backtop


Print 52 comment(s) - last by The Raven.. on Sep 30 at 3:15 PM


Volvo XC60 test vehicle applies brakes to prevent collision with test dummies  (Source: Drive)
Volvo blames collisions on improperly setup dummies

It was just a few months ago when we brought you the first edition of Volvo's "Technology Fail" with the failure of the company's Collision Warning System. In that test, a Volvo S60 test car was supposed to brake in time to avoid a stationary truck with no driver involvement -- instead, the vehicle ended up rear-ending the truck at 30 mph.

Now, Volvo is serving up a second edition of Technology Fail. This time around, Volvo invited the press to witness its Advanced Pedestrian Avoidance System in action. The system is supposed to detect pedestrians in the direct path of the vehicle and apply the brakes to avoid a collision. A Volvo XC60 test vehicle equipped with the new safety feature was sent down a test track towards two dummies (a father and son pairing). Of the 12 demonstrations performed for the press, three of them failed.

In two of the three incidents witnessed, brakes were applied, but not in time to avoid a collision. In one of the collisions, the brakes were not applied at all.

Almost comically, Jonas Tisell, Volvo's Active Safety Systems manager, told Drive that the three collisions were the dummy's fault. "The failure of the test was due to the dummy not being set up properly, therefore it did not give an echo enough for the system," said Tisell. "So the dummy was not relevant in this situation."

It is a bit puzzling to imagine a dummy standing in the middle of the road -- which is approximately representative of a human being doing the same thing -- could possibly be setup wrong, but we'll give Volvo the benefit of the doubt here.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Shift of blame from driver to car manufacturer
By Schrag4 on 9/27/2010 2:42:43 PM , Rating: 4
I can't help but feel that a system like this will ultimately lead to car manufacturers being sued anytime a pedestrian is hit, when ultimately the driver and the pedestrians share responsibility to avoid such collissions. I'm not saying lives wouldn't be saved if this system generally worked 99.9% of the time (which it doesn't), but I'm just afraid that people will FEEL less responsible for their poor driving, or even pedestrians might feel they should be able to cross the street without looking!

The car is a tool. It can do unintended damage. Make it safer, sure, but don't rely on those safety systems to keep yourelf or others safe, instead, use some caution and common sense when using the tool! (or if you're putting yourself in the potential paths of these tools, pedestrians) Suing Volvo would be like suing Craftsman for your smashed thumb when you miss the nail with their hammer. Or like suing Pentel for your mispelled words.




By theArchMichael on 9/27/2010 3:20:41 PM , Rating: 2
I think any tech that can marginally increase the overall safety of a vehicle is worthwhile costs permitting. Also, integrating technology into vehicles is going to be an important step for us as a culture as we move towards the fully automated self-driving car. So as these vehicle automation techs improve over time and people feel more comfortable giving control over to the technology things may change more rapidly.

Also, not sure on this, but I think most times when you are rear-ended... you are the victim... unless you really enjoyed it :-)


By The Raven on 9/28/2010 12:26:56 PM , Rating: 2
quote:
Also, not sure on this, but I think most times when you are rear-ended... you are the victim... unless you really enjoyed it :-)

I was trying to say that that the car manufacturer would get sued by the driver.

And I doubt we are headed for a self-driving car anytime in my baby's lifetime. If we were, we would be using public transportation A LOT more. (I speak from the American perspective).
We are trying to be more fuel efficient (be that gas, solar, wind, coal, etc.) and and a self-driving car is not in those plans.


By Thats Mr Gopher to you on 9/29/2010 12:01:47 PM , Rating: 2
The manufacturer isn't to blame in this situation. The driver in the rear vehicle that is travelling too close or too fast is to blame for the accident. Ask anyone in car insurance, they'll quickly clear that up for you.

The manufacturer may end up getting sued if the system causes some other kind of accident where the driver of the vehicle would normally be at fault.


By The Raven on 9/30/2010 3:15:45 PM , Rating: 2
Yeah you're right. But I still think that people will find a way to hold the manufacturer responsible if there is a 'misfire'. Maybe the rearender will sue the manufacturer the same way people sue cities for causing accidents at 'dangerous' intersections. (I not dismissing such suits but using it as an example)

I'm talking about these suits where the manufacturer gets sued for airbag/seatbelt malfunction. (Barring legislation) they could just leave the feature out and avoid a suit. But I definitely see your point.


