ACC could be smoother

[dbsb3233]

Done about 2000 road trip miles in the Gravity over the last 2 weeks. The pros of the Gravity really shine on road trips. But there's still some things that annoy. One of them is ACC (my Dream Drive version is just the standard version, although I'm upgrading to Premium in a few weeks). It's not as smooth as I've experienced in my previous EVs. For instance, when on a modest incline, I can feel the car doing little surges to keep the speed the same. It's subtle, but noticeable. It's a bit herky-jerky rather than smooth (by comparison our Porsche Macan is smooth).

When a car cuts into the gap, it still slows pretty hard rather than gradual. I know if the gap is too small it has to slow hard no matter what, but even when the gap allows for it, it's still not as smooth and gradual as hoped for. Again, our Macan handles that more smoothly.

I also turned curve slowing off. It's fine on the minimal curves (slows by about 2 MPH), but when the curve is just a little but tighter it slows too much and too harshly (again, not very smoothly). While the idea of curve slowing is nice, I wish it was adjustable where we could select MILD vs STRONG.

The biggest annoyance is still when canceling out of ACC. I keep Regen on Standard. When canceling out of ACC, it's like dropping an anchor. I try to match the speed with the (stiff) accelerator pedal first, but usually you end up going a little light (the anchor drop) or a little heavy (speeding up just before needing to slow down). There's gotta be a better way for the car to help make that transition smoother. Even if you turn Regen off, cancelling ACC slows hard (which makes no sense if regen is supposedly off).

 

[FirstLucidFL]

The biggest annoyance is still when canceling out of ACC. I keep Regen on Standard. When canceling out of ACC, it's like dropping an anchor. I try to match the speed with the (stiff) accelerator pedal first, but usually you end up going a little light (the anchor drop) or a little heavy (speeding up just before needing to slow down). There's gotta be a better way for the car to help make that transition smoother. Even if you turn Regen off, cancelling ACC slows hard (which makes no sense if regen is supposedly off).

Agreed, it’s very annoying. I fix I have found is before you cancel, press the accelerator so you go at least 1mph over what the speed it set to, then hit the X button.

 

[knewman]

Interesting, have you driven an Air? I'd be curious how you feel it compares to ACC in the Air. I've found the ACC in the Air to be the smoothest of any vehicle I've driven. I'd go so far to say it's one of the things I feel they've done a superb job with.

I've always thought dropping out of ACC and trying to match the power input is fun. Kinda feels like a nerf version of rev matching

 

[exxxviii]

The biggest annoyance is still when canceling out of ACC. I keep Regen on Standard. When canceling out of ACC, it's like dropping an anchor. I try to match the speed with the (stiff) accelerator pedal first, but usually you end up going a little light (the anchor drop) or a little heavy (speeding up just before needing to slow down). There's gotta be a better way for the car to help make that transition smoother. Even if you turn Regen off, cancelling ACC slows hard (which makes no sense if regen is supposedly off).

This is the exact same with every BEV I have driven. It seems that nobody has figured out a smooth way to handle cruise disengagement when OPD regen is active.

 

[RandomDesign]

Interesting, I've found the ACC aspect of the Gravity to be very smooth on most everything but braking when traffic slows/stops. It's pretty aggressive with how late it slows and stops. I've never encountered the speed surges you're describing in 6k miles with a lot of ACC highway use. (Adding that I have DD2 Pro)

Cancelling out of ACC with slowdowns has been pretty consistent with every EV I've driven, you just have to be ready for it.

 

[dbsb3233]

Interesting, have you driven an Air? I'd be curious how you feel it compares to ACC in the Air. I've found the ACC in the Air to be the smoothest of any vehicle I've driven. I'd go so far to say it's one of the things I feel they've done a superb job with.

I've always thought dropping out of ACC and trying to match the power input is fun. Kinda feels like a nerf version of rev matching

I haven't driver an Air, just the Gravity. As well as our Macan EV and our previous Mach-E. Both of those were way better dropping out of ACC smoothly. You still have to do the "try and match the right pedal speed first" thing, but it was easier on both of those. 2 big reasons that I can think of...

(1) The regen is way stronger on the Gravity, so when you do "miss" the speed match, it's not nearly as much like dropping an anchor. It's more gentle.

(2) The accelerator pedal in the Gravity is way stiffer than the Macan or Mach-E. Particularly when regen is on. It's easier to dial in an exact speed when the pedal is less stiff. Also feels like less range of play in the Gravity pedal (perhaps because the top speed is so high?).

The stiff pedal is puzzling to me because it's very different with regen on vs off. If you turn regen off during a nice steady drive (like 45 MPH) and keep the pedal pressed the same depth, the car speeds up like 10 MPH. And vise versa. That makes no sense to me. Regen should only have any effect when lifting up on the pedal. Neither our Macan or Mach-E did that. If I turn regen on or off when cruising at 45 MPH (or any speed), the speed doesn't change and the pedal stiffness doesn't change at all. Stays nice and gentle.

 

[dbsb3233]

Interesting, I've found the ACC aspect of the Gravity to be very smooth on most everything but braking when traffic slows/stops. It's pretty aggressive with how late it slows and stops. I've never encountered the speed surges you're describing in 6k miles with a lot of ACC highway use. (Adding that I have DD2 Pro)

I wonder if it's somehow different with DD2 standard? I figured it would do it the same regardless of which DD2 level, but who knows. I will be upgrading to DD2 Premium in a few weeks so hopefully that will improve things.

It's not terrible or anything, but it's just not as smooth as our other EVs have been. Although I really only notice it in the city when there's more traffic on the road, so maybe it's just picking up a car in the next lane and hestiating for a split second. I know that doesn't make any sense but it kinda feels like that. Little hestitations instead of just smoothly adding more power to maintain speed on a slight incline.

I don't notice anything like that on the open road though, with no traffic nearby.

 

[RandomDesign]

I wonder if it's somehow different with DD2 standard? I figured it would do it the same regardless of which DD2 level, but who knows. I will be upgrading to DD2 Premium in a few weeks so hopefully that will improve things.

It's not terrible or anything, but it's just not as smooth as our other EVs have been. Although I really only notice it in the city when there's more traffic on the road, so maybe it's just picking up a car in the next lane and hestiating for a split second. I know that doesn't make any sense but it kinda feels like that. Little hestitations instead of just smoothly adding more power to maintain speed on a slight incline.

I don't notice anything like that on the open road though, with no traffic nearby.

It might just be the difference with having LiDAR in DD2 Pro vs not having it. I'd assume LiDAR helps with ranging to nearby vehicles making it easier for DD2 to determine appropriate speeds relative to them.

 

[Saksak]

• […] The stiff pedal is puzzling to me because it's very different with regen on vs off. If you turn regen off during a nice steady drive (like 45 MPH) and keep the pedal pressed the same depth, the car speeds up like 10 MPH. And vise versa. That makes no sense to me. […]

Perhaps the following will help explain what is going on.

Imagine the accelerator pedal has only three positions.

• Position 0: accelerator is not pressed at all.
• Position 1: accelerator is passed halfway.
• Position 2: accelerator is pressed fully.

What is the expected behavior with full regen?

• Position 0: full deceleration
• Position 1: zero acceleration
• Position 2:full acceleration

What is the expected behavior with no regen?

• Position 0: zero acceleration
• Position 1: 1/2 acceleration
• Position 2: full acceleration

Let’s assume you are keeping the accelerator pressed at position 1 with full regen, which maps to zero acceleration or constant speed. When you turn regen off, position one now maps to 1/2 acceleration so the car speeds up, even though you haven’t moved the accelerator pedal.

It’s impossible to map the same accelerator settings across the various regen levels (off, standard, and high), unless you add dead zones where pressing the accelerator pedal does nothing. Dead zones in the accelerator pedal would feel very unnatural and I don’t think anybody wants that.

My guess is the Mach-E does the exact same thing but you don’t notice it since the regen levels and maximum acceleration is much lower than the Gravity. The mapping in the Mach-E are closer together between the no regen and full regen modes, or perhaps they made the lower portion of the accelerator position diverge while making the upper proportion match.

The latter may not be possible with the Gravity due to the very high deceleration and acceleration ranges. It’s possible that Lucid didn’t try, or they tried and found out it was an even worse pedal feel.

By the way, this has nothing to do with pedal stiffness. Pedal stiffness is a result of mechanical springs and doesn’t change based on regen levels.

 

[dbsb3233]

Perhaps the following will help explain what is going on.

Imagine the accelerator pedal has only three positions.

• Position 0: accelerator is not pressed at all.
• Position 1: accelerator is passed halfway.
• Position 2: accelerator is pressed fully.

What is the expected behavior with full regen?

