Might be plausible that he was just going straight in the gore area (mistakenly thinking it was the lane), and hit the barrier (which is offset to the left) causing it to rotate to the right, then the Mazda hit it and pushed it toward the wall, and then the Audi hit it and pushed it even more into the wall.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Model X Crash on US-101 (Mountain View, CA)
- Thread starter mookhead
- Start date
Might be plausible that he was just going straight in the gore area (mistakenly thinking it was the lane), and hit the barrier (which is offset to the left) causing it to rotate to the right, then the Mazda hit it and pushed it toward the wall, and then the Audi hit it and pushed it even more into the wall.
Tam
Well-Known Member
After the crash, it was then hit by a Mazda first, then an Audi next. So there would be at least those 2 people at the scene.
There were also a gentleman together with 2 other men who pulled the driver out of the seat to a safe distance then he videoed the start of the battery fire.
That sounds like 5 witnesses right there! Ok. May be not because other men might come from Mazda and Audi cars.
NeverFollow
Active Member
I'am sorry for your loss. Thank you for the additional information.
What We Know About Last Week's Accident
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.
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.
From the below picture I noticed that there was at least another car involved.
I did don't find any mention of it, I hope that the driver of this other will be able to provide more information of this tragic event.
https://jalopnik.com/ntsb-opens-investigation-into-deadly-tesla-model-x-cras-1824115008
Attachments
To me, I don't see a scenario where the driver continues straight while on autopilot since this area is familiar ... though it looks like a plausible scenario when viewing the lane from overhead.... but when you imagine it from a driver's perspective, I see a scenario unfolding where he hits somewhat straight while executing a familiar lane change from the left to the right while looking back to the right to check his blind spot at a critical moment. It's possible that he attempted to get auto lane change to work and this delayed a manual change... it's hard to say, but looking back and the barrier being compressed changes the calculation of a driver that is accustomed to seeing that barrier a full 20 feet closer to him in an uncompressed state. It's not enough to explain it, but it's going to be a confluence of events most likely.
Another hypothesis: IF you look at TEG's post where he shows himself in the left most lane just prior to the accident.... imagine if someone in TEG's portion had been a slower driver and that the Tesla had tried to speed up to get around him and then would change lanes to the right after passing the car... in this scenario.... the slower car would hide the barrier as the tesla approached no?
Another hypothesis: IF you look at TEG's post where he shows himself in the left most lane just prior to the accident.... imagine if someone in TEG's portion had been a slower driver and that the Tesla had tried to speed up to get around him and then would change lanes to the right after passing the car... in this scenario.... the slower car would hide the barrier as the tesla approached no?
Last edited:
Sorry about your loss and condolences to his family. My experience is AP will try to detect the lane markers and then centered between them.
There are factors that may affect when the system sees it - e.g. line of sight , direction of the car, speed, and how well the markings are..
The latest firmware released couple of weeks ago are more stable and hardly try to get out of any gap in the lane markings, I am not sure what firmware version your friend had (the phone app should be able to tell) and from pictures shown doesn't seem to have any gap
If the car already entered the gore area for whatever reasons, then it is possible the car will keep on going until the driver realized that's not a lane and naturally make emergency maneuver.. It's 930am in the morning, I am not sure how bad is the sun's glare or whether it is a contributing factor to the system, but human is easy to miss if the car is already inside the gore area.
I hope Tesla will fully cooperate with the authority and disclose all data they have, and eye witnesses will come out and provide any information.
That was the Mazda ("m" in my diagram) that the CHP said was the first to hit the Model X after the initial barrier impact.
Then the Audi ("A" in my diagram) hit it a 2nd time as it came to a rest at the wall.
The Mazda hit it in the #2 lane (and I think pushed it back into the #1 lane), then the Audi (in the #1 lane) pushed it more towards the wall.
I believe that is the Mazda referred to in several of the news reports. That driver was supposedly OK and probably provided an account for the police report.
What is your postulated scenario then? And if your scenario has the Model X traveling in the #1 lane, why would the driver have moved into this lane given his destination was Sunnyvale further south on 101?
How do explain the AP failure experienced at this location by Alset72? Is it possible AP was using a vehicle exiting on the 85 flyover as a reference to clarify poorly resolved lane markings as suggested possible by TEG?
Not necessarily if you’ve driven this a few times. Although it’s generally reverse commute at that time of day (though less than the total gridlock of the other direction it is still bad) even if you intend on staying south on 101 many drivers are attracted to the emptier far left commute lane for as long as you can - and so end up cutting right. Even if you aren’t aggressively focused on speed it’s pretty easy to zone out and follow the left white stripe and end up in the wrong lane.
So far we have no information on whether AP was engaged so let’s not get distracted by that. My biggest concern is whether the exposed rails of the already compressed barrier would guide a car to its demise as a couple others have mentioned.
mkspeedr
Member
My two options -
He was in the 101 south HOV lane and something or AP caused him to swerve into the barrier. The idea of it gradually drifting into the barrier makes sense only if he was not paying attention. He knew the road so he would have seen AP drift off line.