RE: Shift of blame from driver to car manufacturer
By GTVic on 9/27/2010 4:02:06 PM , Rating: 3
Yes, I agree, safety systems should not be implemented for many reasons: fear of legal action, fear of bad drivers getting worse, fear of being afraid, etc.

Abolish all seat belts, they encourage reckless driving!


By Schrag4 on 9/27/2010 5:44:37 PM , Rating: 2
quote:
Abolish all seat belts, they encourage reckless driving!


Seat belts don't slam on your brakes, this system does. Nobody is OK with getting in an accident just because they're more likely to survive. That's not the thrust of my concern. My concern is that people will sue when these devices fail, even though the driver is at least partially at fault (with the rest of the blame falling on the pedestrian, perhaps, but NOT the car).

Oh, and maybe YOU won't drive down a pedestrian even if this device fails, but that doesn't mean a lot of people wouldn't falsely put their faith in the system and pay less attention to the road. I think I'd rather this thing ding at you if you're going to hit something, so you still have to make the final decision to brake, and therefore you still have to pay attention. However, the article clearly states that the goal of these is to get people to agree to relinquish control of their machines to computers, which is something I don't think we're anywhere near ready for, and therefore is a bad idea at this point in time. Just my two cents.


By rcc on 9/28/2010 3:35:49 PM , Rating: 2
quote:
Abolish all seat belts, they encourage reckless driving!


True, but not having them encourages reckless exiting of the vehicle. Usually through the windshield


By Miqunator on 9/27/2010 5:02:23 PM , Rating: 2
"Warning, this vehicle does not have a brain, use your own"

Found that sign somewhere, would probably be cheaper to put that in the car instead...


By Hafgrim on 9/27/2010 2:11:43 PM , Rating: 2
Well dummies arnt exactly very human like considering humans are supposedly 90% water. I would assume they should have used fluidic ballistics gel dummies or dummies with water bags in its clothing something to make them more human like in density if that is whats causeing the sensor failures.




By Brandon Hill (blog) on 9/27/2010 2:19:04 PM , Rating: 1
I believe that the system is radar-based, so would object density really have to be taken into consideration?


By Smartless on 9/27/2010 2:41:11 PM , Rating: 2
Totally true. Besides some people are more dense then others. har har. You know, why should it matter how a dummy is set up when basically ANYTHING in the road should trigger a brake response. Stealth clothes?


By AssBall on 9/27/2010 3:23:27 PM , Rating: 3
So what, it is going to swerve you off the road and cause an accident because there was a big baloon or even a deer that would be much safer to plow over than avoid, but do nothing for a telephone pole that will turn you and your car into biscuits and gravy?

Why doesn't Volvo just save their money and give people driving lessons at the dealership instead.


By Smartless on 9/28/2010 12:13:40 AM , Rating: 2
Well this thing just brakes anyway not steering control. As for running over a deer, you might want to look up the amount of fatalities there are from actually hitting one. They're not soft and squishy.

And for Pete's sake don't we all agree on that? The funny part is the safer people feel in their car the more likely they'll be stupid and get into one. All I'm asking is why is Volvo's test all involve people? I'd actually like to see real statistics where this type of device would actually curb accidents seeing how they failed on identifying a car in front of them the first time.


By AssBall on 9/28/2010 7:07:30 AM , Rating: 2
I think more fatal damage is done avoiding deer than hitting them, statistically. (I don't have any links). A friend of mine hit one going 80mph in her dodge neon at night. It totalled the neon, but since she held it straight, she was fine. I hate to imagine what would have happened if she slammed on the brakes or swerved out of the way in the middle of interstate traffic.


By Smartless on 9/28/2010 2:21:45 PM , Rating: 2
Lol I actually found a few links to one study.
http://www.deercrash.com/states/data.htm

http://www.car-accidents.com/pages/deer-accident-s...

Now let's take these statistics and put them in perspective. Yes 150 deaths a year are a drop in the bucket but even 1 in a 1,000 accidents per fatality means that its not an anomaly since a majority of accidents that happen on the road aren't fatal and that's with another 2 ton vehicle.


By DanNeely on 9/27/2010 9:06:53 PM , Rating: 2
Yes. Or rather how strongly it reflects radar which is somewhat more complex of a question than just density; although broadly speaking denser objects will give stronger returns.

That said, not having confirmed that the dummies were suitable prior to the press conference is a staggering display of incompetence.


By NainoKami on 9/28/2010 9:26:29 AM , Rating: 2
Actually it's closer to 55-60% water for an adult. But interesting point...


