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

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

This site may earn commission on affiliate links.
In your image there is the remnant of the diamond (HOV lane marker) that is well worn down. Someone or something might see that and think it is an arrow to go left, or another white stripe marking the right edge of the lane. Combine that HOV diamond lane marker remnant with the worn down gap in the left hand white lane marker and I could see someone/something drifting from the right HOV lane into the gore area because they were confused into thinking that the lane bent left.

View attachment 290913


View attachment 290920


Here is a video on this from someone. Still does not explain why driver could not stop this.

 
We're most certainly going to have a 72+ page thread on every fatality accident with a Tesla when AP is believed to be involved.

Whether it’s page 7.2, 72, or 720 it still does not change what I stated at the end of 72.

You can discuss forever but the conclusion always stays the same until Tesla or anyone else declares L5 autonomy and the INSURANCE companies agree to the party of liability.
 
This accident hit home for me because for these reasons.
  • I went past the accident site 10-15 min after it happened on the other side of 101. Saw the scene with my own eyes.
  • I ordered my model x few days back, due in June.
  • This route is my everyday commute
  • I was planning to use AP to ease my 1 hr commute.
When I test drove the model x, I already felt this is no where close to what I was thinking.
My first reaction looking at the damage was, this must be autopilot because it is very unusual for somebody to drive into the lane divider at freeway speed and cause that kind of damage

I don’t think I’ll use AP in outside of stop and go traffic. Certainly not above 35-40 mph. And never with family in the car. I think Tesla should restrict the the autopilot on high speeds.

I believe Tesla still recommends using Autopilot on highways instead of city streets in its current state (although I couldn't find anything in a quick search of their website). The one thing you need to keep in mind (as noted multiple times on this thread) is that it's only a driving assistant, and that you still need to maintain control of the vehicle at all times, including being able to take over immediately if it does something unexpected.

There is definitely a period where you will become familiar with its capabilities (and shortcomings), and then realize that after each software update, things could change (such as the behavior during lane changes on the highway). (If anything, looking for potential changes in behavior after software updates are what keep me focused when using it and not getting lulled into a false sense of security!) I would encourage you to try it by yourself (before you include the family) to familiarize yourself, then when you feel comfortable, start using it in more scenarios.

I will try Autopilot whenever it's available, but when I'm on new roads or unfamiliar conditions (like when it claimed to be available on city streets with a center line and parked cars on the right), I tend to have less tolerance for when it drives where I wouldn't drive and I'll steer out of it sooner. Another way to put it--keep Autopilot on a short leash and don't be afraid to take over if you don't like what it's doing. However, Autopilot does a better job than me in the rain, although I still reduce speed as if I'm driving manually. (I've also had one or two occasions on city streets where Autopilot was on, and I thought I was going to have to take over by steering out of it, but it ended up steering exactly where I would have steered--including braking on one occasion--such that it never deactivated until after a maneuver. That's rare, but I've experienced it.)

Now that I've owned a Model X with Autopilot for nearly two years, it's actually painful for me to drive vehicles without Autopilot, especially on long (more than 15-20 minute) trips on highways. As mentioned in other posts, Autopilot greatly reduces fatigue on long road trips because you can spend your time monitoring traffic around the vehicle instead of literally concentrating on keeping the car in the lane. (I guess other vehicles have lane-keeping technology, but I've yet to hear about one that works as well as Autopilot.)

One more thing: AP2 with software version 2018.10.4 is the first release where I now prefer to use AP2 instead of AP1 after using it for real world driving, especially in city driving. There are still a couple specific things that AP1 still does that AP2 can't (like changing lanes in city traffic, and tracking 6 vehicles--two in each lane ahead of you--at all times), but I'm sure AP2 will get those features (and more) in the next year or so. The crash that started this thread was tragic, but I'm still optimistic about the overall future of Autopilot.
 
After seeing all this great user generated reporting, it is pretty darn clear the real issue here is CalTrans. I've always considered that either CalTrans, or a bad actor with paint, could make some very dangerous situations for a car that uses lane recognition for guidance.

CalTrans failed in so many ways.
Worn paint
Seams in road creating confusion
No chevrons or stripes in the gore area
Collapsed Attenuator

The biggest gripe I see against Tesla is an argument that they have not gone far enough to educate their drivers on the limitations of AP.

There is no secret drama here, AP follows lanes, those lanes were iffy in the sunlight, and AP failed and drove into a wall. Would be great if AP can get better somehow, but if I was the jury and had to assign shared liability, it would be 50% CalTrans and 40% Driver, and 10% Tesla.

I am horrified that in a time of budgetary surplus, this intersection that has seen multiple deaths and many many accidents prior to this is in such a state of ill repair. That is the true crime here.
 
ABC7news went to the Mt. View Supercharger to talk to a few people there about whether the accident had shaken their confidence in their Tesla: Tesla drivers still feeling safe despite deadly crash

BTW on the page that linked from their site to the story, ABC had this photo showing the hood and a tire (not sure that's off the Tesla). First time seeing the hood.

View attachment 290934

Is that the left front tire. Can't tell if it says Continental or not. Tread pattern looks close.
Do we know how tall the gore point is to to the top of the hood? Seems kinda strange to have no visible compression damage on the hood at all for such severe crash. Scratching my head.
 
Whether it’s page 7.2, 72, or 720 it still does not change what I stated at the end of 72.

You can discuss forever but the conclusion always stays the same until Tesla or anyone else declares L5 autonomy and the INSURANCE companies agree to the party of liability.

I agree with you in terms of liability when it comes to a L2 car.

The transfer of liability actually starts at L3 though. That's where you can read a book while your car is traveling down the road.

L2 is going to be hotly contested in pretty much every situation where there is an accident. You, and I can say "it's all on the driver" and legally we're correct. But, those that say "It's a difficult ask of a human to remain engaged while the car is doing the driving" do have a good argument. What they seem to be failing on is not realizing that lots of companies do L2 driving, and it's not just Tesla.

L3 will be hotly contested only when the accident happens during the hand off. When the human can't take over because he/she fell asleep.

I don't see much debate about L4, and certainly not with L5.
 
I really hate to say this but I feel like SOME people are actually glad that Mr. Huang lost his life and hope as many people die as possible.

Looking at short sellers and some terrible human beings.

Tesla is in a very unfortunate position of needing to fight a PR war to not be guilty in the court of public opinion.

They already were innocent in the eyes of law and responsibility before Mr. Huang got into his car that day.
 
After seeing all this great user generated reporting, it is pretty darn clear the real issue here is CalTrans. I've always considered that either CalTrans, or a bad actor with paint, could make some very dangerous situations for a car that uses lane recognition for guidance.

CalTrans failed in so many ways.
Worn paint
Seams in road creating confusion
No chevrons or stripes in the gore area
Collapsed Attenuator

The biggest gripe I see against Tesla is an argument that they have not gone far enough to educate their drivers on the limitations of AP.

There is no secret drama here, AP follows lanes, those lanes were iffy in the sunlight, and AP failed and drove into a wall. Would be great if AP can get better somehow, but if I was the jury and had to assign shared liability, it would be 50% CalTrans and 40% Driver, and 10% Tesla.

I am horrified that in a time of budgetary surplus, this intersection that has seen multiple deaths and many many accidents prior to this is in such a state of ill repair. That is the true crime here.
I hope to see a tips type app available at each firmware update, showing video or training driver on AP
 
  • Like
Reactions: e-FTW and timvracer
I really hate to say this but I feel like SOME people are actually glad that Mr. Huang lost his life and hope as many people die as possible.

Looking at short sellers and some terrible human beings.

Tesla is in a very unfortunate position of needing to fight a PR war to not be guilty in the court of public opinion.

They already were innocent in the eyes of law and responsibility before Mr. Huang got into his car that day.
Want to help? Get a tesla.
Telsa going bankrupt? Not today. Not ever.
 
Do we know how tall the gore point is to to the top of the hood? Seems kinda strange to have no visible compression damage on the hood at all for such severe crash. Scratching my head.

Some confusion in terms I think. The gore point is on the ground, it's below the tires and is described/defined by paint lines.

The thing you are trying to refer to is a crash barrier or crash attenuator or impact attenuator.

This TDOT pdf shows a wide variety of those, none are lower than 2 ft tall (apparently the height can be adjusted?). ftp://ftp.dot.state.tx.us/pub/txdot-info/des/crash_cushion_attenuator_chart.pdf
 
Last edited:
It wasn’t Tesla and it wasn’t autopilot and it wasn’t Caltrans for that initial hit. It was 100 percent Mr. Huang.

