"Every man, woman, and child in the U.S. has approximately a one-in-8000 chance of perishing in a car accident every year. Over a decade, that's about one in 800. Given the millions of cars included in the Toyota recalls and the fewer than 20 alleged deaths over the past decade, the alleged fatality rate is about one death per 200,000 recalled Toyotas. Even if all the alleged deaths really are resultant from vehicle defects—highly unlikely—and even if all the worst things people are speculating about Toyotas are true, and you're driving one, and you aren't smart or calm enough to shift to neutral if the thing surges, you're still approximately 250 times likelier to die in one of these cars for reasons having nothing to do with unintended acceleration. So if you can muster the courage to get into a car and drive, the additional alleged risk of driving a Toyota is virtually negligible..."
The rest is in the C&D article referenced, I believe, by the author:
The C&D article (and the ones it points to about their road test) is really good. Because there are more people talking about this than actually know anything. Including in this thread.
The C&D article does leave the door open to the possibility of the brakes not stopping the car. If you pump the brakes, rather than firmly putting them down once, you may lose vacuum, or overheat the brakes in a series of "slow down a little" steps. The vacuum is apparently hard to replenish when the throttle is open.
The advice about turning the car off works sometimes, but not always. Suppose you're in a rental Lexus like the one the CA state trooper died in. Would you know that to kill the engine, you have to depress the off button for 3 seconds continuously? Three seconds is a long time with the throttle wide open.
It seems the key technical mistake on Toyota's part, as pointed out in the C&D article, is to omit an interlock that kills the throttle when the brake is depressed. This is apparently standard on many other cars.
Toyota's damage control on this has been disastrous.
This is definitely a witch hunt. The same thing happened in my ford truck when I misplaced the mat after cleaning it. I simply turned the car off put the clutch in and coasted to a stop at the edge of the highway and fixed the mat. From the reports this sounds like the exact situation for the Toyota cars. From my perspective nothing is wrong with the cars and the whole thing is way overblown.
Many of the Toyota problems had nothing to do with the mats--they are drive by wire and the computer was slammed on the gas without the pedal being depressed. Depending on when this occurs it could be deadly even with an expert driver.
Please forward your proof of this to Toyota and the congressional committee to which Toyota swore this was not the case.
It's tempting to blame drive by wire, since electronics is black magic to most people. Keep in mind that most passenger jets in the sky are fly by wire, these days.
For you to point blank claim this to be true, when it's been flatly denied several times, seems a bit over the top, don't you think?
Complaints of out of control acceleration increased several times over after moving to DbW and some recent complaints have noted that no floor mats were in the vehicle (which makes it hard to blame on floor mats):
Toyota is claiming it's not the DbW system, but they also thought they had this issue fixed a long time ago. It's a hard thing to prove unless you see it yourself (unlike a broken cable like we used to have before DbW).
Complaints of out of control acceleration increased several times over after moving to DbW
Complaints of out of control acceleration increased after 2000. I guess it's the millennium bug. In other words: correlation is not causation. Moreover, no numbers are presented to actually support this correlation. What about 1996-1999? 100 cases would be totally plausible.
Even Woz thinks is the DbW software
With the rather important difference that it doesn't actually cause any problems, because he can just brake and make the acceleration stop.
Really? I think it was an article here (maybe, maybe not, I'm pretty hung over today) that mentioned that the problem was that the primary and backup pedal position sensors' wires could short and max out.
You usually design so that a single point of failure can't hurt you. If one position sensor (they use pots, since they're practically indestructable and not vulnerable to ESD, power transients, etc) fails, you sense the problem with the other sensor. They're not "primary and backup", they're both equally important.
You postulate that both sensors short out at the exact same instant, since any other scenario would power down the vehicle. They did it within milliseconds of each other, and they did it at some intermediate, yet believable level. (They didn't "peg at maximum", as that would also alert the software there was a problem. Instead they pegged, identically, at some believable value.
Can someone explain to me why Congress gets involved in investigating things like Toyota's manufacturing defects or the way college football teams are ranked?
Does Congress have a legal mandate to investigate anything it deems to be interesting/fishy/a potential target for legislation? Has Congress always conducted these types of investigations or is it a modern phenomenon?
cynic: Congress' mandate is to investigate anything that will get votes, or campaign funds.
