Accident while on EAP...

[Freewilly]

OP, back to you, do you typically keep your hands on the wheel when on AP? Admittedly, I don't always, until it nags me. (Sorry so many questions, the former accident investigator in me is curious on this one).

I do normally keep my hand on the lower part of the wheel.

 

[azred]

That's a great point. As we are speculating at this point since the OP dozed....
OP, do you have a habit of keeping your foot near the accelerator when on AP? I have a habit of keeping mine well off, since I have received the warning when accelerating while AP is enabled. Just curious.

I keep my foot close to quickly adjust to sudden, weird decelerations. Situations like two-way
traffic turning safely far in front of my lane and merges are jolting unless I quickly accelerate. Also accelerations after stops are way too slow without a little help from my right foot.

I'll bet his foot was on the accelerator lightly.

 

[Freewilly]

After reviewing the video a few times, it does look like AP was still active.

Vehicle 1 (the OP's vehicle), was still perfectly centred in his lane around a bend in the road. If AP disengaged then vehicle 1 would have ceased to follow the curve in the road and would have proceeded straight into the guardrail rather than into vehicle 2.

Additionally, since vehicle 1's driver (the OP), claims to have dosed off, AP would have begun beeping like crazy before disengaging, thus arousing the OP from his slumber. AP doesn't just disengage without warning.

From the evidence I can see at this time, this looks like a failure of the AP system. The only thing that could change this determination would be additional data showing that the driver of vehicle 1 hit the accelerator before impact. But I don't believe this to be the case as there is no video evidence showing that vehicle 1 lurched forward from an unexpected accelerator hit.



Sorry.. previous Police and private investigator here. These are observations from my professional opinion. I could be wrong with additional information presented.. but I think my determination is correct.

I'm glad that you and many others feel the same way. I've had this problem before (especially after long hours of work and little sleep the night before) and it's one of the main reason's that led to me being a Tesla owner because I don't want to endanger anyone else on the road. I recently read something from the CDC that states that 1 in 25 people on the road doze off behind the wheel. Hopefully, Tesla and other manufacturers can make this technology more reliable going forward.

 

[whatthe2]

After reviewing the video a few times, it does look like AP was still active.

Vehicle 1 (the OP's vehicle), was still perfectly centred in his lane around a bend in the road. If AP disengaged then vehicle 1 would have ceased to follow the curve in the road and would have proceeded straight into the guardrail rather than into vehicle 2.

Additionally, since vehicle 1's driver (the OP), claims to have dosed off, AP would have begun beeping like crazy before disengaging, thus arousing the OP from his slumber. AP doesn't just disengage without warning.

From the evidence I can see at this time, this looks like a failure of the AP system. The only thing that could change this determination would be additional data showing that the driver of vehicle 1 hit the accelerator before impact. But I don't believe this to be the case as there is no video evidence showing that vehicle 1 lurched forward from an unexpected accelerator hit.



Sorry.. previous Police and private investigator here. These are observations from my professional opinion. I could be wrong with additional information presented.. but I think my determination is correct.

Agree with what you have said. (Former Police Officer and Level III Accident Investigator here as well, FWIW).

OP, curious if you can answer the question if you were using NoAP or just AP? I think it makes a difference if the nag alert would have happened in time or not.

 

[Tam]

...NoAP or just AP...

Upon exiting, Navigation On Autopilot would decrease the speed down to zero mile per hour if there's no intervention from driver.

NoA does make some audio clues as it exits the freeway so it might help to wake up the driver.

Driver then has to press the accelerator in order to switch to Autopilot.

Speed/brakewise, NoA does not help in this case because the car was still in middle the ramp and not at the GPS designated 0 mile per hour location (end of ramp) yet.

 

[fbitz777]

Don't you still get TACC even after AP disengages (due to user not responding to nagging)?
TACC should have slowed the car down by itself.

Unless the foot is on accelerator?

 

[whatthe2]

Upon exiting, Navigation On Autopilot would decrease the speed down to zero mile per hour if there's no intervention from driver.

NoA does make some audio clues as it exits the freeway so it might help to wake up the driver.

Driver then has to press the accelerator in order to switch to Autopilot.

Speed/brakewise, NoA does not help in this case because the car was still in middle the ramp and not at the GPS designated 0 mile per hour location (end of ramp) yet.

Are you sure you still don't get AP after NoAP disengages? I thought that was the case. I'm not near my car so can't test, but could have sworn once NoAP disengages, AP kicks in.

 

[buckerine]

The fact that OP was asleep to me points directly to driver error. There are so many things that could’ve gone wrong that were induced by OP. EAP has never failed to stop for me in two Teslas in the exact same scenario as OP’s. On the other hand, OP easily could’ve accidentally disengaged EAP (maybe by inadvertently pressing on the brake) or accelerated (maybe by inadvertently pressing the accelerator). Or maybe EAP wasn’t even on at the time.

EAP has redundancy built into it in terms of the sensors. It’s not solely relying on radar. There’s also the front facing cameras. And the car in front was not stationary either. It was just pretty normal slow traffic. Having driven thousands of miles on EAP, I just can’t see this being an EAP error.

 

[jkoya]

OP - thanks for sharing your videos in the post. Glad to hear no one was injured. I hope you don't have to wait long to get the needed parts for the repairs.

 

[mswlogo]

This is such a simple case for even TACC or NoA.

If you did indeed doze off I suspect you tapped/grazed the brake or throttle and overroad what the car would have done. No chime would have happened. If you just grazed the brake I would think collision avoidance would kick in but it may not prevent impact. If you did hit the throttle, even the slightest, it would drive right on in.

