SHOCKER - Autopilot Buddy Deemed Unsafe

Tawnos

2[H]4U
Joined
Sep 9, 2001
Messages
3,807
So you are implying that software, created by buggy humans, will somehow be more reliable? Consider also that self-driving cars will depend upon hardware. That hardware will break or become degraded by dirt, debris, etc. Those promising self-driving cars are greatly over-simplifying the challenge. At this point in time we flawed human beings are far superior to any self-driving car. I suggest humans will remain superior for quite a long time...

I'm beginning to think your job is "car salesman to the uneducated". You spend more time posting about your hate of self-driving cars or electric cars and how much you hate "progressives" than you do on any other topic here (I just checked to make sure it wasn't because I primarily see your drivel in the front page forum, apparently it's the only place you really spend time posting).

Yes, software is more reliable than us. Whether due to Bayesian models leading to reduction in variance, engineering processes that allow for individual shortcomings to be handled better by many, or just the fact that we have late nights, stressful situations, frustrations caused by the actor-observer asymmetry bias (e.g. "that asshole cut me off!" versus "oh man, I didn't realize that lane was the right one, glad that person is letting me in"), and other such situations that make our bugs not only common, but unpredictable.

You claim to be better than "any self-driving car". If we were to put together a test of that, what would be something that would convince you otherwise or affirm that you are? Would the same test apply to the driving population?
 

Tawnos

2[H]4U
Joined
Sep 9, 2001
Messages
3,807
Not really. Humans are easy to predict, no matter how stupid they are.

Software, on the other hand, is wildly unpredictable. It is that way as we expect software to act a certain way and when it idoes not, we freeze, cuss it out, scratch our heads, reboot our computers, unplug devices, and so on. When software does not act the way we expect it to, it is far more debilitating than watching a human do something stupid. We expect stupid behavior from humans.

Oh look, he/she is crossing 4 lanes of traffice to get to the exit.
Oh look, he/she is tailgating me without concern that my car can stop 30% faster then what they are driving.
Oh look, he/she is on the cell phone drifting all over the place.

None of those things, plus many more, surprise us, unless you are the stupid one not watching what is going on.

Now if sofware drives you into a concrete wall because the mural painted on it fooled it, then that would be surprising.
Or if it takes you cross country because of the snow on the ground making it appear as one large stripe and it is trying to find the edge, that might be surprising.
And so on.....

For all our ills, humans are still better at real time wide scale pattern recognition than computers are. Even stupid humans.

If that's the case, why is it "news" when a self-driving car is in an accident, but the 100+ vehicular deaths per day are not? Expecting stupid behavior doesn't change how dangerous the behavior is, increase our reaction time to it, nor is it an insurmountable engineering challenge to have software react better and more predictably than a human faced with the very situations you mentioned. Not to mention how silly they sound when you consider that such barrier accidents _happen with humans all the time_ and you conveniently ignore them.

What's really got you scared? Do you feel like self-driving cars threaten your agency or freedom in some way?
 

BSmith

[H]ard|Gawd
Joined
Nov 9, 2017
Messages
1,323
Reaching deep there aren't you? You know when you draw a conclusion based on your own assumptions about something said, those conclusions are usually wrong.

Never said humans did not run into barriers. Never said it was an insurmountable challenge. Nothing is impossible until we give up. The accusation of me being scared or threatened, is a poorly drawn conclusion based on your own speculations.

Bosch, the maker of most of the sensor systems for autonomous cars said to do it right would take approximately 1500W@12V of hardware and would barely fit into the trunk of a large sedan. That is today. My own speculation is the hardware to do it right is simply not efficient enough.

Try shoving that into an electric car. Not going to happen. Even a hybrid would struggle and gas mileage would tank. As to the safety, I'll wait to see how well autonomous testing in the real world of bad weather and bad roads is done before I soldify my position. I think it only prudent to do so.

That opinion aside. I'll never own one. I like to drive. I belong to two car clubs and we take road trips all the time. It is a hobby I dearly enjoy and no one is going to take that away from me.

EDIT: Forgot to answer the original question. Why is it news when a self driving car kills someone and it is not news when people do? I have a number of speculations about that.
1) If a self driving klills someone in the simplest of situations where a human would rarely make that mistake, it is going to be news.
2) News is about getting attention/ratings. Jumping on the band wagon of exploiting anything detrimental in new technology is a go to item for news people.
3) We are used to people being killed in automobile accidents. As long as it is someone we do not know, then we no longer care abouit it.

Probably some other points I am missing, but I think that cover a large percentage.