It doesn’t matter if:

1. Mr. Huang was incapacitated

That is the one thing I don't agree with, if he suffered a medical problem and was incapacitated I wouldn't say it was his fault. (Unless it was something like we was diabetic and wasn't managing his blood sugar properly.)
 
  • Like
Reactions: croman and Icer
I really hate to say this but I feel like SOME people are actually glad that Mr. Huang lost his life and hope as many people die as possible.

Looking at short sellers and some terrible human beings.

Tesla is in a very unfortunate position of needing to fight a PR war to not be guilty in the court of public opinion.

They already were innocent in the eyes of law and responsibility before Mr. Huang got into his car that day.
Glad? Cmon now this is a bit overboard.

The only entity putting Tesla in that unfortunate position is Tesla. When they have a big unveil or the CEO makes grand claims about the product... Overpromise/underdeliver label fits. Doesnt matter if a bunch of guys on the internet are arguing about the meaning of the word 'promise' and splitting hairs.

Take the FSD claims and video of guy riding without touching wheel. What would you think seeing that as a potential buyer?
 
That is the one thing I don't agree with, if he suffered a medical problem and was incapacitated I wouldn't say it was his fault. (Unless it was something like we was diabetic and wasn't managing his blood sugar properly.)

For a young man like Mr. Huang I think that distracted is the more likely scenario but legally I am sure it’s all the same.

Say I am driving behind you and had a stroke that made me lose control. I rear end you.

We can’t go well you pay your deductible and I pay mine. I’m still “responsible”. There’s no mitigating circumstances for this type of scenario.
 
Thanks, great effort! It would be interesting to see what would happen if one reattempted this test with a lead-car that encroached into the gore zone before correcting to the right. I would if AP would follow the lead to the right or continue tracking the lane marker to the barrier?

Well, if you wanted to take it to the extreme (Mythbusters style), have the "lead" vehicle being followed by the Tesla (with a minimum following distance setting of "1") not only obstruct the left lane paint from view at the gore point by driving too far to the left of the lane, but actually have the lead vehicle either (a) perform the "idiot maneuver" where they pass a third vehicle (two vehicles in front of the Tesla) in the US-101 HOV lane using the gore (triangular) area where no one is supposed to drive, or (b) switch to the CA-85 HOV lane across the gore area as if it were a lane (starting the lane change at the gore point).

BTW, scenario (a) would be similar to the crash involving the Model S and the fire truck where the vehicle in front swerved too quickly (out of the way of the crash attenuation barrier) for the Tesla to recognize the obstacle in front of it, except the Tesla would be driving in a gore area that looked like a lane at that point. I wonder if the car two ahead of the Tesla would also need to obscure the left lane paint for the US-101 HOV lane (as if were trying to stop the "lead" car from passing, or just due to careless driving) during these maneuvers.

PLEASE DO NOT TRY THIS IN REAL LIFE. The only way it should be ever tested is with the highway shut down and safety experts driving all vehicles and EMTs present in case of an accident, or simulated on a computer (which only Tesla could do).

Again, we have no way of knowing whether this is what happened during the crash that took Walter's life, so I'm looking forward to the NTSB report--even if it takes a year for the preliminary results.

EDIT: I guess scenario (a) would not provide the "unobstructed view" as mentioned in the Tesla blog, so it may not be very likely. Scenario (b) might, though, if the lane change of the "lead" vehicle to the CA-85 HOV lane happened quickly. For scenario (b), I wonder if a second vehicle obscuring the left lane paint (ahead of the "lead" vehicle changing to CA-85 HOV lane, and two ahead of the Tesla) would also make a difference.
 
Last edited:
I agree with you in terms of liability when it comes to a L2 car.

The transfer of liability actually starts at L3 though. That's where you can read a book while your car is traveling down the road.

L2 is going to be hotly contested in pretty much every situation where there is an accident. You, and I can say "it's all on the driver" and legally we're correct. But, those that say "It's a difficult ask of a human to remain engaged while the car is doing the driving" do have a good argument. What they seem to be failing on is not realizing that lots of companies do L2 driving, and it's not just Tesla.

L3 will be hotly contested only when the accident happens during the hand off. When the human can't take over because he/she fell asleep.

I don't see much debate about L4, and certainly not with L5.

It will be Tesla’s or any other autos liability when they are the insured party.

A very good indicator of Tesla’s confidence in FSD is when they offer insurance for sale themselves. This was mentioned in the past.