By rcc on 9/28/2010 3:31:01 PM , Rating: 2
Perhaps if Mr. Tisell is so confident in the system, he should volunteer to stand out there.


i blame
By Gul Westfale on 9/27/2010 2:03:38 PM , Rating: 3
i blame improperly, too. he never learns.




RE: i blame
By Gul Westfale on 9/27/2010 2:04:57 PM , Rating: 2
oh, and:
http://www.youtube.com/watch?v=h_vvI26NnwE
"you attacked me wrong"


RE: i blame
By Motoman on 9/27/2010 2:12:56 PM , Rating: 5
"Volvo blames collisions on improperly"

You're right. That guy is a total douche.


RE: i blame
By Brandon Hill (blog) on 9/27/2010 2:16:59 PM , Rating: 2
Agreed ;-)


RE: i blame
By wht1986 on 9/27/2010 3:42:05 PM , Rating: 2
The poor avoidance signal was due to improper holding of the steering wheel by the driver. - Steve


RE: i blame
By YashBudini on 9/27/2010 9:11:52 PM , Rating: 2
The results are 100% valid and correct.

The dummies were texting. Doesn't get any more real world than that.


I blame Volvo's reaction
By Belegost on 9/27/2010 2:58:08 PM , Rating: 5
I seriously think Volvo has goofed in how they reacted to this, I would have claimed success.

Look at it this way, in testing the system performed perfectly in 75% of cases, and performed partially (at least if the brakes were applied injuries were reduced) in 92% of cases.

More importantly, this is a parallel system meant to augment a human driver. So the probability of total system failure here is now the probability that both the driver and the vehicle systems fail simultaneously, a much smaller probability than either of them failing individually.

I mean, from this perspective 3 out of 4 of the people who would have been hit by a negligent driver could avoid being harmed - sounds like a great thing to me.




RE: I blame Volvo's reaction
By fcx56 on 9/27/2010 3:31:06 PM , Rating: 3
I accidentally downrated you so I had to post to clear that up. I agree, it may be a more dangerous thing to market this to people as having a 100% hit (ha, or maybe non-hit) rate. Not everyone is this dumb but remeber the lady who sued the maker of her RV because no one eve told her she couldn't go in the back for a cup of coffee and let the RV drive itself. I can forsee similar lawsuits being attempted by people that may not understand that no automated crash avoidance system can be trusted without supervision when it's your or another's life at stake


RE: I blame Volvo's reaction
By tmouse on 9/27/2010 3:32:52 PM , Rating: 2
I don't think they have much to be happy about. There is absolutely nothing else on the street to interfere with the signal and the test subjects are fully facing the car. Anything less than 100% is a failure under those ideal conditions. You probably will not get too many people standing in the middle of a road facing oncoming traffic. If it did happen the kid would probably be pulling away and yelling let go of my hand stupid!


The dummy's response
By amanojaku on 9/27/2010 2:19:11 PM , Rating: 3
"How 'bout five cross yo lip?"

RIP, Redd Fox




RE: The dummy's response
By Spuke on 9/27/2010 6:21:24 PM , Rating: 2
This is the big one!! Elizabeth!! I'm coming to join ya honey! I'll be the one with the Volvo shoved up my ass!


Press Release Fail
By The Insolent One on 9/27/2010 3:08:39 PM , Rating: 5
"The failure of the test was due to the dummy not being set up properly"

Since when did Steve Jobs start doing press releases for Volvo?




Wrong dummies
By kontorotsui on 9/27/2010 2:34:51 PM , Rating: 2
They should have used two hot Swedish blond dummies... THEY would have made anybody (alive or otherwise) brake... even if they were not in the path.




RE: Wrong dummies
By Sazabi19 on 9/28/2010 8:20:30 AM , Rating: 2
That didn't stop Tiger did it?


Your holding the phone wrong..
By KentState on 9/27/2010 5:45:34 PM , Rating: 2
Wait, that's the wrong company.




By KentState on 9/27/2010 5:46:27 PM , Rating: 2
You're I mean...


I feel that this sums up the new system perfectly
By arthur449 on 9/27/2010 10:33:17 PM , Rating: 2
9 out of 12 tests were successful.

That means that 75% of the time... it worked every time.




By DarkPhoenix on 9/28/2010 5:35:40 AM , Rating: 2
Yeah, percentages are a beauty for marketting reasons...

But look at the test conditions. They are the most perfect ones for this kind of test and it still failed 25% of time. Now how will the system react to moving people and/or animals. Especially in a crowded street...


