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.
Couple problems with this test vs the crash. One, there's a cut in the video at 37s between when the car is setup with AP set to 0 and when they are moving at 17mph. What they didn't show here is that you also have to scroll the set speed up and then hit the accelerator to make the car move from 0 (as per other video demonstrating this). Hitting the accelerator doesn't make this impossible, it's just another hurtle that would need to be circumvented. Second problem is they didn't slow how fast the car will accelerate in 400 ft (distance from cul de sac to crash) from a standstill. From what other people have said on this forum, after 400ft, the car will likely not be going tremendously fast.
Maybe they left out a part so it's not a complete "how to" guide. Because people might copy them and try it themselves.

Good thing they didn't get a NoA version and show viewers how to program a route and enable the car to navigate itself without anyone in the car. A few more defeat devices might be needed for that one.

They didn't try to connect their test to the 400ft crash and it was a different model so the time to 400ft wasn't necessary.
 
So here's a variant on the most popular theory of what happened:

Owner backs out driveway (maybe garage) with friend in the passenger seat. Owner then hops out of the car and since it's closer, gets in the backseat (or maybe he wanted to experience a launch from the back seat). Passenger friend gets in the drivers seat, and proceeds to launches the car. The car rockets away and the friend is not ready for the acceleration/g forces/curve in the road. Remember, the 2019 Tesla Model S performance is the fastest accelerating cars in the world (or second), the car is INSANE. The driving friend can't negotiate the massive acceleration/speed + curve, and crashes. The driver's door is blocked by a tree, or bend/damanged and won't open. Driver friend then moves to the passenger seat tries that door, unsuccessfully. Owner in backseat is either unconscious, or can't get out due to 12V failure. Smoke/fire overwhelm them both.

This theory has no one switching from front to back seat. It would also seem more likely the friend crashes than the owner as the friend is unfamiliar with the massive acceleration.
 
As compared to any normal automatic transmission vehicle:
Open door
Stretch leg and place foot on brake
Reach in, start engine, and shift to any non-park gear (neutral bring the least interesting unless on a slope)
Remove yourself from vehicle and shut door (bonus points if you lock it first)

No weighted chain need. No need to stay in car. No attempt by the vehicle to stay in a lane.
I thought this was the std procedure used in so many movies that show how to put a body in a car and make it fall from a hill and burst into flames ?!
 
  • Like
Reactions: mrElbe and mongo
I hope they figure out exactly what impact forces and where the force was applied to cause the battery to burn like that. Then I hope Tesla recalls all model S's to fix that problem. I'm afraid to crash into a tree now.

Everything comes down to what the crash speed was, and if the occupants were buckled in.

If the crash speed was low it brings up the possibility that they were doing the "hey, look at the car driving itself with no one in the driver seat" stunt. But, where the person who activated it didn't realize he activated TACC and not AP. So the car goes running off the road, and hits the tree where he's not in a good position to stop it. During the crash they both get injured due to the lack of wearing a seatbelt. Some aspect of the crash causes a fire somewhere, and eventually the car burns up.

If the crash speed was high then its the same thing that's happened in previous high speed crashes where the occupants got trapped, and didn't have time or the physical capability of getting out of the car.

In both cases what's absolutely critical is to know how to get out of your car.

So if anyone is really scared of hitting a tree they should get a window breaker, and they should make sure they're knowledgeable about how their car doors work with zero power for them. To practice getting out of the car.

Regardless of how this crash happened its very likely the two occupants died because they were unable to get out.
 
  • Like
Reactions: jsmay311
Couldn't have been done in this case as the road had no lane lines and AP would not have ever initiated. But enlightening (and scary) nonetheless. Tesla needs to up their game with regard to driver monitoring. I really do hope that this video forces them to.
I don’t see the point of more monitoring. Any car can be made to drive itself it do something stupid. Lock the wheel and put a brick on the accelerator. Move to a passenger seat-enjoy.
We are adults, paying attention to the road is important but at what point you are responsible for your own actions. I think as an adult if you want to do something (even if it’s dangerous) you should be able to do it as long as in no way you cause issues/harm to others.

I feel sad for what happened: death and family suffering. I hope they get time to heal.
 
Everything comes down to what the crash speed was, and if the occupants were buckled in.

If the crash speed was low it brings up the possibility that they were doing the "hey, look at the car driving itself with no one in the driver seat" stunt. But, where the person who activated it didn't realize he activated TACC and not AP. So the car goes running off the road, and hits the tree where he's not in a good position to stop it. During the crash they both get injured due to the lack of wearing a seatbelt. Some aspect of the crash causes a fire somewhere, and eventually the car burns up.
In an emergency, I think the front passenger or even a rear seat passenger (unbuckled) would be able to push the button located at the end of the gear stalk and engage the emergency brakes.
 
What I don’t get is that the two men involved in this tragedy were mature professionals. I have trouble believing that they would try a stunt like this, especially since all the car had was cruise control and auto steer. If it had the beta FSD then maybe but even then, I doubt it would be to punch the car and see what it does. The only thing I can understand is the typical wow demo of the car’s awesome acceleration that the kid in all of us can’t resist. Unfortunately, the car also has a stunning zero to big trouble time.
 
I don’t see the point of more monitoring. Any car can be made to drive itself it do something stupid. Lock the wheel and put a brick on the accelerator. Move to a passenger seat-enjoy.
We are adults, paying attention to the road is important but at what point you are responsible for your own actions. I think as an adult if you want to do something (even if it’s dangerous) you should be able to do it as long as in no way you cause issues/harm to others.

