Model X Crash on US-101 (Mountain View, CA)

[mongo]

The answer is no I think. If he was going to Sunnyvale, the lane which he mostly likely choose to travel, assuming he wanted to travel with the least input required, would be the right HOV lane, of all the 5 lanes possible, coming up to the gore point. This would mean no input was required from him approaching the 85 flyover as that lane continues south on 101.

Agree, that path would also mean the car hits on the right side of the barrier first, likely causing it to spin toward the left side of the barrier.

 

[buttershrimp]

This picture showing a driver getting confused and driving in the lane is very helpful, especially if you imagine a truck in front of the model x or something.

 

[Icer]

Really doesn't look like much of any side barrier metal in that shot. Know it collapses onto itself but that looks super compact if so.

It appears the barrier was not there.
Evidently not repaired after the last incident.
Tesla mentions that in this evenings blog post.
What We Know About Last Week's Accident

Also this:

• "Our data shows that Tesla owners have driven this same stretch of highway with Autopilot engaged roughly 85,000 times since Autopilot was first rolled out in 2015 and roughly 20,000 times since just the beginning of the year, and there has never been an accident that we know of. There are over 200 successful Autopilot trips per day on this exact stretch of road."

It is not mentioned if AP was on or off.

 

[buttershrimp]

The answer is no I think. If he was going to Sunnyvale, the lane which he mostly likely choose to travel, assuming he wanted to travel with the least input required, would be the right HOV lane, of all the 5 lanes possible, coming up to the gore point. This would mean no input was required from him approaching the 85 flyover as that lane continues south on 101.

Also, since I don't know this area, he wouldn't have been the carpool lane correct? Or are there any exceptions for EV's or time of day, etc?

 

[TEG]

Is this the path that is most likely? View attachment 289797

Something like that, but I am just guessing exactly what happened.
I am a bit fixated on the final resting position of the windshield.
How did it get so far to the right? Also the front motor is far to the right.
It makes me think the vehicle was headed from left to right on impact.

 

[buttershrimp]

Something like that, but I am just guessing exactly what happened.
I am a bit fixated on the final resting position of the windshield.
How did it get so far to the right? Also the front motor is far to the right.
It makes me think the vehicle was headed from left to right on impact.

Since the car flipped and landed upright, is it possible that these items were thrown upon rotation?

 

[Bubslug]

I disagree with your post (with respect... I could be very wrong).... I disagree with the 2nd premise... I don't think AP would have taken that angle.... if you look at my video, watch towards the end.... the 2nd scenario... this is what I think happened, but perhaps with a vehicle in front of him, or TACC engaged.

Whether you disagree or not, the wisest course of action, until you and other Tesla drivers know different, would be to assume the scenario I and others have proposed, i.e. AP failure in routine freeway driving, has a high probability of having occurred here, and operate your car accordingly.

 

[TaoJones]

...Please take a look. I base this on my experience of trying to get autopilot to fail in various scenarios. I could be wrong, but I don't think many people can guess much better than me except maybe @TaoJones or some other autopilot addict.

Hmmm? In the face of a tragic situation, this thread, overall, has been incredibly helpful, based in no small measure upon the stellar analysis of @TEG and others. Well done, all.

With that said, having driven that stretch a time or two (thousand) as have many here, I would just add as a reminder with regard to the driver assist features (Autopilot/AP/TACC/AS) that all it takes is one tap of the brake and AP is off. Who's to say yet that TACC was immediately reapplied or if there was a delay in also establishing AS *in the specific conditions at the time*? Then there's the discussion of automatic braking affected by/disabled if one taps/depresses the accelerator pedal - and in momentarily blinding sunlight, well...

Point being, as noted above, we just don't and can't know yet with regard to the circumstances of the accident (pending whatever Tesla is able to recover). I would extend that to any extrapolations from driving in similar conditions. Whatever one's version of AP might do is also going to be subject to adjacent traffic. If the driver had the latest firmware version, for all we know, he got the red hands "Take the wheel now" message and that, too, means AP was then disengaged.

Of all of the issues at hand (including but not limited to road conditions, signage, barrier conditions/neglect, and all of the other things that could have contributed to the accident, not to mention how the vehicle disassembled itself upon impact, and associated discussion of how anyone could possibly survive a crash of that nature (although passing later at hospital, may he rest in peace)), I don't know that the question of whether AP was engaged or not makes it into my top 20 list of concerns.

After reading through and thinking about all of the excellent contributions in this thread, my gut tells me that AP was not engaged at the time of impact, but my $0.02 in this case is arguably not even worth that much.

