Sudden Unexpected Acceleration today

[Cirrus MS100D]

I thought in the Toyotas the floor mat failed and pressed the accelerator to the floor? Then people would ride the brakes until they melted. They fixed it by disabling the accelerator when the brake is applied (as Tesla and probably every other manufacturer now does).

This is correct. There never was anything “wrong” with the Toyotas, except for poor floor mat / pedal design and people not realizing they’d scoot the all weather floor mat on top of the gas pedal, effectively “flooring it” and simultaneously making it more difficult to press the brake pedal aka “pedal entrapment.”

Aside from one tinfoil hat expert, even friggin NASA was brought in and couldn’t conceive of any explanation other than “pedal entrapment” aka “the gas pedal was stuck to the floor.” I completely agree that this type of acceleration is clearly unintended, but there is no mystical software or sensor or unicorn explanation. Even for a Tesla.

 

[mongo]

This is correct. There never was anything “wrong” with the Toyotas, except for poor floor mat / pedal design and people not realizing they’d scoot the all weather floor mat on top of the gas pedal, effectively “flooring it” and simultaneously making it more difficult to press the brake pedal aka “pedal entrapment.”

Aside from one tinfoil hat expert, even friggin NASA was brought in and couldn’t conceive of any explanation other than “pedal entrapment” aka “the gas pedal was stuck to the floor.” I completely agree that this type of acceleration is clearly unintended, but there is no mystical software or sensor or unicorn explanation. Even for a Tesla.

from the testimony in the case, not all the software in the engine controller followed established safety/ redundancy design practices. That left open the possibility that the hw+sw combination along with external factors (radiation flipping bits) could lead to an uncontrolled state.

 

[Knightshade]

This is correct. There never was anything “wrong” with the Toyotas, except for poor floor mat / pedal design and people not realizing they’d scoot the all weather floor mat on top of the gas pedal, effectively “flooring it” and simultaneously making it more difficult to press the brake pedal aka “pedal entrapment.”

Aside from one tinfoil hat expert, even friggin NASA was brought in and couldn’t conceive of any explanation other than “pedal entrapment” aka “the gas pedal was stuck to the floor.” I completely agree that this type of acceleration is clearly unintended, but there is no mystical software or sensor or unicorn explanation. Even for a Tesla.

And sadly the driver in the Lexus crash also apparently didn't know how neutral worked... or how to just turn the car off.

 

[Cirrus MS100D]

And sadly the driver in the Lexus crash also apparently didn't know how neutral worked... or how to just turn the car off.

To be fair, I bet most Tesla owners don’t know that the pressing and holding the park button on the stalk is the e-brake for a similar situation :-/

 

[kavyboy]

I could actually see driver error like this being worse with a Tesla, or EVs in general, due to

...
I would add lack of engine noise. You press a gas pedal to the floor and you hear and feel the engine. You don't get that feedback in an EV, so it reinforces the assumption that your foot is on the brake, not the accelerator. I bet if the car detected a sudden mashing of the accelerator after the car's been moving at less than 5 MPH for a few seconds and played a fake engine sound, it would prevent a few of these incidents (as stupid as that sounds, I know).

 

[Daniel in SD]

And sadly the driver in the Lexus crash also apparently didn't know how neutral worked... or how to just turn the car off.

It was a loaner car. It shows the risks of non-standard UI.
You’ve got to press the power button for 3 seconds or something to turn it off.
I just realized that I have no idea how to turn the Model 3 off... is there a way to do it?

 

[JulienW]

To be fair, I bet most Tesla owners don’t know that the pressing and holding the park button on the stalk is the e-brake for a similar situation :-/

Will this will work while you have the accelerator to the floor. Have you tried it?

 

[Daniel in SD]

To be fair, I bet most Tesla owners don’t know that the pressing and holding the park button on the stalk is the e-brake for a similar situation :-/

What happens when you engage the parking brake at speed? Do the rear wheels lock up? Is it possible to modulate?
I’m the Lexus crash the driver did use the brakes but they burnt out because he rode them to maintain speed instead of coming to a stop.

 

[lolder]

In most if not all current cars, pressing on the brake reduces all accelerator pedal input to zero. Full brake pedal application will always overcome full motor power. If you read the article "Braking Bad" that I linked previously, you will see that pedal confusion is greatest with new drivers and old drivers. I believe it may also be a precursor of dementia. Implementing the brake pedal application to shift into drive interlock decades ago completely eliminated the start-up "unintended acceleration". Some references state that Toyota accelerators had a problem. That was a slightly sticky pedal, not one that would increase acceleration on it's own.