Regardless of the reasons, it would be ill-advised to allow autonmous cars to be released into the wild until they can prove they can handle every normal situation (bad roads, bad weather, poor signage,...) we have to deal with as humans. That is just my opinion.
 
Last edited:

Curl

Limp Gawd
Joined
Jul 13, 2004
Messages
231
One of my friends just took delivery of his Model 3. Although I wouldn't buy one, its pretty neat. As for the autodrive stuff, when he let me drive, the damn thing almost totaled itself when approaching a red light as the car in front of us changed lanes suddenly approaching the stop. Luckily, I didn't trust it enough and was able to brake before we rear-ended the car in front of us.
 

thejokker

Gawd
Joined
Mar 16, 2005
Messages
831
If that's the case, why is it "news" when a self-driving car is in an accident, but the 100+ vehicular deaths per day are not? Expecting stupid behavior doesn't change how dangerous the behavior is, increase our reaction time to it, nor is it an insurmountable engineering challenge to have software react better and more predictably than a human faced with the very situations you mentioned. Not to mention how silly they sound when you consider that such barrier accidents _happen with humans all the time_ and you conveniently ignore them.

What's really got you scared? Do you feel like self-driving cars threaten your agency or freedom in some way?
There are tens of millions of driven cars on the road in America each day. You are assuming the fatal accidents are due to stupidity and not mechanical failure or adverse environmental conditions. You are assuming self-driving cars will be more reliable than human drivers when all across the world computing devices suffer component fails and system crashes. Given a novel situation a human being has a reasonable probability of reacting appropriately whereas a software controlled vehicle will be unable to response without a programmed response to an unanticipated situation.

So what's with this "scared"? Are you twelve? Anyone who has a different opinion must be afraid? You come across as irrationally biased; with an almost religious zealotry towards the inferiority of man vs the superiority of machines. That you lack a healthy skepticism towards an experimental technology in it's infancy tells a story...
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,760

thejokker

Gawd
Joined
Mar 16, 2005
Messages
831
https://www.huffingtonpost.com/laiza-king-/top-15-causes-of-car-accidents_b_11722196.html
Not assuming, just using facts and statistics.
And facts and statistics will show that self driving vehicles will be safer.
Sorry but not a logical conclusion. From the facts you have presented by the HuffPo article you can only draw conclusions about the causes of human driven accidents. There in sufficient data about the causes of self-driving cars. There are a multitude of potential causes of accidents unique to self-driving cars: i.e. Degradation or failure of sensors, Software crashes and errors, etc. You are making broad and unsupported assumptions about a technology that is not only pre-beta but pre-alpha. We are lead to believe by some that self-driving cars are only years away but in fact they may turn out to be decades away.
 

Grimlaking

2[H]4U
Joined
May 9, 2006
Messages
3,250
Has everyone seen Cadillac's new self driving feature? You don't need to touch the wheel but it has eye tracking to make sure you're being attentive. And only works on major highways/expressways. Still seems pretty cool. (Though I wonder how it works with sunglasses.)
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,760
Sorry but not a logical conclusion. From the facts you have presented by the HuffPo article you can only draw conclusions about the causes of human driven accidents. There in sufficient data about the causes of self-driving cars. There are a multitude of potential causes of accidents unique to self-driving cars: i.e. Degradation or failure of sensors, Software crashes and errors, etc. You are making broad and unsupported assumptions about a technology that is not only pre-beta but pre-alpha. We are lead to believe by some that self-driving cars are only years away but in fact they may turn out to be decades away.
I never mentioned a time frame as to when level 5, much less less level 4, self driving vehicles would be the norm. But rest assured that time is coming. https://www.caranddriver.com/featur...f-driving-car-levels-0-to-5-explained-feature
The majority of Humans will never be able to drive better than self driving vehicles.

As for sensor failure, those things will be worked out.
https://theaviationist.com/2018/02/...-take-off-from-guam-and-heres-are-two-videos/
 

thejokker

Gawd
Joined
Mar 16, 2005
Messages
831
I never mentioned a time frame as to when level 5, much less less level 4, self driving vehicles would be the norm. But rest assured that time is coming. https://www.caranddriver.com/featur...f-driving-car-levels-0-to-5-explained-feature
The majority of Humans will never be able to drive better than self driving vehicles.

As for sensor failure, those things will be worked out.
https://theaviationist.com/2018/02/...-take-off-from-guam-and-heres-are-two-videos/
I have a digital subscription to Car and Driver. Elsewhere in that issue of Car and Driver the editors speculates that it may be another 30 years before level 5 cars are readily available for the average driver. Your post about sensor failure only supports my argument. In fact both your links support my position that self-driving cars are decades away. Similarly your "opinion" that humans will never be able to drive better than self driving cars at this point in time is not supported by reality.

