reactions: sjokomelk
NoA does not work on all highways. I think it is only limited access roads. On our trip from Scottsdale AZ to Bryce National Park there was only one 2 hour segment I was able to use NoA (I17). The rest of the trip (5 hours) was mostly on US 89 with NoA not offered.
It can only work on freeways with no traffic lights. Do you mean by highways, those with traffic lights ?
 
IMHO...

If you are going to own a Tesla and be happy, you have to reset your time scale to a couple of years at a time, a time scale of a year or even months won't work out so well for you. You'll find a way to be perpetually miserable.

It's almost hard not to believe that with the investment and assets they have in place today FSD won't become *something* unqiue and pretty great in 2 years or so, but if your hanging on to every update, waiting with bated breath for big news --well, despite anything you heard in the past, that's not going to happen.

Advanced Summon might be about sometime this fall, and that would be quite interesting, I hope, but this is a multi-year game. Our investment was a bet that they'd eventually get it right (on HW3, I sure hope) --but come on now, its a bet. I know Musk said a lot of things, and that can be frustrating, but best to enjoy the coolest and most advanced vehicle on the road RIGHT NOW... and let go of exactly what is coming and when.

Despite being awesome, no one could realistically have bought a Tesla in the last 2-5 years and not thought they were buying very early stage tech, and everything that comes with it. If you did, you were mistaken. I know my 6 month old Tesla will be significantly outdated in 5 years from now (FSD/AP4 HW will not likely be upgradable on my car by then). It's the nature of business, and technology --and this is what we all signed up for.

If you really hate the experience, cut your losses and just sell it. Sit on sidelines for a couple of years and buy something from whoever gets there first --but I think its going to be Tesla, even the analysts are collecting enough information (slowly) to figure out they can actually do this. If your really, really deeply angry, sell it and get a lawyer and see of you can start a class action suit against Tesla --that won't make FSD come any faster (if it got class action status, it might even slow it down further).

But this won't make you happy either --drive and smile friends, you can say you were in the thick of it when you tell your kids and grandkids how cars used to require humans to drive them, and you had one of the first that drove pretty much by itself (or better... : )
 
IMHO...

If you are going to own a Tesla and be happy, you have to reset your time scale to a couple of years at a time, a time scale of a year or even months won't work out so well for you. You'll find a way to be perpetually miserable.

It's almost hard not to believe that with the investment and assets they have in place today FSD won't become *something* unqiue and pretty great in 2 years or so, but if your hanging on to every update, waiting with bated breath for big news --well, despite anything you heard in the past, that's not going to happen.

Advanced Summon might be about sometime this fall, and that would be quite interesting, I hope, but this is a multi-year game. Our investment was a bet that they'd eventually get it right (on HW3, I sure hope) --but come on now, its a bet. I know Musk said a lot of things, and that can be frustrating, but best to enjoy the coolest and most advanced vehicle on the road RIGHT NOW... and let go of exactly what is coming and when.

Despite being awesome, no one could realistically have bought a Tesla in the last 2-5 years and not thought they were buying very early stage tech, and everything that comes with it. If you did, you were mistaken. I know my 6 month old Tesla will be significantly outdated in 5 years from now (FSD/AP4 HW will not likely be upgradable on my car by then). It's the nature of business, and technology --and this is what we all signed up for.

If you really hate the experience, cut your losses and just sell it. Sit on sidelines for a couple of years and buy something from whoever gets there first --but I think its going to be Tesla, even the analysts are collecting enough information (slowly) to figure out they can actually do this. If your really, really deeply angry, sell it and get a lawyer and see of you can start a class action suit against Tesla --that won't make FSD come any faster (if it got class action status, it might even slow it down further).

But this won't make you happy either --drive and smile friends, you can say you were in the thick of it when you tell your kids and grandkids how cars used to require humans to drive them, and you had one of the first that drove pretty much by itself (or better... : )
Well said and I agree. I think this is why many people lease Tesla's actually. But I do think that leases open up some risk for Tesla. If you lease a car for three years and pay for something (FSD) that you never get then someone, somewhere is going to try and get that portion of their money back.
 