Problem solved.
By ctodd on 9/28/2010 10:28:43 AM , Rating: 2
Use real people next time! Let the engineer know in advance that he WILL be the first one.




RE: Problem solved.
By mmatis on 9/28/2010 1:23:57 PM , Rating: 2
Or use politicians and everybody (except the politician) will be HAPPY if the system doesn't work.


no way
By Murloc on 9/27/2010 2:15:34 PM , Rating: 2
this will never be useful. Detecting a person is so difficult, and there are so many cases in which this feature might cause problems.




How about...
By The Raven on 9/27/2010 2:45:40 PM , Rating: 2
SUV detection for pedestrians? I think that would be more imortant since they are the ones who would likely die in such an unfortunate incident. Why are the scientists working backwards on this?

How about the frickin' cure for cancer, Parkinson's, etc.?




By MarkK02474 on 9/27/2010 4:14:13 PM , Rating: 2
It seems Volvo has still gotten better results than defense schemes costing taxpayers billions, like Starwars and Patriot anti-missile systems, or countless other devices and vehicles.

For people who rather do other things than watch the road, we have taxis, planes, trains, buses, and chauffeurs.




Volvo makes videos like this useless
By CZroe on 9/27/2010 6:08:52 PM , Rating: 2
Volvo makes videos like this useless:
http://www.youtube.com/watch?v=Lt4TJMeDMTI

Why bother faking it when Volvo will make a new one every four months? ;)




I'm all for
By Totally on 9/27/2010 6:13:53 PM , Rating: 2
new technology and innovation but I hope this little pet project of theirs never sees the light of day because it just those who get behind the wheel in vehicles equipped with the system another reason to not pay attention. When would this tech every outperform existing tech i.e. your own two eyes?




Why I won't buy a Volvo:
By UNHchabo on 9/27/2010 6:57:01 PM , Rating: 2
How am I supposed to survive the coming zombie apocalypse if I buy a Volvo?

If there are zombies in the middle of the road, then this system would make me a sitting duck, when my Ford would allow me to plow through the horde at my leisure!




Sounds like.....
By MDme on 9/27/2010 7:23:04 PM , Rating: 2
Volvo is taking a page out of the Apple playbook.




This will never work
By DarkPhoenix on 9/28/2010 5:32:19 AM , Rating: 2
As others, I have to say that this will never work and will most likely cause more problems, than actually do any good.

First, the remark about it being the dummy's fault. Well, that guy took lessons from Steve Jobs it seems. The lack of common sense at Apple seems to be contagious or something. Sure, this wasn't a "real" person and Volvo's customer would actually be the driver, not the person below the wheel...but still, trying to be ridiculous as Apple is at damage control, is not acceptable.

Second, about how this will never work and will probably cause more problems, than actually do any good, there's no way that this system is good enough to prevent applying the brakes, on a situation where there really wasn't the need to. And this will cause accidents, when it was intended to prevent them.
As someone else pointed out, a car is a tool and the people driving it should be the ones responsible for their actions, when using it. People are people and they will make mistakes, but adding technology to the mix, that will cause even more mistakes, doesn't seem like a good idea to me. Especially when these test scenarios are in the perfect conditions and the system still fails some times. Imagine when it's a busy street filled with people and/or stray animals...




Back to the lab
By trisct on 9/28/2010 1:16:42 PM , Rating: 2
Volvo had better quit showing off half-baked systems to the press, or they will get bitten when these demos get brought up at the inevitable injury trials. Maybe firing whoever is managing the project or demos would be a good idea...




Hmm...volvo=OCP
By ReclusiveOrc on 9/28/2010 1:45:19 PM , Rating: 2
Hmm... a real life version of ED209 and just as lethal.




75% less casualties is not bad at all
By Zds on 9/30/2010 9:26:32 AM , Rating: 2
I would not laugh too much at them. This kind of protection is meant for cases where nothing else helps, and reducing death/injury toll by 3/4 for cases where driver is not paying attention is pretty amazing in my books.




By BZDTemp on 9/28/2010 2:42:43 PM , Rating: 1
"It is a bit puzzling to imagine a dummy standing in the middle of the road -- which is approximately representative of a human being doing the same thing -- could possibly be setup wrong, but we'll give Volvo the benefit of the doubt here."

Again DT prefers to treat news as entertainment rather than bringing us all the facts. What Volvo is doing will save a lot of people from getting hurt by bad drivers but Volvo is not claiming the system is perfect.




"I modded down, down, down, and the flames went higher." -- Sven Olsen














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki