Slashdot Mirror
Honda Crash Detection System
MImeKillEr writes "MSNBC is reporting that Honda Motor Co. unveiled an early crash-detection system for one of their vehicles. The system is unique in working even before the driver responds. A radar in the front of the car stashed behind the Honda logo detects vehicles within a range of about 300 feet ahead. It then taps the brake and tightens the seatbelt. A buzzer goes off and a light on the dash is illuminated. If the driver responds, the braking power is boosted. If the driver fails to respond, the system kicks in and brakes more while also tightening the seat belt. Unfortunately, Japanese regulations don't allow for the system to fully stop the vehicle."
Unfortunately, Japanese regulations don't allow for the system to fully stop the vehicle
Because what I want is to have less and less operation value, and rely more on technology!
We are slowly devolving into a society that not only has no common sense, but cannot operate anything without help.
No thank you auto stop. I have breaks. I know how to downshift. I'm fine.
I feel much safer knowing the control is in my hands, than an arbitrary machine anyway.
Is it just me?
http://use.perl.org
Am I the only one who thinks 300ft is a mite bit too far to look ahead to get reliable results? What was the old drivers ed rule, one carlength spacing per 10 miles per hour of speed? A typical car is what 20 ft., tops? 300 feet is reasonable then if the vehicle in question is going c. 150 mph.
Since most traffic is less than half of that speed, I can only shudder at the number of false positives this system's going to come up with.
Honda: at least do this: make this system by default only operational when running in cruise control (which at least takes out the cases of heavy traffic false positives).
You're assuming this is an all or nothing, one-time decision. This thing probably re-evaluates the situation 40 - 100 times per second.
I doubt it will jump to any rash decisions. If a car pulls out in front of you a ways a way, the system knows that in 2.4 seconds there will be an impact if things don't change. It taps the brakes (no big deal) and tightens the seatbelts in preparation for that possibility, then reassesses the situation another 100 times before it makes any decisive action.
I think it's great. I'd have it in my car in a second. Honda is one of the best innovators there is in the car industry, and one of the safest and most reliable (yes, I drive a Civic). They wouldn't release something that's half-baked. They simply don't have a history of doing it. They're also about the only company that has a viable upright-walking robot, and have been for 10 years. They have the technology.
90% of the accidents i've seen are from people just SLAMMING on their brakes. They cant stop in time, and careen into the back of another vehicle, or T-bone them.
The solution is more drivers education. You have to learn that you have more than just 1 dimention of freedom, and can change lanes to avoid a collision. More education, Less 'toys' in the car to distract the driver.
-Tim
Is the number of "300 yards! what about rush hour" posts here.
Think about what you've just said guys! Do you REALLY, honestly think that they would release a car that stamps on the brakes when *anything* is in range. Give it some intelligence.. Sheesh.
It only has to save my life during any one of the possible scenarios before it's worth it to me...
I immagine the system would have to have an on/off switch anyway. And that it would have to have some kind of limitations so that when I'm driving around a mountain road i don't take a flying leap over some cliff.
One more reason to keep an eye on your money.
That said I see too many problems with this in terms of many driving conditions resulting in conditions that might look like a potential crash that are really "normal"--and I don't see how such a system could distinguish between the two.
Maybe Volvo doesn't have that technology because it's not safe yet.
Think like a man of action, act like a man of thought.
I just don't think that we're far enough along to have the car make judgements
Couldn't we say this about ANY technology? "I dunno about them newfangled self-powered wagons. Seems dangerous. I don't think we have the technology. We should just stick to old reliable horse drawn wagons."
We have to gain the experience *somewhere*.
Wrong. This is a huge deal if you happen to have seen the problem 125m ahead and are taking evasive action other than braking, for example changing lanes or even swerving. The slightest braking during a sharp direction change can throw your car into a skid.
