Phantom Braking

[sleepydoc]

Maybe it matters only from the perspective that, if there were documented accidents (or more of them?), Tesla would have urgent incentive to fix the issue.

One would think NHTSA would be all over this if there were many documented accidents, and apparently they are not. Tesla PB has been on NHTSA's radar now regarding this since February 16 of this year. 2022 TESLA MODEL Y 5-SEAT SUV RWD | NHTSA

It remains an open investigation. Why? Other issues, such as the heating/cooling system problem, were addressed within a few weeks.

Yeah, I think you're right about accidents creating more urgency. The other issue is that there are actually 2 sets of problems - false AEB activations and the random 5-50 MPH slowdowns. Both of which are generally termed 'phantom braking' but they're quite different in terms of frequency, severity and the degree of danger they impart. Other companies have also had issues with false AEB activations and I suspect it's to some degree a result of trying to make more sensitive AEB systems; decreasing false negatives increases the false positives. The NHTSA has investigated false AEB activations with other companies and I suspect this is their focus with Tesla.

Non-AEB PB events are much less severe and less dangerous but also orders of magnitude more common. I haven't had any accidents but I've had drivers get quite angry with me when TACC essentially brake checks them. I can't say as I blame them - I'd be mad, too. Things it traffic is light PB probably doesn't cause problems beyond being annoying, but if traffic is heavier it can easily cause a downstream accident. A Tesla randomly slows down when no one's expecting it, the car behind it brakes a bit late and hard because it wasn't expecting it and so on. The driver 4 cars back was following too close and looking down at the radio and ends up getting in a fender bender. That will never show up on any statistics but it's a very real possibility.

 

[sleepydoc]

So far I am good with my S as I am keeping the version of code I have now that keeps radar enabled. I may not ever upgrade unless there is some issue that is bad enough to resolve and a later code update fixes it. It is the car I use mostly on the highways.

Our 3 is the city runabout car so really no need for it other than watching my speed in the car on the 30 mph zones. Most of the time the family is in the car in that situation and my wife watches the speedo like a hawk. I also set up the warning chime to ding if the speed is exceeded by enough. I do wish you could adjust the volume on it. Maybe you can and I just haven't found a solution.

I'm like you - I use cruise control constantly in town as a way to avoid speeding. I find it even more necessary with my MY because it's so easy to speed. Over time I've developed the habit of keeping my foot perched above the accelerator to counter PB events. (the fact that I have to do that is telling in and of itself.) Of course the danger here is that when I feel the car slow down I tend to reflexively hit the accelerator and I worry that some day I'm going to hit something the one time the car was actually slowing down for something real.

I will say that with the latest software update (2022.20.17, FSD 10.69.2.2) I've noticed a dramatic improvement such that my foot is actually starting to relax.

 

[Knightshade]

I lived it enough times to know that if I didn't accelerate, I would have been rear ended.

Except nobody- ever- can cite even one such accident happening

So no, you don't "know" that-- you imagine that without any evidence to support it (and quite a lot to contradict it)

Just like my friend how used to tell me driving while stoned wasn't an issue because he was never in an accident... Until he was.

Except "the anecdote of one stoned dude" is not evidence.

The lack of ANY such accidents across millions of Teslas over years kind of is.

Why does this matter? If there’s no accidents it’s not a problem? That’s a pretty low bar.

If there's no accidents it's not a SAFETY problem. It's not DANGEROUS as some folks have said.

Which was the (100% no evidence for it) claims being made. See post I quote above for an example.

It might well be a comfort/driver/other driver annoyance problem- but that's vastly different than the claims to which I was responding.

As you note- other makers systems have had actual NHTSA recalls and class actions over false braking that actually caused accidents--- that isn't this to my knowledge. So the claim Teslas system is somehow the worst- despite actual evidence proving otherwise- continues to baffle.

 

[DrGriz]

Except nobody- ever- can cite even one such accident happening

So no, you don't "know" that-- you imagine that without any evidence to support it (and quite a lot to contradict it)



Except "the anecdote of one stoned dude" is not evidence.

The lack of ANY such accidents across millions of Teslas over years kind of is.





If there's no accidents it's not a SAFETY problem. It's not DANGEROUS as some folks have said.

Which was the (100% no evidence for it) claims being made. See post I quote above for an example.

It might well be a comfort/driver/other driver annoyance problem- but that's vastly different than the claims to which I was responding.

As you note- other makers systems have had actual NHTSA recalls and class actions over false braking that actually caused accidents--- that isn't this to my knowledge. So the claim Teslas system is somehow the worst- despite actual evidence proving otherwise- continues to baffle.

The induction of terror "this is going to cause an accident!" is far more (infinitely more?) common than actual events. And I think that is very interesting. Maybe the software "knows" what it can get away with in a situation and it is well beyond our comfort zone. That is to say, it is also actively avoiding a rear-ending event.

A Tesla driver is more likely to get killed from another driver's road rage than they are from being rear-ended due to PB.

 

[legendsk]

The induction of terror "this is going to cause an accident!" is far more (infinitely more?) common than actual events. And I think that is very interesting. Maybe the software "knows" what it can get away with in a situation and it is well beyond our comfort zone. That is to say, it is also actively avoiding a rear-ending event.

A Tesla driver is more likely to get killed from another driver's road rage than they are from being rear-ended due to PB.

Absolutely! It's annoying and a little worrisome, but AEB is the actually dangerous issue and that hasn't been a problem with Teslas AFAIK, certainly not for me. Thinking about this in perspective makes me feel much better. My passengers and I can just laugh at the car's foibles and carry on.

 

[sleepydoc]

If there's no accidents it's not a SAFETY problem. It's not DANGEROUS as some folks have said.

except by writing your post you are essentially stating "it's not a problem unless it's caused an accident." Not only is that wrong, it's incredibly short sighted. Accident causation is an incredibly low bar to reach. I would hope Tesla (and any other automaker) would strive to do better.

 

[Knightshade]

except by writing your post you are essentially stating "it's not a problem unless it's caused an accident."

No, I'm writing it is not a SAFETY problem and was not DANGEROUS if it does not cause accidents.

because it isn't.

And the claim it WAS one was what I was debunking.

I mean- that's literally the words I used and even cited an example of the type of claim I was debunking...

Here again is what I ACTUALLY said.

If there's no accidents it's not a SAFETY problem. It's not DANGEROUS as some folks have said.

So your selective parsing that ignores words right in the middle of the sentence is... odd is the most polite way I can describe it.


There's plenty of things that are a PROBLEM AT ALL in that they're annoying in every car. But the discussion was about SAFETY problems that cause accidents.

This isn't one.

Stop moving goalposts into a field nobody's even playing in please.

 

[DayTrippin]

Except nobody- ever- can cite even one such accident happening

So no, you don't "know" that-- you imagine that without any evidence to support it (and quite a lot to contradict it)

Keep drinking the Tesla kool-aid. As I mentioned, the only reason I wasn't in an accident, is because I had already experienced PB and was prepared for it and accelerated quick enough to prevent. I don't need to actually be in an accident to know it is dangerous. Just because there isn't one on record doesn't mean it hasn't been the cause of one.

If it is such a wonderful thing to occur, why are were any NHTSA complaint opened up? At last count over 750 complaints. Not to mention looks like a class action lawsuit is coming as well.

Number Of Tesla Owners Who Reported Phantom Braking Rises To 758

The NHTSA has now requested additional information from Tesla to get to the bottom of it, after the number of reports has doubled since February.

insideevs.com

What Is Tesla Phantom Braking? Why Is Tesla Getting Sued?

Tesla's phantom braking problem is a frightening and potentially dangerous malfunction that the company needs to fix ASAP.

www.makeuseof.com

Obviously I am not the only one that finds it an issue. It wasn't a major issue until Tesla jerked the radars and went with a not ready for primetime TeslaVision only option because they likely didn't have enough radar units to make their sales quotas due to logistic issues.

 

[DrGriz]

Keep drinking the Tesla kool-aid. As I mentioned, the only reason I wasn't in an accident, is because I had already experienced PB and was prepared for it and accelerated quick enough to prevent. I don't need to actually be in an accident to know it is dangerous. Just because there isn't one on record doesn't mean it hasn't been the cause of one.

If it is such a wonderful thing to occur, why are were any NHTSA complaint opened up? At last count over 750 complaints. Not to mention looks like a class action lawsuit is coming as well.

Number Of Tesla Owners Who Reported Phantom Braking Rises To 758

The NHTSA has now requested additional information from Tesla to get to the bottom of it, after the number of reports has doubled since February.

insideevs.com

What Is Tesla Phantom Braking? Why Is Tesla Getting Sued?

Tesla's phantom braking problem is a frightening and potentially dangerous malfunction that the company needs to fix ASAP.

www.makeuseof.com

Obviously I am not the only one that finds it an issue. It wasn't a major issue until Tesla jerked the radars and went with a not ready for primetime TeslaVision only option because they likely didn't have enough radar units to make their sales quotas due to logistic issues.

Your second link, covering the lawsuit and just over a week old, reiterates that there have not been any accidents ascribed to PB yet.

Interesting that that remains the case.

Also, it makes the offhand comment that Tesla "adds insult to injury" by having increased the price of full self driving by $3K. As if FSB causes PB.

 

[Knightshade]

Keep drinking the Tesla kool-aid.

You're the one who keeps making claims while being unable to offer any evidence whatsoever to support those claims.

So sounds like some projection on your part here about who isn't thinking rationally.

Just because there isn't one on record doesn't mean it hasn't been the cause of one.

Sure just because there's no proof of bigfoot doesn't mean he's real either- right?


There's millions of Teslas on the road today.

And zero examples of accidents as you keep suggestion should be COMMON if it was actually dangerous

If it is such a wonderful thing to occur, why are were any NHTSA complaint opened up? At last count over 750 complaints. Not to mention looks like a class action lawsuit is coming as well.

As noted- zero actual accidents cited in any- not even the lawsuit


So, to check the math here... you're SURE it's SUPER DANGEROUS.... because... 750... out of over 3,000,000 owners... which is 0.025% of owners... IMAGINE it's dangerous and yet not even THOSE guys have ever had an accident because of it

That's your story, and you think that story makes sense. We got this right so far? Because yikes dude.

 

[2101Guy]

You're the one who keeps making claims while being unable to offer any evidence whatsoever to support those claims.

So sounds like some projection on your part here about who isn't thinking rationally.



Sure just because there's no proof of bigfoot doesn't mean he's real either- right?


There's millions of Teslas on the road today.

And zero examples of accidents as you keep suggestion should be COMMON if it was actually dangerous




As noted- zero actual accidents cited in any- not even the lawsuit


So, to check the math here... you're SURE it's SUPER DANGEROUS.... because... 750... out of over 3,000,000 owners... which is 0.025% of owners... IMAGINE it's dangerous and yet not even THOSE guys have ever had an accident because of it

That's your story, and you think that story makes sense. We got this right so far? Because yikes dude.

This is same as those who allege steering wheel weights cause deaths when used with AP.

Show the evidence of that, too.

 

[Knightshade]

This is same as those who allege steering wheel weights cause deaths when used with AP.

Show the evidence of that, too.

Not sure why you're asking me, as I've never claimed that (cept maybe that one dude who set his up as a claymore with a bag of bbs in front of the airbag)

Nice try at whataboutism though!

 

[2101Guy]

Not sure why you're asking me, as I've never claimed that (cept maybe that one dude who set his up as a claymore with a bag of bbs in front of the airbag)

Nice try at whataboutism though!

Where is the question mark in my post?

Use critical thinking

 

[Knightshade]

Where is the question mark in my post?

You said "show evidence of that" in a reply to my post. Specifically evidence of a claim I never made.

If you were telling someone ELSE to provide evidence of something THEY said then you misposted.

Or it was, as I said, rampant and nonsensical whataboutism.

Use critical thinking

I agree one of us needs to, it ain't me though.

 

[rjpjnk]

Keep drinking the Tesla kool-aid. As I mentioned, the only reason I wasn't in an accident, is because I had already experienced PB and was prepared for it and accelerated quick enough to prevent. I don't need to actually be in an accident to know it is dangerous. Just because there isn't one on record doesn't mean it hasn't been the cause of one.