Wikipedia:
Investigative hearings share some of the characteristics of legislative and oversight hearings. The difference lies in Congress’s stated determination to investigate, usually when there is a suspicion of wrongdoing on the part of public officials acting in their official capacity, or private citizens whose activities suggest the need for a legislative remedy. Congress’s authority to investigate is broad and it has exercised this authority since the earliest days of the republic. Its most famous inquiries are benchmarks in American history: Credit Mobilier, Teapot Dome, Army-McCarthy, Watergate, and Iran-Contra. Investigative hearings often lead to legislation to address the problems uncovered. Judicial activities in the same area of Congress’s investigation may precede, run simultaneously with, or follow such inquiries.
It's not a recent phenomenon at all. For example, the Crédit Mobilier hearings were in 1872.
Thanks for the answer. The difference is that the scandals in the Wikipedia article all directly involve government:
Credit Mobilier: "The distribution of Crédit Mobilier stocks by Congressman Oakes Ames along with cash bribes to congressmen"
Teapot Dome: "control of U.S. Navy petroleum reserves at Teapot Dome in Wyoming and at Elk Hills and Buena Vista in California, were transferred from the U.S. Navy Department to the Department of the Interior"
And Army-McCarthy, Watergate and Iran-Contra are all obviously directly tied to the government.
In those cases it makes sense for Congress to get involved. It's when Congress start to investigate "private citizens whose activities suggest the needs for legislative remedy" that I get concerned. That type of broad power apparently gives them free-reign to investigate anything they don't like.
It seems to me that in the past (15 or 20+ years ago) Congress didn't get involved in non-government related "scandals" like Toyota and college football, I'd be interested if anyone has examples to the contrary.
Seeing as how the government is in charge of setting the safety standards for the nations interstate highways and is responsible (via the NHTSA) for regulating the safety of cars on the road I am wondering how in the world you would not think this is the sort of thing Congress should investigate? If you are looking for a specific justification, examine the commerce clause of the us constitution.
Are we really downvoting facts now? I assume that the people who are downvoting don't think that the commerce clause should be used to grant congress such wide powers. I agree, but you can't just downvote the situation. You have to recognize the problem before you can fix it. So thanks, ryoshu, for reminding me of the part of the constitution that needs to be amended.
Toyota walked into a rotating fan blade. This was a really bad time to give Congress an excuse to concern-troll them, given that the government is now a direct stakeholder in the US auto industry.
It's times like this I feel like someone needs to tell the American people to grow up and stop acting like scared children. They're so easily manipulated by simple fear tactics. Automobiles are incredibly dangerous simply due to the drivers behind the wheel. Design flaws are the least of your worries. We can't seem to muster the same type of concentrated rage against drunk drivers, texting drivers, or any other impairments that make automobiles less safe. We can't even make laws that make automobiles safer by design. Why aren't side air bags a requirement? or roof air bags? More frequent safety inspections? Bad tires or worn out windshield wipers are so much more dangerous than the statistical improbability of a stuck throttle. It really makes me incredibly angry how capable Americans are of ignoring big problems and focusing on the tiny ones instead. We're doomed as a country if this mindset doesn't change and I'm not sure we don't deserve it.
While I agree that the media and politicians are now feasting on Toyotas failures, the fact is that Toyota really botched this. Had they taken the reports seriously earlier none of this would have happened, and lives could have been spared. Apparently they never thought this story could spiral out of control.
People seem to forget that this only happened until after Toyota had repeated recalls, each promising to fix the problem.
Is there a ghost in the machine? I don't know, Woz certainly seems to think so. Replacing floor mats and then gas pedals seems awfully like what happened with the Therac 25 radiation overdoses in the 80s.
(The Therac 25 manufacturer first added an extra hardware switch to the radiation therapy machine and claimed it was "an order of magnitude safer". When more people died, the quick emergency fix was to remove the "up" button from the VT100 console.)
If Congress really wanted to get at the truth, they should have called disinterested third-party engineers to study and get their opinion on this case. Nobody believes Toyota, even if the final facts prove it's correct. Everyone believes the witnesses, even when the engineering evidence often disproves their testimony. It is impossible to come to a scientifically valid conclusion under those two circumstances, which is why many individuals involved in this issue have described the proceedings as "witch hunts."