It is illegal in most (all?) countries to interfere with the driver and/or the car's controls. Why should techology -- with less sensory input and a poorer decision making ability -- be permitted to do this?
i-name =twylite [http://public.xdi.org/=twylite], see idcommons.net
With as many people that tailgate and cut over directly in front of other cars (including just a few feet in front of one they just passed on an empty highway to then just step on the brakes), this should wreek havoc on cruise control systems and also wear through brake pads like sawdust.
At 60 mph, a one second distance gap is 88 feet. So 300 feet is roughly 3.4 seconds. Does anyone even stay that far behind the car in front of them anymore? I remember driver's ed/mva handbook recommending 3 seconds or so in distance, but my observations are that this is rarely more than 1 - 1.5 seconds.
Also, isn't it the last thing people need is a distraction in the event of an emergency. Granted it will take their concentration away from a cell phone or makeup application. How many times has a startled passenger's shrieking caused confusion just enough to distract the driver from the real danger?
It's fairly easy for a radar system to pick something up in front of you, and for a computer to track it. What we're talking about is picking out all the things in your path, and figuring out if you're going to hit any of them.
The trouble the previous poster was referring to is that so much depends on context. For instance, what if I'm in a left turn lane drive directly toward a car in an oncoming left turn lane? We're not going to collide, but does my car know that?
Following closely to prevent someone "cutting in" is dangerous and illegal. If someone enters the space cushion you have left in front of your vehicle, then you have to slow down until you re-establish a safe zone. It's too bad for you if you think you're a road warrior and can't bear the idea of allowing a crappy driver to get in front of you.
Safe driving begins before you get in your vehicle: if you are in a mindset where you need to drive recklessly in order to shave five minutes off your driving time, you've already screwed up. Leave earlier or move closer to work. Take public transportation.
With all the attention paid over the last couple of decades to drunk driving, you would think that people might notice that the real deadly statistics come not from driving drunk but from driving itself. It is time when we Americans need to realize that the numbers of people killed on highways because of aggressive, irresponsible and careless behavior is no longer acceptable just to maintain our carefree, my car represents my Yankee Doodle individuality lifestyles.
The smart thing would be if the 300-ft limit was not hardcoded, but depended on the speed you were traveling. 300 ft seems somewhat reasonable if you're going 80 on the freeway, but really dumb when you're crusing at 30 on residential streets.(actually, I still don't think most people give nearly that much room between cars on the highway, but they give more).
[SIG] It's like putting a moose in the blender -- a recipe for disaster!
To take your example: Honda engineers discover that an oncoming car in a right curve that appears out of "nowhere" (from around the side of the mountain" causes the device to activate unnecissarily. Solution? There are lots of possible solutions. Is the vehicle approaching you at greater than your own speed? If so, it's oncoming traffic, and automatic braking is an inappropriate response (though a warning light/buzzer -may- be.) What is the four-second history of the angle of your car's wheels? If it suggests that you're on a right curve, it would make sense that an oncoming vehicle would appear to be directly in front of you; no alert necessary. Heck, you could even disable the entire system whenever your steering angle is greater than some small angle--it's safe to assume that if you're engaged in a turning maneuver, your attention is on your driving (whereas if you're bombing down the Interstate in Nebraska, it's much more likely that your attention will wander.)
I'm not saying that we shouldn't expect stringent and rigorous development and testing of such systems, but I find it a bit disingenuous to question the ability of such a system to work successfully because one can imagine scenarios where it wouldn't work. There are umpteen-thousand potential scenarios in which it wouldn't "work". That does not mean, however, that Honda cannot produce a system that provides a very real safety enhancement without putting drivers at risk from unnecessary activation.
Airbags, when they first came out, did their job pretty darn well, but there were still a small number of cases--very short people, overly-sensitive triggers, overly-forceful deployment, etc.--where they were problematic. These kinks got ironed out. Today, airbags are really, really, really reliable, they make cars considerably safer, and I'm glad we have 'em. I can see a system such as the one Honda is developing having a similar history--pretty good launch, a few rare but highly-publicized problems, second and third generation systems being excellent safety systems.
Obliteracy: Words with explosions
The distance that the car should be measuring should be based on whatever the current speed of the car is to compensate for the fact that the faster the car moves, the greater the breaking distance, and if the car is barely crawling (as in bumper-to-bumper rush hour traffic), the stopping distance is practically zero.
