Elon: "Feature complete for full self driving this year"

[diplomat33]

Indeed but the problem is that when it violently turns fast it will fling your arm off the wheel You can't possibly have a loose grip and be relaxed in situations with sharp turns. If you watch the autonomy day demo you'll see.

Well, the closest case I've been in has been those very sharp turns in an off ramp. Auto steer can handle those very sharp turns very well at low speeds. And I just hover a hand over the wheel. I don't need to actually hold the wheel at all. I could see the same with "automatic driving on city streets" where you get a few nags before a turn where you can gently tug the wheel but you don't need to hold the wheel during the turn. That would work I think.

 

[rnortman]

In that list, NOA on Highways and Auto Park have been released to the public.

The Auto Park they have released to the public is the same thing you can get from every other manufacturer -- a very basic, very limited supervised self-park based almost entirely on ultrasonic sensors. It works in a tiny fraction of all parking situations, must be carefully supervised, and often doesn't work at all. For it to work you have to be right next to the parking spot, in the right orientation, and the spot must be clearly delineated by two other cars that the Autopilot can park between.

As far as I can tell, Tesla has made no enhancements to Auto Park since its initial release, and it does pretty much what every other manufacturer's park assist features do. I suspect, in fact, that it is based off of an off-the-shelf park assist system from the ultrasonic sensor manufacturer, though I have no direct evidence of this.

This is not what Tesla sold as "self parking" in both EAP and FSD. They sold that it would drop you off at the curb near the front door of your destination building, and then go off into the parking lot and park itself. There is no evidence Tesla is even shooting for that goal at this point, and certainly none that they're working on it or have made any progress.

 

[diplomat33]

Someone got this reply from Tesla back in May. Hopefully, this bodes well for Tesla moving forward with getting to "feature complete" by end of this year.

 

[diplomat33]

The Auto Park they have released to the public is the same thing you can get from every other manufacturer -- a very basic, very limited supervised self-park based almost entirely on ultrasonic sensors. It works in a tiny fraction of all parking situations, must be carefully supervised, and often doesn't work at all. For it to work you have to be right next to the parking spot, in the right orientation, and the spot must be clearly delineated by two other cars that the Autopilot can park between.

As far as I can tell, Tesla has made no enhancements to Auto Park since its initial release, and it does pretty much what every other manufacturer's park assist features do. I suspect, in fact, that it is based off of an off-the-shelf park assist system from the ultrasonic sensor manufacturer, though I have no direct evidence of this.

This is not what Tesla sold as "self parking" in both EAP and FSD. They sold that it would drop you off at the curb near the front door of your destination building, and then go off into the parking lot and park itself. There is no evidence Tesla is even shooting for that goal at this point, and certainly none that they're working on it or have made any progress.

I am sure Tesla will improve Auto Park with AP3 to make it what they promised on FSD. I was merely stating the obvious that the feature has been released. But Tesla always releases a basic version of a feature first and then improves the feature over time.

 

[rnortman]

I am sure Tesla will improve Auto Park with AP3 to make it what they promised on FSD. I was merely stating the obvious that the feature has been released. But Tesla always releases a basic version of a feature first and then improves the feature over time.

OK, but is the current version of self-parking "feature complete"? That's what this thread is about. My position, in case that was not clear, is that it is not feature complete. You had self-park in the feature complete column I believe.

 

[diplomat33]

OK, but is the current version of self-parking "feature complete"? That's what this thread is about. My position, in case that was not clear, is that it is not feature complete. You had self-park in the feature complete column I believe.

Yes, but only because the feature has been released. But I get your point that auto park is not working yet in its fully finished version yet.

 

[rnortman]

Yes, but only because the feature has been released. But I get your point that auto park is not working yet in its fully finished version yet.

I'm not sure you do get my point. My point is that self-park/"park seek mode" is not feature complete and there's no evidence that Tesla is even working on it. This thread is about FSD being feature-complete at the end of this year. As of today they have 6 months to get there, and no evidence that they're working on this, and plenty of evidence that the currently-released version of auto-park is no where near a suitable foundation for the real feature. I would be shocked if they were working on this, because given where they are on the core driving features that everybody pays attention to, I hope all hands are on those problems instead.

I will also point out that self-parking was sold as part of EAP, not FSD. EAP is still not feature-complete, and in 6 months FSD is supposed to be feature-complete?

 

[Az_Rael]

Someone got this reply from Tesla back in May.

I think that person tried to book a retrofit since the six weeks had passed and Tesla cancelled the appointment and said no known dates yet. Most folks over in the retrofit thread have been cancelled and told retrofits aren’t starting. Only 1 person has reported getting a retrofit so far.

FSD Computer Upgrade and Experience

 

[diplomat33]

I'm not sure you do get my point. My point is that self-park/"park seek mode" is not feature complete and there's no evidence that Tesla is even working on it. This thread is about FSD being feature-complete at the end of this year. As of today they have 6 months to get there, and no evidence that they're working on this, and plenty of evidence that the currently-released version of auto-park is no where near a suitable foundation for the real feature. I would be shocked if they were working on this, because given where they are on the core driving features that everybody pays attention to, I hope all hands are on those problems instead.

I will also point out that self-parking was sold as part of EAP, not FSD. EAP is still not feature-complete, and in 6 months FSD is supposed to be feature-complete?

You are confusing terms. Auto Park/Self-Parking is NOT Park Seek Mode. They are two different features.

Self-Park is Auto Park and is what we have now where you approach a spot and you get the P on the screen and the car will park for you. Tesla has delivered on Auto Park/Self-Parking now.

