Any safety system can be defeated, but clearly that is not manufacturers fault and they should not be penalized for that. If someone want to be an idiot nothing is going to stop them except death or jail.Autopilot lost in three categories in the Consumer Reports tests: Keeping the Driver Engaged, Clear When Safe to Use, and Unresponsive Driver.
If you believe the driver engagement metric should be binary and leave zero room for ranking of the systems by difficulty in defeating them, well I doubt many would agree. A system that can be defeated by a simple weight attached to the steering wheel should very clearly be ranked lower than a system that is defeated like this
Consumer Reports used a 1-10 ranking system and that certainly feels appropriate
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
U.S. opens formal safety probe for Autopilot - 2021 Aug 16
- Thread starter Terminator857
- Start date
One thing I that I think is worth thinking about in these emergency vehicle cases is how little attention someone would have to have to run into emergency vehicle on the road. I've never seen one blocking a lane that didn't have the emergency lights on. There are usually cones out and sometimes flares. In my state they also use sign trucks for big accidents on the freeway. The point is you can see them for hundreds of yards or miles even away. That means these people weren't paying the slightest bit of attention for many, many seconds. Plenty of time to take action for anyone that is doing what the AP instructions clearly state. In one of these accounts the driver says, "It happened so fast." BS. Everything happens fast if you're completely oblivious to what's going on around you! I'm sure some of you have seen the video of the girl walking right into a mall fountain because she's glued to her phone. No system can completely fix stupidity. If the NHTSA or any other government agency thinks they can regulate for morons we will never have any FSD cars and that will actually cause MORE deaths because I firmly believe that these systems save more lives than they cost even in there current state.
Yup people can risk their own lives all they want
When idiocy endangers the lives of non-idiots, as in the case with public roads that are shared with idiots, that's when regulators step in and manufacturers need to build in measures to reasonably mitigate risks to the public.
rxlawdude
Active Member
As is the case with COVID-19 mitigation. But "freedom" and all...
Maybe the drivers were paying attention and did watch as the car crashed into the stationary object. Maybe they were led to believe that the car would take some action because "it has avoided things before, I don't know why it didn't stop this time".
If they trust the car to avoid things because "OMG, it is amazing, almost like it's seeing through several cars ahead, I can go hands-free and it relieves me of all the stress of driving", then they are going to be surprised when it fails to avoid things, and they are mostly going to just let it crash because it happened too quickly to figure out that THIS TIME it wasn't going to help.
It's both their fault for trusting it, and the system's fault for being so useless that it can't even tell them "HEY, you know there's a stationary object up here that I'M NOT GOING TO STOP FOR, because I don't stop for stationary objects. In fact, because the car ahead of you moved out of the way I'M GOING TO SPEED UP and just ram that thing. I see it, but I don't care, because you should have read the manual better. I'm not even going to beep or anything. If I stopped here I might get rear-ended and I don't want that, so I'll just ram you into it".
As is the case with COVID-19 mitigation. But "freedom" and all...
Hahah that's opening up a whole other can of worms
But in this situation, if we're thinking regulators will allow a new technology because it swaps out X number of deaths caused by alcohol/fatigue for a lower number of deaths caused by overreliance on the tech or some other new factor... Well they definitely won't do that on beliefs, maybe they would do it on extensive verifiable data and even that would come with all kinds of ethical considerations. But the obvious ideal here is to leverage the strengths of assisted driver tech while reducing the negative impact of any weaknesses.
Last edited by a moderator:
M3P 0-60mph stopping distance on a dry road with the 20" PS4S tyres is around 34-39m. Certainly it's less than 40m. With a M3 SR or LR it will usually be less.I used the numbers I found at link below. The point stands anyway, margins are too small when speed is high. I would love to see some accurate numbers and calculations.
The Importance of Correct Stopping and Following Distances
Rear end collisions are the most common accident type in Australia, so it is important to understand the correct stopping and following distances.ubicar.com.au
I've never let my Tesla drive me through anything as complex as an accident scene. I don't use AP in town either for same reasons. There's no excuse for it IMO.
Good, so your M3P stops in 34m while the Average Family Car smashes into the back of you with its 56m stopping distance. Always rely on your car stopping super quickly in all road conditions and you'll be fine. Perhaps my cynicism isn't warranted, it's good to know the performance characteristics of your vehicle.