Most driving guides recommend that you follow the car in front of you no closer than 2 seconds and I would think that any automatic braking system that kicks in at around that would probably be very appropriate, personally. But 300ft? Even at highway speeds that's almost 4 seconds of distance between cars!!! For in-city driving, it would be absurd.
File under 'M' for 'Manic ranting'
99% of the replies so far may as well be modded -1 Redundant.
Do you people honestly think the Honda engineers aren't bright enough to think of the objections people here came up with 5 seconds after reading the article summary and pouncing on the "Post" button?
The article is light on details, but it still makes the point that this is a collision WARNING system. It doesn't seem to be designed to stop the car or brake to avoid collisions; it's a system that fires off a small warning whenever it detects a potentially dangerous situation - say, if you're dozing off in rush hour traffic and you don't notice the car in front of you is stopped, this'll ideally snap you back to attention.
It doesn't seem that it will brake enough to get you rearended; I'm SURE the Honda engineers can come up with a way to tell the difference between a squirrel, a tree, and an SUV; it's not very difficult to tell which way a vehicle is going, so it's easy to make the system ignore cars going past you in the opposite direction, or cars passing by perpendicularly at an intersection. I don't know the reasons behind the 300 feet range (although I'd imagine the range is dynamic and proportional to your vehicle's speed), but without more information I'll have to assume the Honda people did their research and have some rationale.
There, was that so hard? I'm a couch Honda engineer too now!
ClutterMe.com - easiest site creation on the Net. Just click and type.
Why should techology -- with less sensory input and a poorer decision making ability -- be permitted to do this?
You're assuming an awful lot about the driving ability of the average person out on the roads.
Radar-based tracking is a fairly proven technology. It knows the difference between two different objects, so it starts tracking the car in front of you and should operate fine even if it switches lanes, so long as it's within radar range and whatever side to side, up and down tolerances are programmed.
The global economy is a great thing until you feel it locally.
The problem here is that the majority of people are poor drivers. I blame our driver's education system. Notice, these classes teach very little about vehicle handling or basic mechanical functionality: they teach about the laws and rules of the road.
Now, it's good that Timmy knows the difference between a white line and a yellow line, but that won't help him when he doesn't understand that if he decellerates on a slick curve, the weight of the vehicle will transfer to the front wheels, possibly causing the rear ones to lose traction and induce an oversteer (Timmy spins out and causes an accident).
Or how about proximity? Notice how, when there's a small piece of debris in the road, most drivers give it a good 4-6 feet of berth just because they don't actually have a sense of the boundries of their car.
Driver's education should be rigorous and difficult: not designed so that everyone passes. (How many people -actually- fail Driver's Ed? There was one in my entire high school class of several hundred, and she was, shall we say, half a half-wit ((a quarter-wit?)) ).
[end_rant]
GeekNights!
Late Night Radio for Geeks!
Why should techology -- with less sensory input and a poorer decision making ability -- be permitted to do this?
Because a lot of people have an inexplicable confidence in technology. They think computers are infallible. Airbus jets, for example, are programmed to override the pilot in crisis situations.
"My God...It's full of ads!" -Fry, about the Internet, Futurama
override the pilot in crisis situations
This is a real good thing, in my opinion.
I've flown planes - albeit much smaller planes. There are so many variables, and in a split-second crisis situation I can 100% guarantee that a pilot can not assess enough data to make a rational decision.
A lot of things can happen on an airplane in a very small amount of time. Microseconds can be critical. In many cases it is *impossible* that a human being could see the problem, realize it as a problem, consider options, choose a course of action, and follow through. In some cases - imagine rapid depreasurization or a critical area birdstrike - the time to avoid a fatal catastrophe is virtually instantenous in nature. You have 1.25 seconds to shutdown the left second turbine. The knobby switch for insta kill is 2.5 feet from your nearest hand. Do you feel lucky?
