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

Software Update 2018.39 4a3910f (plus other v9.0 early access builds)

This site may earn commission on affiliate links.
The works well everywhere part is what makes me think/hope that he's shooting for WW release of this instead of just US-only first as the early access release notes said.

Wow, that is some wishful thinking right there. Don't you know that to us Americans "everywhere" means California and New York? But not the rural parts...

Here is what it means: "Our early access testers in the US encountered a great many more bizarre highway interchanges, ramp designs, lane markings, and driving styles than we accounted for, because OMG IT TURNS OUT AUTONOMOUS DRIVING IS SUPER HARD 4REALZ GUYS."

This is going to take longer before we can even roll out in the US, nevermind places with radically different highway designs. I just hope they include blind spot monitoring in V9 across the board -- if even that is giving them trouble right now we're in bad shape indeed.
 
"Extremely difficult to achieve a general solution for self-driving that works well everywhere."

It is that quote right there that confirms even further my thoughts that FSD is several years away. If Tesla is struggling to get "drive on NAV" out the door, which is going to be limited to only certain highways, then we've got a long way to go. Think of your typical daily drive... how many edge cases do you see? Things you may not even think about because for a human they're easily navigated or you're just used to them. A weird intersection here, an partly obscured stop sign there, an unusual traffic light set up, no lines at all on a side road, etc... those things and more are everywhere. I'd love to see FSD become a reality, but I'm not holding my breath. Don't get me wrong I love my Model S and love that Tesla is pushing to make a FSD capable system, but it seems a long way off.

You may be right that FSD is several years away but I think you might be slightly off about the reasons why. Musk's tweet doesn't say that they are struggling to get drive on nav to work on limited highways. Actually, if you watched the videos, drive on nav seemed to work pretty well on limited highways. So the issue is not that they can't even get it right for limited highways. If FSD is years away, it is not, as you say, because they can't even get the limited cases right. The issue is that they want to get it right for more than just limited highways. If you look at Musk's tweet he says it is "Extremely difficult to achieve a general solution for self-driving that works well everywhere.". Note the emphasis in bold. They are looking to expand drive on nav to work everywhere. That is why it is so difficult. So if FSD is years away, it's because they can do the limited cases just fine but need to make it work for all those edge cases too.

If anything, Musk's tweet reveals his ambitious vision again. He is not content to release a self-driving feature in a very limited case like many other auto makers have done. He wants the self-driving feature that "works everywhere".

Musk's tweet also seems to imply, at least to me, what I have been thinking now for awhile. "drive on nav" is intended to be the bucket that all the self-driving features will go into. That is why he calls it a "general solution for self-driving". It is called "drive on nav" or "navigate on autopilot" because it is intended to be that "drive from A to B" that makes FSD work. So right now, it can do auto lane changes on the highway, but I suspect that Musk also wants it to do auto lane changes on other roads as well and eventually be able to make turns at intersections, stop at red lights etc..

The bottom line is that "navigate on autopilot" is a critical feature that has the potential to make or break "highway self-driving" so Tesla needs to get it right.
 
You may be right that FSD is several years away but I think you might be slightly off about the reasons why. Musk's tweet doesn't say that they are struggling to get drive on nav to work on limited highways. Actually, if you watched the videos, drive on nav seemed to work pretty well on limited highways. So the issue is not that they can't even get it right for limited highways. If FSD is years away, it is not, as you say, because they can't even get the limited cases right. The issue is that they want to get it right for more than just limited highways. If you look at Musk's tweet he says it is "Extremely difficult to achieve a general solution for self-driving that works well everywhere.". Note the emphasis in bold. They are looking to expand drive on nav to work everywhere. That is why it is so difficult. So if FSD is years away, it's because they can do the limited cases just fine but need to make it work for all those edge cases too.

If anything, Musk's tweet reveals his ambitious vision again. He is not content to release a self-driving feature in a very limited case like many other auto makers have done. He wants the self-driving feature that "works everywhere".

