Mach-E Hands-Free Active Drive Assist PRICING Announced, OTA Update Available in Q3 2021

[macchiaz-o]

I believe GM and Ford are doing exactly that by restricting it to mapped highways.

Agreed.

Also, by that same argument about latency, all driver should hover their feet over the brake whenever using traditional (or adaptive) cruise control.

I prepare to brake (e.g. hover foot) when the situation looks like it could be needed, and relax my leg otherwise. I'm hoping adaptive cruise control reduces how often this is necessary compared to standard cruises... But I don't know yet.

Steering correction seems more likely to require a low latency response than braking. On roads where cruise control is generally safe to use, you and the ADAS have a lot more time to sense the overall forward path while driving and can get ready to brake. Whereas you may have a lot less time to correct a deadly steering miscalculation (e.g. into a parked vehicle that is mistaken as a lane) or to safely maneuver around road debris or a large pothole.

Sponsored

 

[Nak]

Whenever not using cruise control, I do hover my RIGHT foot over brake unless I'm actively pushing go pedal or brake pedal.

You wrote "Whenever not using the cruise control." So yes, if you take your foot off the gas you're obviously trying to slow down so you should move it to the brake. (Of course it's a whole different discussion with one pedal driving.)

If you really meant when you ARE using cruise control--as your previous post suggests--that's really bad technique for several reasons. First but not foremost, you're training yourself that in an emergency you need to brake. What happens if the best option is to accelerate? You've trained yourself the answer is to brake and that's just what you'll do in an emergency. Second, and this is the big one, it's just not possible to "hover" your foot there for long. So the fact is, whether you are aware of it or not, you occasionally rest your foot on the brake pedal. This is called "riding" the brakes. What you're doing is heating the brake rotors and reducing the capability of the brakes. So your "latency"--"reaction time" is widely used in this context--may be a tiny, tiny bit less, but your stopping distance is still increased over the proper technique: right foot on the gas unless slowing down, then right foot on the brake. Don't heat up your rotors until you need them, that will cause a rather large increase in stopping distance. If you see a situation that worries you enough to hover your foot over the brake pedal, then the proper course of action is don't hover, push the damn thing and slow down.

Hovering over the brake is never correct. If you need to slow down, push the brake pedal. If you don't, keep it on the gas. That way in an emergency it is always the exact same motion you've practice a million times. If your reaction times are so slow that you need to hover your foot over the brake pedal in spite of the drawbacks, it's seriously time to consider giving up your driver's license.

 

[Nak]

Steering correction seems more likely to require a low latency response than braking.

Correct. Also, it is far easier and faster in an emergency to move your foot from the gas to the brake. It requires no finesse , you just have to hit it. With steering, you have to grab the wheel in such a way that you are set up to move it sufficiently far in the proper direction.

That brings up another point. You don't magically know where the brake pedal is without looking. You do know with both mind and muscle memory precisely how far from the accelerator pedal the brake pedal is. It requires no thought whatsoever for your foot to find it's way from one pedal to the other in an emergency. Muscle memory takes care of it and the action takes place remarkably fast in an emergency. If your foot is "hovering" you have no tactile feel for where your foot is at all. Your odds of missing of the brake pedal are substantially increased because the muscle memory you fall back on may put your foot in the wrong place. Unless of course you're riding the brakes. Then you just have to hope what brake capability you've left yourself will be enough.

 

[Nak]

*cough* Lion Air 610 *cough* Ethiopian Airlines 302 *cough*

*cough* Thanks for proving my point. *cough*

Boeing ignored standard design practice learned through tragedy with the 737 Max. It's a longer explanation than I have time for here, but the system used one AOA sensor when they should have followed standard industry practice of triple redundancy given the type of system. *cough* Ignoring lessons learned leads to death.*cough*

 

[Nak]

Tesla has a dangerously inadequate driver monitor system that can be easily defeated. Just about any weight attached to the steering wheel will be detected as a driver’s hand.

Any monitor system can be defeated. It won't be all that hard to defeat Ford's system either. Just like accelerator pedals can be abused. Care to count up the fatalities for each? There will always be dangerous drivers. We can increase the penalty--and we should--for tampering with safety systems, but that won't change the fact that people are going to die on our roads. If you really and truly want to change that, why aren't you pushing for cell phones to become inoperable above 5 mph? How many tens of thousands of people have died because of using cell phones while driving?

 

[Jimrpa]

I'm not sure what the point you're trying to make is? If you'd like we can discuss Air France 447 in depth. At the time of the crash I was flying the A330. I have a bit over 10 years experience flying the A330 internationally. Needless to say I had a very personal stake in understanding both the causes of the crash and how to avoid such an incident myself. I've studied this crash extensively and flown many times with the author of the best book on the subject. If you want to learn more, I can highly recommend his book: Understanding Air France 447

I freely admit I’m “self-loading cargo”. The only times I’ve been on a flight deck was a domestic Quantas flight to Sydney to Alice Springs and Concorde Paris JFK a couple of times. I’m only going by what friends who are Boeing and Airbus captains have told me (one characterized Airbus control inputs as “suggestions”, with the aircraft having final say over whether it would accept and act on those suggestions). A friend of mine who’s a control systems engineer at Boeing tried to give me a more complex answer having to do with control laws. I didn’t follow the entire explanation, but it boiled down to, Airbus aircraft are designed to prohibit control inputs that would cause the aircraft to become unsafe and, since the pitot tube was iced over, the Air France plane didn’t have correct information about itself and prohibited the pilots from taking appropriate action, since it thought their actions would endanger the aircraft. That was the simplistic explanation I was given. I’m sure there’s much more.

sorry to digress and I’ll fly any plane any chance I get over taking the expressway to work every day ?

