Another fatal crash blamed on AutoPilot

[Arizona Willie]

I see the investigating authorities blamed AutoPilot for another fatal crash.
And Tesla wants me to pay $3000 for something that has a fairly high percentage of crashes compared to crashes of ICE cars? Considering how many Teslas are out there with AutoPilot I would believe the percentage of AutoPilot related crashes was pretty high.

Of course, there are no ( as far as I know ) ICE cars with AutoPilot. But they crash a lot.

Which is the main problem with self driving cars. Not the self driving car but the idiot driven cars all around it.

But it appears that AutoPilot is just a beta project and I don't think people should be paying to beta test for Tesla --- with their lives.

 

[VT_EE]

Then don’t buy it. Plain and simple. Meanwhile I will continue to use it properly and enjoy it. Btw-I suggest you also not use adaptive cruise control (on any brand). It is also run by a computer and prone to fail on occasion. In fact, you should not use any product because just about everything out there fails once in a while, sometimes with catastrophic consequences.

 

[Arizona Willie]

Then don’t buy it. Plain and simple. Meanwhile I will continue to use it properly and enjoy it. Btw-I suggest you also not use adaptive cruise control (on any brand). It is also run by a computer and prone to fail on occasion. In fact, you should not use any product because just about everything out there fails once in a while, sometimes with catastrophic consequences.

===========
It's one thing to use something that isn't perfect ( and know it isn't ) and another to be charged thousands of dollars for the priviledge of potentially being the next victim of a malfunction.

PS: I haven't bought it. Tesla would get much more revenue if they sold subcriptions. 1 penny per mile driven using AutoPilot.

 

[EVDRVN]

So autopilot caused the crash and the driver could not avoid it? No.

 

[Knightshade]

I see the investigating authorities blamed AutoPilot for another fatal crash.

They did no such thing.

And Tesla wants me to pay $3000 for something that has a fairly high percentage of crashes compared to crashes of ICE cars?

The rate of crashes on autopilot is significantly lower than without it.

Tesla publishes this data quarterly.

Considering how many Teslas are out there with AutoPilot I would believe the percentage of AutoPilot related crashes was pretty high.

Not sure why since all the facts say the opposite.

 

[Tam]

...Considering how many Teslas are out there with AutoPilot I would believe the percentage of AutoPilot related crashes was pretty high...

Tesla has been publishing Autopilot accidents or "crash-like" event (near misses) since Q3 2018

I myself feel much safer with Autopilot since 2017 for the past almost 36,000 miles.

...But it appears that AutoPilot is just a beta project...

It is definitely a beta. It says so in Owner's manual. It also says so on your car's Autopilot menu under Autosteer and Navigation on Autopilot.

...I don't think people should be paying to beta test for Tesla...

That is a reasonable expectation.

Most companies would either pay or give some kind of incentives for their beta testers.

However, if you make beta testers reach into their pockets to pay up, that should incentivize them to take a minute to think whether this is a wise decision or not.

...with their lives.

Inability to read or follow instructions can be deadly.

 

[Leafdriver333]

all car brands have similar thing to autopilot.
Ford, Chevy, MB, BMW, Audi etc...

 

[qdeathstar]

“facts” are wrong..

 

[Myrddin]

If this is the crash when the Model 3 drove under the truck in Delray Beach, FL last March... Did anyone actually read the NTSB'S (short) official report? It reports that Auto Pilot was ACTIVATED 10 SECONDS BEFORE THE CRASH.

"Released on Thursday, the NTSB’s report stated that, 'The driver engaged the Autopilot about 10 seconds before the collision. From less than 8 seconds before the crash to the time of impact, the vehicle did not detect the driver’s hands on the steering wheel. Preliminary vehicle data show that the Tesla was traveling about 68 mph when it struck the semitrailer. Neither the preliminary data nor the videos indicate that the driver or the ADAS executed evasive maneuvers.'"

The Tesla was traveling 68mph in a 55mph zone.I Don't know about your teslas, but mine warns me that auto steering is unavailable when traveling greater than five miles over what the car recognizes as the local speed limit. 68 mph - 55 mph > the 5 mph window in which the car could have reacted, even if allowing Autopilot a 10 second time period would have allowed adequate time for the Autopilot to wake and then take meaningful action. Implying that Autopilot has some blame in this accident and deathshows a remarkable ignorance of Autopilot and its capabilities... or a deliberate attempt to perpetuate disinformation,

 

[sduck]

This doesn't belong in the User Interface forum. Please take this elsewhere.