Lastly, I agree that it sounds like the feds are only interested in how the vehicle reacted to impact and the fire thereafter. Which is disappointing, given that you'd think they'd be interested in identifying factors that contributed to the accident in order to remediate those as needed.

 

[TEG]

Also, since I don't know this area, he wouldn't have been the carpool lane correct? Or are there any exceptions for EV's or time of day, etc?

The X has/had carpool sticker(s). It was qualified to be in either carpool lane even with only a lone driver.
There are time restrictions for when the carpool lane is enforced. I think the accident may have happened just about the time the carpool restriction was expiring.

Well technically, there was only 1 of 3 carpool stickers mounted. And only the mini-one.
So the vehicle didn't look in full compliance with the sticker guidelines, but I don't think that is really relevant.

The carpool lane opens up for any vehicles starting at 9AM.

 

[mongo]

Something like that, but I am just guessing exactly what happened.
I am a bit fixated on the final resting position of the windshield.
How did it get so far to the right? Also the front motor is far to the right.
It makes me think the vehicle was headed from left to right on impact.

Possibly:

• Traveling left to right
• Left side hits barricade first
• Adds counterclockwise rotation to lateral movement
• Center of pack hits right side of barrier
• Vehicle pivots around pack contract point
• Vehicle tips up on nose and then rotates in the air 270 with center of mass going at a 30-45 degree angle straight/right
• Windshield popped loose due to deformation and is thrown off as car spins
  
• Lands facing to the right
• Gets hit by other car
• Spins up against center divider

 

[TEG]

Possibly:

• Traveling left to right
• Left side hits barricade first
• Adds counterclockwise rotation to lateral movement
• Center of pack hits right side of barrier
• Vehicle pivots around pack contract point
• Vehicle tips up on nose and then rotates in the air 270 with center of mass going at a 30-45 degree angle straight/right
• Windshield popped loose due to deformation and is thrown off as car spins
  
• Lands facing to the right
• Gets hit by other car
• Spins up against center divider

That is about what I was thinking.

 

[buttershrimp]

This helicopter video makes me think he was going from HOV to 101... and the debris field being strewn to the right in that scenario, he would have likely been familiar with the HOV lane from being in the HOV lane and then exited manually, similar to my video at the 30 second mark.... if a truck or large vehicle was in front of him, or if someone was tailgating him and distracting him from his normal lane change timing, it's possible that he would have initiated the change late and not seen the barrier.

 

[Tam]

I'm not sure we are talking about the same thing. I was talking about Traffic Aware Cruise Control which I could see him having on and set to maintain distance between cars. We use TACC often on the highway but not all the other features.

Ok. I see. Technically, TACC is a subset of Autopilot but the behavior is the same: Sometimes it works, other times it doesn't.

TACC uses camera and radar.

After the Florida crash, Tesla wants RADAR to be the primary sensor because for that crash, RADAR did detect the big gigantic trailer tractor fine so it wanted to confirm with the camera.

The camera couldn't recognize the trailer tractor because of its white color paint blended into the bright background so it's clear as far as the camera could perceive: No obstacles.

The RADAR believed the camera's assessment and thought that must be a stationary road sign above so no brakes were issued.

Tesla said that it tweaked the RADAR so it should work better now.

Well, before this incidence, the RADAR allowed the system to slam into a stationary big gigantic red truck at 65 MPH but the driver is fine. So NTSB is sending a team to investigate why the system didn't brake.

I am glad that TACC has been working flawlessly for you but not so if you see all those reports of mishaps.

Others have explained that Tesla RADAR is not designed to brake for stationary objects. If an object registers a speed, it must be an obstacle but if an object is registered as 0 speed, it must be a stationary object.

The trick for RADAR is to discriminate which ones are threatening objects and which ones are not.

That's why Tesla will have TeslaVision from cameras to help RADAR out. But I am not sure when TeslaVsion will be activated!

 

[mongo]

This helicopter video makes me think he was going from HOV to 101... and the debris field being strewn to the right in that scenario, he would have likely been familiar with the HOV lane from being in the HOV lane and then exited manually, similar to my video at the 30 second mark.... if a truck or large vehicle was in front of him, or if someone was tailgating him and distracting him from his normal lane change timing, it's possible that he would have initiated the change late and not seen the barrier.

Seeing this video again, I notice what looks like a deep dent in the lower right side of the barrier, right where the pack might have hit it. Right wheel may have come off before the spin

 

[mattack4000]

How is there not a single witness, or was everyone on their phone and missed it

 

[Bubslug]

Also, since I don't know this area, he wouldn't have been the carpool lane correct? Or are there any exceptions for EV's or time of day, etc?

