Slashdot Mirror
Driver Trapped In Speeding Car At 125 Mph
Hugh Pickens writes writes "The Guardian reports that Frank Lecerf was driving his Renault Laguna in Northern France when the car's speed jammed at 60mph. Then each time he tried to brake, the car accelerated, eventually reaching 125mph and sticking there. While uncontrollably speeding through the fast lane as other cars swerved out of his way, he managed to call emergency services who immediately dispatched a platoon of police cars. Realizing Lecerf had no choice but to keep racing along until his fuel ran out, they escorted him at high speed across almost 125 miles of French motorway, past Calais and Dunkirk, and over the Belgian border. After about an hour, Lecerf's tank spluttered empty and he managed to swerve into a ditch in Alveringem in Belgium, about 125 miles from his home. 'My life flashed before me,' says Lecerf. 'I just wanted it to stop.' His lawyer says Lecerf will file a legal complaint over 'endangerment of a person's life.'"
The article mentioned that the car was adapted with controls for people with disabilities (probably hand controls for the accelerator and brakes).
Not only would this kind of modification introduce another point of failure in the system, the hand controls were probably not debugged and tested to the same degree as the traditional ones.
The car in question was an automatic, so no neutral.
Since when? The N in PRNDL stands for neutral.
Have gnu, will travel.
Details Missing from the quoted article is this bit:
The Frenchman, who suffers from epilepsy and drives a specially-modified car that has controls on the steering wheel to operate the throttle and brake, has filed a legal complaint against the vehicle's manufacturer.
Source here.
Unless Renault did these modifications for him, I doubt he has a chance in hell of winning his suit.
I've never seen a car you couldn't force into Neutral even under heavy acceleration.
Sig Battery depleted. Reverting to safe mode.
The car was modified with disabled-driver controls. It's unclear what options he had available to regain control.
Nope, there's no conventional key. The ignition is entirely computerized. The "key" is a card you stick in a slot and you start (and stop) the engine by pressing a button. Here's the car's dashboard. The thing with the red fob is the "key".
http://www.autotesty.com.pl/fotki/renault/laguna3_gt_20dci_177km/renault_laguna3_20dci_177km_gt_15.jpg
upon the advice of my lawyer, i have no sig at this time
"The brake pedal causing the car to accelerate seems highly unlikely without some major hacking,"
TFA states that the driver was "disabled", so presumably his car was equipped with hand controls. Yes, that's a major hack.
Not the case. I've got one of the fancy new keyless ignition vehicles, and I've tested this.
With the engine running, and with forward motion, three (maybe four) presses in quick succession or pressing and holding the the ignition switch for 2-3 seconds will kill the engine. You need to shift into park and press the brake to start again.
I thought it was interesting that there were two paths that would do this, both of which are a reasonably likely response in a panic situation -- tap the button a zillion times, or try to mash it into the engine compartment.
2009 Nissan Cube, if you care. Or if you don't.
grnbrg.
The article says that while he was unhurt, he did suffer two epileptic seizures. Imagine going through that, twice, at 125mph.
"Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
That's a phenomena specific to diesel engines. Diesel's don't use a spark to ignite the fuel mixture like gasoline engines do, they use the heat from piston compression. Thus, so long as vacuum pressure and fuel supply is maintained, a diesel can continue running without electrical power.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
The same thing happened to a driver in Oz awhile back.
Modern cars contain numerous independent systems which communicate using an internal bus. If one of those systems fails in a way such that it floods the bus with packets, no other system can get a message through.
If you happen to be on cruise-control at that time, there may be no way out of it. The signals from the steering-wheel computer [buttons] or brake won't get to the computer.
Here's some info that came from the Oz incident:
1) Modern cars don't have a direct key-switch - the computer starts and stops the engine. Turning the engine off is not guaranteed to stop the car. (This was tried in the Oz case.)
2) Some cars do not have direct shift capability; ie - it's "shift by wire": the shifter tells the computer what gear to be in. (Admittedly, I've never seen one, don't know if it's true.)
3) A driver is not strong enough to stop the car against the engine, especially since the engine can down-shift to get more power. Some "mythbusters"-style experimenters disagree with this statement, but their conclusions don't track with these incidents. Also, consider that the driver may be female, young, elderly, out-of-shape and otherwise incapable of braking with the full force of an "average" human driver.
I used to write the software for aircraft instruments, and one thing the hardware should always do is "fail safe". If you have a remote sensor such as a switch, in this case the brake light switch, you always have some mechanism to determine whether the wire is broken. If the remote sensor is on a communication bus, you always look for a "heartbeat" packet saying that the remote sensor is working properly. If something fails, the default action is to go out of cruise-control.
Car software is not safety certified (as aircraft systems are), and perhaps they should be. This will become more important as cars get smarter, and will be critical for self-driving cars.
This sort of event is convincing me even more that I want three pedals in my car. Press the clutch and your problem is solved. No electronics can fail because the clutch in a manual transmission car is controlled by you, with your foot, mechanically.
The article mentions he's epileptic and the car is modified for disabled drivers. I'm guessing its got an automatic transmission. When it mentioned he had two seizures during the situation, I'm actually wondering if he was having a seizure and the whole time and depressing the accelerator without even realizing it.
Found another story about this indecent. It indicates that "...which the Weeknotes was customized in light of Lecerf's epilepsy, with the gas and brake controls moved to the steering wheel)." from USA today http://www.usatoday.com/story/money/cars/2013/02/14/newser-wild-ride-france-drive/1919139/. The Manual (Found at http://www.globalcars.com.au/site/uploads/Renault_Laguna_ENG.pdf) does indicate that the automated version has an Neutral gear. It also indicates that shifting to neutral and then pressing the ignition should stop the engine. There is also a park for the automatics... I suppose if they moved the gas and brake they may have moved the gear-shift (thought it's not stated). If that was the case and the disability controls were malfunctioning (likely) then your screwed. Of course you could throw the keys out the window From the manual, page 2.5 titled "Starting/STOPPING THE ENGINE (continued)". "If the card is no longer in the passenger compartment when you try to switch the engine off, the message “card absentlong press” appears on the instrument panel: press button 1 for longer than two seconds"
Yeah yeah, it "happened" to him 3 times and his driving license was cancelled since 2004 over speeding tickets. But sure, this is the car manufacturer fault if your modified car (gas and brake operated from the steering wheel) has a strange behavior.
Sig (appended to the end of comments you post, 120 chars)
Uh, guys... we can probably stop trying to troubleshoot with all the obvious stuff like turning the car off, shifting to neutral, parking brake, etc.
From the article:
A Renault technician had been on the phone with police throughout the chase trying to help but couldn't come up with a solution.
Note that seizures range from symptoms as minor as deja vu or a brief lapse in awareness (that you might not even know you had) to full-body thrashing and flailing with the potential for both physical and mental injury.
The article doesn't specify what he went through. My guess is that it was toward the middle of the spectrum: too small, and he might not have even been aware he had a seizure; too large, and he probably wouldn't have survived the ordeal.
(Other people have already brought up the possibility that the seizures were responsible for the problem, so I'll leave it at that.)
Actually that's not entirely true. I do a lot of high speed driving (on closed road courses in racecars) and have been in many similar situations. A driver can definitely come to a controlled stop from ~60 (or even a bit higher, although 125 is stretching it) by scraping into most common metal or concrete barriers. It's certainly feasible if you're calm, do it intentionally, understand what's going on, and steer appropriately. However, in this particular scenario there were probably other mitigating factors, even if 125mph was an acceptable speed against the barriers in question:
A) If the accelerator signal was actually stuck on, and the brake signal dysfunctional (as appears to be the case according to TFA), you could shed some speed that way, but eventually you'd get slow enough that the car's torque would overcome the remaining friction and you'd just be grinding along at a constant speed. If that speed were ~25mph or less, it might've been worth trying to jump out of the car at that point (if the barrier was on the passenger side), but if it were higher you'd just be making your situation worse. You could try to turn into it a bit harder at that point to increase friction, and I'd guess the first fallout would be blown front tires, which definitely puts further vehicle control in doubt and the engine's still trying to go all out at that point...
B) If the driver wasn't in his right frame of mind and a driver of decent skill (for normal road drivers), he could've freaked out in such a situation and made matters worse by bouncing off or turning in to sharp, either of which would lead to the car spinning and possibly flipping.
But really, I still think there's something wrong with the facts in the article, they just don't jive. I can understand this isn't a Toyota (totally the fault of deranged consumers) situation, due to the fact that the car was modified with some assistive gas/brake inputs for the disabled that may have gone crazy and actually caused the gas/brake behavior described, but...
What I can't understand is how the owner, the police, and the supposed Renault Engineer (likely it was some local mechanic the cops called, and likely there was miscommunication involved? I don't know...) couldn't figure out how to turn off the engine. A car can definitely be safely stopped on a regular road with the engine off, although it will take a while to slow down without brakes. I suspect one fact that must be missing from TFA is that the driver was the disabled person the system was installed for and couldn't operate the normal controls of the car, but there still should've been a way to shut that car down. Either way, the mfg/installer of the assistive device is probably the primary culprit here, not Renault or all drive-by-wire cars in general.
Airplane manufactures do it for their fly by wire systems so that the pilot always stays in control, even when the system would rather beg to differ on the matter.
If I recall correctly, this corresponds to somewhat of a philosophic difference between Airbus and Boeing. From what I read a few years ago, Airbus absolutely limits what the pilot can do - he/she can not make the plane do something the computer doesn't approve of. Boeing, assumes the pilot knows best, and allows the pilot to 'override' the system (do things with the controls that seem unwise to the computer). Boeing's POV is that the computer may be wrong, and/or the situation may not be one the computer is ready for.
I did a bit of Googling 'airbus and boeing philosophy' and found many interesting links. Boeing still insists on classic controls, which require the pilot to act like a pilot instead of automating everything (even though it's automated). And the autopilot automatically disengages as soon as the pilot takes the controls. Airbus philosophy is to automate everything to avoid human error - but slashdotters generally know that computers are only as smart as their programs, and are _never_ as adaptable as their programmers.
It's easier to be a result of the past, but more fun to be a cause of the future! http://www.spacefinancegroup.com/
I'm from Europe and in all the cars I've owned or driven, which were mostly manual, turning off the key never engages the steering lock. The steering lock is engaged only when you remove the key.
That said, the car being a Renault Laguna and presumably a rather recent one it most likely comes with Renault's keycard ignition. Basically the key looks like a fatter credit card, and it goes into a slot. The car starts by pressing a Start button. In that case, the card is blocked inside the slot while the car is in motion, so it can't be removed at all.