Final report on Autopilot issued by @NHTSAgov & feature names

[scottf200]

@elonmusk tweets 19Jan2017:
1: Final report on Autopilot issued by @NHTSAgov is very positive https://t.co/KsOZSrr3l9
2: Report highlight: "The data show that the Tesla vehicles crash rate dropped by almost 40 percent after Autosteer installation."

https://static.nhtsa.gov/odi/inv/2016/INCLA-PE16007-7876.PDF

I was not aware of the underlined names.

"These updates have included changes to improve TACC, AEB and Autosteer performance, as well as adding new driver assistance safety features, such as In-Path Stationary Object (IPSO) braking and Pedal Misapplication Mitigation (PMM). In September 2016, Tesla released its 8.0 firmware update which included revisions in the driver monitoring strategy, as well as several enhancements to AEB, DBS, and TACC performance."

"As part of Tesla’s 8.0 over-the-air (OTA) software update in September 2016, Tesla revised the timing of the hands-on warnings and added a feature that takes away the Autopilot driving feature for the remainder of the drive cycle if the driver fails to respond to the alerts adequately (known as an “Autopilot strikeout” – Figure 6)."

 

[chillaban]

There's a lot of other fascinating data too. I'm still reading through the report, and will update with more things I find (emphasis mine):

If FCW is OFF, the driver will still get a Brake Capacity Warning (BCW) when driving in Traffic-Aware Cruise Control (TACC) mode (see Section 3 below). BCW alerts the driver when the closing speed to a lead vehicle may be too great to avoid a collision with the standard TACC deceleration limits

The system requires agreement from both sensor systems to initiate automatic braking. The camera system uses Mobileye’s EyeQ3 processing chip which uses a large dataset of the rear images of vehicles to make its target classification decisions. Complex or unusual vehicle shapes may delay or prevent the system from classifying certain vehicles as targets/threats

ODI’s analysis of Tesla’s AEB system finds that 1) the system is designed to avoid or mitigate read-end collisions; 2) the system’s capabilities are in-line with industry state of the art for AEB performace through MY 2016; and 3) braking for crossing path collisions, such as that present in the Florida fatal crash, are outside the expected performance capabilities of the system.

and that neither [Tesla or a Mercedes C-class] vehicle effectively responded to a realistic appearing artifical “target” vehicle in the SCP or LTAP scenarios

Finally, to paraphrase, the NHTSA's conclusion is that the crash is due to a period of at least 7 seconds of driver distraction, and Joshua's last input was a TACC speed change 2 minutes prior to the crash. If there's anything to take away from this report, it's the (intuitive) lesson that traffic conditions can change from uneventful to deadly in a matter of seconds. Had he looked up 8 seconds before the crash, he would likely have seen nothing threatening.

So…. keep that in mind when evaluating the temptation of looking away from the road while using AP. I know I've been guilty before of looking away for a couple seconds while using AP.


EDIT: Also, while 8.0 added In Path Stationary Object collision avoidance, there is no mention of Lateral Turn Across Path avoidance

 

[chillaban]

Reading more into the footnotes, I found these lines interesting too:

While drivers have a responsibility to read the owner’s manual and comply with all manufacturer instructions and warnings, the reality is that drivers do not always do so. Manufacturers therefore have a responsibility to design with the inattentive driver in mind. See Enforcement Guidance Bulletin 2016-02: Safety-Related Defects and Automated Safety Technologies, 81 Fed. Reg. 65705.


An unreasonable risks[sic] due to owner abuse that is reasonably foreseeable (i.e., ordinary abuse) may constitute a safety-related defect. See United States v. Gen. Motors Corp., 518 F.2d 420, 427 (D.C. Cir. 1975) (“Wheels”).

It does seem like the NHTSA is trying to warn (automakers in general) that you cannot just hide behind blanketing warning statements in the manual, and egregious deficiencies in level 2 ADAS implementations can still be classified as safety defects.

Of course, it did not find any such deficiencies in Autopilot.

 

[chillaban]

Other questions include whether the reduction in crash rates is actually due to Autosteer itself

Let me guess: No.

But that's not what the NHTSA data is trying to show. Simply put, comparing when AP1 HW was a decorative camera and radar antenna, versus a later date when FCW and AEB alongside TACC/AP were activated, serious collisions went down 40%.

That sounds believable to me, and has more to do with FCW/AEB than Autosteer.

 

[scottf200]

Let me guess: No.
But that's not what the NHTSA data is trying to show. Simply put, comparing when AP1 HW was a decorative camera and radar antenna, versus a later date when FCW and AEB alongside TACC/AP were activated, serious collisions went down 40%.
That sounds believable to me, and has more to do with FCW/AEB than Autosteer.

The ability to see two cars ahead is a big deal for mitigating damage, accidents, serious injuries, IMO. What other companies cars implement this capability? I used AP (AP1) on my X even around the suburb 40-45 MPH roads (with both hands on wheel) and loved when it was watching two cars ahead. Lot of distracted drivers.