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.
That said, I still consider an "active" steering decision by Autopilot to go outside of a lane a type 2) event, separate from "passive" events like the Joshua Brown incident, fire trucks - and others where lane simply ends at an obstacle - where mere predictable lane following have resulted in disaster through no fault of Autopilot... that I categorized as type 1) above. The Mountain View event certainly seems to be an active event where Autopilot failed.

There is a high dependence on the exact sequence of events/ positioning of vehicles in the Mountain View accident.
With a following distance of 1, the worn right gore line, and a leading vehicle, it is questionable whether what AP could see and what it crossed met the criteria of a lane (realizing of course that there was a lane line which is discernable with ground truth data or sufficient field of view).

Side thought/ realization: at T-8, the car was going 65 and following a lead vehicle at T-3 it was reported to be accelerating from 62 MPH. That means the lead car was either going slower than 65 (and the Tesla slowed to match), or the lead car slowed down in that time span.

Possibly, the lead vehicle made a maneuver that necessitated a lower than limit speed and/ or the Tesla was closing on the vehicle, in which case the following distance may have been even less...

Made up senario:
Tesla following vehicle in 101 fast lane.
Lead vehicle slows to cross gore
AP slows but still gets closer to lead vehicle, this reduces camera field of view of the road.
AP now follows lead car to the left over the worn right gore line (limited view).
AP sees the left gore line and stops following the lead car.
Lead car continues on 85 offramp.
 
Made up senario:
Tesla following vehicle in 101 fast lane.
Lead vehicle slows to cross gore
AP slows but still gets closer to lead vehicle, this reduces camera field of view of the road.
AP now follows lead car to the left over the worn right gore line (limited view).
AP sees the left gore line and stops following the lead car.
Lead car continues on 85 offramp.

This is certainly possible. But there are enough videos to see AP2 tends to turn (center?) towards a gore when it gets confused by the lane markings even when there are no cars around - including in that particular Mountain View location - that I believe we know now this incident could have happened even without a lead car.

IMO it is still possible the lead car wasn't in the gore area at all in the Mountain View case, that the Model X slipped there by itself as it lost the left lane markings. The report fits this scenario if mentioning a lead car is simply a red-herring - it may have remained in its lane all the time and the Model X was recognizing it as a lead car (so technically following a lead car), but still may have decided to leave the lane into the gore and stopped "following" as the lead car faded from view. Tesla saying there were 5 seconds of unobstructed view (retrieved images from car?) could fit this scenario too. Just speculation, of course.

In any case, slipping into the gore by AP2 has been shown on video even without lead cars, so this seems to be something AP2 can do. It seems the later versions of AP2 seem to "err" towards taking an unsafe steering action in these situations, instead of keeping a neutral direction and alerting the driver for example when it gets confused. I think one current speculation is this has been a result of the new(ish) lane centering feature for wide lanes?

So, IMO this certainly differs from events where simple passive lane following ends up in a disaster...

I wonder if part of the answer isn't simply that AP2 should be far more aggressive in handing back control (or just sounding an alert) where it gets confused in some way. Has it gotten too confident? I remember when AP2 was new, that is certainly how many city street/intersection tests ended, with audible bongs as it lost its way - that was better than trying to do something it wasn't capable of doing... (I only use AP2 on divided roads these days, but I was testing back then.)
 
Last edited:
So this accident was referred to as a "perfect storm" of things going wrong... This included:

#1: Low morning sun ahead making it hard to see for driver and cameras.
#2: Following distance=1 is perhaps too close for this kind of traffic pattern / road conditions.
#3: "Lead car" apparently improperly enters gore area possibly guiding auto-steer into the same area.
#4: Poor lane markings including lack of warnings in the gore area.
#5: Wide / long gore area looks very much like a lane.
#6: It appears that the driver wasn't paying close enough attention to the roadway to realize the error and failed to manually correct it.
#7: Automatic Emergency Braking doesn't stop for objects like that.
#8: Gore point smart cushion was damaged and not repaired, and did not do normal energy absorption on impact.
#9: This particular type of crash cushion seems particularly bad to impact when it was not repaired. It is more like a knife edge in that case.
#10: It appears that the Tesla hit it at "just the wrong place" so it sliced between the Tesla's energy absorbing structures.
#11: It sliced through just enough to crunch the edge of the battery pack leading to a fire in the pack.

It's unfortunate that so many things lined up "just wrong" and this accident occurred. I do think one of the most important factors (besides the driver not paying attention) was the state of the crash attenuator. Had it been reset, the driver probably would've walked away from the accident even if all the other factors remained the same.

Just this morning, I was reading this thread and the poster mentions falling asleep and hitting a"crash cushion" and surviving the impact:
I actually fell asleep right before the toll booth in Emporia and woke up mid air after hitting the crash cushion and rolling the car several times. Thank God I didn’t kill anybody or myself. If those superchargers would have been installed in Emporia I could have just drove straight there and got some juice and went straight down to Wichita.

Now in no way shape or form am I blaming Tesla or autopilot for this by any means! I just simply fell asleep about a mile before the toll. My point is that if that supercharger was up and running I wouldn’t have had to drive those extended hours just to be able to make it in one shot. I had just drove from Florida to KC so I was already tired. The reason my car was in the shop Was because they told me that my rebuilt the battery had came back and I needed to hurry and have them put it in otherwise they would loan it out to somebody else also the doorhandle LED quit again and I had them replace it. When I picked the car up at the door handle was rubbing and there was a shorted wire in the door. That’s the reasoning behind me having to wait around all day so they could attempt to fix it. I know this is getting off topic but I’m just throwing it out there as an example why we need a supercharger in Emporia.
 
  • Like
Reactions: bhzmark
It's a tragedy that the Mt. View crash barrier took over 10 days to be reset by the regional CalTrans workers.

Over 10 days NOT being reset... NOT in time for the next impact.

In other words they didn't reset it until someone else hit it.

Hard to know how many other impacts there hit a non reset barrier. We know there were other fatalities at that location.
 
CDDF6731-6AB5-43F9-8159-B64378657438.png
So they finally painted chevrons in the gore area there. That is some progress although we could hope for even more markings.
 
Hi,

well, it's an improvement on what it was before.

But seen from Europe, one wonders what a gallon of white paint costs in California??

Over here, the chevrons are twice or thrice as many, and at least the lengthwise limitation line to the right side of the gore area could also have been painted, while they had the truck out.

But OK, it's better than nothing, and maybe the paint comes from China, and has been impacted by the latest Trump tariff increases ;) !

Kind regards from Northern Italy,
Krouebi
 

Attachments

  • upload_2018-7-22_18-1-2.png
    upload_2018-7-22_18-1-2.png
    82.6 KB · Views: 43
  • Informative
Reactions: FlatSix911
Why are we still digging up this old thread? Please lay it to rest and let the man (and Tesla's legal team) rest in peace (for now).

Why are you reading this thread if people posting to it bothers you so much? Posting “shut up” isn’t useful or nice.

Thank you, @TEG for posting the pic of the chevrons. It is nice to know that even Caltrans can be shamed into action, even though it looks like there is still a lane line they still haven’t repainted. I think it disgusting that they left this kill zone without proper marking or a functioning attenuation barrier.