 

[Oldschool496]

To be fair, I bet most Tesla owners don’t know that the pressing and holding the park button on the stalk is the e-brake for a similar situation :-/

I've done this by accident for a very brief moment by accident going 50+ nothing happened, I was thankful. It must take over one second for them to engage. Does anyone know how long you need to press to activate emergency braking and are there special conditions, speed, hold it on to keep it activated, if not how to release it? etc?

 

[JulienW]

In most if not all current cars, pressing on the brake reduces all accelerator pedal input to zero....

...and in an EV (not just Tesla) pressing the brake disengages power to the motor(s). So we have nothing to worry about.

 

[Knightshade]

It was a loaner car. It shows the risks of non-standard UI.
You’ve got to press the power button for 3 seconds or something to turn it off.

That's been the standard interface for all toyota/lexus push button start vehicles for over 15 years now.

Not only that- HIS car that he had dropped off for service was a 2006 IS250.

Which uses the same push button start/stop as the 2009 ES350 he was given as a loaner.

So no, nothing non-standard there at all from his normal car.


And regardless of that- the way you shift such a car into neutral hasn't changed in many, many, many, decades.

Move the shifter on the floor up one notch from D to N.

Done.


It's awful about the folks who died in the crash- but the driver had multiple, simple, decades-known, ways to deal with the accelerator getting stuck under the floor mat that all worked the same way on his own car as on the loaner car he was in.

 

[jelloslug]

We'll see.

In the meantime no reason to insult or jump to conclusions.

99.997% of the time, "unintended acceleration" is the result of pushing the wrong pedal. Disagree all you want, but it's reality.

 

[GregRF]

Oh, perhaps. I was half-listening to a Radiolab episode called "Bit Flip" I think, I might even have the manufacturer wrong.

Manufacturer was right, but that podcast did a bad job of exploring the whole case. A much better podcast on it was the Revisionist History podcast Blame Game.

 

[Cirrus MS100D]

For those who haven’t done it in your Tesla, FFS, go out and do it where it’s safe! It takes about a second or two to engage the parking brake, which slowly brings the vehicle to a stop. It doesn’t cause the car to skid out of control and it should obviously only be necessary in an emergency where the brake pedal isn’t working for some reason (e.g. your ex triggered a remote evaporation of your brake fluid)

Here’s an old example, but virtually nothing different with the current S/3/X

 

[lolder]

Pushing for several seconds on the "P" button brings the car to a maximum effort violent stop.

 

[lairdb]

Just for laughs, I'll throw out an alternative theory: not pedal mis-application, but gearshift misapplication.

I've had several close calls when I move the shift lever to switch from R to D or vice versa, and (for one reason or another) the vehicle ignores it. (Maybe I wasn't sufficiently stopped. Maybe I was stopped, but didn't have my foot on the brake. Maybe etc. etc. several possible reasons the logic could arrive at "ignore that control input".)

Since the tactile feedback suggests to me that it worked, and there's no audible feedback that it didn't, and the only visual feedback is the D/P/R label on the display that I'm not looking at -- I then (briefly) travel in an unexpected direction.

There are some things I love about soft controls and firm controls over hard controls -- but the UX designers need to be careful about feedback. Sound, haptic, exception feedback, etc.

 

[ALSETJC]

I believe some of these instances can also be attributed to putting the car into regular cruise control by accident or forgetting that it is engaged .

 

[Dan203]

Just for laughs, I'll throw out an alternative theory: not pedal mis-application, but gearshift misapplication.

I've had several close calls when I move the shift lever to switch from R to D or vice versa, and (for one reason or another) the vehicle ignores it. (Maybe I wasn't sufficiently stopped. Maybe I was stopped, but didn't have my foot on the brake. Maybe etc. etc. several possible reasons the logic could arrive at "ignore that control input".)

Since the tactile feedback suggests to me that it worked, and there's no audible feedback that it didn't, and the only visual feedback is the D/P/R label on the display that I'm not looking at -- I then (briefly) travel in an unexpected direction.

There are some things I love about soft controls and firm controls over hard controls -- but the UX designers need to be careful about feedback. Sound, haptic, exception feedback, etc.

There are videos showing that you can actually switch from R to D, or vice versa, at low speeds and it'll actually take. So I don't think this issue would be caused by any software in the vehicle ignoring the input. Perhaps there is something wrong with your physical shift control. If this happens a lot you should take it in and have a tech check it out.

 

[Dan203]

I believe some of these instances can also be attributed to putting the car into regular cruise control by accident or forgetting that it is engaged .

As soon as you hit the brake any type of cruise control should automatically disengage. So if they were truly pressing the brake then that wouldn't be an issue.