Tesla's autopilot, on the road

[GRA]

IEVS:

Despite Adversity And Impeding Issues, Tesla Autopilot Will Prevail

https://insideevs.com/despite-adversity ... l-prevail/

This originally comes from Evannex, which I generally consider almost a shill for Tesla given its constant boosterism of the company (and dependence on its existence to sell their products), but judge the article for yourself on the merits.

[GRA]

I can't wait to hear all the same complaints about ProPilot, since it works very similar to AP and you are allowed to use it on similar roads (including those with intersections and traffic lights).

If it causes similar accidents, you certainly will hear those complaints. As it appears to be similarly unrestricted as to when and where it can be used as A/P is, I expect we'll start to hear about some in the not too distant future. From reading reviews, it's lane-keeping ability suffers similar problems to A/P.

[Oils4AsphaultOnly]

I can't wait to hear all the same complaints about ProPilot, since it works very similar to AP and you are allowed to use it on similar roads (including those with intersections and traffic lights).

If it causes similar accidents, you certainly will hear those complaints. As it appears to be similarly unrestricted as when and where it can be used as A/P is, I expect we'll start to hear about some in the not too distant future. From reading reviews, it's lane-keeping ability suffers similar problems to A/P.

Go test drive a tesla. Try it out for yourself first before you stuff your foot any further in your mouth.

[GRA]

I can't wait to hear all the same complaints about ProPilot, since it works very similar to AP and you are allowed to use it on similar roads (including those with intersections and traffic lights).

If it causes similar accidents, you certainly will hear those complaints. As it appears to be similarly unrestricted as to when and where it can be used as A/P is, I expect we'll start to hear about some in the not too distant future. From reading reviews, it's lane-keeping ability suffers similar problems to A/P.

Go test drive a tesla. Try it out for yourself first before you stuff your foot any further in your mouth.

Why would driving a Tesla affect my views on how A/P can cause accidents, when it (or any other assistance system currently available) obviously can? After all, the whole point of such assist systems is to remove me from some of the driving. If I'm not solely driving the car, then I'm not solely responsible for any accident that A/P causes. Same goes for Propilot or any other such system. The reviews of Propilot indicate that it too can wander in the lane or cross double yellow lines (especially on sharp curves) , just like A/P has been documented to do. As I've written repeatedly, I have absolutely zero intention of trusting my life to any L2 or L3 system, and will wait until L4 has been well proven in service before I will do so.

Meanwhile, I think this is the best comparo of A/P and Supercruise I've read, and while I think Supercruise is the winner for the typical driver, I still wouldn't trust my or anyone else's safety to either of them: http://www.thedrive.com/tech/17083/the- ... ad-to-head

[Oils4AsphaultOnly]

If it causes similar accidents, you certainly will hear those complaints. As it appears to be similarly unrestricted as to when and where it can be used as A/P is, I expect we'll start to hear about some in the not too distant future. From reading reviews, it's lane-keeping ability suffers similar problems to A/P.

Go test drive a tesla. Try it out for yourself first before you stuff your foot any further in your mouth.

Why would driving a Tesla affect my views on how A/P can cause accidents, when it (or any other assistance system currently available) obviously can? After all, the whole point of such assist systems is to remove me from some of the driving. If I'm not solely driving the car, then I'm not solely responsible for any accident that A/P causes. Same goes for Propilot or any other such system. The reviews of Propilot indicate that it too can wander in the lane or cross double yellow lines (especially on sharp curves) , just like A/P has been documented to do. As I've written repeatedly, I have absolutely zero intention of trusting my life to any L2 or L3 system, and will wait until L4 has been well proven in service before I will do so.

Meanwhile, I think this is the best comparo of A/P and Supercruise I've read, and while I think Supercruise is the winner for the typical driver, I still wouldn't trust my or anyone else's safety to either of them: http://www.thedrive.com/tech/17083/the- ... ad-to-head

Because right now it feels like I'm arguing with a virgin about the joys of sex. Everything you've read indicates that it can be messy, traumatic, and fraught with all sorts of peril. But reality is a huge world of difference that you'd have to experience to understand.

[Oils4AsphaultOnly]

Why would driving a Tesla affect my views on how A/P can cause accidents, when it (or any other assistance system currently available) obviously can? After all, the whole point of such assist systems is to remove me from some of the driving. If I'm not solely driving the car, then I'm not solely responsible for any accident that A/P causes. Same goes for Propilot or any other such system. The reviews of Propilot indicate that it too can wander in the lane or cross double yellow lines (especially on sharp curves) , just like A/P has been documented to do. As I've written repeatedly, I have absolutely zero intention of trusting my life to any L2 or L3 system, and will wait until L4 has been well proven in service before I will do so.

Meanwhile, I think this is the best comparo of A/P and Supercruise I've read, and while I think Supercruise is the winner for the typical driver, I still wouldn't trust my or anyone else's safety to either of them: http://www.thedrive.com/tech/17083/the- ... ad-to-head

Because right now it feels like I'm arguing with a virgin about the joys of sex. Everything you've read indicates that it can be messy, traumatic, and fraught with all sorts of peril. But reality is a huge world of difference that you'd have to experience to understand.

To take this analogy one step further, those who engage in the misuse of autopilot are essentially sado-masochists. They get enjoyment out of the deviant behaviour and usually hurt themselves and others in the process, which sometimes ends up in death (granted the ratio of accidental death is higher with autopilot than with auto-asphyxiation, but the analogy is quite apt).

[EVDRIVER]

The AP experts seem to be those that don't own the car. Interestingly, you can read most comments from that group and instantly tell they have no experience using AP but they seem to be the experts.

[jlv]

Everything you've read indicates that it can be messy, traumatic, and fraught with all sorts of peril. But reality is a huge world of difference that you'd have to experience to understand.

+1

[GRA]

Why would driving a Tesla affect my views on how A/P can cause accidents, when it (or any other assistance system currently available) obviously can? After all, the whole point of such assist systems is to remove me from some of the driving. If I'm not solely driving the car, then I'm not solely responsible for any accident that A/P causes. Same goes for Propilot or any other such system. The reviews of Propilot indicate that it too can wander in the lane or cross double yellow lines (especially on sharp curves) , just like A/P has been documented to do. As I've written repeatedly, I have absolutely zero intention of trusting my life to any L2 or L3 system, and will wait until L4 has been well proven in service before I will do so.

Meanwhile, I think this is the best comparo of A/P and Supercruise I've read, and while I think Supercruise is the winner for the typical driver, I still wouldn't trust my or anyone else's safety to either of them: http://www.thedrive.com/tech/17083/the- ... ad-to-head

Because right now it feels like I'm arguing with a virgin about the joys of sex. Everything you've read indicates that it can be messy, traumatic, and fraught with all sorts of peril. But reality is a huge world of difference that you'd have to experience to understand.

The reality is that several people have died and others have been injured in A/P driven cars, when the accidents wouldn't have happened if the cars had been driven by an alert and engaged driver. I've been driving over 40 years, and have yet to cross a centerline except when I intended to (to enter a driveway). Current semi-autonomous systems do so all too frequently to provide peace of mind. As crossing the centerline into oncoming traffic is one of the top three causes of serious or fatal accidents (the others being road departure and failure to yield), why on earth would anyone think that engaging a system that will increase the chance of that happening is a safety improvement?

Similarly, I am able to recognize stopped vehicles in front of me and take appropriate action, rather than assuming that they are part of scenery, ignoring them and plowing into them at high speed. And so on. But only if I'm engaged and alert, and anything that allows and encourages me to not be either increases rather than decreases the risks at the current state of AFV development.

[GRA]

The AP experts seem to be those that don't own the car. Interestingly, you can read most comments from that group and instantly tell they have no experience using AP but they seem to be the experts.

Plenty of people who have used A/P and similar systems have exactly the same reservations I do, including many of those involved in developing AVs. Or do you not consider as experts the Mobileye or Tesla people (to name two groups directly involved with A/P) who tried to get Elon to restrict A/P for safety reasons? There are many others not involved with Tesla or any of the other companies trying to introduce semi-autonomous systems who have the same concerns. "Driverless" includes arguments from those on both sides, but the authors are unquestionably in favor of going direct to full autonomy for safety reasons (the hand-off problem). One of them was part of the Cornell team who competed in the DARPA challenges - is he not an expert?