Likewise the future has an annoying tendency to disappoint those who believe they can predict it. Self driving cars may one day be a reality but it is far from an absolute certainty.
 

dreamwriter

Limp Gawd
Joined
Jun 22, 2018
Messages
264

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
We went over this extensively right after this accident, but here is new news and it looks really bad.

If what I researched and believed was true was accurate, then what is happening now looks like Uber is throwing their employee under the bus.
https://abcnews.go.com/GMA/News/video/bodycam-video-reveals-details-deadly-uber-driving-car-56105010

Please keep in mind, just a day or so before this accident, the AZ. Governor had just signed new regulations butting a little tighter control on autonomous car testing. Arizona signed up to allow Uber and others a lot of room for testing in the State and but very little in the way of demands on the companies involved.

There is another threat here on the accident with several links to statements made by Uber, etc, that support what I have said here.

BTW, the video I linked to says that investigators claim that if the employee had been paying attention she could have hit the brakes and stopped the car 42' short of hitting the woman. I checked earlier, a car going 40MPH, the approx speed of this vehicle, it would travel over 50' a second. So what the investigator is saying is that if the employee had failed to react within less than the first possible second then she could not have avoided killing the woman. But they say they are considering charges, didn't say if they were charging the driver or Uber.
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,760
I have a digital subscription to Car and Driver. Elsewhere in that issue of Car and Driver the editors speculates that it may be another 30 years before level 5 cars are readily available for the average driver. Your post about sensor failure only supports my argument. In fact both your links support my position that self-driving cars are decades away. Similarly your "opinion" that humans will never be able to drive better than self driving cars at this point in time is not supported by reality.

Likewise the future has an annoying tendency to disappoint those who believe they can predict it. Self driving cars may one day be a reality but it is far from an absolute certainty.
I recommend you sit in your lawn chair and practice yelling at people to get off said lawn.
  • The New York Times, Oct 9, 1903, p. 6.
That the automobile has practically reached the limit of its development is suggested by the fact that during the past year no improvements of a radical nature have been introduced.
  • What can be more palpably absurd than the prospect held out of locomotives traveling twice as fast as stagecoaches?
    • The Quarterly Review, March, 1825.
Hence, if it requires, say, a thousand years to fit for easy flight a bird which started with rudimentary wings, or ten thousand for one which started with no wings at all and had to sprout them ab initio, it might be assumed that the flying machine which will really fly might be evolved by the combined and continuous efforts of mathematicians and mechanicians in from one million to ten million years--provided, of course, we can meanwhile eliminate such little drawbacks and embarrassments as the existing relation between weight and strength in inorganic materials.

TLDR
The future may not be for you.
 

dreamwriter

Limp Gawd
Joined
Jun 22, 2018
Messages
264
Throwing the employee under the bus...What else is new. People that are indifferent to their fellow citizens being treated as next to nothing might wake up to the same fate themselves.

In a civilzed society it is possible for everyone to propsper.
 

Nobu

Supreme [H]ardness
Joined
Jun 7, 2007
Messages
4,932
We went over this extensively right after this accident, but here is new news and it looks really bad.

If what I researched and believed was true was accurate, then what is happening now looks like Uber is throwing their employee under the bus.
https://abcnews.go.com/GMA/News/video/bodycam-video-reveals-details-deadly-uber-driving-car-56105010

Please keep in mind, just a day or so before this accident, the AZ. Governor had just signed new regulations butting a little tighter control on autonomous car testing. Arizona signed up to allow Uber and others a lot of room for testing in the State and but very little in the way of demands on the companies involved.

There is another threat here on the accident with several links to statements made by Uber, etc, that support what I have said here.

BTW, the video I linked to says that investigators claim that if the employee had been paying attention she could have hit the brakes and stopped the car 42' short of hitting the woman. I checked earlier, a car going 40MPH, the approx speed of this vehicle, it would travel over 50' a second. So what the investigator is saying is that if the employee had failed to react within less than the first possible second then she could not have avoided killing the woman. But they say they are considering charges, didn't say if they were charging the driver or Uber.
No, what they are saying is if the employee was paying attention, even accounting for reaction time, after seeing the woman and applying brakes the car would have stopped 42' short of the woman. If the woman had come into view a second later, neither the driver nor the system could have stopped in time, but that's not what happened. Whether this happened in a "self-driving" car or not, the driver would be charged with (though not necessarily found guilty of) distracted driving. Whether they were found guilty would likely depend on whether the distraction was caused by legal driving activity (like checking your mirrors), illegal driving activity (like talking on your phone or watching a movie), or simple negligence.
 