Actual American political debate hardly happens anymore. What we have now is theater. The mentality of the LCD American voter is simply abysmal. You can see it from the low or sloppy intellectual content of the "product" which is presented to them by politicians, journalists, and other media workers.
It may be a witch hunt, but I have a 2008 Prius that has had a number of issues with brakes after potholes etc -- going for about one second without having much effect.
It has happened to me once when I went over a pothole and my wife a number of times (she drives more).
I'm in Canada and my local dealer can't reproduce the problem yet, but put yourself in the position of a Toyota owner who is no longer confident in being safe and who would get less than half the purchase price for a car with < 70,000 KM on it.
The Honda dealer doesn't want Toyotas as tradeins; they can't sell them.
Can you feel the anger?
I don't want to have to buy another car; I want to work on my startup. But I don't want to worry about my family's safety or buying some used car.
Toyota should be forced to buy back their crap products at some reasonable percentage, since these problems remind me of "heisenbugs" in code...
Or to put it more succinctly: just because there are stupid and/or dishonest people complaining about bugs in a product does NOT mean there are no bugs in said product or that all claims are baseless.
I would bet any sample of 3000 complaints about the Ford Pinto or any product had at least some stupid and some dishonest ones.
It would be very difficult to evaluate Toyota stock. This kind of media events can warp public opinion about a company for a long time. I think it would be very hard to put number to this. I would bet that even if tomorrow the government comes out and says "oops, my bad, everything is fine" you wouldn't see a sudden reversal.
Instead, we were treated to Dr. David Gilbert of Southern Illinois University, also a guest of Mr. Kane's, who claimed to have found how Toyota's electronic system could totally malfunction, creating a runaway car—and claimed he'd found the error in less than four hours. Spoiler alert: Dr. Gilbert was assigned this work by Kane's safety advocacy firm, with at least partial funding by trial lawyers.
Here, too, is a problem: Dr. Gilbert said he relayed the results of that test and his concerns directly to Toyota. In short order Toyota looked into Dr. Gilbert's claims and found them not to be valid in terms of creating unintended acceleration. Then, to the company's surprise, it watched his appearance with Brian Ross on ABC News this past Monday night, Feb. 22.
According to Toyota, it now appears that Dr. Gilbert had done something completely different in order to get a Toyota Avalon to accelerate under its own power. Toyota offered to evaluate Dr. Gilbert's Avalon, with ABC in attendance, and see what he did electronically to cause it to accelerate.
Additionally, Toyota is fairly adamant that Gilbert's "test evaluation" on ABC News was not the original "discovery" he relayed to them on Feb. 16.
The professor apparently hooked up an "external bypass device" to a brand new Toyota Avalon, and then complains that the computer should sense the artificial short he introduced into the system and disable the acceleration system. Hardly the most reliable test.
1. to a certain degree, the US is punishing Japan due to their recent actions in the bond markets. This is a very important problem for a very small amount of people, but it makes great headlines. Do you think the US government is genuinely interested in resolving this properly from an engineering perspective?
2. that 100% "drive by wire" systems should not be allowed...there should be a big red "holy shit" kill switch on the dash of modern cars that kills power to everything except steering and brakes
So, imho, its a very serious problem, that affects very few people (in the big scheme of things), and there is a very simple solution, for going forward technology at least, if not existing models. But do you guys get the feeling that this is what the conversation is actually about? Something seems a bit odd to me.
But then I'm one of those conspiracy theorists that think when you're running 15% budget deficits and have real (as measured using historical methods) unemployment of 20%, and many bond issues are taken down 50% by the federal reserve (literally printing money to buy your own debt), that you will start to observe unusual things happening in the marketplace. The adherents of Occam's Razor would imply that that I am simply crazy, becasue the simplest answer is always the correct answer. So I guess thats it.
I had the exact same reaction when I saw this.."congress? why is congress getting involved in this?"
If people are upset with how their Toyota cars are operating, then no one stands to lose more than the Toyota company itself in lost revenue. So what additional incentive or pressure could congress apply here except increased publicity? Let's see...we lose a billion dollars, or get chewed out by some old guys for a few hours, which sounds worse?
