SF Bay Accident Surveillance Footage

Exclusive: Surveillance Footage of Tesla Crash on SF’s Bay Bridge Hours After Elon Musk Announces “Self-Driving” Feature — The Intercept

Musk has said Tesla’s problematic autopilot features are “really the difference between Tesla being worth a lot of money or worth basically zero.”

apple.news

 

[jabloomf1230]

NHTSA confirms FSD beta was engaged and may have caused accident:

The Tesla Model S That Caused a Pileup in San Francisco Had Full Self-Driving Engaged

Investigation reveals that the driver of the Tesla Model S that caused an eight-vehicle pileup in San Francisco activated FSD seconds before the crash

www.autoevolution.com

At least cite the original CNN article which states:

The Tesla's driver told authorities that the vehicle's "full self-driving" software braked unexpectedly and triggered the pileup on Thanksgiving day. CNN Business was first to report last month the driver's claim that "full self-driving" was active.

The NHTSA has not released any findings based on data from the car's computer. For now, we will have to take the driver's word for it that FSD was engaged.

 

[diplomat33]

At least cite the original CNN article which states:



The NHTSA has not released any findings based on data from the car's computer. For now, we will have to take the driver's word for it that FSD was engaged.

The autoevolution site says that the info came from investigation report. So CNN was merely citing the report:

"According to the investigation’s report cited by CNN, the controversial driver-assist software was activated roughly 30 seconds before the crash."

 

[enemji]

What if there was a kid on the lane1?

 

[MP3Mike]

The autoevolution site says that the info came from investigation report. So CNN was merely citing the report:

"According to the investigation’s report cited by CNN, the controversial driver-assist software was activated roughly 30 seconds before the crash."

The point is that there is no report yet. The only "report" available, is the data from Tesla on ADAS related crashes, which is not an investigation.

 

[jabloomf1230]

Ok, Google masters of TMC, find the NHTSA report on the crash. I couldn't find it on their website and CNN only cites the driver as proof that FSD was engaged.

I'm not saying FSD didn't contribute to the crash. As I said previously in this thread, let's wait for the official report.

 

[diplomat33]

The point is that there is no report yet. The only "report" available, is the data from Tesla on ADAS related crashes, which is not an investigation.

Maybe autoevolution.com is wrong then because it says there is an "investigation report".

 

[Dan D.]

Maybe autoevolution.com is wrong then because it says there is an "investigation report".

theintercept.com/2023/01/10/tesla-crash-footage-autopilot has a link to the CHP investigation report. While not from the NHTSA it is a government investigation report.

"P-1 claimed V-1's Full Self-Driving Capability malfunctioned at the time of the crash."
P-1 being the driver of the first Vehicle V-1.

"P-1 stated V-1 was in Full Self Driving mode at the time of the crash, I am unable to verify if V-1's Full Self-Driving Capability was active at the time of the crash. On 11/24/2022, the latest Tesla Full Self Driving Beta Version was 11 and is classified as SAE International Level 2. SAE International Level 2 is not classified as an autonomous vehicle. Under Level 2 classification, the human in the driver seat must constantly supervise support features including steering, braking, or accelerating as needed to maintain safety. If the Full Self Driving Capability software malfunctioned, P-1 should of manually taken control of V-1 by over-riding the Full Self Driving Capability feature."

"He was driving V-1 on I-80 eastbound traveling at 50 miles per hour in the #1 lane. V-1 was in Full Auto mode when V-1 slowed to 20 miles per hour when he felt a rear impact."
"On 11/27/2022 at approximately 1330 hours, I contacted P-1 via telephone to clarify his statement. He related to me in essence the following: He was driving V-1 on I-80 eastbound in Full Self Driving Mode Beta Version traveling at approximately 55 miles per hour. Prior to the Yerba Buena Island tunnel entrance V-1 moved from the #1 lane to the #2 lane. When V-1 was in the tunnel, V-1 moved from the #2 lane into the #1 lane and started slowing down unaccountably. When V-1 was about 20 miles per hour, he felt a rear impact."

"V-1 was traveling at approximately 55 miles per hour. P-1 claims V-1's Full Self Driving mode was active and malfunctioned. In any event, V-1's left signal activated, V-1's brakes activated and V-1 moved into the #1 lane slowing to a stop. P-2 observed V-1 stopping and applied V-2's brakes. V-1 made an unsafe lane change (21658(a) California Vehicle Code) and was slowing to a stop directly into V-2's path of travel. This caused the front of V-2 to collide into the rear of V-1 (A.O.I. #1). P-2 did not have enough time to perceive and react to V-1's lane change."

edited to add further quotes.

 

[jabloomf1230]

theintercept.com/2023/01/10/tesla-crash-footage-autopilot has a link to the CHP investigation report. While not from the NHTSA it is a government investigation report.

"P-1 claimed V-1's Full Self-Driving Capability malfunctioned at the time of the crash."
P-1 being the driver of the first Vehicle V-1.

"P-1 stated V-1 was in Full Self Driving mode at the time of the crash, I am unable to verify if V-1's Full Self-Driving Capability was active at the time of the crash. On 11/24/2022, the latest Tesla Full Self Driving Beta Version was 11 and is classified as SAE International Level 2. SAE International Level 2 is not classified as an autonomous vehicle. Under Level 2 classification, the human in the driver seat must constantly supervise support features including steering, braking, or accelerating as needed to maintain safety. If the Full Self Driving Capability software malfunctioned, P-1 should of manually taken control of V-1 by over-riding the Full Self Driving Capability feature."

"He was driving V-1 on I-80 eastbound traveling at 50 miles per hour in the #1 lane. V-1 was in Full Auto mode when V-1 slowed to 20 miles per hour when he felt a rear impact."
"On 11/27/2022 at approximately 1330 hours, I contacted P-1 via telephone to clarify his statement. He related to me in essence the following: He was driving V-1 on I-80 eastbound in Full Self Driving Mode Beta Version traveling at approximately 55 miles per hour. Prior to the Yerba Buena Island tunnel entrance V-1 moved from the #1 lane to the #2 lane. When V-1 was in the tunnel, V-1 moved from the #2 lane into the #1 lane and started slowing down unaccountably. When V-1 was about 20 miles per hour, he felt a rear impact."

"V-1 was traveling at approximately 55 miles per hour. P-1 claims V-1's Full Self Driving mode was active and malfunctioned. In any event, V-1's left signal activated, V-1's brakes activated and V-1 moved into the #1 lane slowing to a stop. P-2 observed V-1 stopping and applied V-2's brakes. V-1 made an unsafe lane change (21658(a) California Vehicle Code) and was slowing to a stop directly into V-2's path of travel. This caused the front of V-2 to collide into the rear of V-1 (A.O.I. #1). P-2 did not have enough time to perceive and react to V-1's lane change."

edited to add further quotes.

Like I said. Let's wait for the official report from the Feds.

 

[Dan D.]

Like I said. Let's wait for the official report from the Feds.

NHTSA Level 2 ADAS Incident Report ID# 13781-4338 seems to match the accident conditions.

Data from the CSV indicates a 2021 Tesla Model S collision Nov (no day stored) 2022 at 20:39 in San Francisco. The CHP accident report says the collision was at 12:39PST or 20:39UTC. Are they converting time to UTC in which case the times match. Initial reporting was from Telematics (the car itself) and Media.

Precrash speed was 7mph. Data is from EDR, Telematics. Collision damages match the CHP accident report. Enough data points seem to match to indicate this is the correct report, but annoyingly they don't list the full VIN, nor does CHP, and NHTSA does not report on the day of the month. Much of the report text and some other fields are REDACTED due to Tesla's request. Very useful NHTSA & Tesla, thanks.

For this report it indicates ADAS was engaged. Which means Tesla Level 2 systems were engaged.

 

[MP3Mike]

For this report it indicates ADAS was engaged. Which means Tesla Level 2 systems were engaged.

That just said it was engaged sometime in the 30 seconds leading up to the collision. So it could have been disengaged during the whole time the videos cover.

 

[Dan D.]

That just said it was engaged sometime in the 30 seconds leading up to the collision. So it could have been disengaged during the whole time the videos cover.

If Automation System Engaged? = ADAS
The Reporting Entity’s report of the highest-level driving automation system engaged at any time during the period 30 seconds immediately prior to the commencement of the crash through the conclusion of the crash.

We can see the NHTSA has a report ID# 13781-4338 going on the crash, and Tesla has reported it twice, once on the day and once on a followup. All the other useful information, such as the narrative, is REDACTED by Tesla. Though they always do this, so it's not suspicious in itself, just unhelpful.

 

[stopcrazypp]

Ok, Google masters of TMC, find the NHTSA report on the crash. I couldn't find it on their website and CNN only cites the driver as proof that FSD was engaged.

I'm not saying FSD didn't contribute to the crash. As I said previously in this thread, let's wait for the official report.

As others mentioned it was just an incident report as part of the general standing order on ADAS systems:

Standing General Order on Crash Reporting | NHTSA

NHTSA has issued a General Order requiring the reporting of crashes involving automated driving systems or Level 2 advanced driver assistance systems.

www.nhtsa.gov

I don't know why the news sources don't link it.

 

[AlanSubie4Life]

If Automation System Engaged? = ADAS
The Reporting Entity’s report of the highest-level driving automation system engaged at any time during the period 30 seconds immediately prior to the commencement of the crash through the conclusion of the crash.

Right, so it may not have been.

But not sure why that is important here, of course. All that matters is that it was engaged in the period just prior to the collision for this sort investigation to proceed according to the NHTSA standing order.

But anyway we’ll see. Details will come in due time. I would be modestly surprised if AP/NOA/FSD/FSDb were engaged at the moment of collision.

Engaged and slowed down vs. disengaged and coasted down, FSDb (incorrectly) vs NOA, doesn’t really matter. Those are just details of what happened with the failure of the automation interface. In the end the fix/learnings will be applied to all the systems to the extent possible.

 

[bradtem]

Right, so it may not have been.

But not sure why that is important here, of course. All that matters is that it was engaged in the period just prior to the collision for this sort investigation to proceed according to the NHTSA standing order.

But anyway we’ll see. Details will come in due time. I would be modestly surprised if AP/NOA/FSD/FSDb were engaged at the moment of collision.

Engaged and slowed down vs. disengaged and coasted down, FSDb (incorrectly) vs NOA, doesn’t really matter. Those are just details of what happened with the failure of the automation interface. In the end the fix/learnings will be applied to all the systems to the extent possible.

Alas, there are several explanations, and you want to know whether it as disengaged and when.

1. Did Nav-on-AP initiate the lane change and do it without confirmation from the driver? (This will happen if the driver is driving in "constant wheel torque" mode, which is how I drive with AP.)
2. Did AP get disengaged by wheel torque or brake pedal at any time prior to the impact?
3. Was the lane change possibly initiated by driver turn signal while AP was on?

 

[EVNow]

As others mentioned it was just an incident report as part of the general standing order on ADAS systems:

Standing General Order on Crash Reporting | NHTSA

NHTSA has issued a General Order requiring the reporting of crashes involving automated driving systems or Level 2 advanced driver assistance systems.

www.nhtsa.gov

I don't know why the news sources don't link it.

Because that would reduce the shock value of the "news" report.

 

[BitJam]

In situations like this before, Tesla has quickly commented, when the telemetry data is in their favour.

Here is a counter-example where it took ten months for the exonerating data from the onboard computer to be made public:

Autopilot was not engaged in 2022 Tesla Model S crash that killed 3 in Newport Beach

The Tesla Model S Plaid was traveling in "great excess" of the speed limit when it slammed into construction equipment on Pacific Coast Highway May 12, Newport Beach Police Lt. Eric Little said.

www.latimes.com

The crash was May 12 2022 and the data was released March 18, 2023.