Self-Driving Uber Car Racks Up First Kill
POSTED BY: JEWELSTAITEFAN
UPDATED: Thursday, September 19, 2024 15:05
VIEWED: 20804
PAGE 5 of 7
And Kim Kommando listed the criminal and driving convictions of that driver.
Plus, reported that Waymo requires intervention of Safety Driver an average of every 5,600 miles. For Uber, the comparable figure is 13 miles.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Uber has reportedly discovered that the fatal crash involving one of its prototype self-driving cars was probably caused by software faultily set up to ignore objects in the road, sources told The Information.
Specifically, The autonomous programming detects items around the vehicle and operators fine-tune its sensitivity to make sure it only reacts to true threats (solid objects instead of bags, for example). Unfortunately, the car's software was supposedly set too far in the other direction, and didn't stop in time to avoid hitting bicyclist Elaine Herzberg.
www.engadget.com/2018/05/07/uber-crash-reportedly-caused-by-software-t
hat-ignored-objects-in/
The Joss Whedon script for Serenity, where Wash lives, is Serenity-190pages.pdf at www.mediafire.com/folder/1uwh75oa407q8/Firefly
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:She was not riding her bike, right?
Originally posted by second:
Uber has reportedly discovered that the fatal crash involving one of its prototype self-driving cars was probably caused by software faultily set up to ignore objects in the road, sources told The Information.
Specifically, The autonomous programming detects items around the vehicle and operators fine-tune its sensitivity to make sure it only reacts to true threats (solid objects instead of bags, for example). Unfortunately, the car's software was supposedly set too far in the other direction, and didn't stop in time to avoid hitting bicyclist Elaine Herzberg.
www.engadget.com/2018/05/07/uber-crash-reportedly-caused-by-software-t
hat-ignored-objects-in/
A poorly designed system, which allows sensitivity to be "adjusted" off spectrum, could mistake an unmounted bicycle in motion for transparency.
A poorly designed system, yet one which was granted authorization to run loose on the roads, without a qualified driver, in the state. Whoever authorized that should also be charged, along with the company.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
Originally posted by second:
Uber has reportedly discovered that the fatal crash involving one of its prototype self-driving cars was probably caused by software faultily set up to ignore objects in the road, sources told The Information.
Specifically, The autonomous programming detects items around the vehicle and operators fine-tune its sensitivity to make sure it only reacts to true threats (solid objects instead of bags, for example). Unfortunately, the car's software was supposedly set too far in the other direction, and didn't stop in time to avoid hitting bicyclist Elaine Herzberg.
www.engadget.com/2018/05/07/uber-crash-reportedly-caused-by-software-t
hat-ignored-objects-in/
I really hope Uber was trying to give a simplistic explanation for their software, because that does not sound like the proper way to build a machine learning algorithm. You don't have sensitivity to objects. You get the system to learn how to identify what an object is. Besides something the size of a person and a bike should definitely not be assumed to be a non-solid object. That "sensitivity" would have to be almost entirely disabled at that point. I mean by size that's extremely similar to a motorcycle. You can't just plow through those like they're a bag.
Uber probably should not be in the self driving business. They just don't have the resources to develop a good system. They're burning cash like crazy and that is not a good situation to be investing in a lot of research and development.
NOTIFY: N | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:And wasn't she a pedestrian anyhow, ignoring their Spin?
Originally posted by Zeek:Quote:
Originally posted by second:
Uber has reportedly discovered that the fatal crash involving one of its prototype self-driving cars was probably caused by software faultily set up to ignore objects in the road, sources told The Information.
Specifically, The autonomous programming detects items around the vehicle and operators fine-tune its sensitivity to make sure it only reacts to true threats (solid objects instead of bags, for example). Unfortunately, the car's software was supposedly set too far in the other direction, and didn't stop in time to avoid hitting bicyclist Elaine Herzberg.
www.engadget.com/2018/05/07/uber-crash-reportedly-caused-by-software-t
hat-ignored-objects-in/
I really hope Uber was trying to give a simplistic explanation for their software, because that does not sound like the proper way to build a machine learning algorithm. You don't have sensitivity to objects. You get the system to learn how to identify what an object is. Besides something the size of a person and a bike should definitely not be assumed to be a non-solid object. That "sensitivity" would have to be almost entirely disabled at that point. I mean by size that's extremely similar to a motorcycle. You can't just plow through those like they're a bag.