thejokker

Gawd
Joined
Mar 16, 2005
Messages
831
I recommend you sit in your lawn chair and practice yelling at people to get off said lawn.
  • The New York Times, Oct 9, 1903, p. 6.
That the automobile has practically reached the limit of its development is suggested by the fact that during the past year no improvements of a radical nature have been introduced.
  • What can be more palpably absurd than the prospect held out of locomotives traveling twice as fast as stagecoaches?
    • The Quarterly Review, March, 1825.
Hence, if it requires, say, a thousand years to fit for easy flight a bird which started with rudimentary wings, or ten thousand for one which started with no wings at all and had to sprout them ab initio, it might be assumed that the flying machine which will really fly might be evolved by the combined and continuous efforts of mathematicians and mechanicians in from one million to ten million years--provided, of course, we can meanwhile eliminate such little drawbacks and embarrassments as the existing relation between weight and strength in inorganic materials.

TLDR
The future may not be for you.
I find your childlike naivete amusing... You're just like those bozo's from the 1950's who were "positive" everyone would have a flying car by now.

42BE654F00000578-4736484-image-a-2_1501169742891.jpg

I have some prime real estate in the Florida Everglades if you are interested...
 

Grimlaking

2[H]4U
Joined
May 9, 2006
Messages
3,250
And there go's the disussion. now it's all.. "My news article says x." "My news article says Y" "Damn you bob" "Screw you bill" *BANG* *BANG* --thud-- --thud-- "Now how about some coffee" *groans*
 

tangoseal

[H]F Junkie
Joined
Dec 18, 2010
Messages
9,097
The problem with Tesla Autopilot is that it wasn't designed for full autonomy from point to point. It was designed to have the driver pay attention. Anything that bypasses the driver paying attention is dangerous as shit.

As I'm a regular driver and with my son in my truck that thing I want is my son to die because an isiot driving a Tesla hits us because he was too lazy to look at the road.
 

BSmith

[H]ard|Gawd
Joined
Nov 9, 2017
Messages
1,323
https://www.huffingtonpost.com/laiza-king-/top-15-causes-of-car-accidents_b_11722196.html
Not assuming, just using facts and statistics.
And facts and statistics will show that self driving vehicles will be safer.

Did you see the source for the information? Not exactly what I would call an objective credible source. If that is what you call "using facts and statistics" to support your summation, then good luck with that doctorate thesis.

Probably should retitle that article "What kind of car accidents cause more law suits than any other?"

I am not saying there is not any truth to the list, and it would not surprise me if many of them were not in the NHTSA statistics. Personally, I just do not see much credibility in your support data.

Generally speaking, I do not find the huffington post to be the best source of unbiased information.
 
Last edited:

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
No, what they are saying is if the employee was paying attention, even accounting for reaction time, after seeing the woman and applying brakes the car would have stopped 42' short of the woman. If the woman had come into view a second later, neither the driver nor the system could have stopped in time, but that's not what happened. Whether this happened in a "self-driving" car or not, the driver would be charged with (though not necessarily found guilty of) distracted driving. Whether they were found guilty would likely depend on whether the distraction was caused by legal driving activity (like checking your mirrors), illegal driving activity (like talking on your phone or watching a movie), or simple negligence.


I don't think you see it.

According to the earlier reports, the employee was not a driver, not a backup driver, not an emergency driver, not driving. Two employees were normally assigned to the vehicle, one to monitor the vehicle, one to do data-entry on the vehicle data. The employee had stated that recently, Uber had been short people so one employee was required to both monitor the vehicle and do the data entry, the employee reported that she was doing data entry at the time of the accident. Arizona agreed to these procedures to allow the testing of these vehicles.

If this past information was accurate, how do you charge a driver who isn't a driver?

I can see you charging a company and I can see you charging the State. But I do not see how you but a human into an impossible situation and charge them for failing to do what they were not responsible for doing.

Again, this is all based on previous reporting related to this accident, the links are in the other thread, too easy to search and read for yourself.

But now it seems the government is going to set charges although this article doesn't specifically say who is being charged, and it looks like Uber is making noises like they are willing to throw this employee right under the bus.

And I understood perfectly well what the investigators are saying and I communicated it perfectly well. They said in simplest terms, the car could have been stopped 42' short of impact, but a car travels over 50' in 1 sec, meaning that even 1 second late on reaction time and the woman would still have died. Try again telling me I am wrong.
 

Nobu

Supreme [H]ardness
Joined
Jun 7, 2007
Messages
4,932
I don't think you see it.