There are 2 HOV lanes at the gore point, not one. The left hand HOV lane starts as a fork where the southbound Bayshore Freeway widens out about 3.5 miles north of Freeway 85. Again, following the simplest scenario, the Model X was on Autopilot, and in the HOV lane 4 miles north. When the HOV lane(s) doubled 3.5 miles north, the Model X, stayed in the lane it was in, which at that point became the #2 lane (and still an HOV marked lane). It stayed in the #2 lane until the gore point, where because of a bad sun angle, and worn lane markings at the gore point, it followed the wrong solid white line, into the barrier. The assumptions behind this scenario require no mistakes by the driver (other than inattentiveness), or other drivers, only unfavorable conditions for resolving lane control by AP, which we know was true.

 

[22522]

This helicopter video makes me think he was going from HOV to 101... and the debris field being strewn to the right in that scenario, he would have likely been familiar with the HOV lane from being in the HOV lane and then exited manually, similar to my video at the 30 second mark.... if a truck or large vehicle was in front of him, or if someone was tailgating him and distracting him from his normal lane change timing, it's possible that he would have initiated the change late and not seen the barrier.

I don't see how the debris field ends up so close to the barrier with a left to right transition. Also, the disoriented car seems more like something I would do.

Really paranoid about bridge abutments and don't try anything if I see a rising road in front of me. Being a little late for a left exit might be something I would have tried before this tragic event.

 

[buttershrimp]

I don't see how the debris field ends up so close to the barrier with a left to right transition. Also, the disoriented car seems more like something I would do.

Really paranoid about bridge abutments and don't try anything if I see a rising road in front of me. Being a little late for a left exit might be something I would have tried before this tragic event.

It's his daily commute that made me think this. That he would have intended to stay on the 101 and would have been familiar with the area.

 

[buttershrimp]

There are 2 HOV lanes at the gore point, not one. The left hand HOV lane starts as a fork where the southbound Bayshore Freeway widens out about 3.5 miles north of Freeway 85. Again, following the simplest scenario, the Model X was on Autopilot, and in the HOV lane 4 miles north. When the HOV lane(s) doubled 3.5 miles north, the Model X, stayed in the lane it was in, which at that point became the #2 lane (and still an HOV marked lane). It stayed in the #2 lane until the gore point, where because of a bad sun angle, and worn lane markings at the gore point, it followed the wrong solid white line, into the barrier. The assumptions behind this scenario require no mistakes by the driver (other than inattentiveness), or other drivers, only unfavorable conditions for resolving lane control by AP, which we know was true.

Again, I have no way to know, so no offense if I am wrong.... but this scenario seems the least likely to me. You are describing a situation where AP excels, the sun is not as relevant as people think. Consider an old video of mine.... which is old firmware.... the second half... and watch on mute.

 

[buttershrimp]

The driver was personally known to me. This is a terrible, terrible tragedy and is awful for his family. He leaves behind a wife and two young kids who I'm sure are all in a lot of pain right now.

I am not a Tesla owner, but I found this thread while doing my own research into what could have gone wrong. I've read this entire thread and a lot of people are focusing on the barrier and/or the fire following the crash. Since I learned yesterday that it was Walter involved in the crash, I've been obsessing over the circumstances that could have caused his car to drive straight into the barrier. I'm sure a better barrier could have lessened the damage, but I'm interested in the idea that Autopilot could have contributed to the crash. Knowing Walter, he was not an aggressive or risky driver, and I doubt that this could have happened if he was in full control.

I don't have any more information about the crash than anyone else here. It's been mentioned here before that the driver was an Apple employee. That is correct. I can also confirm that he worked out of Apple's Sunnyvale office, and he was on his way to work at the time of the crash. He should have been continuing straight on 101-S at this junction, so there is no reason he should have been as far left to impact the barrier.

I had a thought/hypothesis yesterday that I would like some opinions on. Not being a Tesla owner personally, I'm not too familiar with exactly how AP works, so you please tell me if this is feasible:

Could it be possible that if AP was engaged, the car was following another vehicle left of the gore point to the 85 ramp? Perhaps the driver realized late that he was in the wrong lane, and signaled the car to change lanes to the right to stay on 101. But, due to the white line markings in the gore point, could the car have moved right into the gore point thinking that it was in a lane? This is total speculation on my part. I'm looking for opinions from those of you who are familiar with the AP technology on the likeliness of this scenario.

The electrek article was helpful... Tesla's role in the scenario is yet unseen, but this picture makes me realize that this event would have likely been survivable had the barrier been in place. For what it's worth, this accident has made me start calling in damaged guard rails around Austin to 311