Uber probably should not be in the self driving business. They just don't have the resources to develop a good system. They're burning cash like crazy and that is not a good situation to be investing in a lot of research and development.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
Originally posted by second:Quote:People driving have 1.02 fatalities per 100 million vehicles miles traveled in 2014. Uber has 1.00 fatalities per thousand vehicle miles. There is much room for improvement at Uber. Maybe next time Uber could apply the brakes? Or steer around the pedestrian? Or beep the horn? Flash the high beams on the headlights? Or do something other than drive straight toward the pedestrian at a constant speed in the same lane like a railroad train on tracks.
Originally posted by Wishimay:
They ALREADY have better accident per mile records than PEOPLE do, as Google started testing its first self-driving car in 2009.
ONE fatality from then till NINE years later?? It's impressive. Not enough to make me go near it yet, but still impressive.
https://en.wikipedia.org/wiki/Transportation_safety_in_the_United_Stat
es#Road_safety
Maybe the Uber Computer was not paying sufficient attention because it was watching youtube when it killed the pedestrian. Who knows?
Quote:Looks like pedestrian to me.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Heard a news item today.
Statement by some nondriven auto maker, maybe uber.
Said their self driving cars are not intended to avoid crashes, or killing humans.
What kind of retards are in charge of these places?
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
The most difficult to build is the perception module, says Sebastian Thrun, a Stanford professor who used to lead Google’s autonomous-vehicle effort. The hardest things to identify, he says, are rarely-seen items such as debris on the road, or plastic bags blowing across a highway. In the early days of Google’s AV project, he recalls, “our perception module could not distinguish a plastic bag from a flying child.”
According to the NTSB report, the Uber vehicle struggled to identify Elaine Herzberg as she wheeled her bicycle across a four-lane road. Although it was dark, the car’s radar and LIDAR detected her six seconds before the crash. But the perception system got confused: it classified her as an unknown object, then as a vehicle and finally as a bicycle, whose path it could not predict. Just 1.3 seconds before impact, the self-driving system realised that emergency braking was needed. But the car’s built-in emergency braking system had been disabled, to prevent conflict with the self-driving system; instead a human safety operator in the vehicle is expected to brake when needed. But the safety operator, who had been looking down at the self-driving system’s display screen, failed to brake in time.
The cause of the accident therefore has many elements, but is ultimately a system-design failure. When its perception module gets confused, an AV should slow down. But unexpected braking can cause problems of its own: confused AVs have in the past been rear-ended (by human drivers) after slowing suddenly. Hence the delegation of responsibility for braking to human safety drivers, who are there to catch the system when an accident seems imminent. In theory adding a safety driver to supervise an imperfect system ensures that the system is safe overall. But that only works if they are paying attention to the road at all times.
www.economist.com/the-economist-explains/2018/05/29/why-ubers-self-dri
ving-car-killed-a-pedestrian
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:This was the Safety Driver without a Driver's License?
Originally posted by second:
The most difficult to build is the perception module, says Sebastian Thrun, a Stanford professor who used to lead Google’s autonomous-vehicle effort. The hardest things to identify, he says, are rarely-seen items such as debris on the road, or plastic bags blowing across a highway. In the early days of Google’s AV project, he recalls, “our perception module could not distinguish a plastic bag from a flying child.”
According to the NTSB report, the Uber vehicle struggled to identify Elaine Herzberg as she wheeled her bicycle across a four-lane road. Although it was dark, the car’s radar and LIDAR detected her six seconds before the crash. But the perception system got confused: it classified her as an unknown object, then as a vehicle and finally as a bicycle, whose path it could not predict. Just 1.3 seconds before impact, the self-driving system realised that emergency braking was needed. But the car’s built-in emergency braking system had been disabled, to prevent conflict with the self-driving system; instead a human safety operator in the vehicle is expected to brake when needed. But the safety operator, who had been looking down at the self-driving system’s display screen, failed to brake in time.
The cause of the accident therefore has many elements, but is ultimately a system-design failure. When its perception module gets confused, an AV should slow down. But unexpected braking can cause problems of its own: confused AVs have in the past been rear-ended (by human drivers) after slowing suddenly. Hence the delegation of responsibility for braking to human safety drivers, who are there to catch the system when an accident seems imminent. In theory adding a safety driver to supervise an imperfect system ensures that the system is safe overall. But that only works if they are paying attention to the road at all times.
www.economist.com/the-economist-explains/2018/05/29/why-ubers-self-dri
ving-car-killed-a-pedestrian
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
News said a Tesla self-driving car crashed into a PARKED Police cruiser, totalling the cop car.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME