IMHO, this is a common problem in the automotive and software areas. In 1963, my parents purchased a Chevrolet passenger car with a 6 cylinder engine. It was not "fly by wire", but had an annoying problem. When you attempted to hold a steady speed, the vehicle would constantly "buck" - surge forward and back. We had the new vehicle back to the dealer dozens of times. Always, the "fix" was just a few days away. Finally a "special part" was ordered to fix the problem.. It was installed, and the problem was just as bad as ever. This time however, any acceleration produced dense black smoke from the exhaust. Additionally, the gas mileage went down by about 8 miles a gallon (that's a LOT). A high school student at the time - i finally was "allowed" to look at the problem. I found that the "magic part" consisted of a main metering jet in the carburator that was 7 steps rich. I replaced it with the standard jet, and mileage returned to normal - but the bucking continued. After a few weeks of experimenting, i found that the bucking was caused by springs in the distributor that were too weak. As engine speed increases, it is necessary to advance the spark timing to burn the fuel before it is exhausted. In the 1963 era, this was done with "bob weights" in the distributor body, which were restrained by small springs. As the speed increased, centripital acceleration forced the weights outward against spring tension, thereby advancing the spark. The springs installed by the factory were considerably weaker than they should have been. NOBODY at the factory had ever put one of these distributors on a distributor machine (common equipment in a "speed shop") to verify that the advance curve was appropriate. The problem persisted until 1970. So the vast automotive industry could not solve a problem that a high-school junior solved in 3 or 4 weeks in 7 years. Don't hold your breath for Toyota to come up with a solution. Cheers.
This is true to a degree. The brakes must be applied firmly to bring the vehicle to a complete stop and stall the engine without delay. If the brakes are "ridden" to merely reduce or control the vehicle speed, the brakes will fairly quickly overheat and fail due to loss of "coefficient of friction" of the braking material. Nearly all friction material becomes more slippery as it heats up. The exception is "metallic" linings used for police and racing use. Metallic linings become less slippery as they heat up. These linings are not stocked by common auto parts stores, you have to obtain them through special sources. I have a vehicle equipped with these - they wear like iron - never wear out. The down side is that you almost have to be an elephant to stop the car when they are cold - particularly in the winter. Without power assist (which my equipped car does not have), i suspect that a woman would be incapable of stopping the car with cold brakes. I am its only driver - so that doesn't matter.
The other point is that a sudden power application - when you are not expecting it - at a critical time, could cause loss of control. An example would be suddenly decending a steep exit ramp from a freeway on wet pavement. You saw the exit sign at the last moment - not knowing the area - and exited at a somewhat higher than prudent speed, but still under control. Then you discover that the exit ramp has a decreacing radius curve (poor design). If the vehicle suddenly goes to full throttle at this point - i suspect most drivers (including myself) would be hard pressed to maintain control. Firm application of the brakes on wet pavement could cause the vehicle to continue straight on - into a retaining wall or other obsticle, rather than tracking a curve. Cheers.
chanj0, driver training is certainly not uniform throughout the U.S., and I don't know the statistics, but many states require such training before a young person can get a driver's license.
One problem may be that except in rare cases, the focus of such training is learning how to drive a vehicle under normal traffic conditions rather than abnormal conditions, and of course learning the traffic laws. There is a driving school in my city (all 3 of my children learned there) which teaches collision avoidance and driving in adverse conditions, like how to recover from a rear wheel skid. They have specially modified vehicles in which the instructor can push a button and force the car into a skid.
Perhaps more driver education should teach collision avoidance and learning how to deal with emergencies and malfunctions (a tire blowout on the highway comes to mind), but I do not believe this is only a U.S. problem.
I recently took out a new Toyota Sienna van for a test drive at my local Toyota dealer.
I was taken aback by my inability to maintain a steady speed. It seemed the throttle control was noisy, in that the vehicle would either surge ahead or hold back. When I turned on the cruise control the vehicle maintained a smooth steady speed. The salesperson with me simply denied there was a problem.
It appears that Toyota still has an accelerator sensor problem - perhaps whiskers caused by lead free solder or noisy pots.
It's unbelievable that after all this time and publicity, Toyota seems unable to solve the problem.
@zeeglen: AFAIK, lawyers don't want anyone who actually knows anything about the underlying issues on the jury. Each side can challenge juror selection to reject jurors they think might be biased against their side of the story and it's part of the game that is played before trial. In addition, jury duty is one of those things most folks try to duck if they have an excuse.
The worst examples are probably medical malpractice trials, and why malpractice insurers do their best to settle out of court, since they'll have to pay assessed damages.
I've read the follow-up article (Toyota Case: Single Bit Flip That Killed). Did the analysis - based on simulations performed - provide an estimate on the probability that such an event was the cause in this case? Or more generally, how frequently such task death-related unintended accelerations might be expected to occur overall? Showing that something could happen is of course not the same thing as proving that it actually did...
Blog Doing Math in FPGAs Tom Burke 24 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...