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

So SilkySmooth never happened now waithing on SomethingSpecial

This site may earn commission on affiliate links.
You don't think surviving family will have a case for a massive settlement? Not toast with the money, but blood on their hands for sure.

Mm... I don't disagree with a fatality being bad due to an AP2 failure, but Tesla seems to have a tendency at getting away with this stuff. They're the cool kids. Like Apple used to be, who got away with the antenna issues with some mansplaining...

Tesla will pull some logs and say nothing happened, wave some arms about how the driver is responsible anyway and any legal issues are settled behind closed doors, never to be heard from again.
 
  • Informative
Reactions: oktane
My experience (soon to be 6 months) with AP2 is similar to above. Last iterations have better steering control with less ping-pong. But, lane identification is poor and to many phantom events.

As I see it, the pattern recognition has really low specificity. Sensitivity seems high though but declining with the last two-three firmwares. Therefore I want a sensitive&specific update.
 
Still awful running 17.28.whatever
IMG_7929.JPG
IMG_7928.JPG
IMG_7927.JPG


These last ones were right before AS drove me over the line toward the barrier and my wife screamed at me.

IMG_7919.JPG
IMG_7920.JPG
 
I was on !-40 Friday in the passing lane following a panel truck and could not see around it to watch a quarter mile ahead as I normally do. Traffic was fairly heavy, we were traveling at 65 and I couldn't get around the truck. All of a sudden, EAP slammed on the brakes for a short period and then released them. I looked at the truck in front of me and its break lights had finally come on. I could see it was breaking hard and skidding, and move to the shoulder to avoid hitting the car in front, finally coming to a complete stop as we had come to a construction area slow down. I had plenty of time to come to a controlled stop due to EAP's early detection of the unseen (by me) stopped cars. My brake lights came on well before those of the truck giving the cars behind me longer to react, so I did not get rear ended.

EAP had observed that the cars in front of the truck (which I couldn't see) had come to almost a complete stop, and took appropriate action.
 
I was on !-40 Friday in the passing lane following a panel truck and could not see around it to watch a quarter mile ahead as I normally do. Traffic was fairly heavy, we were traveling at 65 and I couldn't get around the truck. All of a sudden, EAP slammed on the brakes for a short period and then released them. I looked at the truck in front of me and its break lights had finally come on. I could see it was breaking hard and skidding, and move to the shoulder to avoid hitting the car in front, finally coming to a complete stop as we had come to a construction area slow down. I had plenty of time to come to a controlled stop due to EAP's early detection of the unseen (by me) stopped cars. My brake lights came on well before those of the truck giving the cars behind me longer to react, so I did not get rear ended.

EAP had observed that the cars in front of the truck (which I couldn't see) had come to almost a complete stop, and took appropriate action.
Something similar happened on my drive home from viewing the eclipse in Nebraska yesterday. Most traffic was doing 80 mph or so on I-76. I was behind a large pickup truck that I couldn't see around. My car started slowing down all of a sudden and then the large pickup in front of me quickly changed lanes to avoid hitting a much older truck that was probably doing close to 60 mph. Nice to know it is still detecting two cars ahead, even if it doesn't display both on the IC.
 
  • Informative
Reactions: SlicedBr3ad
It really is quite unusable now. I was using it quite regularly in May/June, not exactly sure at which update but things have definitely gotten worse. Car darts towards the sides too often, its dangerously close to the edge of the lanes, especially not seeing cars right next to it. I have not engaged Auto-steer in almost a month now and feel its very unsafe.

Not quite sure how or why things went south but there is seemingly no action being taken on getting it sorted out. All because of the 3 I suppose, I do hope the focus on the catch up updates and parity kicks off sooner than later and we really have all the things promised by Fall/Winter
 
Model S - AP1 (9/30/2016 purchased new). The AP still will swerve from one side of lane to the other intermittently and will occasionally swerve to take an Exit on an un-commanded basis (hopefully, I won't be reported for DUI). The dynamic cruise control is much more reliable but it will occasionally brake for no apparent reason (will watch for over-passes to see if there's a correlation). Conversely, the dynamic cruise control on my wife's Lexus is 100% perfect as is the Blind spot monitor (which doesn't work at all in the MS - although it was a feature on the web site at time of purchase). Maybe there's hope that these problems will be fixed in a future FW update along with the web browser, homelink, etc..
 
FWIW my AP2 works great on my commute, even in the construction crossover lanes.

I'm guessing that Teslas parting with their former head of AI and hiring of Karpathy has something to do with delays in development. I know how hard it can be to find and keep just regular, but good software developers so I can be patient with them as they get going again. It must be super frustrating to have progress in this area delayed though. Also dangerous to the company, for all the reasons posted here
 
Been thinking about this a long time...my AP1 car has had some scary moments, but is approaching "silky" in most scenarios . But my suggestion to everyone with AP cars is to leave it off and drive your very best, diligently. Before you get fired up about my opinion, read why I think this is best course of action:

As I understand AP now, it's uploading data to AI in the cloud, parsing the data, sending data to the cars to improve their accuracy and behavior. Given that (at least for AP2 cars), which is the better way for us to behave? Scenario A: If we allow EAP to do it's thing (it makes its occasional misjudgments and we jerk the wheel back, accelerate or brake frantically), it only learns that it's done something wrong but it didn't see how to pass through that area correctly. Scenario B: However, if we drive the car unassisted (but EAP is still watching) and pass through a "dangerous" spots like narrow bridges, over oddly painted crosswalks, under bridge overpasses, etc, without incident, the car will learn that behavior and thus behave better in those "trouble" spots for the current MX/MS and 1000s of M3 drivers soon to pass that spot.

