Volvo Safety Demo Goes Poorly
Lanxon writes "At a demonstration of Volvo's new collision warning system in Sweden this week, Wired got first-hand experience (and video) of what happens when it goes badly wrong. The new Volvo S60, due for release later this year, was fired out of Volvo's testing tunnel at around 30MPH, and the collision detection system should have kicked in, bringing the car automatically to a halt before hitting the truck in its path. It didn't. Instead, the brand new car ploughed into the back of the truck in front of us, and indeed the world's press who had gathered in Sweden to see the collision detection system in action."
A crash course in safety.
YEEAAAAAHHHHHHH!
Distributed Denial of APK: It takes 15 seconds to reply to him anonymously, but wastes tons of his time if we all do it.
Much like regular Volvo drivers, it was oblivious to everything around it.
The article (I read it) said "around 30 MPH", but the car had "35 KMH" printed on it (which is approximately 22 MPH).
Are the reporters bad with numbers or did they forget the metric system?
This is why I'm afraid of cars in the future. They'll have all these safety features, people will forget how to drive (even the little bit they know) and rely on the car, and things will go wrong.
FTA:
"had a human been driving, he or she would have noticed the system was not operating correctly"
And they would have ignored it. Like every check engine light in the world that no one cares about.
This article actually doesn't belong in the Idle section. This is a very serious issue with wide-ranging ramifications, i.e. the automation of our automobiles (hah!). I have said before, and I will say it again, we have gone past the point where adding any more automation to the car will do any good, instead of improving the driver's skills.
A train is the only safe car on autopilot.
Stupidity is its own reward.
An excellent demonstration...of why automation should never attempt to take control of a car. Software errors, hardware failures, unreliable sensor technology, and an endless supply of unforeseeable situations mean that automation simply cannot be reliable.
Example: Our car has collison sensors - in heavy snowstorms, they warn us continually of imminent crashes with snowflakes.
Enjoy life! This is not a dress rehearsal.
I didn't know Toyota had bought out Volvo.
(sorry, had to)
It was a perfect demonstration of what happens when the one piece of safety equipment which could have prevented this was replaced by an inferior technology: the part they left out was AN ALERT DRIVER. I dearly love how car companies try to find more and more exotic ways to relieve the driver of his one BASIC RESPONSIBILITY while driving a car: DRIVING THE CAR. And it fails, of course, as I could only have hoped it would. Just wish I could have seen it in person. Until computers are truly smarter than we are, (and being able to count faster doesn't count,) they will not be effective at replacing us. Next time you think a computer is smarter than a person, ask it who the better captain was, Kirk or Picard. Ask it to defend it's response. If it can do that, MAYBE it is as smart as a person. (If, without being pre-programed to do so, it replies "Why are you wasting time with such a stupid question?" and it actually knows WHY it's a stupid, time wasting question, it will have demonstrated achievement of sentience. Bonus points if it replies "Neither, Malcolm Reynolds!")