It seems really silly to watch politicians chastise a CEO and talk down to him. Good engineering (especially on a massive scale like that) is really hard. If they think they can do better then no one is stopping them (members of congress) from starting their own car company to try and compete. Otherwise, I'm not sure why they feel qualified to criticize.
Something else that bugged me is the quote from one of the witnesses which said 'Shame on you Toyota for being so greedy'. So...you're saying they intentionally pissed off all their customers as some sort of plan to make more money? I'm so confused by this mentality people have that greed in corporations is a bad thing. Yes, you could argue they sacrificed long term quality for a short term savings, but this is hardly greedy, it's just bad business practice that leaves you with LESS money.
This may be unrealistic, but I really wanted to see the Toyota CEO reply to congress with some sort of statement like, "We won't be appearing at any hearings. We are a private company and don't answer to government. We only answer to our customers. So if any of our customers have questions here is a toll free hotline. Members of congress, if you are a Toyota customer, feel free to call as well, but you're role in government won't be taken into account in how we respond."
The Mythbusters tried to slam a car into reverse at speed. What they discovered, is that there are safety mechanisms to prevent this from happening in automatics.
Thank god! I have heard only from people who buy into the witch hunt (and an excellent name for it, that is). I am really very glad to see I'm not the only person who realizes most of it is ridiculous and overblown. Hell, even if the gas pedal corroded AND the floor mat held it down, has not one of these drivers even though to reach down and PULL the goddamn thing up?
I pray people realize this is folly... I don't know what we're gonna do if stupidity brings down the pillar of the auto industry, through no fault of it's own.
Witch hunt is a great word. When I watched a CNN clip of one woman giving evidence about her "broken" Lexus I felt a very strong analogue to Arthur Miller's The Crucible.
I don't disagree that this has been overblown, but in fairness it would be fairly difficult to reach down and pull the gas pedal up while the engine is raging and the vehicle is undergoing significant forward acceleration, while the driver is presumably trying to avoid colliding into other vehicles and objects.
Of course, they should just shift it to neutral and let it redline.
This whole Toyota debacle makes me wonder if competing car companies buy cars made by other manufacturers and poke at them in the hopes of having something just like this happen.
"As noted before, brakes always win out over engines, even at full throttle; that has been tested and proved many times in the past 20 years, including recent Car and Driver tests on Toyotas."
This is not necessarily true. It's entirely dependent on the brakes involved and the engine involved; any given vehicle could go either way. I personally have witnessed steel brake rotors melted and dripping off the vehicle; the brakes had locked up at highway speed and the driver had continued driving, reporting afterwards that the vehicle seemed a little sluggish but not excessively so, so he hadn't stopped. For that vehicle: the brakes lost. For other vehicles, perhaps ones with dinky engines and big fat brakes: the brakes may win. It's certainly not the sort of thing where one can say "brakes always win".
Car and Driver's article ( http://www.caranddriver.com/features/09q4/how_to_deal_with_u... ) seems to assume that Toyotas don't cut the engine power when the brakes are applied, but their own tests belie that. I would guess that Toyota's software cuts the power LESS than the competition's software does. But if you're stopping from 70mph with and without the throttle at full, and you notice a minimal difference - as C&D did - I can guarantee you that the throttle is being reduced when the brakes are applied. Maybe not to zero, but reduced.
Absolutely. I used to own a V-8 Ford pickup; if you happened to hit the brakes and the gas at the same time the power of the engine would completely overwhelm the brakes and the truck would (start to or continue to) move. If you accidently left the parking brake on, you couldn't discern a difference while driving it. Coming from a state where many people own pickup trucks, I was taught that the engine will overpower the brakes, every time. Now I own a small car with exceptional brakes that I'm sure would stop in that situation, but that's just the exception that proves the rule - it depends on how big your engine is.
I bet there is a mechanical/electrical issue that is blown way out of proportion, and human error in some of these issues.
Once I accidentally slammed on the gas thinking it was the brake. (I got confused switching from a manual transmission to an automatic.) It was scary...the car jumped a curb and my driver's ed teacher was furious. He slammed his brake and it stopped the car.
If you are in a car moving along and you have to kill the engine, or the engine dies, get on the brakes as fast as you can and don't let go. You can ease up on the brakes almost to the point of letting go of the pedal, but don't come off all the way.
The reason is the brakes will maintain the power-assist vacuum from the engine, even when the engine is later switched off, until they are released completely. The 'power-assist' strength will remain for tens of seconds, at least. I've tried this many times before. I can't guarantee it works on modern cars, but it sure did work on my '86 Toyota with standard hydraulic vacuum-assist brakes.
The vacuum assist is just that - an assist. The brakes still work if you don't have it, you just have to put some extra force on the pedal. It isn't even that much force; I've driven older vehicles that don't have any assist at all and it's not that much harder to push the pedal.
I hold the conspiracy theory that these issues are being so aggressively sought out in an attempt to bring Toyota's image down to the level of American auto manufacturers.
I wholeheartedly agree. The sheer number of failures that have to occur, the low level of action on the drivers' parts required to overcome the problem (put it in neutral! Use your brakes! Use your parking brakes!), the speed and intensity of the government's response, and the fact that GM is so unpopular combine to make the entire debacle suspect.
While it is easy to say "put it in neutral", this is not quite a simple as it once was. With the popular serpentine shift patterns on modern cars a lot of people are surprised to find that neutral is no longer just a spot on the transition path between park and drive. To take a frequently cited example, the ES350 whose stuck accelerator killed a CHP officer and his family has a shift pattern that makes getting into neutral a bit more tricky than you suggest (particularly when you are shooting down the road at over 140 feet per second.) The driver response to these incidents may factor in to some of the problems and the fact that Toyota is the company with a problem may have politicized some of the government response, but to suggest that the solution was as simple as you claim reveals a profound ignorance of the facts at hand.
Putting into neutral isn't a necessity, any downshift in gears will increase the torque of the engine and thus decrease the maximum velocity. Even if you can't get it into neutral, getting it into 1st or 2nd will still get you slowed down to a more manageable speed.
The major failure of all automobile manufacturers is that they haven't implemented a fail safe mechanism in the event of engine failure at velocity. If they had it would literally be as simple as turning the engine off and the brakes deploying at a reasonable (IE not maximum) pressure.
The fact of the matter is that every vehicle except consumer automobiles fail safe from aeroplanes to tractor trailers (power failure leads to de-pressurizing of the air compressor and the brakes apply harder as the air level decreases). Toyota shouldn't be being hauled up, they all should be for risking civilian lives. IMO you accept certain risks when you pilot an aircraft, train, or transportation vehicle, however these are all typically far safer than the vehicles readily handed to consumers. That just is not right.
Getting it into a lower gear would mean the engine's current RPM is at a level low enough for the transmission to allow you to shift it into a lower gear. You can throw an automatic into "1" at 80mph and it won't comply.
I think it's probably because people would assume that a car with a big button that says EMERGENCY STOP or something like that is more likely to need an emergency stop button than a car that doesn't have one.
Though I guess they already have that in the form of "emergency brake". Do you suggest they implement electronics to perform these functions when the e-brake is depressed?
Considering you haven't presented air-tight evidence for your position, I think you should consider another issue.
The increased use of electronic control in automobiles means that cars now have the problem that software has had for a while; real but difficult to reproduce bugs. When the control of the car has gone away from physical devices like rods and wires, not only is it harder to find the control but any problem that exists just seems creepier. This is another that I suspect is driving the current reaction.
What bugs me is that people don't keep things in perspective. I see people all the time claim they want to go back to the way cars were (more mechanically simple, or whatever). What they seem to forget is that cars broke down a hell of a lot more often back then.
In fact, they broke down so often that it wasn't such a big deal. Now we expect them to work flawlessly and when something goes wrong it somehow seems worse.
The newer Toyotas all have push button start. Apparently, it's not very intuitive, unless you have a lot of experience building PCs with ATX power supplies, that you need to hold down the start button for 3-4 seconds to turn off the engine when it is in drive.
It's almost like Microsoft designed these cars: "What? I push Start to shut down the car/computer?"
It is coupled with RFID chips in your key. I actually find it very convenient on my Hyundai. I can walk up to the car door and open it, it unlocks automatically because the RFID key is in my pocket. I can sit down and start the car and take off without ever removing my keys from my pocket.