IMHO...

If you are going to own a Tesla and be happy, you have to reset your time scale to a couple of years at a time, a time scale of a year or even months won't work out so well for you. You'll find a way to be perpetually miserable.

It's almost hard not to believe that with the investment and assets they have in place today FSD won't become *something* unqiue and pretty great in 2 years or so, but if your hanging on to every update, waiting with bated breath for big news --well, despite anything you heard in the past, that's not going to happen.

Advanced Summon might be about sometime this fall, and that would be quite interesting, I hope, but this is a multi-year game. Our investment was a bet that they'd eventually get it right (on HW3, I sure hope) --but come on now, its a bet. I know Musk said a lot of things, and that can be frustrating, but best to enjoy the coolest and most advanced vehicle on the road RIGHT NOW... and let go of exactly what is coming and when.

Despite being awesome, no one could realistically have bought a Tesla in the last 2-5 years and not thought they were buying very early stage tech, and everything that comes with it. If you did, you were mistaken. I know my 6 month old Tesla will be significantly outdated in 5 years from now (FSD/AP4 HW will not likely be upgradable on my car by then). It's the nature of business, and technology --and this is what we all signed up for.

If you really hate the experience, cut your losses and just sell it. Sit on sidelines for a couple of years and buy something from whoever gets there first --but I think its going to be Tesla, even the analysts are collecting enough information (slowly) to figure out they can actually do this. If your really, really deeply angry, sell it and get a lawyer and see of you can start a class action suit against Tesla --that won't make FSD come any faster (if it got class action status, it might even slow it down further).

But this won't make you happy either --drive and smile friends, you can say you were in the thick of it when you tell your kids and grandkids how cars used to require humans to drive them, and you had one of the first that drove pretty much by itself (or better... : )

Very correct, though I suspect that HW4 will be designed as a slot-in replacement for current cars, as HW3 has been.
 
My bet is that HW4 will come with some changes to the sensor suite. I'm not going to wade into the quagmire of suggesting what the changes will be, but I'll bet they'll change.
Really the only 100% safe bet would be that HW4 involves a process shrink. Saves power, saves space, and in theory could increase speeds. Added features beyond shrinking the die are anyones guess though.
 
I predict more redundancy in sensor suite will be needed for true driverless autonomy in all situations and all locations (true L5).
E.g. two or three rear-facing cameras per side, two or three side-facing cameras per side, two or three front-facing radars, etc.
I was listening to a podcast about airplanes' autopilot systems and they added sensors until there was triple redundancy to be truly safe. And that's still just AP with human pilot supervision.
 
I predict more redundancy in sensor suite will be needed for true driverless autonomy in all situations and all locations (true L5).
E.g. two or three rear-facing cameras per side, two or three side-facing cameras per side, two or three front-facing radars, etc.
I was listening to a podcast about airplanes' autopilot systems and they added sensors until there was triple redundancy to be truly safe. And that's still just AP with human pilot supervision.

To be fair, it's a lot easier for a car to pullover in the event of a camera failure than for an airplane to land.
 
To be fair, it's a lot easier for a car to pullover in the event of a camera failure than for an airplane to land.

Agreed. Conversely, it's a lot worse if a car crosses lanes onto oncoming traffic, fails to stop on time at a traffic light, etc. versus a plane not following its exact course or speed for a few seconds. Pros and cons to each I guess.
 
Agreed. Conversely, it's a lot worse if a car crosses lanes onto oncoming traffic, fails to stop on time at a traffic light, etc. versus a plane not following its exact course or speed for a few seconds. Pros and cons to each I guess.

I think those scenarios would only be possible if the car had a camera failure and didn't sense it had a camera failure. Otherwise it wouldn't continue driving long enough to cross into oncoming traffic or run a red-light, it would just pull over.