According to the earlier reports, the employee was not a driver, not a backup driver, not an emergency driver, not driving. Two employees were normally assigned to the vehicle, one to monitor the vehicle, one to do data-entry on the vehicle data. The employee had stated that recently, Uber had been short people so one employee was required to both monitor the vehicle and do the data entry, the employee reported that she was doing data entry at the time of the accident. Arizona agreed to these procedures to allow the testing of these vehicles.

If this past information was accurate, how do you charge a driver who isn't a driver?

I can see you charging a company and I can see you charging the State. But I do not see how you but a human into an impossible situation and charge them for failing to do what they were not responsible for doing.

Again, this is all based on previous reporting related to this accident, the links are in the other thread, too easy to search and read for yourself.

But now it seems the government is going to set charges although this article doesn't specifically say who is being charged, and it looks like Uber is making noises like they are willing to throw this employee right under the bus.

And I understood perfectly well what the investigators are saying and I communicated it perfectly well. They said in simplest terms, the car could have been stopped 42' short of impact, but a car travels over 50' in 1 sec, meaning that even 1 second late on reaction time and the woman would still have died. Try again telling me I am wrong.
You're right. Wouldn't surprise me if they charge them anyway, though.
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
You're right. Wouldn't surprise me if they charge them anyway, though.


We do have a trial system so someone is probably getting charged. It's possible to charge more than one person, for instance, they could charge Uber with the crime and the employee as an accessory, or vice verse. Guilt is another issue entirely for which I am thankful.
 

Tsumi

[H]F Junkie
Joined
Mar 18, 2010
Messages
13,389
I don't think you see it.

According to the earlier reports, the employee was not a driver, not a backup driver, not an emergency driver, not driving. Two employees were normally assigned to the vehicle, one to monitor the vehicle, one to do data-entry on the vehicle data. The employee had stated that recently, Uber had been short people so one employee was required to both monitor the vehicle and do the data entry, the employee reported that she was doing data entry at the time of the accident. Arizona agreed to these procedures to allow the testing of these vehicles.

If this past information was accurate, how do you charge a driver who isn't a driver?

I can see you charging a company and I can see you charging the State. But I do not see how you but a human into an impossible situation and charge them for failing to do what they were not responsible for doing.

Again, this is all based on previous reporting related to this accident, the links are in the other thread, too easy to search and read for yourself.

But now it seems the government is going to set charges although this article doesn't specifically say who is being charged, and it looks like Uber is making noises like they are willing to throw this employee right under the bus.

And I understood perfectly well what the investigators are saying and I communicated it perfectly well. They said in simplest terms, the car could have been stopped 42' short of impact, but a car travels over 50' in 1 sec, meaning that even 1 second late on reaction time and the woman would still have died. Try again telling me I am wrong.

Hopefully she retains a copy of her original job description. If it doesn't say safety driver or takeover in the case of an emergency in there, then she has a fighting chance of not being held liable.
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
Hopefully she retains a copy of her original job description. If it doesn't say safety driver or takeover in the case of an emergency in there, then she has a fighting chance of not being held liable.

There is that other mitigating factor for which I am a little perplexed as to why they are talking about charges. The woman who was killed was jaywalking. I can't say all the time but I do believe that frequently when someone is killed while jaywalking no charges are filed unless the driver is found impaired or negligent in some way. So yes, if it was in fact the employee's job to do date entry while the vehicle is operational and unless they can prove she was using her phone or something else instead, then it could be a tough sell in court.

Of course the wrongful death lawsuit is another thing entirely.
 

Jim Kim

2[H]4U
Joined
May 24, 2012
Messages
3,760
Did you see the source for the information? Not exactly what I would call an objective credible source. If that is what you call "using facts and statistics" to support your summation, then good luck with that doctorate thesis.Probably should retitle that article "What kind of car accidents cause more law suits than any other?"I am not saying there is not any truth to the list, and it would not surprise me if many of them were not in the NHTSA statistics. Personally, I just do not see much credibility in your support data.Generally speaking, I do not find the huffington post to be the best source of unbiased information.
Feel free to come up with something that disputes what they posted. Hint: I'll save you the time, it does not exist.
Sorry my link was not up to your journalistic standards of excellence.
 

BSmith

[H]ard|Gawd
Joined
Nov 9, 2017
Messages
1,323
Feel free to come up with something that disputes what they posted. Hint: I'll save you the time, it does not exist.
Sorry my link was not up to your journalistic standards of excellence.

My journalistic standards have nothing to do with the incorrect use of data. You are correct, the data you want does not exist unless you draw it from a source which was not designed to support the argument, but instead was reinterpreted to fit some other agenda.
 
Top