Three days over 6 months to get a motorcycle license in Oz (NSW), after you have completed the road rules test, which you may take at your leisure.
Day 1, you are required to demonstrate you are able to ride a bike before you're given your L's (Learner plates/license). 3-6 months later, you're back for another two days worth of obstacle avoidance, emercengy breaking and general "how not to get dead" theory and practise. Gruesome video footage of people who fuck it up is optional. You are then tested on your emergency skills, plus a standard road ride, before being issued a Provisional license. This allows you to do upto 80km/h, upto 0.02BAC (one standard drink), and three whole points. You get to wallow in your lameness for a year or three - if you fuck up, you're off the road. If you don't, you get an unrestricted license. By this time, you have real experience under your belt, are now 20+ years old(er
Opportunity knocks. Karma hunts you down.
I will leave the driving to me. This sounds even more absurd than the Linked Braking System that Honda uses on some of its motorcycles. That took 3 or so revisions to make it reliable, and un-noticable.
;-)
Imagine the number of brake checks in your commute to work it it goes off anytime a car is slowing down within 100m of your front bumper. Where will the fuse for this be located again?
While I agree with many of your statements, there is such a thing as going too far.
Hell, GPS tracking of vehicles would, if it reduced traffic deaths by a few percent, would be well worth the loss of privacy.
No. Sorry, but the privacy bit is too big. It's still possible to do it, but not with a total loss of privacy please.
I'm not worried that I'll be the cause of an accident. But I'm scared shitless of your driving, because you are, in all probability, one of the drivers who is a threat to my continued well-being.
I used to be an incredibly shitty driver... oh, sure, I thought I was a good one, and could drive fast, yet safely, etc... but then I got married and started driving more safely. Partly because my wife made it quite clear that she didn't like my driving (yes you did hon), but mostly because I couldn't live with myself if I got in an accident and hurt her. And to all you dweebs thinking that it's sappy and crap - yeah it is, and if you ever really love someone then you'll understand it at that time. I know I didn't until it happened to me.
Driving more safely now, I'm amazed at just how many utterly shitty drivers there are out there... and I also know that just about every single thing that I see them doing is the same crap I used to do. It's not safe. You are not some god of driving. You are causing problems by tailgating and making other drivers uncomfortable around you.
Yeah, I still speed, but not like I used to. I no longer consider the speed limit as a posted minimum, and I'm a lot more wary of rapid lane changing and weaving. Tailgating is just abysmally stupid. Not that I recognized these things as bad back when I was driving like that, and not that I actually expect any of the moron drivers out there to become enlightened based on reading this... but maybe in a few years they'll also look back on their driving habits and realize what idiotic punks they really were. I do.
All of that said, as long as the system works well (which is a huge caveat), I'm all for it... this probably would've saved my sister a great deal of physical therepy when some stupid driver decided to cross over the double yellow line in rush hour traffic and hit my sister's brand new Honda at 45 mph (which is the speed both cars were going... in opposite directions). My sister was fortunate to live through the accident, thanks to modern safety systems, but she wasn't able to drive for nearly 6 months afterwards due to injuries. And I know she was lucky.
If both cars had been equipped with such a system maybe each car would've only been doing 30 mph at impact instead... which would've been less than half the impact force. Sounds good to me.
That's the argument people always used to use with seatbelts... what if I were trapped inside the car and it was burning, and I couldn't get out because of the seat belt?
You just have to do a cost/benefit analysis... as long as the new technology saves more lives than it endangers, it's worth pursuing... you can always improve it to reduce the false positives.
Doug
Venn ist das nurnstuck git und Slotermeyer? Ya! Beigerhund das oder die Flipperwaldt gersput!
Regular driving exams, say every three to five years: great idea.
Graduated licensing programs: great idea.
Mandatory driver training: great idea.
Stop right there. You were on a great roll there. We have hideous safety statistics in this country precisely because we give out driver's licenses in Crackerjack boxes. And we respond by lowering speed limits, which doesn't really work.
You want safe roads at any cost? Really? OK, simple: National 15 MPH speed limit, enforced with severe jail time. Or maybe death. Didn't think so.
