Tesla runs red light into a semi-trailer truck

Darn near did it myself last night. 65 MPH interstate. I'm doing 50-55 in the slow lane because it is dark and raining very hard. All of a sudden a cluster of red lights flashing up ahead as I round a curve. A car behind me. And I have to sort out what to do in an instant through the glare with my eyes desperately trying to pick a light level to begin functioning from. Oh yea the emergency truck was all lit up as were about 5 other cars. Now with 60 years of driving I got through it but the pucker factor was extreme. No cruise control involved. Luck.

 

has anyone asked the emergency vehicle workers if they would prefer the risk with typical smartphone users that aren't driving vs Tesla autopilot ? (you know, one out of 6000+ collisions can be blamed on the technology, all the rest of them can be blamed on humans making bad decisions)

has anyone asked emergency vehicle workers if they would consider pulling people over at off ramps instead of 3 feet away from 75 mph traffic fill of smartphones that aren't driving and Tesla autopilots ?
(you know, safety and all)

has anyone considered that 1/6,000 collisions (none of these are accidents) is a pretty dang good ratio to improve the hit/miss expectations ?

 

60 years of experience for your appropriate actions, and a huge helping of luck that the driver behind wasn't following too closely... or texting

 

You are repeating yourself. Cell phone use is outlawed in most areas, so by default it falls under 'making bad decisions'.

 

in the USA, smartphone use is not outlawed in most areas. the technology were talking about is the Tesla auto pilot system, and from what I've read here, it's 1/6000th as bad at driving as humans are, when considering emergency Vehicle collisions.

I doubt that smartphone use accounts for 5,999 of those collisions....

but Tesla is only accountable for one of them.


again, if traffic stops were not made in the side of the interstate, but instead carried out on the side of an exit ramp, or so far as a gas station, the chance of being struck by traffic traveling at 70+ mph would be incredibly low.

 

No.

You're assuming two things. The first is that this is the only time a Tesla on autopilot has ever hit a vehicle parked on the side of the road. The second is that Tesla's on autopilot have driven the same number of hours/miles as all other cars put together.

 

no - it's a comparitive estimation of how many times per year one manufacturer or another is involved in accidents - not manufacturer ever hitting another in all of history. Equally important, hitting a vehicle, whether parked or in motion is an accident & thus germane to the topic.
This notion is a lot like ev's catching on fire versus gas cars catching on fire. It makes the news when it's the new technology, not the fires that are ubiquitously happening on regular gas cars. But since we seldom here of car's gas fires, the media makes it sound like it's mostly electric cars catching on fire.
.

 

Which is an entirely useless metric. For example, none of the cars I've built have ever been involved in an accident. Does that make them the safest cars?

To put this another way, the Concorde was the safest plane around until it had its first accident, at which point it became the least safe commercial airliner - from one accident.

 

so sorry - giving readers the benefit of the doubt that they can understand a manufacturer of one car does not equal a manufacturer of hundreds of thousands or millions of cars from which we can extrapolate statistics of accident frequencies per millions of miles ..... but should your single manufactured car ever rack up tens or hundreds of millions of miles, sure, we can put that in the metric.

.

 

Tesla's rack up something like 1% of the miles driven in the US. If you don't compare on the bases of miles driven or interactions or something to make up the difference in use, then the metric is useless. Sorry you didn't get that.

 

I often drive on this freeway on-ramp on 101. It is relatively new and allows cars to merge onto two different freeways: 85 or 101. It is very confusing and poorly marked. The signs say that one lane takes you to one freeway, and the other lane takes you to the other freeway. But when you drive around the turn you find yourself in a middle lane! Where does that take you? Do I need to change lanes in the next two seconds? The crash barrier where the Tesla crashed is often crushed, which shows that human intelligences have just as much trouble with this on-ramp.

Google Maps

 

According to google street maps, that's been there since at least 10 years. It has pictures from as far back as may 2011.

The difference between humans having problems with that on ramp and Self driving cars having problems, is that once it's reported to Tesla as a problem, they should have mapped it into their software so that ALL Teslas would no longer have problems in the future. Instead, there have been reports of multiple people having to take control at that intersection.

The human analogy is: If my friend tells me that he almost blew that intersection, I become hyper aware of the problem when I use that off ramp.

