Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Tesla Autopilot 2

Status
Not open for further replies.
GregLocock

GregLocock

Automotive
Apr 10, 2001
23,127
1
38
Orbiting a small yellow star
I think it's worth breaking out Tesla from Uber, both the hardware and software are different.

So, the second crash of a Tesla into a stationary firetruck looks like more than a coincidence. Even if Autopilot wasn't engaged (the police say it was) Automateic Emergency Braking should have stopped this.

From tesla's website

Standard Safety Features
These active safety technologies, including collision avoidance and automatic emergency braking, have begun rolling out through over-the-air updates

Automatic Emergency Braking

Designed to detect objects that the car may impact and applies the brakes accordingly


Side Collision Warning

Warns the driver of potential collisions with obstacles alongside the car


Front Collision Warning

Helps warn of impending collisions with slower moving or stationary cars

So questions that need to be asked, are which of these were fitted to the crash vehicle? AEB is widely available on other cars, but according to Tesla forums it is possible that it was remotely disabled. According to one user you can set AEB to warn only. That is bizarre choice of UI design.

Anyway, I think so far there have been three AP collisions with large objects on the road in front of the car in good viewing conditions, the idiot with the truck crossing the road, and two fire trucks.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Replies continue below

Recommended for you

Sort by date Sort by votes
Of the systems I've seen, there are significant limitations to the ability to auto-braking based on approach speed. I think the one from Volvo is delta-V of 30mph max; I know I saw a demo for it and they had really big targets and rather low approach speeds.

"In tests in which the vehicle traveled at under 30 miles per hour, systems designed to prevent crashes successfully avoided collisions in 60% of the time. Meanwhile, automatic emergency brakes designed to only reduce crash severity were able to completely avoid crashes in only 33% of test scenarios, according to AAA."
I see a lot of articles based on the same AAA test in 2016.

It seems like the tech is perfectly fine to prevent fender-benders in stop and go traffic and as the delta-V departs from that it becomes increasingly unreliable across the board.

This is an ad from Audi touting 40 km/hr limits:
 
Upvote 0 Downvote
What that means is that the average crash severity will increase. Not because things are getting worse, but simply because the bottom end of the curve gets cut off. That can be seen, correctly as far as I can tell, as an overall improvement or it could be seen as a degradation because per centage wise the severe crashes increase. One perception is damning while the other is encouraging.
 
Upvote 0 Downvote
Reading the Tesla forums is interesting. If AP is activated then AEB should never be necessary, AP is supposed to offer a kindler gentler more predictable driving pattern. AEB is designed to avoid false positives, and basically if it detects an unavoidable collision jams on the brakes until the speed has dropped by 25mph, or the steering wheel is touched (that's a failure mode I'd investigate) or the accelerator is touched, or the target disappears. AEB does have a speed limitation, which varies, but it seems like at least some versions work below 85 mph.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Upvote 0 Downvote
davidbeach said:
One perception is damning while the other is encouraging.
Click to expand...
The cost savings to insurance companies to write off less fender-bender damage still makes it worth it? My insurance company will pass the savings on to me, right? :/

STF
 
Upvote 0 Downvote
3DDave: "It seems like the tech is perfectly fine to prevent fender-benders in stop and go traffic and as the delta-V departs from that it becomes increasingly unreliable across the board."

davidbeach: "What that means is that the average crash severity will increase. Not because things are getting worse, but simply because the bottom end of the curve gets cut off."

If drivers become accustomed to, or expect, their car to brake for them, severe crashes could become more common, as well. These systems need to function more than a third or even two-thirds of the time, since people WILL rely on them, whether they should or not.

Edit: "...function more than a third..." not "...function a third..." Sorry for any confusion.
 
Upvote 0 Downvote
It's unclear whether that will actually happen; I would have thought the same about ABS, but people don't seem to be getting into those kinds of crashes. I think that we're close to the limit of what our normal "chicken" limits are.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list
 
Upvote 0 Downvote
For that matter, will AVs be aware of wet, icy, oil spill or other low friction conditions and reduce speed, respectively increase following distance accordingly, as an attentive and prudent human driver would?
Or are they going to operate according to nominal mu conditions at all times?
If they are going to dynamically adjust the assumed mu, I'd be interested in the algorithms for that.
As a human driver, I have many data I monitor to adjust my assumed mu downward from the nominal...: ambient temperature, appearance of the road, weather history of the preceding several hours, presence of vehicles in the ditch on the side of the road or otherwise appearing to have lost control due to insufficient traction; not to mention experience on certain patches of road, e.g. near intersections, where mu is noticeably below par.

"Schiefgehen wird, was schiefgehen kann" - das Murphygesetz
 
Upvote 0 Downvote
There was a good television documentary program about highway safety and its history. Although cars are far safer than years ago, fatality rates have stopped going down. One expert speculated that the reason was that roads have been straightened and drivers feel so safe in their cars that they are driving faster. The result is that accidents have become more severe with more deaths. He said that the greatest true safety measure would be to put a 12" spike in the steering wheel pointed at the driver's chest. Drivers would drive a lot more safely.
 
Upvote 0 Downvote
I'm relatively certain that texting are other "distracted driving" situations including "distracted pedestrians" are the chief cause of increased deaths. The 12" spike would do nothing, they pretty much build cars that way for decades. I really don't think people get into cars and and think, this car is so safe I can drive like an idiot.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.
 
Upvote 0 Downvote
I had thought I posted this graphic recently, but couldn't find it. It shows that the fatality rate per capita is still on a downward trend, and what's really flattened out is the deaths per Vehicle Mile Traveled (VMT), whose increase is somewhat matching the population rate. Therefore, what's flattening the death/VMT is the increased exposure to driving, i.e., we're driving more and farther, which increases the probability of death per year.

US_traffic_deaths_per_VMT%2C_VMT%2C_per_capita%2C_and_total_annual_deaths.png


TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list
 
Upvote 0 Downvote
On a related note to this discussion, I have a car which has a limited emergency braking system (speeds <19 MPH). Above that speed, it will warn me of a collision. While driving the other day, and following at a safe distance, the car in front of me smacked a pothole which created a spray of water and aggregate. Curiously, the radar system interpreted this as an object and warned me of a collision with it. Now this tech in my car was in the early days for Mazda (2014). And they have since expanded its capabilities as they have worked some of these issues out. But this "blip" I experienced really served to illustrate the challenges that exist in getting to Level 2 and higher automation.
 
Upvote 0 Downvote
irstuff,

If the graph had continued on a steady path from 1985 to 1995 the US would start seeing resurrections of the dead about the year 2005. Fortunately this will be avoided for the time being.
 
Upvote 0 Downvote
152 ft is poor for a medium size car, and the difference to Tesla's claim of 133 is probably statistically significant (repeatability is about 3 ft from memory). CU say "The Tesla Model 3’s 152 feet is 21 feet longer than the class average of 131 feet for luxury compact sedans". There are plenty of ways of gaming the braking test, and CU's test procedure is not ideal. A much more repeatable test is the AMS one where the braking starts at 66 mph and the distance from 60 to rest (roughly speaking) is the quoted figure. CU's procedure relies on the force buildup.

There's also the old favorite of how grippy your track is (we measure ours).

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Upvote 0 Downvote
Those Teslas weigh at least 1000 pounds more than a ICE vehicle with the same seating capacity. All those batteries. There's only so much brake disk you can squeeze into a 21" wheel. Does that explain the braking distance?

IRStuff: What is the source of that graph? I'd like to see the graph of "Fatalities per billion VMT" plotted on a logarithmic scale. I don't think that kind of number can actually trend to zero. By definition I think it can only trend to smaller and smaller fractions.

STF
 
Upvote 0 Downvote
SparWeb - that may be a little of it, but dry braking is primarily a function of tires, usually. There are effects on subsequent stops, where the brake rotor size is important, but from 60 mph you are well within the thermal limits of the brakes.

CR has a huge effect on new car purchases, at least for ICE cars. I imagine Musk will be blustering away about this on Twitter for a few days, and then they'll work with CR to find out what is going on. I saw some nasty creep from Tesla claiming it didn't matter because it could be solved by a software upgrade. Um, no, almost certainly not.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Upvote 0 Downvote
"Regenerative braking (using the electric motors as generators) should help (a bit) with braking distances."

That might actually be what's screwing it up. Longer stopping distances are not the result of inadequate braking power. They can all apply enough force to lock up the brakes. It's either a delay in fully applying the brakes (after the pedal is pushed) or the brakes are being released more than they should be during braking.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

MartinLe
  • Locked
  • Question
Tesla &quot;autopilot&quot; disengages shortly before impact 9
8 9 10
Replies
187
Views
371
IRstuff
IRstuff
GregLocock
  • Locked
  • News
First Tesla Autopilot deaths court case 1
2
Replies
20
Views
48
IRstuff
IRstuff
JohnRBaker
  • Locked
  • News
Tesla's giant 'recall'...
2
Replies
23
Views
24
Alistair_Heaton
Alistair_Heaton
JStephen
  • Locked
  • Question
Tesla Autopilot Update 3
2 3
Replies
54
Views
121
MacGyverS2000
MacGyverS2000
IRstuff
  • Locked
  • Question
Tesla's are apparently pretty tough
Replies
4
Views
9
TugboatEng
TugboatEng
Back
Top