If the driver isn’t engaged he surely isn’t responsive. And if he isn’t responsive how can he be engaged?
that isn’t what I am arguing at all. I am saying they are also all the same in that regard, but Consumer Reports chose to ding Tesla for jt singularly.
Your post is mostly empty elocution…
Last edited:
In CR's assessment of Driver Engagement, Tesla received 3/10 along with Audi, Buick/Chevy, BMW, Nissan/Infiniti, Land Rover, VW, Porsche, and Volvo, and Mazda ranked last with 2/10, so I'm not sure where they singularly dinged Tesla. Cadillac received 7/10, and the next highest score was 4/10 shared by a bunch of other brands.
If you're referring to the test CR performed to see if Autopilot could be tricked into operating without a driver in the seat, that's a separate thing and was in response to YouTube videos of Teslas driving around with no driver but someone in the passenger seat or back seat. Tesla was dinged for that because it became a big story, you weren't seeing it with other brands even though it was clearly doable.
Unresponsive Driver is about how the system reacts if a driver stops responding period, from falling asleep or having a medical emergency or whatever else. If that happens, what alerts does the system throw out? How far will it travel without a response and what action will it take? That's how Unresponsive Driver was ranked.
You're correct that an unresponsive driver will also not be engaged, but a disengaged driver can still be responsive and measuring engagement includes stuff like how well the system keeps you focused what the car is doing and what's happening on the road in front of you, how it notifies the driver that they need to take over to deal with something the system can't handle, etc.
Long story short, there's overlap between the two things but also clear differences that need to be evaluated separately.
Last edited by a moderator:
The ironic thing is that from AI day marketing video (intro) the driver has a grip on the wheel that is in no way sufficient for an immediate evasive maneuver. Only 1 hand, hand placed at 7 o'clock, letting wheel slip through hand. Tesla continues with ambigous communication.
- Driver's foot was not over the brake pedal either. You can see the pedal move from the FSD inputs at some times
- Car drives left without using its turn signal at an intersection, even though the car in front is using its turn signal. Then the Tesla signals right before completing its left turn
- Gets to a 4-way stop sign first. Waits too long so the black car decides to go out of turn. Tesla starts going but then has to stop
- Some nice driving down Main St. Los Altos
- Goes through a green light but then slows to 3mph at the far end of the intersection where the cyclists are standing. Odd
- An oncoming left-turning silver SUV turns badly across the Tesla which has no trouble stopping at the stop sign and avoiding the SUV
- Tesla then crosses the stop-line while turning right at the stop sign. Too far. A pedestrian runs in front of the car but the Tesla was too far forward. Tesla is at 0mph but continues to turn the steering wheel as the pedestrian is in front of car, could be uncomfortable to see that
- Overall looked like a fairly typical FSD Beta drive so at least they didn't cherry-pick it. Made some mistakes and hesitated a bit. Handled the various situations well enough. No disengagements, no interventions
monteitis
Member
Like the Holocaust and other more recent events. Dumb bad hugely out ways the dumb good.
Looks like theres a new report of a car on Autopilot crashing into emergency crews, this one in Orlando. Or at least the driver is claiming the car was on autopilot
:quality(70)/cloudfront-us-east-1.images.arcpublishing.com/cmg/YNYPBUUOKRGRBKD73TWPIVRQME.jfif)
www.wftv.com
SEE: Tesla on autopilot crashes into FHP cruiser on I-4
FHP said the driver of the Tesla told investigators that the car was on “autopilot” at the time of the crash.
www.wftv.com
rxlawdude
Active Member
Another case where the driver was not attentive as he/she are clearly instructed and warned to be.Looks like theres a new report of a car on Autopilot crashing into emergency crews, this one in Orlando. Or at least the driver is claiming the car was on autopilot
SEE: Tesla on autopilot crashes into FHP cruiser on I-4
FHP said the driver of the Tesla told investigators that the car was on “autopilot” at the time of the crash.
Nothingburger.
Additional fodder for the regulators who will likely be pushing for more robust driver monitoring tech to help ensure engagement when these systems are being used
100% of these surely boil down to lacking attentiveness, you‘d need to be very checked out to ram into stopped emergency vehicles
Similar threads