It's also very convenient for opening the trunk without a key as well. It has proximity sensors by the driver and passenger doors, as well as the trunk.
You can, but then it's stuck in drive/neutral. Seeing that the brakes and neutral are still operable, it should not be an issue. I don't even know why people are being advised to turn off the ignition. Shifting to neutral while braking reduces the load on the engine.
> Wouldn't shifting to neutral make the engine rev up really high? Does the computer limit the RPMs to prevent damage?
Yes, the computers do.
However, if you're going to have an accident, engine damage is pretty low on the list of concerns, even if you ignore the fact that accidents cause engine damage.
Switching the engine off (if you can do it fast enough) should slow you down faster though. If you stay in gear with no ignition, the engine drag will brake you a bit.
> Switching the engine off in drive or neutral does not appear to lock the steering.
Turning the engine off with the key need not lock the steering, but it will lock the steering if you turn the key past off to lock.
In a run-away engine scenario, are you likely to stop turning the key in a position that you don't use very often? Or, are you likely to turn it as hard as possible?
There may be an interlock that is supposed to keep the key from locking the steering while in drive or neutral, but do you trust it? (What about manual transmission cars? I know that I can lock my steering while in gear. I forget whether it requires clutch-in though.)
Most cars here in Europe are stick shift though. A stick shift car can't go berserk on you, you simply press the clutch pedal, as you do anyway when braking, and the acceleration is gone.
Hacker News is great. I had been thinking this exact same thing but then realized I was being ridiculous. However, isn't it a little scary to realize how no one is in control of us? I noticed this when I was riding on Bart. Everyone acts a particular way based on what other people expect them to act like. We all operate with so much order and yet no one is telling us to. Just thought this was cool.
The first concern of most news networks today is ratings/viewer numbers. They pick stories that will attract the average person, and thus you have Missing White Woman Syndrome. It's also well documented that certain networks (i.e. FOX News) have stuck to a schedule of talking points handed to them by various political entities.
They also love "counter-intuitive" stories, so it's easier to sell a story about how a company that seemed to be doing really well is doing something bad instead of a bad thing happening to a mediocre company that nobody likes much anyway.
Really? You really believe that? All I ever see on the news is death, tragedy and drama. If they showed us something we didn't expect we wouldn't watch.
I believe there is a distinction between acting in a way to meet expectations, and acting in some way to gain attention. You seem to be speaking of the latter in your last post, and of the former in the one prior.
I've been beginning to suspect this as well, mainly because of all the media coverage this has been getting.
If this theory is true, it's saddening because Toyota's image is completely tarnished in many people's eyes, and they won't be able to recover from this for years (if ever).
I don't buy that. The US has a big interest in the Japanese economy that goes well beyond the auto industry. I think it's just media fear mongering. If you do own a Toyota it's hard to ignore so they're going to milk it for every cent it's worth.
The real question is, just how in-control is the computer. And how much has to go wrong such that uncontrollable acceleration happens and nothing can override that acceleration such that no matter what you do the computer makes the car go into "drive" and accelerate to the max, without responding to other parts of the system like the breaks and such.
The accelerator pedal would have to be immovable, frozen in place.
The ignition switch would have to lock or otherwise malfunction
The shift lever or clutch or automatic shift lever would have to malfunction
The brakes would have to overheat (does not really happen much in a production car). It is not 100% true that brakes will always overpower the engine, but the cases in which they don't are the ones with massive 700hp engines with big 'ol turbos and stock drum brakes, and in that situation it's the owner's responsibility.
There would also have to be no runaway truck exits.
If you were creative and knew something about cars, you could pull one of many fuses from the fuse box. On cars I've serviced, the computer fuse is always in the driver kick panel, so just rip off that panel and start yanking fuses while watching the road.
It should be noted that the e-brake on fwd cars is not your best choice (though still a decent choice); while a locked up wheel provides drag, it does not apply as much drag as a rolling wheel with maximum braking (which can easily be obtained cause of ABS). Also at extreme speeds you run the risk of a spin, which puts you possibly at risk for a roll.
> The accelerator pedal would have to be immovable, frozen in place.
In recent Toyota vehicles the pedal isn't connected to the throttle--it's drive by wire. In theory the computer could continue accelerating regardless of the gas pedal's position.
I've designed systems like this. You use redundant potentiometers to sense throttle position. If they don't match within 5%, you power down.
You also toggle bits in your code. One in the main loop, another in any interrupt service routine that you are dependent upon. These bits then go to a GPIO so that the GPIO toggles. If the ISR freezes or the main loop freezes, the GPIO quits toggling. You then have an hardware, RC-Diode type circuit, that stays high if it's input toggles, but goes low, if it's input goes low and stays low, or goes high and stays high. The output of this circuit drives a relay, so that if the CPU quits toggling the GPIO, the relay clicks off. The output of that relay is a dead-man's switch that powers down the vehicle. The end result is a hardware fail-safe if the CPU misbehaves. There are other tests to confirm that the relay isn't frozen, that the RC-Diode circuit isn't frozen, etc.
So, in theory, if a high-school sophomore designed the vehicle, you might have a problem. Otherwise, I think you don't give the engineers that design these systems enough credit.
... And yet there are out of control cars without the pedal being stuck.
"Well, I have many models of Prius that got recalled, but I have a new model that didn't get recalled. This new model has an accelerator that goes wild, but only under certain conditions of cruise control. And I can repeat it over and over and over again--safely."
"This is software. It's not a bad accelerator pedal. It's very scary, but luckily for me, I can hit the brakes," he said.
This is more believable to me, as the "set-point" for cruise control is not dependent upon throttle position, but instead, is remembered by the computer. There's nothing to check against to insure believability.
More suspect, in my opinion, is that in a cruise control, you have a feedback system, based on car speed. The car speed is almost certainly redundantly sensed, so that's no worries, but the feedback loop itself could potentially go oscillatory if there were other variables introduced that hadn't been designed for. Those variables could be pretty subtle. For example, maybe the gear motor that you use to mechanically control the engine gets sourced from another distributor, and they give you a better one, that has more torque. Perhaps that throws off the stability analysis that you had done. Tons of things could change somewhere between the 500K's car and 1M's car you produced, lots of different vendors and permutations could come into play that could throw off the stability of a cruise control, I would think.
EDIT: They're probably using feedback in the control motors as well (servos), so that's a non-realistic example, but it illustrates the problem.
To be fair, Woz's account seems to be limited to the case where someone uses the cruise control above 80 MPH, far from what the control loop was optimized for. I didn't see anything in what he wrote that could possibly have any bearing on the cases being examined.
Brain-dead cruise control behavior is nothing new. My '92 Porsche 968 would cheerfully redline the engine if you disengaged the clutch with cruise active.
Still, this doesn't account for the case where the CPU and the watchdog signal is working, but the algorithm isn't. Pure software malfunctions, have happened on commercial airliners, where there are quite stringent FAA requirements regarding how the software is developed and maintained. So it wouldn't be surprising if it happens in automotive systems, where you might not have the redundancy of multiple CPUs running concurrently.
I'm not sure if automotive systems are held to any certification standards. Maybe someone working in this field could answer that?
If you were creative and knew something about cars, you could pull one of many fuses from the fuse box. On cars I've serviced, the computer fuse is always in the driver kick panel, so just rip off that panel and start yanking fuses while watching the road.
Unfortunately, this is probably considerably more dangerous than driving while on your cellphone.
Being distracted while driving is less safe than simply driving--given that your car is in good working order. If you have a stuck accelerator, you're going to crash if you can't fix it. I'll take the distraction and possible loss of reaction time over the sudden and guaranteed death of flipping a curve at 120+, personally.
Please consider the entirely of your own scenario. Bending down awkwardly and taking your eyes off the road to get at fuses under the dash could well kill you if you are in an out of control speeding car.
Most of the other actions suggested can be taken while in standard driving posture. Those will be much safer than reaching for fuses under the dash.
It's not the first solution. It's just a last-ditch effort to try. In other words, it "could well kill you". I'd much rather take the "COULD" route than doing something that will darn sure kill me, i.e. speeding uncontrollably.
The rest is in the C&D article referenced, I believe, by the author:
http://www.caranddriver.com/news/car/10q1/toyota_recall_scan...