 

[Jimrpa]

My opinion... If I were building personal vehicles intended for general road usage:

Hands free should not be implemented until the manufacturer can reasonably guarantee that the vehicle will not make the wrong decision at the worst time and with too much latency needed for a fully aware, hands off driver to recover from the situation.

Until then, hands on wheel and eyes on road, so that you, the aware and active driver, are ready to correct or takeover as needed.

I sincerely hope Ford's hands free driving will only be able to be enabled in the scenario I just described. If they do otherwise, that will be irresponsible, too.

Define “reasonably guarantee”? What you may consider a “reasonable guarantee”, I may consider an overburdensome reach that ensures no degree of automation ever becomes available? Consider - we have adaptive cruise control, yet I’d better be paying attention because it could disengage for some reason and my car could plow into the car ahead. I have blind spot warning, but I’d still better be checking my blind spots before changing lanes. And how about glancing at the backup camera before pulling out of the garage?

im not happy with Tesla’s way of doing it - I’d like to see SOME kind of objective performance standards established, but they should be realistic and achievable within the current state of technology.

 

[Jimrpa]

When I use ACC (and I use it frequently - big fan) I rest my foot on the floor near both the accelerator and the brake so, if the situation arises, I can respond with the appropriate input. I also have the following distance set to max. Sure, that means that, sometimes people may cut in and the system will brake, but it also gives me extra time to react if the system does do something stupid. Assistance systems don’t relieve me of my primary responsibility to operate the car safely.

 

[Nak]

Airbus captains have told me (one characterized Airbus control inputs as “suggestions”, with the aircraft having final say over whether it would accept and act on those suggestions). A friend of mine who’s a control systems engineer at Boeing tried to give me a more complex answer having to do with control laws. I didn’t follow the entire explanation, but it boiled down to, Airbus aircraft are designed to prohibit control inputs that would cause the aircraft to become unsafe and, since the pitot tube was iced over, the Air France plane didn’t have correct information about itself and prohibited the pilots from taking appropriate action, since it thought their actions would endanger the aircraft. That was the simplistic explanation I was given. I’m sure there’s much more.

This is a fun subject for me, but to fully explain it would at the least take a couple of beers, LOL. The Airbus flight control system is different from previous control systems. Under normal law--there are 3 laws and two sub laws in the second law--moving the stick fore or aft demands a load factor, not elevator movement. With a 747 or a piper cub if I pull back on the yoke or stick I am demanding the elevator to move. With an Airbus in normal law I'm demanding some load factor. If I let go of the stick I'm demanding 1 g, so the nose should stay where it is: level, 5 degrees nose up, whatever. If I pull back a bit I am now demanding say 1.1 G and the nose will track up. If I move the stick left I am not moving the ailerons, I'm demanding a certain roll rate. There are limits as to what the controls will do. at some nose up angle, pulling further on the stick will do nothing. At a given bank angle the plane will stop rolling even if I keep the stick hard over. That's a pretty simplistic explanation, but it gives a general idea of how the controls function.

As far as AF 447, there were a lot of contributing factors. Two of the biggies are we didn't train a lot for unreliable airspeed and AOA indications and we didn't train a lot in recovering from deep stalls without power. It's not that the controls didn't let the AF 447 pilots recover, it's that the pilots failed to recognize the deep stall due to iced over sensors and the control safety systems didn't save them because of said ice. Further complicating the issue was the application of high power simply made the stall worse due to the low mounted engines pitching the nose up. Prior to this accident we really didn't do any training demonstrating the difficulty you could have recovering from a stall if you added power. Oh, there were a few instructors that demoed it, but it certainly wasn't a big part of the syllabus. Mostly it was ingrained in our souls to max out the power in the event of a stall. Again, this is a HIGHLY simplified version of events. If you're at all interested, the book I linked to is fascinating and informative.

With the 737 Max crashes, the safety system did in fact actively fight the pilots attempt to save the aircraft. The system existed because the placement of the engines on that aircraft could make it exceptionally difficult to pitch the nose over in the event of a stall. So, yes, the system was absolutely needed. However, Boeing chose to save a few bucks rather than follow industry standard practices. Normally, a system like that would use input from three sensors. If one disagreed with the other two it's data would be disregarded. If all three disagreed then the system would deactivate and warn the pilot he/she was on their own. The Boeing system relied on input from just one AOA sensor. If that sensor failed--a not too uncommon event--the system could activate and drive the nose down. That's exactly what happened in the two crashes. The pilots tried to raise the nose while the safety system forced the nose down. Tragically, the pilots didn't even know about the system because boeing chose to hide the existence of the system from pilots. They did this because they didn't want the FAA to require training in the system, which would add cost to the purchase of the airplanes.

