Tesla Puts Blame On Missing Barrier For Deadly Model X Crash

MAR 28 2018 BY ERIC LOVEDAY 86

Late last night, Tesla published a blog post with some details on the recent deadly Model X wreck.

More Info – Video From Immediate Aftermath Of Tesla Model X Crash, NTSB Investigation

According to the automaker, the severity of the crash was largely due to a missing safety barrier on the highway.

Tesla states:

“The reason this crash was so severe is that the crash attenuator, a highway safety barrier which is designed to reduce the impact into a concrete lane divider, had either been removed or crushed in a prior accident without being replaced. The following image shows what the barrier looked like when the crash attenuator was in proper condition, and what it looked like the day prior to the crash, based on dash cam footage from a witness of the accident who commutes daily past this location. We have never seen this level of damage to a Model X in any other crash.”

The automaker is currently working with authorities in an attempt to retrieve logs from the computer within the vehicle, but has been unable to do so as of right now.

Regarding the potential for blame to be placed on Autopilot for this incident, Tesla states:

“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.”

Additional details in Tesla’s blog post below:

What We Know About Last Week’s Accident

The Tesla Team March 27, 2018

We were deeply saddened to learn that the driver of a Model X vehicle involved in an accident last Friday passed away. Safety is at the core of everything we do and every decision we make, so the loss of a life in an accident involving a Tesla vehicle is difficult for all of us. Earlier this week, Tesla proactively reached out to the authorities to offer our assistance in investigating.

While we do not yet know what happened in the moments leading up to the accident, and we do not yet have any idea what caused it, here is what we do know:

Due to the extensive damage caused by the collision, we have not yet been able to retrieve the vehicle’s logs.

We are currently working closely with the authorities to recover the logs from the computer inside the vehicle. Once that happens and the logs have been reviewed, we hope to have a better understanding of what happened.

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.

The reason this crash was so severe is that the crash attenuator, a highway safety barrier which is designed to reduce the impact into a concrete lane divider, had either been removed or crushed in a prior accident without being replaced. The following image shows what the barrier looked like when the crash attenuator was in proper condition, and what it looked like the day prior to the crash, based on dash cam footage from a witness of the accident who commutes daily past this location. We have never seen this level of damage to a Model X in any other crash.

Tesla battery packs are designed so that in the rare circumstance a fire occurs, it spreads slowly so that occupants have plenty of time to get out of the car. According to witnesses, that appears to be what happened here as we understand there were no occupants still in the Model X by the time the fire could have presented a risk. Serious crashes like this can result in fire regardless of the type of car, and Tesla’s billions of miles of actual driving data shows that a gas car in the United States is five times more likely to experience a fire than a Tesla vehicle.

It is worth noting that an independent review completed by the U.S. Government over a year ago found that Autopilot reduces crash rates by 40%. Since then, Autopilot has improved further. That does not mean that it perfectly prevents all accidents — such a standard would be impossible — it simply makes them less likely to occur.

Out of respect for the privacy of our customer and his family, we do not plan to share any additional details until we conclude the investigation.

We would like to extend our deepest sympathies to the family and friends of our customer.

Categories: Crashed EVs, Tesla

Tags: , ,

Leave a Reply

86 Comments on "Tesla Puts Blame On Missing Barrier For Deadly Model X Crash"

newest oldest most voted

Looks like the road crew had time to re-attach the black/yellow hashed signs and clean the scene up from the previous event. One might think that having the extra parts on hand to rebuild the whole barrier wouldn’t be a stretch.

As for the wreck, If you google “car ripped in half in accident” you could spend days scrolling through the list of results.

Sadly, it happens. Too often.

There is only so far you can bend the rules of physics. That short crash barrier would do very little to help.

R.I.P.

One might also start to ask if a barrier is being hit a lot, is there a road design issue at that point?

“That short crash barrier would do very little to help.”

Really? I’d think it would have made all the difference in the world.

Why do you think it wouldn’t have helped much?

If you’re not being sarcastic, you are clearing lacking in basic physics education.

Break it down for me.

20′ of extra cash structure will offer a great substantial reduction in deceleration forces. That reduction would almost certainly saved this guy’s life.

Right? Where is my basic physics education failing us?

Thanks.

It’s how parachutes work. They only slow down the deceleration to acceptable limits. 120 MPH to 0 in an inch and zero time vs. 800 or so feet in about 4 seconds.

See this for info on crash attenuators:
http://workareaprotection.com/attenuators/

You can note that the cost and time required for resetting these attenuators after an accident is pretty minimal. There is no excuse for these not being reset as part of the normal highway cleanup and investigation – immediately after each accident.

For those that argue that this is more expensive, I would argue the contrary. It should be cheaper because you don’t need to send a 2nd crew later to re-close lanes and do the work. And let’s not forget the pending litigation cost to the taxpayers. Resetting crash attenuators immediately as standard policy is a no-brainer.

I think you mean acceleration. Slowing down, or speeding up, are both acceleration, in the realm of physics.

Both are correct. Deceleration is generally used to mean speed is decreasing though that is not mandatory. Both acceleration and deceleration were used in several of the physics texts I had in college and university.

The “short”barrier he is referring to the one in the second picture that looks to be less than 2 feet long. That would do very little in such an impact. The original one in the first picture is 15-20′ long, and would have reduced the impact significantly. No telling if the driver would have still survived, but his odds would have been much better.

Ahhhh. That makes sense. Thanks!

It looks to me, the most recent picture, shows the attenuator is still there and just collapsed from a recent accident and wasn’t reset yet. You can still see the track system and the reflector looks damaged with cones. There is a small possibility it was reset in between the time of the picture and the time of the Tesla collision. I still believe if used properly, these auto pilot measures are safer for the driver. I see too many people that fail to follow instructions all the time. So many people feel rules don’t apply to them so the tech is only as good as the user.

Rebuilding the crash barrier takes time, and that creates a window of tragic misfortune for repeat collisions at the same point of impact. Unfortunately the barrier is nothing more than a concrete wall abutment (a blunt stone knife) with a veneer of aluminum cladding.

This particular crash barrier is designed to be reset easily and quickly. But if you peruse Google Street view data, it’s clear that the barrier spends a lot of time compressed instead of extended.

It can be rebuilt as fast as Caltrans is motivated to rebuild it.

I drove by the intersection yesterday 3/27 (4 days after the accident) and they have already completely replaced the barrier to the original full 20 feet.

I’m sure all the coverage around the accident has motivated them more than usual

The lack of the barrier explains why the accident was so bad. It doesn’t explain why the car crashed into it in the first place.

Good spot to commit suicide…

I am not sure I am missing something, but Tesla’s blog post explains why the car was severely damaged hitting the barrier …. Whether should have been there or not can be discussed and I am sure will be debated. But it’s completely irrelevant to the question WHY the car hit the barrier.

Tesla made a point that their cars with AP made that trip many times … which really leaves two ways to explain it. Either the AP was off and the human driver failed, or AP was on and the AP failed.

Which one is it then?

With the car as incinerated and obliterated as it is, I highly doubt Tesla will be able to get log data out of the computer, which is why they posted “hey, it works every day, all day long, don’t assume because it was our car that autopilot was involved.”
However, I can see news sources foaming at another “autopilot kill” story potential. Vultures!

Right … but, Tesla certainly knows whether the AP was on or not (they don’t need the box for that) … yet they choose to not clearly confirm that.

I am not sure about you, but I sure as hell would not be comforted by the fact that AP handles the piece of road many times a day … yet, this car managed to wreck itself in single vehicle accident …. if I was driving Tesla and used AP.

Vultures? … not sure what they have to do with that … this is about common sense, nothing else.

If they don’t have any computer remains from the car, to pull the logs from, I’m not sure why you think Tesla would know, the AP status, real-time. The AP “phones home” with periodic updates about roads, etc. But I’m pretty sure that’s not real-time streaming.

That said, it would be nice to have the last time stamp and status of the car, when it DID last phone home. If it was in AP 1 minute before the crash, that doesn’t bode well, if the car hadn’t uploaded any data in that last 30 mins or so, then it’s hard to say..

Given the NTSB is involved, there is no way, this sort of details will not eventually get out, and I don’t see any company (Uber, Waymo, or Tesla) outright lying about what data they have. So my sense is we’ll get the real details a few months from now, with the NTSB report.

I was not talking about real time streaming, just a time stamp of AP status change.

AP = on … system phones home with the change … AP=off … system phones home with the change yet again …. would be my blunt expectation.

But, I am no expert, and I understand that perhaps they don’t want to conclusively say either way without BB recovered.

Yet, it doesn’t mean that they could not confirm already whether AP was off or on.

My guess is the logs are NOT offloaded in a “phone home” communications based on real time events, otherwise every change in the car’s status is going to generate a network request, likely a TCP handshake to start a session, etc. – That’s what I mean about “real-time” aspect of the communications, or streaming the event data records as they occur Knowing what I do about data collection, and polling, my gut instinct is that the system generates local log files of “interesting events”, of both the car status, and AP shadow mode (when a driver corrected the system), and these are likely uploaded based on some period (every 10 mins, every half hour, etc), when the car checks in, looks for any updates that it needs, etc. Hence my comment about when it “last checked in” – I honest don’t know that check in interval, but that’s the timestamp I’m curious about relative to the accident… I really, really doubt that each AP On, AP Off, event is uploaded from the car in real-time as these events occur, that creates a scaling nightmare for large numbers of cars talking to their “cloud”.. Bottom line, did the car just do… Read more »

As I recall, Tesla says that its cars do actively “phone home” right after there is any incident which Autopilot interprets as a near-accident. Otherwise, my guess would be that a Tesla car only sends data to Tesla HQ on a set timetable (maybe once or twice an hour?), or when the data buffer is full.

It seems very, very unlikely to me that every Tesla car “phones home” every time someone activates or deactivates AutoSteer. (I think that is what you meant; Tesla Autopilot is always active and in “shadow mode” in cars it’s installed in, even when certain driver assist features are deactivated. You can’t actually turn Autopilot off.)

mxs:

“…Tesla certainly knows whether the AP was on or not (they don’t need the box for that)”

What would lead you to think so?

Tesla cars are not in constant, real-time communication with Tesla’s headquarters at all times. (The bandwidth needed for that would be prohibitive!) Periodically the cars do a wireless (and very probably compressed) “data dump” to Tesla’s data banks when being driven, but even that is only selected info. Anyone who knows much about computer-to-computer communications should understand that there is a very low probability that there was any direct data connection to Tesla Inc. at the time of the accident.

Regardless of whether CALTRAN is liable for not fixing the barrier in a timely manner (seems like they are), the outstanding question is “Was AP active at the time of the crash?”. Any mention of that is conspicuously absent from Tesla’s statement. We need to know HOW the driver ended up crashing into the barrier.

Agree on all points, which is why I think it’s a bad move for Tesla to point out how many times/day AP drives by this location without incident.

My guess is Tesla simply doesn’t know if AP was engaged. But their release makes it sound like it was; if it wasn’t, then why would Tesla mention it, except to try to blame the driver?

This is a case where a company should have said less. Either say nothing or just a minimal “we’re working with authorities to pull the data from the wreck and will issue a more detailed statement when we have that information”.

A simple “we are deeply saddened by the loss/we are cooperating with authorities/will update further once the investigation is finished” statement was all that was needed.
Instead, they use most of the statement to blame the barrier and defend AP.

I guess with the stock tanking like it has been lately, Elon felt he needed to divert attention from AP.

Yelp lets say something while covering our ass https://www.google.com/amp/s/www.cnbc.com/amp/2018/03/28/teslas-elon-musk-may-have-fallen-in-love-with-the-wrong-thing.html while our stocks are beginning to tank and credit is tanking

bro1999 continued his “concern troll” comments:

“…the outstanding question is ‘Was AP active at the time of the crash?’. Any mention of that is conspicuously absent from Tesla’s statement. We need to know HOW the driver ended up crashing into the barrier.”

WOW! You are actually complaining that Tesla, quite properly, did not do what several of you serial Tesla bashers did: Jump to a conclusion about Autopilot before the data was actually in. Good for Tesla, and thank goodness they have higher ethical standards than you!

However, contrary to your myopic assertion, Tesla certainly implied that the accident was most likely not the fault of Autopilot, by pointing out that Tesla cars drive past that point about 200 times per day, with no accidents previously happening. Altho that does not rule out some low-probability “corner case” where Autopilot caused the accident, it most certainly does point out the reality that most likely, this accident was the result of human error.

Duh.

If 99 kids go down a water slide, then the 100th gets decapitated, that means the 100th kid must have done something wrong right? Because 99 kids did it “right” before.
http://m.scmp.com/news/world/united-states-canada/article/2139239/water-slide-decapitation-caleb-schwab-10-leads

+1. Yes the big issue is whether AP was active or not. I’m assuming that if AP is active you can over ride it. Not sure if that turns it off or not.

In a way, pointing out how many times AP worked on that stretch makes it worse if it was on. Basically it says it works until it doesn’t, which pretty much means it doesn’t work.

I guess it’s possible that removing the barrier misled AP. That would also not be good.

I’m sure at some point we’ll have an answer.

“In a way, pointing out how many times AP worked on that stretch makes it worse if it was on. Basically it says it works until it doesn’t, which pretty much means it doesn’t work.”

WOW again! One “concern troll” supporting another.

You could say the same about air bags: “It works until it doesn’t, which pretty much means it doesn’t work.”

No, that’s not at all what it means. Even if a system doesn’t work perfectly every time — just like air bags — it might be well worth using because — again, like air bags — you’re much safer with it than without it.

Duh. Again.

As if the Tesla fanboi set would EVER admit there might be a real concern about a Tesla.

Or as if Tesla itself would ever do anything but point fingers away from itself when such incidents occur.

This damage appears to have occured durring a three car collision that included the Model X, and what ever the interactions between the three cars the Tesla ended up sidways with its right side facing the barrier. The Tesla ended up sliding sideways into the compressed barrier in what apears to be at least an impact speed of 70 mph or so. The impact apears to be centered between the right front wheel and the A pillar. This riped through the car and tore the front of the car including the front seats and the front of the battery pack completly off as the momentum of the back two thirds of the car finished the job and continued down the road another 40 odd feet or so. I do not think any standard car could have absorbed such forces any better. This is an educated guess from someone that has done many accitent investigtions with cars. I am so sorry for their loss in this tragic accident

Perhaps one of the three cars changed lanes and “touched” the Tesla Model X, and that resulted in this crash.

Unfortunately does not look to be the case … this appears to be a single vehicle accident causing three car damage. X started the whole thing hitting the barrier … that’s my understanding from the on-line reports.

Quote
The impact caused Huang’s vehicle to catch fire, the CHP said. Moments later, an approaching Mazda and Audi hit the 2017 Tesla.
Unquote

No. The Tesla was in the car pool lane and the other two cars were on the main roadway. The Tesla veered into the barrier and whatever was left of the car landed in the left lane of the main highway on the other side of the barrier and in front of the Mazda and Audi.

” … landed in the left lane of the main highway on the other side of the barrier …”

Are you saying that this Tesla Model X was airborne (after it hit the barrier)?

Portions of the car likely did. The front wheels were still on the barrier where the crash happened. The rest of the car somehow made it around to to the other side. With that much weight and momentum, you’re no longer talking about a single rigid object. Metal can twist like putty.

“The impact apears to be centered between the right front wheel and the A pillar. This riped through the car and tore the front of the car including the front seats and the front of the battery pack completly off…”

I think you must be describing a different accident. The front seats are still in the car and the battery pack – although severely damaged – also remained on the car.
The driver was pulled by Good Samaritans from the front seat which was still inside the car. Also, if the car hit near the right front wheel, how did 90+% of the car end up to the right of the barrier on the Hwy 101 side?

Those are the 2nd row of seats in the six seat configuration that you see in the image. There is no front left.

The view from the front is hard to get a good perspective on. Find at any of the photos taken from the side – the seats are exactly in the correct position for front seats – even with the B pillar. Also, the seats appear to have the movable headrests which are consistent with the front seats of a Model X.

See 0:40 mark in video. Those are the front seats.

https://youtu.be/T3RNISfwa4E

(⌐■_■) Trollnonymous

So we still don’t know if AP was on……

Correct. Does it matter?

If it was on, it drove into the barrier due to poor lane markings. If it was off, the driver drive into the barrier likely for the same reason.

The NTSB had already found miles covered with AP on are safer.

NTSB commented on safety with AP on vs. AP off???

I read the report and don’t remember anything like that. I do remember NHTSA saying accident rates dropped 40% after AP came out, but that’s standard for AEB.

People use the NTSB’s finding so loooooosely that it’s ridiculous. Tesla has not disclosed it, not even to NTSB … nobody knows exactly what sort of Tesla’s miles traveled those are and what part of AP’s system was on at the time.

“NTSB commented on safety with AP on vs. AP off???”

No, NTSB commented on a ~40% lower accident rate after installation of AutoSteer (not Autopilot), as opposed to not being installed. NTSB would have no way of knowing whether or not AutoSteer was actually active at the time of an accident, unless Tesla told them. Logically, that means the accident rate with AutoSteer active is even lower than a 40% reduction would indicate, since it’s obviously not active all the time.

https://electrek.files.wordpress.com/2017/01/tesla-autopilot-crash-rate.png?w=728&h=513

I don’t know why you keep trying to equate AutoSteer with an ABS (Automatic Braking System). Those are two entirely separate things, with entirely different functions.

That’s even more irritating than seeing so many people in write “Autopilot” in comments when they mean “AutoSteer”. Autopilot doesn’t steer a Tesla car, and you can’t turn it off. AutoSteer does, and you can turn it off.

“Autopilot doesn’t steer a Tesla car, and you can’t turn it off.”

You can’t turn it off? I’m confused. When Tesla released their statement referring to Autopilot being “engaged”, doesn’t that imply that AP engaged = on and AP disengaged = off?

Well, it might not matter to you, who is obviously very willing to put yourself into hands of a system which can fail if the road does not look like the one in the system’s brain, but it matters to everyone else who does care.

I cannot believe that people just say … does it matter??? And add the false narrative that NTSB already said that AP is safer than some imaginary bad driver. They made that claim without willing to disclose how they came to the conclusion … was the AutoSteer on or off?? AP is just too brought …. so how many miles of what kind of miles has AP really done so far? Nobody knows (except Tesla who is not disclosing this sort of thing), including you.

mxs continued his Tesla bashing:

“…add the false narrative that NTSB already said that AP is safer than some imaginary bad driver. They made that claim without willing to disclose how they came to the conclusion … was the AutoSteer on or off??”

That is, as they say: “Not even wrong”. Most of your Tesla bashing FUD here isn’t even relevant to reality.

1. NTSB reported on the comparative accident rates in Tesla cars before and after the installation of AutoSteer.

2. The NTSB report was about AutoSteer, not Autopilot.

3. It’s not a false narrative. The NTSB did not claim to know if AutoSteer was active or not at the time of any accident.

4. I don’t think there is any great mystery as to how the NTSB would know of AutoSteer was installed in any Tesla car or not. Tesla would certainly have those records.

Now, what you could point out is that, since the NHTSB doesn’t know whether or not AutoSteer is active at the time of a crash, simple logic strongly suggests that AutoSteer must actually give even better protection than the 40% lower rate indicates! But of course, that wouldn’t fit your FUD agenda here, now would it? Hmmm?

https://electrek.files.wordpress.com/2017/01/tesla-autopilot-crash-rate.png?w=728&h=513

Nick,
It certainly matters to current Tesla owners. It is THE main question on everyone’s mind over on TeslaMotorsClub.com forum among nervous Tesla owners.

It is not hyperbole to suggest it may also matter for the survival of the company. Tesla lost $25/share yesterday and is down $22 more as of 1 PM ET today.

If Tesla stock becomes cheap enough, Tesla becomes a takeover target by buyers with less noble intentions than Musk.

would be kind of funny if zuckerberg took over to spite elon. never gona happen, just would be ironic.

“…It is not hyperbole to suggest it may also matter for the survival of the company. Tesla lost $25/share yesterday and is down $22 more as of 1 PM ET today.”

It certainly is hyperbole, or more precisely FUD, to suggest that there is a serious question about Tesla’s survival just because the highly volatile stock price fell today to what it was about a year ago. Most likely in another few days, it will bounce right back up near what it was a few days ago. At least, that’s what the history of this highly volatile stock suggests.

Looks like a good time for Tesla short-sellers to exit their position and take a profit… or at least, to minimize their losses!

Tesla goes out of it’s way to create a blog post and provide many examples of how AutoPilot enabled cars have driven on this road BUT would not acknowledge whether AutoPilot (which version) was operating (or not) at the time of the crash.

Sometimes it is what is not said that SPEAKS VOLUMES.

The victim was not some young impulse driven person and driving into a fixed object is not normal or a regular occurrence on dry pavement.

Tesla states: “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.”

This is actually quite useful in estimating the likelihood that AP was engaged. Tesla should also have data for trips per day on the same stretch without active AP enabled. It then becomes simple math.

Example: Tesla data shows average of 120 trips per day without AP and 240 trips with AP, then the average Tesla driver has a 66.7% chance of using AP on this stretch of highway.

Of course, this only gives a rough probability- nothing definitive in this particular case. I’d be interested in knowing the number – as an rough indication of the likelihood that AP was enabled.

Obviously, the crash barrier was hit previously, and was shortened to the point of being ineffective. Hitting that concrete just beyond it would do an INCREDIBLE amount of damage to any car. In fact, in most ICE cars, it would probably flip them and kill the occupants.

I am surprised that there were no sand/water barrels in front for temp purposes.

BUT, we still need to know if autopilot was on. If not, then this is not a big deal ( ICE cars would be in far far worse shape; back seat occupants would have survived this). But if AP was on, then there will need to be some explanations.

And those ICE cars would have burst in to flames in spectacular fashion in the air!!

Unsubstantiated theories aside I do think it’s ridiculous how long it takes to often repair these safety barriers. When they’re damaged in an accident the person at fault has to pay to repair them, or more than likely their insurance company does. Why does it take so stinking long for them to get repaired when their sole job is to help keep everyone on the road safe? There were empty water drums at a spot by me for months when all they needed to do was bring up a water truck to fill the ones that weren’t damaged and a couple new bins to replace the ones that were.

I’m not saying the state or somebody should be sued but seriously they need to have a quicker turn around time in general to fix these things.

Oh, I bet all kinds of parties will be sued, including the party responsible for the road maintenance, city etc.

This is another lawsuit to watch … with the one regarding Uber. A lot of eyes will be on those two … I say two, because I really expect that Tesla would have state already if the AP was not used at the time. Speculation does not help anyone, including themselves. They want the box, so they can better prepare to explain how the AP crashed …

Anyways, will have to wait a little longer.

“…I really expect that Tesla would have state already if the AP was not used at the time.”

Reasonable people would expect Tesla to wait until the facts were in before issuing a statement regarding that issue.

Only conspiracy theorists and FUDsters seem to believe that those “in charge” must know everything simply because they’re in charge.

How can we expect CalTrans to make these sort of repairs ACROSS THE ENTIRE ROAD NETWORK quickly and routinely?

Mind you this is a FIXED object and not a pothole or road sign down. People expect perfection far too often in a society where people do not want their taxes to fund this level of support.

(⌐■_■) Trollnonymous

CA raised the taxes just for road and highway repair/maint.
http://www.sacbee.com/news/politics-government/capitol-alert/article147437054.html

Yet the roads are still all phukedup.
The filthy politicians have about 60% of those funds earmarked for other projects not related to road and highway repairs.

I blame CA for the lack of repairing that barrier.

“People expect perfection far too often in a society where people do not want their taxes to fund this level of support.”

I agree, the level of support necessary for the kind of instant response called for by some in previous comments about this accident, is far from realistic or reasonable. For those barriers that involve large barrels filled with sand, to have them immediately replaced during the normal cleanup after an accident would require crews and trucks standing by, filled with empty barrels and sand, as though they were firemen in a fire station waiting for a call.

This would not be a reasonable way to spend taxpayer money.

On the other hand, I agree that a collapsible barrier on a busy highway remaining collapsed two weeks after the previous accident, is a sign that better maintenance is needed. If the highway is that busy, then maintenance crews should be busy on that highway when needed.

200,000 cars in the same direction every day?

“Over 200”

HAHA!

I’m usually willing cut a little slack here, but Tesla’s explanation this time sounds to me like:

“I was driving along, minding my own buisness and then a TREE jumped out of nowhere and hit me. Besides, the Bark was too dark (or in Tesla’s case, too white, given the car’s propensity to drive under white stopped semi’s), and the wood is TOO HARD.”

Well, let’s not forget Teslas with autopilot on hitting stopped big red firetrucks, stopped big orange street cleaners (China), or sideswiping walls, etc.

Of course, it’s NEVER the fault of Tesla or autopilot, just ask Must or the fanbois.

These systems should not be tested in beta mode by customers on public highways, but of course, that’s the cheap way to do things.

For Tesla Haters cultists, it doesn’t matter how many lives Tesla AutoSteer has saved, nor that the NHTSB says that Tesla cars with AutoSteer installed have 40% fewer accidents than Tesla cars without.

Tesla Hater cultists will always focus only on the tragic accidents, never on all the lives saved, regardless of whether or not Autopilot (or more properly, AutoSteer) had anything to do with any particular accident or not.

The haters are always happy to claim that Autopilot was at fault, even when there is absolutely no evidence of that. If a family in China sues Tesla, claiming that “Autopilot” (actually AutoSteer) was controlling the car, why naturally the Haters all pretend to believe that, even when the family refuses to allow Tesla to examine the wreck! Why, of course the family “knows” that AutoSteer was active, because… well, just because Tesla is a big company with deep pockets!

I think the Tesla Hater cultists must have a very low opinion of themselves. Obviously they care nothing about honesty or hypocrisy!

“The haters are always happy to claim that Autopilot was at fault…”

Can you cite even one instance in this thread where someone claimed that Autopilot was at fault?

There are basically 3 questions here. 1. Why did the car hit the barrier, regardless of whether the barrier is properly attenuated. Whether it was on autopilot is the critical piece of information. 2. Why did the car disintegrate? Sure, other cars do too when it hit the pole, trees or relatively blunt surfaces at high speed. Tesla supposes to be one of the safest vehicles out there. The battery pack didn’t disintegrate after collision but rest of the front are gone. Yes. They are designed to absorb energy by breaking down in an event of violent crash. But driver didn’t survive anyway. At this much disintegration, would things such as air bag even work since they are no longer attached to the vehicle. What was the speed at the time of crash. 3. This is more to Cal Tran policy. We have accidents here in CA all the time and we don’t expect Cal Tran to instantly fix every single barrier in the case of crash. However, what is their policy and is this particular location meeting the policy? Granted, we can’t pad down every roadway or surfaces to make sure vehicles never crash into it. But if this… Read more »

Not to say this is the cause in this case, but typical Silicon Valley driving at an exit can be best described as “last minute Louie” Everybody waits until the last possible second to “juke” into the exit lane. Happens on my commute several times every day. If the driver did this, or another driver juked in front of this driver causing him to overreact, this could easily have been the cause. There were enough witnesses that this would come out in the report, whenever that is released. Hopefully we’ll know soon.

The article title is unnecessarily inflammatory. Tesla does NOT blame the barrier for the crash. You need to change the title of this article. They identify the barrier as a factor in the severity of the crash.

Definitely agree with this point. There is a difference between reasons for the crash and reasons for the severity of the crash.

@ Eric Loveday

Can somebody please make a drawing of that location (with all the lanes and barriers, etc.).

That would make it easier for us to understand what people are trying to say/explain regarding this horrific accident.

Thank you

The media has treated this story like no one has died in a car crash before. The lack of deaths in Tesla crashes should probably be the bigger story, but it’s not. It didn’t stop us from buying an X on Sunday, and I’m pretty sure it’s not going to create some wave of cancellations for the Model 3. Teslas remain quite safe, as long as you don’t drive them into solid concrete at 70mph, drive them under a semi, etc.

Looks like a very appropriate reply from Tesla Inc. on the subject, and I see it uses some of the same points I made in the discussion in comments to the previous IEVs article on this subject. For example, the fact that many, many Tesla cars must have passed this point on the road under control of Autopilot + AutoSteer, within the past two weeks following the collapse of that crash attenuator, so Occam’s Razor suggests it’s rather unlikely to be the fault of Autopilot.

Not impossible, but unlikely, and Occam’s Razor shaves in the direction of human error.

Poster ‘Bubslug’ invokes Occam’s Razor over at TMC website with an opposing view. See post 563 here:
https://teslamotorsclub.com/tmc/threads/model-x-crash-on-us-101-mountain-view-ca.111505/page-29

I’ll give you the first point for at least spelling Occam correctly. After your done, make sure you accuse him of being an anti-Tesla troll. 🙂

It goes without saying that Occam’s Razor is merely a guideline/screening tool that can be applied in differing interpretations. I have never seen it applied in a serious root cause investigation.

When you’re playing spelling Nazi, it’s probably best not to ignorantly use “your” where the possessive form should be used. [sic] Dontcha think?

Fair enough Roger. You got me.

BTW – I was not attempting to play spelling Nazi – but was rather trying to be somewhat gracious and humorous by offering up that Pu Pu got the spelling correct. Pu Pu perceives me as an enemy troll and I was merely trying to set a light and conciliatory initial tone. It apparently failed. Oh well.

I certainly didn’t mean to suggest that Occam’s Razor is the preferred method of accident investigation! It’s merely a useful tool in scientific reasoning, allowing people to focus in on the most likely cause or causes of something, rather than the least likely. The simplest explanation — that is, the one requiring the fewest assumptions — is usually the correct one; but not always! I have seen known serial Tesla bashing, short-selling FUDsters posting to TMC, but the frequency is thankfully pretty low. Certainly lower than it is for IEVs comments! But I see no sign that “Bubslug” is a Tesla basher. He’s merely unfamiliar with the scientific method, and does not understand how to apply Occam’s Razor. His scenario involves no less than 4 assumptions, whereas the scenario where human error is at fault requires only one. Occam’s Razor definitely does not shave in his direction, despite his claim. My point about Occam’s Razor is that if we are going to speculate in advance of the facts coming out, which people always do despite the fact it’s not rational to do so, then Occam’s Razor is the method giving the best probability for arriving at the correct conclusion from… Read more »

That’s fine. But if you are intellectually honest, you need to concede that there is speculation occurring on both sides in this thread. You seem to be just fine with those who speculate in defense of Tesla but launch ad hominem attacks against those that speculate that Tesla may be at fault.

Your analysis isn’t sound. If “human error ” requires just one assumption then “AP error” just requires one. You have to advance a specific reason for the human error.

The obvious possibility for an AP error is that the missing barrier was interpreted by AP as opening up a new lane and it moved the car into a non-existent lane. Or AP may not have been engaged and the driver just fell asleep. Who knows.

He’s perfectly correct in pointing out your intellectual dishonesty.

DonC, I concur with your analysis. In a different thread early on, Pu-Pu incorrectly invoked Occam’s Razor to rule out use of Autopilot because the crash damage indicated the driver hit the barrier at a speed above the upper limit for Autopilot engagement. He has since backed away from that earlier assessment. Although there might have been some logic behind his original conclusion, it clearly had nothing to do with Occam’s Razor. Occam’s Razor is intended to rule out possible scenarios that involve a sequence of events that as a whole make the scenario less likely. For example, the following sequence is one of literally hundreds of scenarios that conceivably could have happened: the driver was distracted by a wasp buzzing around inside the vehicle. While attempting to swat the insect, he accidentally disengages Autopilot and also knocks his coffee mug on the ground. The driver attempts to pick the mug up off the ground – assuming AP is still engaged. The driver sits up in time but is disoriented and blinded by the sunlight (now hitting his eye below the sun visor) and cannot take the necessary evasive action to direct the vehicle out of the path of the… Read more »