You make a few great points here. Driver training and licensing in this country is a joke. I don't have his statistics handy, but there is an ER doctor in Southern California who is tracking the DMV records of a group of drivers who participate in performance driving schools (NASA and SCCA street schools, Open Tracks, AutoX schools and the like). He's seeing better than 90% reductions in both accident and moving violation rates. Ninety Percent! From better driver training. Not automagic systems that drive for you, just having a human that can actually control the machine.
Black boxes reporting accident data: great idea.
There are very real privacy concerns here. And very real property right concerns. It's not that having good data in a real accident is bad, it's the legal environment surrounding such data in the country that is horrifying. We have an environment where speed limits are set for political (I don't want them going fast near *my* house) and revenue reasons, not actual safety and engineering reasons. Yet exceeding those artificial speed limits is prima facie evidence of fault in any situation.
Automatic safety systems: great idea.
Maybe. Have you ever been in a situation where avoiding the accident required accelerating? How do you think the brake grabbing systems described here are going to react?
Photo radar: great idea.
If it were actually being used to enhance safety in places where the speed limits are set rationally, yes. But they're not. They're used to enhance revenue in places where the speed limits are set arbitrarily.
Let's look at a related issue, one that based on your comment is near and dear to your heart: Red light cameras. There have been numerous cases over the past couple of years of municipalities reducing yellow light duration to increase revenue. In Fairfax County, VA, cameras were installed at one intersection because of high incidences of red light runners. The cameras were catching an average of 52 events a day. Increasing the duration of the yellow from 4s to 5.5s reduced that number from 52 to less than 1. Engineering fixed the problem, not enforcement.
Hell, GPS tracking of vehicles would, if it reduced traffic deaths by a few percent, would be well worth the loss of privacy.
Do I *have* to quote Franklin?
I'm not worried that I'll be the cause of an accident. But I'm scared shitless of your driving, because you are, in all probability, one of the drivers who is a threat to my continued well-being.
I hear you. The average joe out there can't drive. The solution is to *teach them to drive*. It really is that simple.
Let's get our streets safe.
Through training and safety engineering, yes. Trying to idiot proof the roads and cars isn't going to work. Reducing the idiocy of the average driver will. (And does!)
You have violated Robot's Rules of Order and will be asked to leave the future immediately.
You didn't quote an important part of the comment above yours.
It's assuming a lot that the person is going to be changing lanes or swerving? That seems like about the most likely response... either that or just jumping on the brakes at full, which is why we have ABS; to handle panic responses from people who really have no business driving to begin with.
Now, this system definitely has less sensory input. It has only one sense - depth. We have five, at least two of which apply to driving in every situation; vision and touch. Hearing usually comes into play as well, unless you have the radio up so loud that you can't hear anything else. (This is generally only a problem when the windows are up.) And your brain is necessarily more complex than this auto-braking system.
It has been shown that some technologies that do things for the driver can be beneficial to yahoos who don't know how to drive. ABS is one of them. However it doesn't do anything on its own, it only comes into play when you actually do something. This system does not, and as such it is fairly worthless, and will most likely cause more harm than help. Another one is skid control, unless you are driving in a way that you probably shouldn't be on the street (Though my car is designed to slide, and I do it basically every time I drive, in a quite controlled fashion) skid control is a great thing. Of course, it builds on ABS, and is really just an extension of same. But you see, those things are good because they help you maintain traction and therefore control. The best thing about ABS is that you can slam on the brakes as hard as you can, getting "maximum" stopping force (ABS actually stops longer than a skilled driver with manual brakes of the same size) while still being able to steer.
So if your argument is that people might not be steering to avoid something, well, those people shouldn't be driving at all. This invention will harm both those who can and those who cannot drive. The only people it will help are those who are asleep, and I think we need to work on more technologies to keep them awake instead.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The trouble the previous poster was referring to is that so much depends on context. For instance, what if I'm in a left turn lane drive directly toward a car in an oncoming left turn lane? We're not going to collide, but does my car know that?
Perhaps the emergency system does not kick in until the car can calculate that a collision is inevitable or at least come very likely. In your left turn example, this would not be the case, because both of you are travelling at speeds where you have plenty of time to break or turn before hitting each other.
But if you came speeding into the intersection, the car could calculate that even if the driver would break, and/or turn, there would be no way to avoid a collision. Time to tighten the seat belt and try to slow down.
Generally speaking, I am surprised to see how negative Slashdotters are to new technology, especially before knowing much about how it works. The assumption always seems to be that the implementation will be completely useless, dangerous and insecure.
Of course there will be troubles to get this to work properly. Honda will solve those problems, and put the technology in their expensive cars. If it is efficent and cheap enough it will go mainstream. That's all there is to it, pretty much.
Tor
Nonsense. It stands, or at least, can stand for 'rice burner,' indicating that it comes from an area with high yields of rice. If referring to a region based on suitability for certain food crops is 'racist,' then we've got problems.
Vintage computer games and RPG books available. Email me if you're interested.
You are forgetting that when you are turning a corner, you are turning the steering wheel. You would think that a system like this would know which direction your are travelling! If it didn't, it would be a pretty dumb idea.
kyjello is too damn smooth to make a signature.
The system is a lot smarter than you seem to be willing to grant it.
Why does everyone hear assume that a complex auto-breaking system is going to be built by idiots that do not even consider simple situations?
While I am sure that this device might cause problems in complex situations, the simples ones should already be accounted for.
excitingthingstodo.blogspot.com
with that ability, or lackthereof comes liability, blame
who do you blame when this system fucks up.
Im against automated anything with cars. Do you know how often small electrical sensors go bad in cars and start to cause it to run unreliably. Up till now it meant your mixture control got fucked up and you started running lean, now it could mean your car suddenly decides to lock them up doing 85 on the highway.
People drivers aren't perfect. But its easy to counter their imperfections. When driving, a good driver is aware of everything around them and can tell when someone is not in command of their vehicle. You cant tell when some automagic sensor is about to fail.
Im sorry. I havent done the research admittedly, but I dont think traffic accidents are quite bad enough to say we're assuming an awful lot about the driving ability of the average person out on the roads. Id say youre underestimating it. Drunk driving. Now theres a crazy real problem, and no system like this could fix it.
Maybe the car should have a radar at the back as well, taking into account what will happen if it suddenly brakes. A predictive system should probably turn on the brake lights (although no actual braking is taking place) just in case it really needs to break soon right after.
What time is it/will be over there? Check with my iPhone app!
We've got plenty of safety critical technology aids in cars already. Witness ABS brakes, automatic transmissions, cruise control, etc. They all all aid the driver (in most cases) but don't drive for him. This system starts to drive the car instead. Would you want the computer to jerk the wheel on you? I think controlling the throttle (cruise control) is a bad enough idea. Anything that requires the driver to concentrate less on the path and speed of the vehicle is bad, imho..
Actually, I worry that a system like this could make the driver LESS effective in avoiding the crash than alerting him to it. A sudden acceleration of the body is FAR more disconcerting than a noise or light going off. The time it takes the brain to understand why it just got jerked is probably more than it'd be to just realize the situation and take corrective action with the aid of a buzzer or something. This is moreso when the driver isn't paying attention. If the driver is actively monitoring the impending situation and the car takes action prematurely, he's likely to be less surprised. But if he's already paying attention, then you don't need this stupid thing anyway. An unfocused driver isn't going to be expecting a jerk like that, and is likely to spend more time in the "what the heck was that" mode than "move my limbs we'regunnadie!!!" mode. I'm not a physiologist, I'm just guessing..
Focussing the radar seems like an easy solution in which there's no need to associate it with GPS or road conditions. Simply target it based off of your steering wheel rotation.
If the steering wheel is pointing to the right, focus the radar to the right as that's where the collision would most likely happen providing it was you doing the hitting.
Now if it was someone going to ram you from behind could it temporarily boost the speed for a second?
The real danger with this system in my mind is with rear end collisions. Sometimes braking is not a good option because of tailgaters, and the article doesn't mention anything about monitoring someone behind you. I've avoided about eight accidents in the past few years, all from tailgaters that would have hit me if I didn't swerve over in time to avoid them...
Here's the exact scenario I envision. You are driving along, leaving some space in front of you like a good driver. Suddenly, a Jerk(TM) pulls in right in front of you - and because the cars up ahead are slowing down now (which is what made the Jerk(TM) jump lanes in the first place) your cars auto-brake freaks out and slams on the brake.
Of course, what the auto-braking did not know was that Tailgater(TM) was driving (literally) right behind you. And now his bumper meets yours in an unholy union.
But I guess I few more rear-end collisions is a small price to pay for a safer car to drive. Or something like that. Wait till they try this in America, that's all I can say! Have them try out the car for a day in Boston or someplace like that and then see what they think about auto-braking.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Perhaps because many of us have seen it fail miserably despite glowing claims?
7 November 2006: The day Americans realized corruption and incompetence weren't addressing 11 September 2001
exactly my point. How about a couple more confusing scenarios?
1. Some dork cuts you off on the freeway and steps on gas. You don't want or need to break even though the idiot is only 20 feet in front of you.
2. Wet pavement, your friend is tail-gating you. Now you neighbor's cat runs across the street. You would rather run over the stupid thing (and remove stupidity from the gene pool as an added benefit) than get rear-ended by your friend.
3. You are merging into a tight spot on the freeway. If you push the breaks, you are likely to "clip" the car in front of which you are trying to merge.
4. Some dork didn't bother looking when merging/fell asleep/whatever-else and is drifting into your lane on the freeway. Behind you is a semi truck. The only way (besides honking) to avoid a collision is to downshift into second gear and step on gas even though the car in front of you is closer than 300 feet.
5. Your example of the left-lane turn is very good too. It's a classic "looks like a head-on" scenario.
Also, sudden breaking under some circumstances (steep turn, slippery pavement, poorly distributed load in the vehicle, etc) drastically reduces responsiveness and maneurability of the vehicle.
Jobs? Which jobs?
Generally speaking, I am surprised to see how negative Slashdotters are to new technology, especially before knowing much about how it works. The assumption always seems to be that the implementation will be completely useless, dangerous and insecure.
I may not speak for all SlashDotters, but perhaps our negative attitude comes from having seen the little man behind the curtain a few too many times. Although I am a web application developer, I still do my taxes with paper forms, and refuse to sign up for receiving bills by email instead of regular mail. I'm perfectly willing to order t-shirts and books on the web, but for the important stuff, I want to see paper documents. And there are few things more important to me than the performance of the vehicle I'm riding in.
Perhaps it is the typical SlashDotters familiarity with the concept of probability that makes us suspicious. What percentage of false-positives would be acceptable for a car that applies its own brakes?
Also, I just keep picturing some driver getting creamed by a truck while looking back, asking: "Who's pulling on my seatbelt?"
On the bright side, pay attention to the technologies that SlashDotters are excited about. Linux, Open Source, Macs, TiVOs, MySQL, PHP -- good stuff!
Mike van Lammeren
It will challenge your head, your brain, and your mind.
Depends on the driving habits of your local city.
.01 MPH faster than the one they were in originally.
Here in Los Angeles, I always try to maintain a safe distance and keep several car lengths between me and the car I am following.
Unfortunately for most drivers, just one "car length" means "I can change lanes and drive here" and cut you off.
