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

[UPDATED] 2 die in Tesla crash - NHTSA reports driver seat occupied

This site may earn commission on affiliate links.
The Daily Mail also has an aerial image of Dr. Varner’s home also annotating the path. Their article states Dr Varner was in the back seat.


From the aerial photo it looks to me like being at Dr Varner’s home way at the top of the cul de sac around the bend anyone standing outside at the house likely would not been able to see the vacant wooded lot where the car ended up at.
 
Last edited:
  • Like
Reactions: Microterf
The Daily Mail also has an aerial image of Dr. Varner’s home also annotating the path. Their article states Dr Varner was in the back seat.

He says witnesses have told him that the pair were driving on autopilot.

'We have witness statements from people that said they left to test drive the vehicle without a driver and to show the friend how it can drive itself,' he said

🤣🤣🤣🤣
 
  • Like
Reactions: electricar
We don’t know if the wives and BIL went in the house after the guys were ready to drive off. We don’t know they didn’t call 911 too. Doesn’t mean the Commander was lying, just knew what info had been passed on to him. I think the relatives involvement in their deaths is more highly unlikely than not.

Good points. The "drive" was about 5 seconds, so if anyone of the witnesses were outside it's unlikely they don't know what happened. But they could watch the car from inside the house. They could have called 911 too, but it seems to take a relatively long time for the firefighters to learn that it is a car fire. The MOST unlikely part of this story is that Tesla started to drive itself from a complete stop with a person seating in the back. We drive Teslas, we know it is not possible (not talking about Summon here :)) unless there is some kind of software or hardware hack. Which makes the relatives involvement more likely IF Harman tells the truth.
 
Man....I'm tryna get FSD beta in my car. Will Tesla owners stop ACTING UP and getting into wrecks every week and giving us bad publicity please?! I fully understand that these instances are not FSD/Autopilots fault but the media and those who don't have a Tesla don't understand what FSD/Autopilot is. All they hear is "Tesla in a car crash = Autopilot killed people!!!"
 
I'm guessing: Owner lets friend test drive. Unfamiliar driver has a "pedal misapplication" and floors it into a tree. Driver ends up in back seat for some reason - several have been presented here. Police representative is talking out his... butt. It's Occam's Razor in spades.
My thoughts exactly. So much nonsense here and on the internet. This car has red brake calipers so it is a performance car. If they put AP on it would take some crazy and stupid effort.
 
Last edited:
  • Like
Reactions: Microterf
How fast would a Model S be going if autopilot is allowed to accelerate from a standstill in ~380 ft? Would it be fast enough to cause such an accident?

View attachment 655467


Let's assume they bypassed all safety checks--seat-belt buckled behind, autopilot thought the curb was a line, autopilot turned on, then set to 0 mph, car was stopped with autopilot still "on", then increased to 60 mph or something if the system allows (car is now primed, but not moving), driver climbed into the back, and used a yard stick or equivalent to hit the accelerator to kick it off. The car would then accelerate at it's natural pace up to the curve. I wonder if this would even be fast enough to cause such an accident.

I also find it odd anyone would try to demonstrate auotpilot on a road that is only 0.3 miles long with a stop sign at the end. Did he think autopilot would stop at the stopsign? (it won't) Or he would use the yard stick/equivalent to hit the break? Or get his passenger to decrease the set speed on the steering wheel? Maybe he thought he had FSD but didn't? It just seems so far fetched/unlikely.
Your distance is correct. You can add another 50ft to the back of the cul de sac, and another 50ft if you start up the driveway, so it's around 400-500ft. If they actually did enable AP in the method you suggested, it would correlate with the witness statements. Since the two men were talking about AP - as their wives have said - then they may have been playing around, and may have mistakenly believed the car would stop.

Elon's tweet is important as he is trying to control the narrative saying no Autopilot and no FSD. He may be correct but he does say "Data logs recovered so far". Therefore what he tweeted may be incorrect once all the data has been analyzed.

If the car WAS on Autopilot and the car DID have Traffic Light and Stop Sign Control, then the owner was showing it off but it failed to work. That would explain why they conducted the test and why they expected it to stop. It does not excuse the ridiculous nature of the test, that is all on them. I'm not convinced there wasn't someone in the driver's seat despite witness statements - it does seem stupid to be conducting the test from the back - but the people on the scene would know better. You might be right too that they thought they had the FSD package, or had just watched some Youtube video showing something and went off to try it.
 
Your distance is correct. You can add another 50ft to the back of the cul de sac, and another 50ft if you start up the driveway, so it's around 400-500ft. If they actually did enable AP in the method you suggested, it would correlate with the witness statements. Since the two men were talking about AP - as their wives have said - then they may have been playing around, and may have mistakenly believed the car would stop.

