Accident while on EAP...

[SMAlset]

I actually see acceleration near the end. That should have been the point that the chimes start announcing end of EoP. Definitely looks like foot on accelerator to me. After all, it had happily been following for awhile. Not enough curvature to lose the sonar, vision, or road position.

It looked to me like there might have been a slight acceleration and maybe that would mark when OP nodded off. Hard to say what your foot would have done during that time. Very glad to hear that both of you were okay and kudos for taking responsibility and sharing this with us. Glad you weren't rear ended. With your hood flying up I'm not sure how you managed to judge where to pull over to the right out of traffic. Learning from others is why we are all here.

Rewatched the video a few times and the Model 3 seems pretty much centered in the lane throughout and seemed to be verified by the side cameras as well. The Forrester however was not centered at that point but to the right, so as it moved to the right on the curve it caused the centered Tesla to hit it more on the left side than centered on its bumper, which I think would have happened instead if the Forrested stayed centered. BTW I noticed at the very beginning of the clip that the speed limit was 55mph on the off ramp.

BTW did you have music playing?

 

[David L]

In my experience, AP is overly cautious about a lead vehicle that's moving out of the lane. It waits until the vehicle has completely cleared the lane before accelerating.

If Autosteer is enabled and the OP completely ignored the nags, the car would have come to a complete stop. The following is from p. 74 of the Model 3 manual. "If you repeatedly ignore hands-on prompts, Autosteer displays the following message and is disabled for the rest of the drive. If you don't resume manual steering, Autosteer sounds a continuous chime, turns on the warning flashers, and slows the vehicle to a complete stop."

Based on my experience with AP, I vote for foot on the accelerator overriding AP braking. I hope that Tesla replies with details of what happened, because I'd like to understand for my own safety when using AP.

OP, thank you for sharing.

 

[jcasetnl]

Or...AP was on when car initially slowed for the car in front, then sped up, but OP missed the nags and it shut off.

If you ignore the nags, I believe the vehicle stops on its own, correct?

 

[MP3Mike]

If you ignore the nags, I believe the vehicle stops on its own, correct?

Unless your foot is on the accelerator then it keeps going, and steering. (Or at least it used to, I haven't seen that tested lately.)

 

[JasJ]

If you ignore the nags, I believe the vehicle stops on its own, correct?

Yes. try it on an empty road. Eventually car stops and hazards go on.

 

[OPRCE]

Tesla is working hard at it but when will it graduate from a beta status is still unknown.

My best guess is FSD on HW3 will lose the nags and become L3 in Jan.2023 at earliest, if regulators are shown that from 90mph it reliably and safely deals with stationary obstacles, plus a wide variety of other common hazard scenarios in a standardised AV Driving Test which remains to be defined. Clearly just quoting some carefully curated statistics will never cut it.

Regarding probable reason for OP's crash, I see this sequence of events:

1. Driver fell asleep in AP, presumably no touching of brakes or accelerator
2. Radar [?] tracking of lead car erroneously merged into the second car ahead, AP thinks it has more space and tries to close the gap
3. AEB failed because lead car was partially offside in curve

Prediction: Tesla will clam up on this one, offering no retrieved data to OP but rather "reward" his exemplary honesty with a cold shoulder.

Thanks for posting this OP and hope you have fully comprehensive insurance!

 

[Fernand]

Not to break this sherlockian train of thought, but much as I admire the OP's confession, what's the current thinking with that old insurance mantra: Never Admit Fault?

Someone I know had a minor road "incident", and here he was, pre-loaded with instructions never to admit fault, though he had winged a passing car. The driver wanted badly for him to at least agree it wasn't her fault, and everything decent in him screamed to agree and tell her it was his fault. But the lawyerly mantra force was strong in him, and although he said he'd pay for the repair out of his pocket, he hemmed and hawed and never admitted fault. I think he was lucky that she worked with him in spite of his being, in her eyes, a jerk. Are insurers still telling everyone never ever to admit fault?

Can we just say "it's beta"?

 

[whatthe2]

Not to break this sherlockian train of thought, but much as I admire the OP's confession, what's the current thinking with that old insurance mantra: Never Admit Fault?

Someone I know had a minor road "incident", and here he was, pre-loaded with instructions never to admit fault, though he had winged a passing car. The driver wanted badly for him to at least agree it wasn't her fault, and everything decent in him screamed to agree and tell her it was his fault. But the lawyerly mantra force was strong in him, and although he said he'd pay for the repair out of his pocket, he hemmed and hawed and never admitted fault. I think he was lucky that she worked with him in spite of his being, in her eyes, a jerk. Are insurers still telling everyone never ever to admit fault?

Can we just say "it's beta"?

I was never (and still am not) a fan of "Don't ever admit fault". If you do something wrong, own up to it like the OP did.

 

[immolated]

I don't see how i'm "surely mistaken".

I reviewed the evidence provided thus far, and the video does not indicate even the slightest increase in speed as would indicate that the driver of vehicle 1 pressed the accelerator.

I could be mistaken.. but.. "surely mistaken"... idk about that.

Do you have any additional insight/ evidence to provide?

Yes, next time you're driving, set the TACC to 40mph, and press the accelerator 1cm. The car does not accelerate, it stays at 40mph. It only starts accelerating if you push the pedal far enough to the normal 40+mph pedal deflection. And as soon as you press the pedal even a few millimeters, you get the warning message "Accelerator pressed, cruise will not brake" (text only not audio)

So the driver had the pedal pressed enough to sustain ~30mph, but not exceed whatever speed he was going. You can even recreate this next time you drive (with TACC on, press the pedal a little bit and approach a stopped car at a light, tell me what happens)

 

[Runt8]

Not to break this sherlockian train of thought, but much as I admire the OP's confession, what's the current thinking with that old insurance mantra: Never Admit Fault?

Someone I know had a minor road "incident", and here he was, pre-loaded with instructions never to admit fault, though he had winged a passing car. The driver wanted badly for him to at least agree it wasn't her fault, and everything decent in him screamed to agree and tell her it was his fault. But the lawyerly mantra force was strong in him, and although he said he'd pay for the repair out of his pocket, he hemmed and hawed and never admitted fault. I think he was lucky that she worked with him in spite of his being, in her eyes, a jerk. Are insurers still telling everyone never ever to admit fault?

Can we just say "it's beta"?

Never really understood this. Insurance is there for when we screw up. If I cause an accident I’d rather admit fault and have my rates go up than weasel out of it and potentially have the innocent party see increased rates.

 

[jsmay311]

If you ignore the nags, I believe the vehicle stops on its own, correct?

Yes. try it on an empty road. Eventually car stops and hazards go on.

That’s what I thought too.

Why so many earlier posts in this thread suggesting it just turns off without slowing the car?

 

[McFlurri]

Yes, next time you're driving, set the TACC to 40mph, and press the accelerator 1cm. The car does not accelerate, it stays at 40mph. It only starts accelerating if you push the pedal far enough to the normal 40+mph pedal deflection. And as soon as you press the pedal even a few millimeters, you get the warning message "Accelerator pressed, cruise will not brake" (text only not audio)

So the driver had the pedal pressed enough to sustain ~30mph, but not exceed whatever speed he was going. You can even recreate this next time you drive (with TACC on, press the pedal a little bit and approach a stopped car at a light, tell me what happens)

Your telling me that while the driver was asleep, he PERFECTLY pressed the accelerator to make the vehicle stay at EXACTLY the same speed as he was travelling before? That seems highly improbable.

I am well aware of the "Accelerator pressed, cruise will not brake" warning. I use it all the time when getting on to a highway.

My concern/ issue I have with you is that you are making statements based off of no factual evidence. "So the driver had the pedal pressed enough to sustain ~30mph". You don't know that. none of us do.

My statements are based off of the evidence provided thus far. You may have a theory, but please then say, "it's my belief that x,y, z happened", rather that saying "the driver did this".

 

[Freewilly]

It looked to me like there might have been a slight acceleration and maybe that would mark when OP nodded off. Hard to say what your foot would have done during that time. Very glad to hear that both of you were okay and kudos for taking responsibility and sharing this with us. Glad you weren't rear ended. With your hood flying up I'm not sure how you managed to judge where to pull over to the right out of traffic. Learning from others is why we are all here.

Rewatched the video a few times and the Model 3 seems pretty much centered in the lane throughout and seemed to be verified by the side cameras as well. The Forrester however was not centered at that point but to the right, so as it moved to the right on the curve it caused the centered Tesla to hit it more on the left side than centered on its bumper, which I think would have happened instead if the Forrested stayed centered. BTW I noticed at the very beginning of the clip that the speed limit was 55mph on the off ramp.

BTW did you have music playing?

I had a podcast on.

 

[Freewilly]

My best guess is FSD on HW3 will lose the nags and become L3 in Jan.2023 at earliest, if regulators are shown that from 90mph it reliably and safely deals with stationary obstacles, plus a wide variety of other common hazard scenarios in a standardised AV Driving Test which remains to be defined. Clearly just quoting some carefully curated statistics will never cut it.

Regarding probable reason for OP's crash, I see this sequence of events:

1. Driver fell asleep in AP, presumably no touching of brakes or accelerator
2. Radar [?] tracking of lead car erroneously merged into the second car ahead, AP thinks it has more space and tries to close the gap
3. AEB failed because lead car was partially offside in curve

Prediction: Tesla will clam up on this one, offering no retrieved data to OP but rather "reward" his exemplary honesty with a cold shoulder.

Thanks for posting this OP and hope you have fully comprehensive insurance!

I do! My deductible is only $200 as well, so it won't hurt too much financially.

 

[TMThree]

It does seem very odd that autopilot didn't stop...

I wonder if it disabled itself and chimed. OP didn't hear it and then it rammed into the other car. As he said, it's weird the car didn't turn on hazards and park instead of doing this..

 

[NCDoc]

Literally this exact same scenario happened to me yesterday. On an off ramp that was going around a curve and the lane was wide like that one. Car in front of me was hugging the inside curve, going slow like in this footage. My car started to accelerate and acted like it was going to try to pass that car on its left, but there wasn't nearly enough room. There was about as much room as in this video. If I hadn't taken over I'm pretty sure it would've hit.

For reference, I have the latest update pushed by mobile service last week (2019.8.3).

 

[OPRCE]

Literally this exact same scenario happened to me yesterday. On an off ramp that was going around a curve and the lane was wide like that one. Car in front of me was hugging the inside curve, going slow like in this footage. My car started to accelerate and acted like it was going to try to pass that car on its left, but there wasn't nearly enough room. There was about as much room as in this video. If I hadn't taken over I'm pretty sure it would've hit.

For reference, I have the latest update pushed by mobile service last week (2019.8.3).

Good evidence that OP was in all probability still in AP when he struck the car in front. Somehow in the curve the radar track got lost or melded with another signature. Not sure why visual recognition would have simultaneously failed though, but maybe it is not even used for TACC/AEB?

 

[Nocturnal]

OP. Tesla will NOT comment or respond to you on this out of fear of legal action. Hopefully they did pull the data and will determine any flaw that may have caused this.

 

[Joelgjr]

Guys - we've been talking about inadequacies of AEB for years over on the Model S forums.

1) autopilot DOES NOT DISENGAGE ITSELF EVER. If the driver ignores the nags, It will either a) bring the car to a complete stop with the flashers on (you can see this on youtube in 1,000 videos) or b) run forever if it thinks your hand is on the wheel. Autopilot NEVER disengages itself, which is the inherent difference between it and other systems in other cars.

Only one thing could have happened - autopilot was on when he hit the car and Tesla Vision failed to see the vehicle. There are no guarantees with AEB.

 

[Runt8]

Guys - we've been talking about inadequacies of AEB for years over on the Model S forums.

1) autopilot DOES NOT DISENGAGE ITSELF EVER. If the driver ignores the nags, It will either a) bring the car to a complete stop with the flashers on (you can see this on youtube in 1,000 videos) or b) run forever if it thinks your hand is on the wheel. Autopilot NEVER disengages itself, which is the inherent difference between it and other systems in other cars.

Yes, definitely agree with this. Autopilot doesn't just turn off.

Only one thing could have happened - autopilot was on when he hit the car and Tesla Vision failed to see the vehicle. There are no guarantees with AEB.

There are several possible things mentioned in this thread that could have occurred. The simplest is that the OP had his foot on the accelerator.