So basically you have to keep hitting the brakes because asshats decide to use your buffer zone because they decided that your lane is going
I tried every decent and legal way I could think of to resolve the issue w/the business before I rented the chicken suit
This could also encourage laziness in the part of the driver as he is conditioned not to brake until the car starts braking for him. Remember Pavlov's dog? Same premise works on humans as well. The worst thing about this though is that these vehicles will be driven on roads that see ice, freezing rain and snow conditions. Touching the brakes unexpectedly in these conditions can easily cause a vehicle to go out of control. This is nothing like driving on dry roads at all, and requires much more skill on the part of the driver. Since I live in Minnesota, a state renowned for it's winters and bad weather, this is not an idle concern. Vehicles with brake systems that engage without the driver pressing the brake pedal first should be banned from public roads for safety's sake. I am not referring to brake assist feature in some cars that helps push down the brake pedal when panic braking on behalf of the driver is detected. I say this all as someone who has been in a very severe accident where such a system in the vehicle behind me just might have prevented the accident (rear ended at freeway speed by full size truck).
you are cruising on the free-way at speed, the car in fron off you slams on its brakes. You realize that you are folloing too close( just like everyone else), and quickly check the lanes to the left and right, taping the accelerator you slide to the right and avoid the hazard, giving the driver behind you ample time to break and avoid the same obstacle....NOW the same scenario, only as you scramble for an exit to the left or right, your car begins to brake by itself making a lane change MUCH harder than it would be if you were at traffic speed or slightly faster. Granted this is probably a fairly rare happening but there are a LOT of options and complications to deal with, and I for one would not feel comfortable driving in a vehicle which did not respopnd EXACTLY as I asked it to.
errr....umm...*whooosh* *whoosh* Is this thing on ?
"While I am sure that this device might cause problems in complex situations, the simples ones should already be accounted for."
But we mortals can already, most of the time, handle simple situations while driving.
It's the complex ones that are complex, and that's where this car is going to fuck up and perform an incorrect decision.
As someone who spends about 2 hours in his car every day, I hear you. I've never been in an accident, but I've had some very close calls and my reflexes kind of suck. For example, one time a duck landed on the highway right in front of me, and I nearly creamed an Oldsmobile trying to dodge it. Stupid, yes, but that was the decision my brain made. I'm trying to deprogram myself by occasionally whispering "death to all lower life forms" as I drive. Next time, the duck is roadkill, but I still believe that the highways would be safer if my judgments were removed from the equation.
The fact is, even the best driver among us as moments of inattention, bad judgments, irrational desires to gun it and chase down the bastard who swerved in front of you, and incomplete situational awareness.
The system would be difficult to implement on a car-by-car basis. Whatever problem the computer has to solve, it would be infinitely simpler if each car knew what the car next to it was about to do, and could warn neighboring cars about the drastic measures it expected to be taking. Networking would also have advantages because it would enable a higher degree of coordination: routing traffic away from potential jams, allowing groups of cars to drive "in formation," and clearing the road for emergency vehicles.
But that would take a massive investment, and I don't see any easy way to switch over. But a non-networked system might be useful. Even a mediocre system would be useful as a punishment for chronically bad drivers, or for people who aren't capable of driving themselves.
Another thing that's been bothering me: Traffic lights. Even with sensors, they're painfully stupid at times. An ideal system would not only sense the single car idling at the red light, but also the pack of fifteen cars heading towards it on the cross street. That way, the system might wait until after the pack has driven through to switch the light. A vision system to implement this wouldn't have to be very sophisticated, and it would speed up the commute, use less gas, and save wear and tear on brakes.
Eventually, I'm sure cars will drive themselves. We'll all be nervous at first, but I don't think a system would be allowed on the road until it was at least as safe as a better-than-average driver. But even if switching fully over would lead to the same 50,000+ deaths that we're racking up now, we would be better off.
You want the truthiness? You can't handle the truthiness!
There's traffic density that would make clog up modern highways for years, but its all flow-controlled, so you go 120MpH with only inches between cars, so your trip takes half the time.
The problem with this is that everyone has to have the same car performance-wise. That might be achievable. There could be standards mandated to enter the particular roadway... but then everyone using these cars/trucks will have to keep them maintained and performing in the same fashion. This probably isn't achievable. One guys car breaks down suddenly, and the whole system crashs... Literally, as in, into each other. Bob's car has a tire fly off, the 600 cars behind him are going to have a rough time if they are travelling at 120 with scant inches between themselves, no matter how smart the highway is.