05-30-2018, 05:17 PM
|
#241 (permalink)
|
Corporate imperialist
Join Date: Jul 2011
Location: NewMexico (USA)
Posts: 11,268
Thanks: 273
Thanked 3,571 Times in 2,835 Posts
|
All I want to know is are the self driving cars crashing more than or less than human driven cars?
__________________
1984 chevy suburban, custom made 6.5L diesel turbocharged with a Garrett T76 and Holset HE351VE, 22:1 compression 13psi of intercooled boost.
1989 firebird mostly stock. Aside from the 6-speed manual trans, corvette gen 5 front brakes, 1LE drive shaft, 4th Gen disc brake fbody rear end.
2011 leaf SL, white, portable 240v CHAdeMO, trailer hitch, new batt as of 2014.
|
|
|
Today
|
|
|
Other popular topics in this forum...
|
|
|
05-30-2018, 05:48 PM
|
#242 (permalink)
|
Human Environmentalist
Join Date: Aug 2010
Location: Oregon
Posts: 12,819
Thanks: 4,327
Thanked 4,480 Times in 3,445 Posts
|
Quote:
Originally Posted by MetroMPG
So, two more cases of Model S crashes on autopilot have happened:
- One ran into the back of a stationary fire truck at just shy of 60 MPH. The "driver" admitted whe was using her phone for an extended period, and hit the brakes less than 1 second before impact. The car's adaptive speed control had actually accelerated just before the crash.
- The other ran into the back of a stationary police vehicle.
This is the 3rd case I know of with Teslas autopiloting themselves into stationary emergency vehicles at speed.
But so far, the cars are only harming their "drivers". (Their "non-drivers"? "Beta testers?")
All hell's going to break loose the first time a Tesla autopilots itself into killing an innocent person. I have a feeling it'll generate a lot more attention than the Uber crash.
|
Tesla had better make their system better at seeing plainly obvious obstructions soon, or they will have a very difficult time building public confidence back up. So far it appears the video only approach to obstical detection is failing them. They likely need to add SONAR/RADAR/LIDAR to improve obstacle detection.
|
|
|
05-30-2018, 06:02 PM
|
#243 (permalink)
|
Batman Junior
Join Date: Nov 2007
Location: 1000 Islands, Ontario, Canada
Posts: 22,534
Thanks: 4,082
Thanked 6,979 Times in 3,614 Posts
|
oil pan 4: Musk is already on record saying autopilot is statistically much safer than the average American driver, measured by fatalities per mile.
But I haven't seen the details of the numbers behind the claim (e.g I've read one suggestion that the comparison includes motorcyclists. Not sure if it compares only to contemporary cars, or all vehicles, which would have less passive crash protection, and possibly higher crash rates for other reasons.)
|
|
|
06-01-2018, 11:35 AM
|
#244 (permalink)
|
lurker's apprentice
Join Date: May 2008
Location: the Perimeter
Posts: 942
PlainJane - '12 Toyota Tacoma Base 4WD Access Cab 90 day: 20.98 mpg (US)
Thanks: 504
Thanked 226 Times in 173 Posts
|
Autopilot doesn't have to match human driver accident rates -- it has to do vastly, enormously better than humans. No other level of performance will result in it being successfully mainstreamed.
|
|
|
06-01-2018, 05:24 PM
|
#245 (permalink)
|
Corporate imperialist
Join Date: Jul 2011
Location: NewMexico (USA)
Posts: 11,268
Thanks: 273
Thanked 3,571 Times in 2,835 Posts
|
Now there should be enough vehicles driven enough hours to make a statistical comparison.
I'm expecting robo driving to be better than human drivers in all normal conditions.
__________________
1984 chevy suburban, custom made 6.5L diesel turbocharged with a Garrett T76 and Holset HE351VE, 22:1 compression 13psi of intercooled boost.
1989 firebird mostly stock. Aside from the 6-speed manual trans, corvette gen 5 front brakes, 1LE drive shaft, 4th Gen disc brake fbody rear end.
2011 leaf SL, white, portable 240v CHAdeMO, trailer hitch, new batt as of 2014.
|
|
|
06-01-2018, 06:01 PM
|
#246 (permalink)
|
Master EcoWalker
Join Date: Dec 2012
Location: Nieuwegein, the Netherlands
Posts: 4,000
Thanks: 1,714
Thanked 2,247 Times in 1,455 Posts
|
Hey Google...
https://www.theverge.com/2018/5/31/1...-driving-fleet
62,000 more self driving vans. They haven't lost heart, that's for sure.
__________________
2011 Honda Insight + HID, LEDs, tiny PV panel, extra brake pad return springs, neutral wheel alignment, 44/42 PSI (air), PHEV light (inop), tightened wheel nut.
lifetime FE over 0.2 Gigameter or 0.13 Megamile.
For confirmation go to people just like you.
For education go to people unlike yourself.
|
|
|
06-01-2018, 06:09 PM
|
#247 (permalink)
|
Not Doug
Join Date: Jun 2012
Location: Show Low, AZ
Posts: 12,241
Thanks: 7,254
Thanked 2,234 Times in 1,724 Posts
|
Waymo also teamed up with Fiat-Crystler.
Why Waymo, why?!
|
|
|
06-07-2018, 04:25 PM
|
#248 (permalink)
|
Batman Junior
Join Date: Nov 2007
Location: 1000 Islands, Ontario, Canada
Posts: 22,534
Thanks: 4,082
Thanked 6,979 Times in 3,614 Posts
|
Model X autopilot crash: some NTSB details
NTSB revealed preliminary findings.
Notable:
- During the 60 seconds prior to the crash, the driver’s hands were detected on the steering wheel on three separate occasions, for a total of 34 seconds; for the last 6 seconds prior to the crash, the vehicle did not detect the driver’s hands on the steering wheel.
- At 8 seconds prior to the crash, the Tesla was following a lead vehicle and was traveling about 65 mph.
- At 7 seconds prior to the crash, the Tesla began a left steering movement while following a lead vehicle.
- At 4 seconds prior to the crash, the Tesla was no longer following a lead vehicle.
- At 3 seconds prior to the crash and up to the time of impact with the crash attenuator, the Tesla’s speed increased from 62 to 70.8 mph, with no precrash braking or evasive steering movement detected.
That last point is a doosie.
https://www.scribd.com/document/3812...11-Preliminary
|
|
|
The Following 2 Users Say Thank You to MetroMPG For This Useful Post:
|
|
06-07-2018, 05:07 PM
|
#249 (permalink)
|
Human Environmentalist
Join Date: Aug 2010
Location: Oregon
Posts: 12,819
Thanks: 4,327
Thanked 4,480 Times in 3,445 Posts
|
Love the "just the facts" reporting.
The driver is responsible for having set the speed 10 MPH above the limit and for not paying attention.
CDOT is responsible for constructing an interchange that is confusing to both people and machines as evidenced by the repeated collisions by both.
Tesla is responsible for not being able to detect the danger that would be plainly obvious to an alert human.
I still think it would be in manufacturers best interest to have designated safe zones where autonomous features are allowed to operate, or at minimum define the unsafe or unknown zones where their systems are not sophisticated enough to ensure a reasonable level of safety.
A smart system would notice patterns where human intervention was required, and mark that zone as unsafe for autonomous travel.
|
|
|
The Following 3 Users Say Thank You to redpoint5 For This Useful Post:
|
|
06-07-2018, 05:39 PM
|
#250 (permalink)
|
Master EcoModder
Join Date: Aug 2012
Location: northwest of normal
Posts: 28,734
Thanks: 8,157
Thanked 8,938 Times in 7,380 Posts
|
That's a great idea. You could release it under a Creative Commons license.
Would it be demarcated by a wi-fi hotspot or GPS waypoint? Or both?
Roadside QR code?
__________________
.
.Without freedom of speech we wouldn't know who all the idiots are. -- anonymous poster
____________________
.
.Three conspiracy theorists walk into a bar --You can't say that is a coincidence.
|
|
|
|