 

[ClaudeMach-E]

Any monitor system can be defeated. It won't be all that hard to defeat Ford's system either. Just like accelerator pedals can be abused. Care to count up the fatalities for each? There will always be dangerous drivers. We can increase the penalty--and we should--for tampering with safety systems, but that won't change the fact that people are going to die on our roads. If you really and truly want to change that, why aren't you pushing for cell phones to become inoperable above 5 mph? How many tens of thousands of people have died because of using cell phones while driving?

The only downside of this would be that you can't use it for Nav.

 

[macchiaz-o]

I’d like to see SOME kind of objective performance standards established, but they should be realistic and achievable within the current state of technology.

This exists... Here is an example:

https://www.mobileye.com/responsibility-sensitive-safety/

 

[DBC]

One reason Super Cruise continues to be rated much higher then Tesla’s system by CR and others is the vastly superior ratings it gets for Driver Engagement, which includes a Driver Monitor system appropriate for the system. Something Tesla does not have.

Very true. The other great advantage of SuperCruise, and what Ford's Hands-Free system will have, is digital mapping. With digital mapping the vehicle can recognize when there is something unexpected in the road and alert the driver to possible danger.

In some ways it's unfair to compare these hands-free systems to what Tesla offers with FSD since FSD is not a hands-free system.

 

[Jimrpa]

This is a fun subject for me, but to fully explain it would at the least take a couple of beers, LOL. The Airbus flight control system is different from previous control systems. Under normal law--there are 3 laws and two sub laws in the second law--moving the stick fore or aft demands a load factor, not elevator movement. With a 747 or a piper cub if I pull back on the yoke or stick I am demanding the elevator to move. With an Airbus in normal law I'm demanding some load factor. If I let go of the stick I'm demanding 1 g, so the nose should stay where it is: level, 5 degrees nose up, whatever. If I pull back a bit I am now demanding say 1.1 G and the nose will track up. If I move the stick left I am not moving the ailerons, I'm demanding a certain roll rate. There are limits as to what the controls will do. at some nose up angle, pulling further on the stick will do nothing. At a given bank angle the plane will stop rolling even if I keep the stick hard over. That's a pretty simplistic explanation, but it gives a general idea of how the controls function.

As far as AF 447, there were a lot of contributing factors. Two of the biggies are we didn't train a lot for unreliable airspeed and AOA indications and we didn't train a lot in recovering from deep stalls without power. It's not that the controls didn't let the AF 447 pilots recover, it's that the pilots failed to recognize the deep stall due to iced over sensors and the control safety systems didn't save them because of said ice. Further complicating the issue was the application of high power simply made the stall worse due to the low mounted engines pitching the nose up. Prior to this accident we really didn't do any training demonstrating the difficulty you could have recovering from a stall if you added power. Oh, there were a few instructors that demoed it, but it certainly wasn't a big part of the syllabus. Mostly it was ingrained in our souls to max out the power in the event of a stall. Again, this is a HIGHLY simplified version of events. If you're at all interested, the book I linked to is fascinating and informative.

With the 737 Max crashes, the safety system did in fact actively fight the pilots attempt to save the aircraft. The system existed because the placement of the engines on that aircraft could make it exceptionally difficult to pitch the nose over in the event of a stall. So, yes, the system was absolutely needed. However, Boeing chose to save a few bucks rather than follow industry standard practices. Normally, a system like that would use input from three sensors. If one disagreed with the other two it's data would be disregarded. If all three disagreed then the system would deactivate and warn the pilot he/she was on their own. The Boeing system relied on input from just one AOA sensor. If that sensor failed--a not too uncommon event--the system could activate and drive the nose down. That's exactly what happened in the two crashes. The pilots tried to raise the nose while the safety system forced the nose down. Tragically, the pilots didn't even know about the system because boeing chose to hide the existence of the system from pilots. They did this because they didn't want the FAA to require training in the system, which would add cost to the purchase of the airplanes.

Thanks! I really appreciate the detailed (although I understand still simplified) explanations!

 

[eastern refugee]

Question. I have the ford lane keeping system on my current car. All it does is vibrate the steering wheel if you go over the line. Will this be the same with the MME???

 

[JTK44]

Question. I have the ford lane keeping system on my current car. All it does is vibrate the steering wheel if you go over the line. Will this be the same with the MME???

Yes and No:

If you are on cruse control you also have lane centering:

I already have this on my 2018 Edge Sport: though technically not "hands free" in reality for short periods of time, up to 30 seconds, it is: while in adaptive cruise control you can take your hands off the wheel, reach for a bottle of water for example, open it, take a drink and the car will remain in the lane. The adaptive cruise control also has distance control so braking and acceleration is also done automatically.

Later on, (mid 2021?) if you have purchased the hardware, there will be OTA update to make the adaptive cruise control legally "hands free" on certain roads.

Hope this helps.

Sponsored