Park Seek Mode is a different feature. That's where the car drops you off and can find its own parking space. They have not delivered Park Seek Mode yet.

And EAP has been feature complete for awhile now.

 

[wcorey]

I don't think anyoe thinks Tesla will get to robotaxi (lvl 5) next year.

Except for the folks that took rides after Autonomous Day.

 

[OPRCE]

I merely mentioned the weight, not endorsing it, as some people trust it to that extent and DO want to watch Harry Potter movies. Please explain how. The definition I gave from Google said nothing about 99%. Is that an EU thing? What I gave you as first hand is accurate. I think when they get approval to go GA, not beta, it'll be level 4 if not 5. They are planning on a robo-taxi fleet next year. Am I buying stock in anticipation? No. Again, you're talking about the car's performance in Switzerland, not US. It may well lag.

1. Some people are suicidally deluded that their adding weights to the the wheel magically upgrades AP to L3, which enables them to enter for a Darwin Award (and maybe watch half a movie).

2. No, the 99% is not an EU thing, just my remark on where I feel AP is not yet re. Highway L3.

3. I was talking about AP's performance in USA. In Europe it is lagging by ~6 months and Autosteering power was recently reduced to comply with UN/ECE regulations.

 

[OPRCE]

If Tesla expands NOA to include non-highways so that NOA does not need to disengage at an off ramp, but is able to keep going on the local road, then that would be L4/5. That is what Tesla is aiming for.

Not true. When a L3 car prompts the driver to take over, the driver MUST take over.

I think you might be confusing L3 with L4. If the car can self-drive everywhere and the driver does not need to take over even when prompted, that is L4.

By the way, here is a video of Audi's Traffic Jam Assist:

Check the 6:30 mn mark. Basically, the system just gives the driver a warning to hold the wheel again and bumps the driver out of Traffic Jam Assist. Traffic Jam Assist does NOT continue to operate outside of it's ODD.

TJA is not L3, that would be TJP, which the video mentions will arrive sometime in the new A8.

 

[diplomat33]

TJA is not L3, that would be TJP, which the video mentions will arrive sometime in the new A8.

Yes, thank you. I confused the two systems. I figured it out after I posted the video.

Although, I did watch a video for Traffic Jam Pilot. It appears to be the same as Autopilot except hands free and limited to low speeds. Tesla could do L3 now on Autopilot, in the limited case of stop and go traffic.

 

[electronblue]

Tesla could do L3 now on Autopilot, in the limited case of stop and go traffic.

Only if they allowed no driver supervision.

 

[diplomat33]

Only if they allowed no driver supervision.

Well, yeah. That is implied in L3.

If you are on a highway in a traffic jam situation (cars detected around you, speed below 30 mph), AP/NOA could give you a green message that "hands free driving is allowed". You get no nags. When the car detects that you are leaving a traffic jam situation, you would get a warning "hands free driving ending. Please keep hands on wheel" and you would get nags again. That would quality as L3 I believe.

 

[OPRCE]

I am distressed on the apparent lack of clarity on whether a Tesla can detect the side of a barn door (side of an 18 wheeler) and not ram it. Hopefully they no longer have that issue. If that's the case they should announce it.

Best to assume the problem persists until a 3rd party demonstrates otherwise.
Same goes for highway cut-out into stationary vehicle.

 

[DigAdrenaline]

Really? I have virtually no problem with it on interstates and other limited access roads. Several people I know of that traveled cross country said the car literally did 99% of the driving. One thing I will say To get the the nearest theater there is a 19 mile interstate trip. The default behavior of NoA is to not be in the slow (entrance/exit) lane so the first thing it tries is to change lanes. Unfortunately I-395 is a two lane road so unless I want to do 10 over speed limit I have to stay in slow (entry/exit) lane. That's easy enough to do by deselecting auto lane change.

Agreed. Having driven from Calgary, Canada to Texas, Florida and California now, I believe AP is 99% or better. I’d give NoA 95%. I am thrilled with how well it works and look forward to anything even close to this in CityNoA, since AP generally/mostly works in the city already.

I expect Advanced Summon will be released and the cars will move at 3 km/h, but if that increases by a km/h per 1 million kms driven, that’s likely still going to mean reasonable speeds in less than a year.

Kev

 

[Daniel in SD]

Well, yeah. That is implied in L3.

If you are on a highway in a traffic jam situation (cars detected around you, speed below 30 mph), AP/NOA could give you a green message that "hands free driving is allowed". You get no nags. When the car detects that you are leaving a traffic jam situation, you would get a warning "hands free driving ending. Please keep hands on wheel" and you would get nags again. That would quality as L3 I believe.

They could but then Tesla would be responsible when this happens:
streamable.com/c5hbu
I would think they would also have to be able to deal with road debris as well. I do feel like level 3 on limited access highways should be possible with current technology.

 

[Daniel in SD]

Best to assume the problem persists until a 3rd party demonstrates otherwise.
Same goes for highway cut-out into stationary vehicle.

I wouldn't be surprised if both of those scenarios are dealt with properly 90% of the time. It's best to assume the problem persists until Tesla says it doesn't. They're the only ones with a large enough sample size to make that determination.

 

[Eno Deb]

Lex Fridman's research says the opposite. You would have to provide evidence to prove him wrong - not just anecdotal observations.

It actually doesn't. His paper is much more nuanced. Note e.g. this sentence:

"Also, if system reliability were to improve significantly over time, the levels of functional vigilance observed here may well change."