 

mannnnnnn this conversation went straight from interesting to impossible quick.

oh well, I guess once everyone agrees that we need to know the tire manufacture, miles on tires, sunroof usage, smartphone vs cellphone, and glasses vs no glasses them and only then can we use a ratio to compare Tesla's hitting emergency vehicles vs non Tesla's.

or something, I'm really not sure at this point.


while we're waiting, it sure would be nice if the non emergency stops (you know, the ones where a driver receives a citation) were modified. it would REALLY nice if both the revenue generators and the revenue contributes would agree that stopping next to 55-75mph traffic to hand each other pieces of paper _might_ be more safely done at an off ramp or (gasp) a gas station. I seriously don't understand , if emergency vehicles are collided with so often, why the tickets are still being issued on the side of high speed interstates, instead of the next available exit ramp/ gas station. it would be so much safer for everyone involved.

 

it's really not as complex as some here continually desire to make it - & then change issues to another oppositional point to deflect. There's a thread with links here on PC that has already discussed this issue ad nauseam.
Auto accident actuaries base accidents on a per million mile basis. Cars with NO driver-assist run into something at roughly 3 accidents per million miles. But 2 accidents per million miles are related to driver-assist cars that run into something. Even then, it doesn't mean it's completely the fault of the driver assist system, as the operator is still responsible. So - if paying attention, the operator must intervene if the system fails .... just like airline captains have to do when the situation is more then the airliners' autopilot system can handle.
.

 

you've been hitting that rant pretty hard, there.

Why would the traffic still be going full rate when there's emergency activity on the shoulder? I mean I can't claim that I've seen this consistently done properly anywhere, but at least we have slow down/slide over rules in PA with some positive effect.

Also consider: It's possible that the gas station and other taxpaying retail owners have already spoken, and that's why you're seeing fewer police traffic stops in front of their establishments.

And from a driving skill standpoint: Drivers always need to be ready to manage activity on the road shoulder, whether it is tumbleweeds or pavement repair or somebody getting ticketed.

 

The most recent Source shows a much lower instance;

Tesla with Autopilot engaged approaches 10x lower chance of accident than average car: here's the data - Electrek
I was recalling older data that was not so optimistic .... but that was on earlier software versions. Here's some older data gathered up by insurance carrier showing the same thing, only not as old of data as I had originally come across years ago;

Tesla Cars 6 Times Less Likely to Have an Accident

.

 

Hah! Well that shows my age--I've lived here over 60 years (back when Silicon Valley was just a bunch of fruit orchards) so 2011 is recent to me.

 

1 just mentioning common sense actions that will prevent collisions. take out the entire possibility of a collision by not being right next to 55-75mph multi lane traffic, and poof, the collisions decrease.

2 to give the attentive drivers some credit- multi lane highways with each lane full of vehicles don't always afford a prefect line of sight to the shoulder. turns and hills amplify this problem. large trucks etc ... but then we have the rest of the drivers that are playing with phones instead of paying attention...

3 when possible, I change lanes to provide a cushion of safety between my 4-5ooo pound vehicle that is traveling 55-75mph vs the unprotected human walking on the shoulder. sometimes, traffic does not allow that nicety. eliminate the roadside danger, and the traffic doesn't need the extra , and sometimes impossible to follow , rules

4 they would be entirely within their rights to do so... but to date the attitude has always been "all of you drivers need to respect mah athoritayyy, and yours too my traffic stop, instead of me taking the traffic stop to a safer location"

5 100% agreement. it doesn't hurt to eliminate huge speed differentials also.

6 tumble weeds don't have a consciousness or the ability to make decisions about direction of travel. road crews have the option of using very large, very heavy (dump trucks with portable collision barriers on the back come to mind) vehicles to help prevent inattentive drivers from slamming into their pedestrian working crews.

emergency (non elective) crews love ambulances, tow trucks and fire trucks, don't have much of a choice, if people need rescuing efforts.

emergency (ok ok it's not really an emergency if you're just handing out pieces of paper that require acceptance and payment/court , for driving violations) vehicles that stop people in the side of the road, not under emergency conditions, could very easily continue to the next exit ramp and be far away from the high speed traffic.

it's really just that easy.

 

agreed, all around