Elon's tweet is important as he is trying to control the narrative saying no Autopilot and no FSD. He may be correct but he does say "Data logs recovered so far". Therefore what he tweeted may be incorrect once all the data has been analyzed.

If the car WAS on Autopilot and the car DID have Traffic Light and Stop Sign Control, then the owner was showing it off but it failed to work. That would explain why they conducted the test and why they expected it to stop. It does not excuse the ridiculous nature of the test, that is all on them. I'm not convinced there wasn't someone in the driver's seat despite witness statements - it does seem stupid to be conducting the test from the back - but the people on the scene would know better. You might be right too that they thought they had the FSD package, or had just watched some Youtube video showing something and went off to try it.
Elon said he didn't pay for FSD package. So he wouldn't have the Stop light and stop sign control.
 
I wonder if any of the surrounding houses have surveillance cameras... and if Harman wants to watch the records before making conclusions about Teslas driving themselves without a driver, but he obviously wasn't aware the police cars have cameras when he run the red light... ;)
 
Good points. The "drive" was about 5 seconds, so if anyone of the witnesses were outside it's unlikely they don't know what happened. But they could watch the car from inside the house. They could have called 911 too, but it seems to take a relatively long time for the firefighters to learn that it is a car fire. The MOST unlikely part of this story is that Tesla started to drive itself from a complete stop with a person seating in the back. We drive Teslas, we know it is not possible (not talking about Summon here :)) unless there is some kind of software or hardware hack. Which makes the relatives involvement more likely IF Harman tells the truth.
Nobody needs to be lying. The relatives only said the guys went off for a drive. The brother-in-law only said he thought nobody was in the driver's seat. Therefore none of them were watching and it's a big house so they probably had no idea and didn't hear the crash. The crash would have been a crunching thump slightly around a curve 500ft away. Doubt it was all that noticeable or people would have called 911 for the crash. Most likely the car caught fire due to battery damage rather than any immediate explosion. All the other statements are consistent with the accounts. I don't sense any cover-up or lies.
 
  • Like
Reactions: zecar
How fast would a Model S be going if autopilot is allowed to accelerate from a standstill in ~380 ft? Would it be fast enough to cause such an accident?

The car left the roadway by 380 feet which means, given your measurements, it was likely out of control by 300 ft. How fast can a Model S go in 300 ft? Potentially, given a 0-60 of 2.5s, it could reach 99 MPH by the time it traveled 300 ft. Now, were they doing 100 MPH? Who knows. I would kind of doubt it.

But the road angle there changes by 51 degrees in about 194 feet. A car traveling about 50 MPH would be pushing a little over .8G in that turn so I would think a car (with driver) should still be able to negotiate that turn and stay on the road at 50 MPH. Point being, I suspect we are theoretically in the 50-100 MPH range which means... they were hammering it.

Mike
 
  • Like
Reactions: ZsoZso and Big Earl
Elon said he didn't pay for FSD package. So he wouldn't have the Stop light and stop sign control.
Elon may be wrong too. People under oath later have also said that they "mis-spoke" their initial comments. The car could have come with FSD but it was not recently purchased. If Elon wants to commit the statement to an affidavit and say "The car never had Autopilot engaged and the car did not have any FSD features" then his statement can be taken as fact.

edit: Elon is also wrong in his second point: "Moreover, standard Autopilot would require lane lines to turn on, which this street did not have" as others have shown that AP can be enabled on streets like that one in certain circumstances.
 
Last edited:
The Daily Mail also has an aerial image of Dr. Varner’s home also annotating the path. Their article states Dr Varner was in the back seat.

That aerial image is incorrect. The car didn't make it that far down the road.
The Daily Mail also has an aerial image of Dr. Varner’s home also annotating the path. Their article states Dr Varner was in the back seat.


From the aerial photo it looks to me like being at Dr Varner’s home way at the top of the cul de sac around the bend anyone standing outside at the house likely would not been able to see the vacant wooded lot where the car ended up at.
The Daily Mail wasn't quite right with their areal image. It's even shorter than what they have:
1618949064199.png


1618948957037.png


Note the circled objects (somewhat hard to see) are a tree on top, fire hydrant on the left, and manhole on the right.
1618949149776.png


Here's a picture of the same tree, hydrant and manhole cover with the tire tracks into the woods.
1618948817171.png


The other angle toward the cul de sac:
1618948859717.png
 
How fast would a Model S be going if autopilot is allowed to accelerate from a standstill in ~380 ft? Would it be fast enough to cause such an accident?