Be cool if the video could provide a few stats. Autopilot, speed, brake or throttle active. It sure could save a lot of guessing. I’m sure Tesla can sync it up and tell you.

 

[fbitz777]

Something strange:
You car is perfectly centered (as opposed to the one in front which is a little bit off to the right)...hard to do without AP on.

Only scenario; AP is ON but OP has the foot on accelerator.

 

[jeffro01]

Examples like this show very clearly that Tesla is a loooonnnnggg way away from actual FSD... This should have never happened and while I applaud the OP for taking responsibility and ownership, this is one of those select few instances where Tesla should cover all costs. Every supposed driver assistance feature failed, that simply can't happen in the FSD world and it sure as hell shouldn't be happening with the current state of AP.

Unless somehow the OPs foot was on the accelerator peddle, TACC alone should have prevented something like this 100% of the time.

Come on Tesla... You really want people to believe FSD is "right around the corner" and you can't even reliably break in stop and go traffic which is literally the easiest scenario out there? Happy I didn't waste my money on that FSD upgrade while it was discounted, yikes...

Jeff

 

[whatthe2]

Something strange:
You car is perfectly centered (as opposed to the one in front which is a little bit off to the right)...hard to do without AP on.

Only scenario; AP is ON but OP has the foot on accelerator.

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.

 

[mswlogo]

Examples like this show very clearly that Tesla is a loooonnnnggg way away from actual FSD... This should have never happened and while I applaud the OP for taking responsibility and ownership, this is one of those select few instances where Tesla should cover all costs. Every supposed driver assistance feature failed, that simply can't happen in the FSD world and it sure as hell shouldn't be happening with the current state of AP.

Unless somehow the OPs foot was on the accelerator peddle, TACC alone should have prevented something like this 100% of the time.

Come on Tesla... You really want people to believe FSD is "right around the corner" and you can't even reliably break in stop and go traffic which is literally the easiest scenario out there? Happy I didn't waste my money on that FSD upgrade while it was discounted, yikes...

Jeff

I have nothing against the OP.

But you have more faith in someone that was sleeping at the wheel than basic ACC?

There is just no evidence that NoA failed. Car is guilty until proven innocent.

Let’s wait for Tesla report, assuming we get it.

 

[mswlogo]

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.

It should still slow down or stop if the car is “out of control”

 

[whatthe2]

It should still slow down or stop if the car is “out of control”

Not disagreeing, but why would it? If the OP misses the final nag and AP turns off, would the car slow down or stop? I honestly don't know as I've never tried it. And if so, how quickly does that happen? Can't imagine it comes to an abrupt halt.

 

[technoz]

Thanks for sharing glad your ok. Can't wait to hear the verdict from the mothership.

 

[mswlogo]

Not disagreeing, but why would it? If the OP misses the final nag and AP turns off, would the car slow down or stop? I honestly don't know as I've never tried it. And if so, how quickly does that happen? Can't imagine it comes to an abrupt halt.

The car does have collision avoidance. Many cars have it. And many cars apply it differently. Depends partly on the speed. Many systems are trying to minimize bodily injury. So it might only slow down. I know some systems that would do a complete stop even with your foot on the throttle. It is not clear, to me, what the model 3 should do in that exact situation. But if it was “out of control” I would expect it to at least slow down. But, like many systems, if it thinks you are in control (pushing throttle) it will not step in. That is my hunch.

BTW if I recall correctly. You can even have the throttle BELOW set speed and it will obey. I could be wrong about that, but it knows if your foot is “active” on the throttle and can be above or below to override the set speed. It’s only when no pedals are touched it controls speed. I know I’ve resumed TACC and the car speeds up AFTER I remove my foot from the throttle.

 

[whatthe2]

The car does have collision avoidance. Many cars have it. And many cars apply it differently. Depends partly on the speed. Many systems are trying to minimize bodily injury. So it might only slow down. I know some systems that would do a complete stop even with your foot on the throttle. It is not clear, to me, what the model 3 should do in that exact situation. But if it was “out of control” I would expect it to at least slow down. But, like many systems, if it thinks you are in control (pushing throttle) it will not step in. That is my hunch.

Valid point, don't know either. I am thinking AP disengaged just before impact, but why it wouldn't slow down, haven't a clue.

 

[immolated]

After reviewing the video a few times, it does look like AP was still active.

Vehicle 1 (the OP's vehicle), was still perfectly centred in his lane around a bend in the road. If AP disengaged then vehicle 1 would have ceased to follow the curve in the road and would have proceeded straight into the guardrail rather than into vehicle 2.

Additionally, since vehicle 1's driver (the OP), claims to have dosed off, AP would have begun beeping like crazy before disengaging, thus arousing the OP from his slumber. AP doesn't just disengage without warning.

From the evidence I can see at this time, this looks like a failure of the AP system. The only thing that could change this determination would be additional data showing that the driver of vehicle 1 hit the accelerator before impact. But I don't believe this to be the case as there is no video evidence showing that vehicle 1 lurched forward from an unexpected accelerator hit.



Sorry.. previous Police and private investigator here. These are observations from my professional opinion. I could be wrong with additional information presented.. but I think my determination is correct.

You are surely mistaken as I'm sure the logs will show. The chances of him dozing off and touching the accelerator are FAR greater than the chances AP failed, AEB failed, and TACC failed. All you have to do is touch the accelerator a few millimeters for the message that pops up saying "Cruise will not brake", but it won't be enough to accelerate faster than the speed TACC was set to.