The question becomes "What's the minimum number of functioning sensors the vehicle needs to safely pull over?" And that's where redundant sensors might come into play. The car might have trouble knowing where the shoulder is without a right repeater, but maybe it could use the B-pillar cam to make up for that.
 
I think those scenarios would only be possible if the car had a camera failure and didn't sense it had a camera failure. Otherwise it wouldn't continue driving long enough to cross into oncoming traffic or run a red-light, it would just pull over.

The question becomes "What's the minimum number of functioning sensors the vehicle needs to safely pull over?" And that's where redundant sensors might come into play. The car might have trouble knowing where the shoulder is without a right repeater, but maybe it could use the B-pillar cam to make up for that.

Front and rear wide angle will show the shoulder clearly, though not the portion currently beside the car - and the car will have to know how far beyond the camera position the car extends, of course.
 
I think those scenarios would only be possible if the car had a camera failure and didn't sense it had a camera failure. Otherwise it wouldn't continue driving long enough to cross into oncoming traffic or run a red-light, it would just pull over.

The question becomes "What's the minimum number of functioning sensors the vehicle needs to safely pull over?" And that's where redundant sensors might come into play. The car might have trouble knowing where the shoulder is without a right repeater, but maybe it could use the B-pillar cam to make up for that.
Theoretically, if they get good at using the cameras to generate a 3D point cloud representation of the surrounding area, they can store this and use it to handle camera failure. Perhaps something like 100m of stored road data would be enough to pull over. Obviously this wouldn't cover moving objects like other cars, but those can actively accommodate a disabled vehicle, just they way they do now if someone has a blow-out or other severe mechanical problem that causes a car to have to pull over right away. Not guaranteed to be perfect, but might be OK in many cases. (I would like to see at least two cameras cover every angle, though. And maybe corner or at least rear-facing radar.)
 
Theoretically, if they get good at using the cameras to generate a 3D point cloud representation of the surrounding area, they can store this and use it to handle camera failure. Perhaps something like 100m of stored road data would be enough to pull over. Obviously this wouldn't cover moving objects like other cars, but those can actively accommodate a disabled vehicle, just they way they do now if someone has a blow-out or other severe mechanical problem that causes a car to have to pull over right away. Not guaranteed to be perfect, but might be OK in many cases. (I would like to see at least two cameras cover every angle, though. And maybe corner or at least rear-facing radar.)
Exactly, computers have perfect memory. I suspect they would do way better than a human after getting hit in the face and blinded.
 
  • Funny
Reactions: favo
Theoretically, if they get good at using the cameras to generate a 3D point cloud representation of the surrounding area, they can store this and use it to handle camera failure. Perhaps something like 100m of stored road data would be enough to pull over. Obviously this wouldn't cover moving objects like other cars, but those can actively accommodate a disabled vehicle, just they way they do now if someone has a blow-out or other severe mechanical problem that causes a car to have to pull over right away. Not guaranteed to be perfect, but might be OK in many cases. (I would like to see at least two cameras cover every angle, though. And maybe corner or at least rear-facing radar.)
I also think this technique could/should be used when auto-parking the car, so it can be aware of low curbs and things the cameras can't see when up close and the ultrasonics might miss altogether (poles, et al.). In fact, they could use this as a pseudo 360 camera view and show the point-cloud on the display when manually parking. (Mostly useful in the front for low curbs, but maybe on the side when parallel parking.)
 
  • Informative
Reactions: pilotSteve
My bet is that HW4 will come with some changes to the sensor suite. I'm not going to wade into the quagmire of suggesting what the changes will be, but I'll bet they'll change.
No doubt in my mind. I feel like they would have made changes when introducing the FSD chip if Elon had not promised full self driving with the current sensor suite.

Tesla has locked themselves in jail because if new sensors are introduced, many prior owners will be angry at being left behind.
 
  • Like
Reactions: patrick80132