He was in the 85 south HOV exit using AP and didn't notice he was about to take the wrong HOV lane. This would have only happened if he was not paying attention. I have seen many people become to reliant on AP and forget they should be driving.
It looks to me like he was in the 85 south HOV lane but that is not the lane he would need to take to get to work. That is the part that does not make sense to me.
He was in the 101 south HOV lane and something or AP caused him to swerve into the barrier. The idea of it gradually drifting into the barrier makes sense only if he was not paying attention. He knew the road so he would have seen AP drift off line.
He was in the 85 south HOV exit using AP and didn't notice he was about to take the wrong HOV lane. This would have only happened if he was not paying attention. I have seen many people become to reliant on AP and forget they should be driving.
It looks to me like he was in the 85 south HOV lane but that is not the lane he would need to take to get to work. That is the part that does not make sense to me.
Tam
Well-Known Member
According to today's Tesla blog:
What We Know About Last Week's Accident
It does not confirm nor deny that they know the AP status of the car.
They only say they'll know better when they can get to the car's physical log.
They're silent about getting the car's status streaming wirelessly and continuously up to the time of impact when the antenna were destroyed at that time which prevented any further transmission of remote data logging.
So may be as @verygreen said they know the AP status, but I guess they just don't reveal it just yet!
Last edited:
Looking at the picture of the car closely, I find it odd that the A pillars show very little buckling given the severely of impact and forces required to detach the front. Also, if the car did impact fairly straight on into the collapsed smart cushion, wouldn't there be a lot more damage to the interior (seats and etc) given the forces involved to detach the front?
Is it possible the car hit the barrier at a high slip angle while trying to correct the position in the lane?
As a structural engineer, the damage sustained by this car really puzzles me...
Is it possible the car hit the barrier at a high slip angle while trying to correct the position in the lane?
As a structural engineer, the damage sustained by this car really puzzles me...
I am thinking it was almost like a zipper. Maybe the smart cushion was holding onto the left front and the right front was being pushed into the road by the kinetic energy of the back part.
The front end looks splayed apart. Like the windshield and dashboard just fell out when the two sides came apart.
The front end looks splayed apart. Like the windshield and dashboard just fell out when the two sides came apart.
If a Caltrain car flies off the (rail)road and into an embankment, I would definitely blame Caltrain. But I assume you meant Caltrans.
If a car flies off of a road and into an embankment, the answer is "maybe". Roads have to be designed under the assumption that rain, ice, potholes, blowouts, etc. will cause cars to behave in ways that diverge significantly from the ideal image of a car going straight down the center of a lane. They must be as safe as they reasonably can be under those conditions, too, or else the roads aren't safe. This isn't to say that they have to be safe against somebody doing something completely absurd (e.g. extreme speeding), mind you; it really depends on the details.
Those aren't cuts. That's the edge of the original concrete roadway. In a cost-cutting measure, they used asphalt to add lanes in the median of an existing concrete road. That's the border between the concrete and the asphalt you're seeing. And it is sealed because it probably would have problems with water undercutting the road bed if it weren't.
Unfortunately, I have—not on the current version, but in a previous version. I was either on southbound 101 or 237 in Sunnyvale (I forget which), and I hit my right turn signal to indicate that I was about to exit. I forgot that autosteer was engaged until the moment when it dutifully began to steer me onto the shoulder. So yes, it is quite possible.
water can freeze, probably wouldn’t happen here(yet). Also if the water leaked out you might not know. Sand barrels can also get water in them potentially and also freeze.
Crazy - looks like a scene in a Michael Bay movie.
So they know about the other 85k times AP was used and 20k YTD, but they don't know about this one at ~9:30am from San Mateo to Mountain View?
Strange indeed.
As an engineer do you think it’s plausible that the car did a front somersault (as reported) over the leading edge of the battery (the low down but foremost really strong point). With the front bumper beam / motor embedded in the barricade the front end would be pulled off while it somersaulted? The front end would have been designed to be much stronger in compression than tension.
That's definitely plausible, but I am not sure that explain the fairly clean shearing of structure at the a-pillar position. It takes a lot of energy to break metal like that, especially on a unitized structure. Any in-plane loads (compression or tension in the lengthwise direction of the car) to break the front that violently should result in significant surrounding structure damage. To break a unitized metal structure cleanly, I would think it'll have to be a load normal to the plane of the structure.
I just drove this today, the left most carpool lane is preferable when the speed differential is large. I find it more comfortable to have the #2 carpool lane to my right, even when it is empty, when going at high speed; you are less likely to be cutoff by a car jumping into the lane from the slow moving traffic. During my drive this afternoon it was clear that most drivers also preferred the left most carpool lane, so much so the right carpool lane was almost empty. This accident was on my mind and I moved to the #2 carpool lane a little earlier than I probably normally would have. But then I had to go into into hyper vigilant mode since I was zooming by the slow non-carpool traffic. I found myself slowing down because the speed differential was so large.
Similar threads