Musk's tweet also seems to imply, at least to me, what I have been thinking now for awhile. "drive on nav" is intended to be the bucket that all the self-driving features will go into. That is why he calls it a "general solution for self-driving". It is called "drive on nav" or "navigate on autopilot" because it is intended to be that "drive from A to B" that makes FSD work. So right now, it can do auto lane changes on the highway, but I suspect that Musk also wants it to do auto lane changes on other roads as well and eventually be able to make turns at intersections, stop at red lights etc..

The bottom line is that "navigate on autopilot" is a critical feature that has the potential to make or break "highway self-driving" so Tesla needs to get it right.
Well said!
 
Anyone know if Firetruck Super-Destruction mode has been successfully removed from v9 yet?

IOW, should the use of all 3 forward-facing cameras be able to alleviate what is AFAIK essentially the problem caused by using a radar sensor unable to discriminate with sufficient resolution between hard stationary objects lying in the planned vehicle path and those to the side or overhead of same?

Or is Tesla more likely to need to upgrade the Continental radar sensor for AP3.0 and safe/reliable FSD?

Or is this a fundamental design flaw unlikely to be overcome with any radar/camera/ultrasonic sensor suite in the next 5 years?
 
Anyone know if Firetruck Super-Destruction mode has been successfully removed from v9 yet?

IOW, should the use of all 3 forward-facing cameras be able to alleviate what is AFAIK essentially the problem caused by using a radar sensor unable to discriminate with sufficient resolution between hard stationary objects lying in the planned vehicle path and those to the side or overhead of same?

That is correct, so far Autopilot has used mostly the radar for TACC and radar has this inherent limitation (at least the radar they're using, and automotive radar in general really).

Or is this a fundamental design flaw unlikely to be overcome with any radar/camera/ultrasonic sensor suite in the next 5 years?

Well it's a fundamental design flaw but it can in theory at least be overcome with the cameras. It's just really hard to overcome with the cameras reliably; it's a big problem. Now if they had lidar it would be much easier (still hard, but easier). They may need to upgrade the processing power even on non-FSD cars in order to get an adequate vision solution that can both stop for stopped cars and not have excessive false positives on signs and overpasses. Also they will clearly need a lot more time to develop the vision systems; there has not been much progress made on this recently. If anything I think we have regressed slightly on false positives; since the high-profile accidents (slamming into fire trucks) I think they have turned the dial to be more sensitive, leading to more false positives and phantom braking.
 
  • Informative
Reactions: OPRCE
Yes, but not realistic feature sets.

In June, V9 added the first FSD features. In August, it was down to "the next level of autonomy." At release, it's "no additional autonomy."

It's easy to hit dates if you just keep deleting features. Of course, he missed the date by a month too, even when it was only 2 months away.

Called it.
 
Watched a video on coma.ai with George explaining their newest release of self driving software and he mentioned that the radars on most of the cars filters out stationary objects so he has to rely on vision to detect stopped objects and avoid hitting them. It is apparently not easy so he doesn't have that working right yet. He talks about changing the cars radars to some newer version. Tesla has the advantage of 3 forward facing cameras so it should be able to obtain distance information to stationary objects and discriminate between objects that are near the path of travel and those that are in the path of travel. This would fix the phantom braking and running into firetrucks.
 
Watched a video on coma.ai with George explaining their newest release of self driving software and he mentioned that the radars on most of the cars filters out stationary objects so he has to rely on vision to detect stopped objects and avoid hitting them. It is apparently not easy so he doesn't have that working right yet. He talks about changing the cars radars to some newer version. Tesla has the advantage of 3 forward facing cameras so it should be able to obtain distance information to stationary objects and discriminate between objects that are near the path of travel and those that are in the path of travel. This would fix the phantom braking and running into firetrucks.

They've had stereo vision for over a year. Only AP1 seems to slam into firetrucks. 2.0 lusts for gorepoints because it couldn't properly identify navigable road paths. 2.0 is improving, 1.0 seems stagnant.
 
  • Like
Reactions: chillaban