 

[darth_vad3r]

I see the investigating authorities blamed AutoPilot for another fatal crash.

Where did you see this? Source? I saw a report that didn’t blame Autopilot, maybe you saw a different one? Blaming Autopilot when it is on is like blaming the engine if it is on. The driver is responsible. Seriously. This is like blaming air conditioning because it was on.

And Tesla wants me to pay $3000 for something that has a fairly high percentage of crashes compared to crashes of ICE cars?

High percentage? How do you define this? Like 50% is high? Less but not ‘less’ enough?

Considering how many Teslas are out there with AutoPilot I would believe the percentage of AutoPilot related crashes was pretty high.

What makes you believe this? What’s “pretty high”, 2%? Do you believe there are a “pretty high” percentage of plane crashes too because those get covered extensively in the media as well?

Of course, there are no ( as far as I know ) ICE cars with AutoPilot. But they crash a lot.

Since “Autopilot” is a Tesla-only term (as far as I know) and Tesla doesn’t make ICE cars, then I would agree that there are no ICE cars with “Autopilot”. There are ICE cars with TACC, lane keeping auto steer, like Nissan’s “ProPilot Assist” which they offered in the ICE 2018 Nissan Rogue SUV (their best selling model in the US).

Which is the main problem with self driving cars. Not the self driving car but the idiot driven cars all around it.

Ya, like truck drivers making unsafe turns across traffic, but also the ones behind the wheel of Autopilot not paying attention. Seems those are two parties to share the blame in the latest accident.

 

[Helmuth]

I don’t think there was a single accident which was autopilots fault.
The driver is responsible! It’s like blaming a dumb cruise control if you rear end someone, or other ASSISTANT features.
AP is a feature for a more relaxed driving experience. Which it is IF you use it where and how it should be used!
I drive hundreds of miles every day. And I use it on the interstate all the time.
I use it between Orlando and West Palm Beach without disengaging it once! Close to three hours of driving without a single nag or disengagement (using NoA)!
Autopilot has its KNOWN limitations. Just don’t blame it if it hits a stationary object in front of it (or even half way in your lane), if it changes lanes in front of a speeding car coming from behind you, or if it doesn’t react for a tractor trailer crossing the street ahead of you!
I heard stories about drivers reading books and working through emails while using autopilot! And even if the car detects the drivers hand (while reading) on the wheel it wouldn’t be Tesla’s fault if it hits something.
Please pay attention and be attentive while using a great feature while it gets further developed and don’t be included in one of the futures (bad) headlines blaming autopilot for something it can’t handle yet...

 

[Arizona Willie]

If this is the crash when the Model 3 drove under the truck in Delray Beach, FL last March... Did anyone actually read the NTSB'S (short) official report? It reports that Auto Pilot was ACTIVATED 10 SECONDS BEFORE THE CRASH.

"Released on Thursday, the NTSB’s report stated that, 'The driver engaged the Autopilot about 10 seconds before the collision. From less than 8 seconds before the crash to the time of impact, the vehicle did not detect the driver’s hands on the steering wheel. Preliminary vehicle data show that the Tesla was traveling about 68 mph when it struck the semitrailer. Neither the preliminary data nor the videos indicate that the driver or the ADAS executed evasive maneuvers.'"

The Tesla was traveling 68mph in a 55mph zone.I Don't know about your teslas, but mine warns me that auto steering is unavailable when traveling greater than five miles over what the car recognizes as the local speed limit. 68 mph - 55 mph > the 5 mph window in which the car could have reacted, even if allowing Autopilot a 10 second time period would have allowed adequate time for the Autopilot to wake and then take meaningful action. Implying that Autopilot has some blame in this accident and deathshows a remarkable ignorance of Autopilot and its capabilities... or a deliberate attempt to perpetuate disinformation,

============
I don't bad mouth the car out of any hostility towards Tesla ( other than their prices ), I love my Tesla.

I've only had it 3 months and I'm still like a kid who just got his learner's permit. Looking for any reason to drive it.

But I also try to be realistic and not let my love for my car shut out reality and AutoPilot is NOT ready for prime time.

 

[Knightshade]

============
I don't bad mouth the car out of any hostility towards Tesla ( other than their prices ), I love my Tesla.