If it is such a wonderful thing to occur, why are were any NHTSA complaint opened up? At last count over 750 complaints. Not to mention looks like a class action lawsuit is coming as well.

Number Of Tesla Owners Who Reported Phantom Braking Rises To 758

The NHTSA has now requested additional information from Tesla to get to the bottom of it, after the number of reports has doubled since February.

insideevs.com

What Is Tesla Phantom Braking? Why Is Tesla Getting Sued?

Tesla's phantom braking problem is a frightening and potentially dangerous malfunction that the company needs to fix ASAP.

www.makeuseof.com

Obviously I am not the only one that finds it an issue. It wasn't a major issue until Tesla jerked the radars and went with a not ready for primetime TeslaVision only option because they likely didn't have enough radar units to make their sales quotas due to logistic issues.

Agreed with being prepared. When I drive my Model Y (No FSD) I am always prepared to intervene manually when PB hits. It happens so frequently that it is the only way I feel safe driving. I have come to understand the kinds of conditions that tend to trigger it, and this has been helpful. It is almost never a problem on major highways, fortunately.

So in my personal experience, to drive a Tesla (at least mine) responsibly and safely one must accept the phantom braking is just part of life. It will occur, and we must always be prepared to intervene to prevent accidents or at least to be courteous to other drives who need to deal with our cars abruptly slowing down for no obvious reason.

Now as for Phantom AEB, that's an entirely different story. Fortunately only happened once.

 

[Bad Horse]

Here is another anecdotal data point:
I just got done with a 800 mile round trip on a variety of roads, although most of the miles were done on a major US interstate (90). We have a 2018 Model 3 with the latest version of FSD Beta.
Worst phantom braking per mile that I've ever experienced since owning, as well as being part of Beta for over a year now. Two lane highways were consistently bad as they usually have been for me for most versions, but what really surprised me was the interstate performance. Very very hard phantom braking on 80 mph interstates are not fun and make the drive pretty miserable. In the display during PB the car would show that a person or car would be right in front of me even through no one was obviously there. Happened most frequently while going over a hill with vehicles out a quarter mile or so in my lane.
I ultimately did a hard reset then camera recalibration at a Supercharger stop. Oddly the phantom braking went totally away once I could use just basic autopilot but not FSD yet. It makes me think that it reverts to a much earlier version of a stable autopilot build while the cameras are still fully recalibrating. I did this two different times on this trip and experienced the same thing. Once FSD was ready (even if unchecked) it went back to the same PB issues.
#pleasegiveusdumbcruisecontrol

This is almost exactly how I would describe my experience, except the "person" or "vehicle" would both times be far, far down the road, quarter mile seems about right-ish

 

[Bad Horse]

Knightshade arguing about other people's experience not being dangerous (and I can tell you whole heartedly I would've been rear ended many a times had it been a busy road) with that xkcd comic clip is either the height of accuracy or lack of self awareness as one will come on this stupid, stupid site

 

[Knightshade]

It's that accuracy one.

Especially since there's no empirical evidence to support the DANGEROUS claim, and 3 million plus Teslas never having an accident from it directly disproving the claim.

Again "feeling" it's dangerous, and then literally every bit of measurable data proving it's not, are different things.

 

[Tronguy]

And, over this past weekend, the spouse and I drove our EAP, but not FSD Model Y to Massachusetts and back, running TACC/LK all the way. I drove about 75% of the time; no phantom braking, period, over about 500 miles, interstates, and some divided-highway local roads. The car was built in fall of 2021 and likely doesn't have RADAR.

Now, the 2018 Model 3 (which we didn't take) has FSD-b on it; and that's all about the testing. You want weird behavior? You want this car. If brakes at the sight of a pedestrian looking at the car (kidding.. a bit), brakes for no good reason on 4-lane roads and interstates, and You Don't Want To Know how it handles turns. (Actually, mostly OK, but, depending upon the day, Whoa Nellie!). But that's in FSD-b land, where one expects oddball behavior.

 

[2101Guy]

Again "feeling" it's dangerous, and then literally every bit of measurable data proving it's not, are different things.

Agreed. Same with steering wheel weights.

Someone needs to PROVE they are dangerous