View attachment 655467


Let's assume they bypassed all safety checks--seat-belt buckled behind, autopilot thought the curb was a line, autopilot turned on, then set to 0 mph, car was stopped with autopilot still "on", then increased to 60 mph or something if the system allows (car is now primed, but not moving), driver climbed into the back, and used a yard stick or equivalent to hit the accelerator to kick it off. The car would then accelerate at it's natural pace up to the curve. I wonder if this would even be fast enough to cause such an accident.

I also find it odd anyone would try to demonstrate auotpilot on a road that is only 0.3 miles long with a stop sign at the end. Did he think autopilot would stop at the stopsign? (it won't) Or he would use the yard stick/equivalent to hit the break? Or get his passenger to decrease the set speed on the steering wheel? Maybe he thought he had FSD but didn't? It just seems so far fetched/unlikely.
In theory, it could get to over 100 mph before hitting the tree in Ludicrous mode. On TACC/AP it would be much slower of course. I can go try to measure TACC acceleration in my M3 now. The story just doesn't make sense.
 
  • Like
Reactions: byeLT4
edit: Elon is also wrong in his second point: "Moreover, standard Autopilot would require lane lines to turn on, which this street did not have" as others have shown that AP can be enabled on streets like that one in certain circumstances.
Please reproduce the AP STARTING to work in a residential area without lane marks. The only video I saw was from Sergio-the-Ford guy on one very specific street.
 
Of course Tesla’s log reports would be subpoenaed. Happens in accidents by insurance companies, attorneys or policing authorities. When I read Mr. Herman’s statement I kind of laughed thinking he wanted to sound tough. Now does he have the authority to be the investigating agency here? I thought that federal agencies had jurisdiction over local into conducting the investigation. Anyway the logs will get turned over and eventually we’ll see the results published hopefully resolving a lot of questions.
Your distance is correct. You can add another 50ft to the back of the cul de sac, and another 50ft if you start up the driveway, so it's around 400-500ft. If they actually did enable AP in the method you suggested, it would correlate with the witness statements. Since the two men were talking about AP - as their wives have said - then they may have been playing around, and may have mistakenly believed the car would stop.

Elon's tweet is important as he is trying to control the narrative saying no Autopilot and no FSD. He may be correct but he does say "Data logs recovered so far". Therefore what he tweeted may be incorrect once all the data has been analyzed.

If the car WAS on Autopilot and the car DID have Traffic Light and Stop Sign Control, then the owner was showing it off but it failed to work. That would explain why they conducted the test and why they expected it to stop. It does not excuse the ridiculous nature of the test, that is all on them. I'm not convinced there wasn't someone in the driver's seat despite witness statements - it does seem stupid to be conducting the test from the back - but the people on the scene would know better. You might be right too that they thought they had the FSD package, or had just watched some Youtube video showing something and went off to try it.

he did not have fsd so no stop light control but he crashed before making it to the stop sign anyway....

In my experience if you take a turn too fast in autosteer the car will automatically slow down.
 
How fast would a Model S be going if autopilot is allowed to accelerate from a standstill in ~380 ft? Would it be fast enough to cause such an accident?

View attachment 655467


Let's assume they bypassed all safety checks--seat-belt buckled behind, autopilot thought the curb was a line, autopilot turned on, then set to 0 mph, car was stopped with autopilot still "on", then increased to 60 mph or something if the system allows (car is now primed, but not moving), driver climbed into the back, and used a yard stick or equivalent to hit the accelerator to kick it off. The car would then accelerate at it's natural pace up to the curve. I wonder if this would even be fast enough to cause such an accident.

I also find it odd anyone would try to demonstrate auotpilot on a road that is only 0.3 miles long with a stop sign at the end. Did he think autopilot would stop at the stopsign? (it won't) Or he would use the yard stick/equivalent to hit the break? Or get his passenger to decrease the set speed on the steering wheel? Maybe he thought he had FSD but didn't? It just seems so far fetched/unlikely.

Following this train of thought and to show how unlikely use of Autopilot was. Using the uniform acceleration formula d = 1/2 ( Vf + Vi ) × t, say the driver accelerated to 100mph from 0 (probably didn't get up there, but likely over 60mph). From Car and Driver 0-100 takes 6.4 seconds in a P100D. Plug that into the formula and it would cover 470 ft to hit 100mph, pedal to the floor. Even if they only hit 70mph that likely occurred right around the bend in the road. Autopilot doesn't come close to accelerating at that rate.

There was likely loss of control of the car at a high rate of speed, and the driver tried to get out after the accident only to find he couldn't. Climbed into the back seat and succumbed to the flames.