I've only had it 3 months and I'm still like a kid who just got his learner's permit. Looking for any reason to drive it.

But I also try to be realistic and not let my love for my car shut out reality and AutoPilot is NOT ready for prime time.

Not sure what prime time means in this context- but for this exact accident it was being used in a location the owners manual specifically says it's not intended to be used so it's pretty tough to blame the system for user error.

 

[darth_vad3r]

Autopilot is ready for “prime time”. You are confusing FSD for AP.
You are supposed to watch the road with AP but can relax your mind a bit. It makes a massive difference in fatigue for long trips and reduces slight hassle of maintaining speed/follow distance manually in traffic.

It’s not designed to avoid surprise stationary objects that radar normally ignores.

When that works FSD will be ready for “prime time”

 

[Helmuth]

I wouldn’t take my eyes off the road for road debris alone!
It happened just yesterday that there was a entire bumper in our lane! In the middle of the exit ramp where AP does a good job.
It doesn’t see tires, traffic cones, bumper, ladders, pot holes, animals crossing,...
And I don’t want our Tesla being damaged driving distracted while using autopilot in its current state.
But I love using it where it’s doing a great job (while supervised) and I enjoy arriving relaxed at the destination. Even after hours of “driving” or “supervising”...

 

[electrictorque]

============
I don't bad mouth the car out of any hostility towards Tesla ( other than their prices ), I love my Tesla.

I've only had it 3 months and I'm still like a kid who just got his learner's permit. Looking for any reason to drive it.

But I also try to be realistic and not let my love for my car shut out reality and AutoPilot is NOT ready for prime time.

The current state of Auto pilot is ready for what it is. Every time you turn it on the car shows you a message telling you to be alert and ready to take over at anytime.

I use it every day, knowing it's limitations and I'm loving it every time. Yes I have taken over many times, yes I don't trust it 100% but I have been witnessing tremendous improvements since the first time I used it.

 

[SiNiSon]

Auto-Pilot is Beta and it says so everywhere... Tesla has consistently said early adopters will pay more and get less.. so not sure what the issue here is.. they want people to pay a premium for a good product that will only get better.. and that part of getting better is paying more and getting less... for now... none of this is new and if it is you have not been paying attention.

============
I don't bad mouth the car out of any hostility towards Tesla ( other than their prices ), I love my Tesla.

I've only had it 3 months and I'm still like a kid who just got his learner's permit. Looking for any reason to drive it.

But I also try to be realistic and not let my love for my car shut out reality and AutoPilot is NOT ready for prime time.

 

[acoste]

The rate of crashes on autopilot is significantly lower than without it.

Tesla publishes this data quarterly.




Not sure why since all the facts say the opposite.

There is no evidence that support Autopilot being safer than human drivers.

This is what Tesla says in their safety report:

"In the 1st quarter, we registered one accident for every 2.87 million miles driven in which drivers had Autopilot engaged. For those driving without Autopilot, we registered one accident for every 1.76 million miles driven. By comparison, NHTSA’s most recent data shows that in the United States there is an automobile crash every 436,000 miles."

First, there is no reason to compare all cars vs Tesla with autopilot since the demography is not the same.

Second, Tesla doesn't say anything about their comparison method. The only reasonable method would be if they compared the miles driven by Autopilot versus the miles driven manually on the same road under the same conditions (rain, night, fog, etc). Since their method is not disclosed it can't be accepted as facts. The way they wrote it it just means they counted the total miles with and without Autopilot regardless of the conditions.
Independent third party data would be the best.

Third. This does not belong to this topic but I stumbled upon it in the safety report.
" That’s why Tesla vehicles are engineered to be the safest cars in the world. "
Now this isn't proven either. Nor IIHS nor EuroNCAP have tested the Model 3. NHTSA doesn't test many foreign cars, and their methods are favorable to EVs.

 

[darth_vad3r]

This does not belong to this topic but I stumbled upon it in the safety report.
" That’s why Tesla vehicles are engineered to be the safest cars in the world. "
Now this isn't proven either. Nor IIHS nor EuroNCAP have tested the Model 3. NHTSA doesn't test many foreign cars, and their methods are favorable to EVs.

You don’t have to prove it, it’s a statement of opinion. They aren’t saying they ARE the safest cars in the world, they are saying (in their opinion) they ENGINEER them to be the safest cars in the world.