So, unless I misunderstand the way Tesla EAP/AI system is functioning, I think we should be teaching the system ourselves, as good experienced drivers, and not allowing EAP in its nascent stages so be figuring it out by itself. We often note that EAP is "like driving with a teenager"...very few of us would set the cruise at 80 for a teenage driver in dense urban freeway traffic and then look down at our phones, would we? In fact, we'd have had our pre-teens watching our driving behavior for years before they get behind the wheel. Then we'd start them slowly, with some off-street training, and THEN ease into public driving, and THEN more complex scenarios. The magical thing about EAP is that it's leapfrogging the human model, but we shouldn't expect it to parse an incredibly complex environment immediately without proper training. I'm guilty too...my de facto behavior has been to set EAP near the speed that I would drive myself, based on how quickly I want to get somewhere, but would I do that with a teenage driver?? No...I'd advise them to go slower, watch them like a hawk, ready steer at any time.

So I think we've been approaching EAP the wrong way, and my suggestion has a broader Tesla community benefit: stop using EAP, but let it watch you drive. This TMC (and TMC community should consider harnessing the wisdom of a crowd of good Tesla drivers to more rapidly improve EAP, so we can all enjoy a rock-solid EAP (and FSD) sooner, and possibly prevent catastrophic incidents). Some will say "I paid for it so I demand that it work now". I paid for it too, understanding fully the caveats that came with it, but now I have an opportunity to help improve the technology just by doing what I was already doing before?!? AFAIK, no other brand owners can say their car gets better as they drive it.

And of note to the OP, my AP1 car has improved substantially since I bought it, and further still since Musk's Silky Smooth comment. I had expected his comment would only apply to AP2 cars, but I'm impressed with the smoothness of the system now. But I don't look away for more then a moment and my hands are always ready to catch that "teenage swerve" FNAR.
 
Been thinking about this a long time...my AP1 car has had some scary moments, but is approaching "silky" in most scenarios . But my suggestion to everyone with AP cars is to leave it off and drive your very best, diligently. Before you get fired up about my opinion, read why I think this is best course of action:

As I understand AP now, it's uploading data to AI in the cloud, parsing the data, sending data to the cars to improve their accuracy and behavior. Given that (at least for AP2 cars), which is the better way for us to behave? Scenario A: If we allow EAP to do it's thing (it makes its occasional misjudgments and we jerk the wheel back, accelerate or brake frantically), it only learns that it's done something wrong but it didn't see how to pass through that area correctly. Scenario B: However, if we drive the car unassisted (but EAP is still watching) and pass through a "dangerous" spots like narrow bridges, over oddly painted crosswalks, under bridge overpasses, etc, without incident, the car will learn that behavior and thus behave better in those "trouble" spots for the current MX/MS and 1000s of M3 drivers soon to pass that spot.

So, unless I misunderstand the way Tesla EAP/AI system is functioning, I think we should be teaching the system ourselves, as good experienced drivers, and not allowing EAP in its nascent stages so be figuring it out by itself. We often note that EAP is "like driving with a teenager"...very few of us would set the cruise at 80 for a teenage driver in dense urban freeway traffic and then look down at our phones, would we? In fact, we'd have had our pre-teens watching our driving behavior for years before they get behind the wheel. Then we'd start them slowly, with some off-street training, and THEN ease into public driving, and THEN more complex scenarios. The magical thing about EAP is that it's leapfrogging the human model, but we shouldn't expect it to parse an incredibly complex environment immediately without proper training. I'm guilty too...my de facto behavior has been to set EAP near the speed that I would drive myself, based on how quickly I want to get somewhere, but would I do that with a teenage driver?? No...I'd advise them to go slower, watch them like a hawk, ready steer at any time.

So I think we've been approaching EAP the wrong way, and my suggestion has a broader Tesla community benefit: stop using EAP, but let it watch you drive. This TMC (and TMC community should consider harnessing the wisdom of a crowd of good Tesla drivers to more rapidly improve EAP, so we can all enjoy a rock-solid EAP (and FSD) sooner, and possibly prevent catastrophic incidents). Some will say "I paid for it so I demand that it work now". I paid for it too, understanding fully the caveats that came with it, but now I have an opportunity to help improve the technology just by doing what I was already doing before?!? AFAIK, no other brand owners can say their car gets better as they drive it.

And of note to the OP, my AP1 car has improved substantially since I bought it, and further still since Musk's Silky Smooth comment. I had expected his comment would only apply to AP2 cars, but I'm impressed with the smoothness of the system now. But I don't look away for more then a moment and my hands are always ready to catch that "teenage swerve" FNAR.
It appears that you have bought into the great apocryphal tale of EAP shadow mode and learning, and you have provided us an apologetic.

I don't particularly find useful your chastisement of my (and others) usage of EAP. I have MANY times passed through areas that EAP under one iteration handles successfully, yet under another it has regressed. Tesla is using us as Alpha testers.
 
Just some good-natured ribbing:

Nothing is more necessary than a needed necessity...

No auto-wipers... major first world problem...

I get where you're coming from though :)

Can one live without autosensing wipers? Certainly.
Should one live without autosensing wipers on a $100,000 2017 car when consumer-priced vehicles (Fords and Chevys) have had autosensing wipers that work flawlessly for over ten years? This answer is so obvious that it practically makes me quake with anger. At this point, it's much more the principle of the thing than the convenience it would provide — i.e., if a luxury-car maker cannot even provide me with a promised pedestrian feature (which I've paid for) that the competitors have been providing routinely for over a decade, why should I trust it to do anything else correctly? Rain-sensing wipers are not a deal breaker in and of themselves, but this ludicrous bush-league result gives one pause about the entire company.