I feel sad for what happened: death and family suffering. I hope they get time to heal.

I don't have any issue with adding the requirement that the seat has to be occupied before TACC/AP will engage. I think that just makes sense, and removes what I'd call casual abuse. So it forces anyone really trying to hack the system to go through a bit more effort. Now I'm not an advocate of immediately slowing down the car the second the seat monitor says a person isn't there. People adjust themselves quite a bit on a long trip, and this would be just be annoying.

Now I am a huge advocate for moving from the torque sensor to a proper driver monitoring setup. This is not because the torque sensor is easy to defeat, but the torque sensor is annoying. Plus its ill-suited for the job for FSD on City Streets. My belief is Tesla is already planning on augmenting the torque sensor with the interior camera in the Model 3/Y, and the new S/X.

European regulation will also require proper driver monitoring. So its pretty much inevitable for any new L2/L3 vehicle.
 
So if anyone is really scared of hitting a tree they should get a window breaker, and they should make sure they're knowledgeable about how their car doors work with zero power for them. To practice getting out of the car.
Also know if you have tempered glass or laminated glass and which windows have which type. Laminated side windows are going to be harder to escape from.
 
  • Like
Reactions: S4WRXTTCS
In an emergency, I think the front passenger or even a rear seat passenger (unbuckled) would be able to push the button located at the end of the gear stalk and engage the emergency brakes.

I agree in terms of capability, but people often fail moments.

I've had countless times in my life where I failed the moment when that moment required something I wasn't prepared for.

Back in the old days when the emergency brake was a lever it was pretty easy, but now days its always some silly button somewhere.
 
No, but you can make it a little harder to be a clown. You should not be able to defeat the driver monitoring with a wheel weight. And the seat weight sensor going to zero should immediately sound alarms, hit the flashers, and bring the vehicle to a controlled stop.
So, then, what should be able to defeat the driver monitoring system in a "safe" car? :)
 
I agree in terms of capability, but people often fail moments.

I've had countless times in my life where I failed the moment when that moment required something I wasn't prepared for.

Back in the old days when the emergency brake was a lever it was pretty easy, but now days its always some silly button somewhere.
Yes, it would definitely require knowledge of the button, quick reflexes, calm wits and perhaps some good dexterity to be able to push the button in time.
 
I don't have any issue with adding the requirement that the seat has to be occupied before TACC/AP will engage. I think that just makes sense, and removes what I'd call casual abuse. So it forces anyone really trying to hack the system to go through a bit more effort. Now I'm not an advocate of immediately slowing down the car the second the seat monitor says a person isn't there. People adjust themselves quite a bit on a long trip, and this would be just be annoying.

Now I am a huge advocate for moving from the torque sensor to a proper driver monitoring setup. This is not because the torque sensor is easy to defeat, but the torque sensor is annoying. Plus its ill-suited for the job for FSD on City Streets. My belief is Tesla is already planning on augmenting the torque sensor with the interior camera in the Model 3/Y, and the new S/X.

European regulation will also require proper driver monitoring. So its pretty much inevitable for any new L2/L3 vehicle.
Having a weight sensor is not a bad idea, but if you are going though the trouble as in CR video of adding the weight to the wheel, changing the speed to 0, leaving the seatbelt, moving to passenger seat. It’s quite logical to bring a bagpack with you and put it on the driver seat to replica the weight.

I agree wheel torque move is annoying.
 
  • Like
Reactions: DaveG_NJ
Everything comes down to what the crash speed was, and if the occupants were buckled in.

If the crash speed was low it brings up the possibility that they were doing the "hey, look at the car driving itself with no one in the driver seat" stunt. But, where the person who activated it didn't realize he activated TACC and not AP. So the car goes running off the road, and hits the tree where he's not in a good position to stop it. During the crash they both get injured due to the lack of wearing a seatbelt. Some aspect of the crash causes a fire somewhere, and eventually the car burns up.

If the crash speed was high then its the same thing that's happened in previous high speed crashes where the occupants got trapped, and didn't have time or the physical capability of getting out of the car.

In both cases what's absolutely critical is to know how to get out of your car.

So if anyone is really scared of hitting a tree they should get a window breaker, and they should make sure they're knowledgeable about how their car doors work with zero power for them. To practice getting out of the car.

Regardless of how this crash happened its very likely the two occupants died because they were unable to get out.
I put the thing about hitting a tree to be funny. A couple of you guys took it seriously. 😳 I'm actually scared of rapid deceleration. 😩

But I am really concerned that the battery should not explode/burn. If a passenger can survive the crash and try to get out... the battery should survive too without catching fire.
 
  • Disagree
Reactions: qdeathstar
Consumer Reports just posted a video of how it could have been done (not an S, though)....

The correct headline for that is ..

"You can hack Tesla to drive without anyone in driver's seat. Wait till I show you how to use a knife to stab yourself".
 
Now I am a huge advocate for moving from the torque sensor to a proper driver monitoring setup.
...
European regulation will also require proper driver monitoring. So its pretty much inevitable for any new L2/L3 vehicle.
At some point there will be only so much they can monitor. If people keep fooling the system then we may just have to have severe penalties like DUI for the worst cases.
 
Here we go.... the driverless Teslas meet brainless mad printer.

I looked at the bio of the author of the "story".... should we be surprised?

"Rebecca Elliott covers Tesla and car technology for The Wall Street Journal. She previously wrote about oil and gas for the WSJ, based in Houston, and local government and politics for the Houston Chronicle. "