• Position 0: full deceleration
• Position 1: zero acceleration
• Position 2:full acceleration

What is the expected behavior with no regen?

• Position 0: zero acceleration
• Position 1: 1/2 acceleration
• Position 2: full acceleration

Let’s assume you are keeping the accelerator pressed at position 1 with full regen, which maps to zero acceleration or constant speed. When you turn regen off, position one now maps to 1/2 acceleration so the car speeds up, even though you haven’t moved the accelerator pedal.

It’s impossible to map the same accelerator settings across the various regen levels (off, standard, and high), unless you add dead zones where pressing the accelerator pedal does nothing. Dead zones in the accelerator pedal would feel very unnatural and I don’t think anybody wants that.

My guess is the Mach-E does the exact same thing but you don’t notice it since the regen levels and maximum acceleration is much lower than the Gravity. The mapping in the Mach-E are closer together between the no regen and full regen modes, or perhaps they made the lower portion of the accelerator position diverge while making the upper proportion match.

The latter may not be possible with the Gravity due to the very high deceleration and acceleration ranges. It’s possible that Lucid didn’t try, or they tried and found out it was an even worse pedal feel.

By the way, this has nothing to do with pedal stiffness. Pedal stiffness is a result of mechanical springs and doesn’t change based on regen levels.

That 3rd set of assumptions is incorrect. Actually both the 2nd and 3rd need qualifying. Pressing halfway down when starting from a stop will accelerate in both scenarios (whether regen on or off). Once up to speed and cruising, both will maintain a steady speed on a flat road (zero acceleration). That's just normal highway cruising. The car doesn't just keep accelerating up to 155 MPH when the pedal is halfway.

Regen on or off should have zero impact on a steady press. Only when lifting up.

I don't have the Mach-E anymore to do such a test, but I do have the Macan and I can tell you 100% that turning regen on/off during a steady cruising speed (manual pedal press) has zero impact on the speed. Not light impact (from lighter regen), but zero. Nor should it. By definition, regen is only supposed to occur when letting up on the pedal.

Maybe that's took some kind of perfect matching by Porsche. I don't know, but I don't see how it would even need it. You simply don't do anything different until the pedal starts to lift. Then you trigger regen slowing in proportion to the degree of lift. And if regen is off, you don't (coasting, or mimic coasting if the motors are always engaged).

The only reason I can see to have different pedal mapping for regen on vs off is if regen needs more pedal range to effectively modulate it. Stronger regen needs more ability to modulate it. That's my best guess at why it's so different in the Gravity, that the combination of a higher top end speed (155), limited pedal play range, and very strong regen requiring more pedal play to modulate forced them to map regen on to a deeper press. Boxed in with more to squeeze in the box, so to speak. Probably also why it's a little harder to modulate speed in the Gravity. A 1 MPH change takes 1mm press on the Mach-E or Macan but 0.8mm on the Gravity (just theoretical numbers to illustrate the point), so it's harder to pinpoint.

 

[Saksak]

That 3rd set of assumptions is incorrect. Actually both the 2nd and 3rd need qualifying. […] The car doesn't just keep accelerating up to 155 MPH when the pedal is halfway. […]

The only reason I can see to have different pedal mapping for regen on vs off is if regen needs more pedal range to effectively modulate it.

Technically, the accelerator position is for torque delivered and you are correct in that acceleration/deceleration depends on speed. Increasing speed requires more torque to maintain speed due to drag. We cannot accelerate forever at a given pedal position.

You identified the reason for the different pedal mapping with regen on vs off - the pedal range needs to cover different torque delivery profiles. Here is another way of looking at it. What is the zero torque point with regen off vs. on?

• With regen off, the zero torque point is always at zero pedal press. Any amount of pedal press applies positive torque (accelerate until drag equals power).
• With regen on, the zero torque point is at some non-zero pedal position. This is what enables one-pedal driving. The pedal is being pressed, yet the torque is zero. Releasing the pedal slightly applies negative torque, slowing you down.
• This zero torque position in the full-regen case always maps to some positive torque in the no-regen case. Also, the negative torque positions don’t exist in the no-regen case. The mappings must be different between regen and no-regen.

An illustration (curves are not real, example only):

There is one additional nuance. The mapping is dynamic with regen on. The zero torque point changes based on speed (only for regen on, not for regen off). This is why the car applies positive torque/acceleration even with a slight pedal press when stationary vs. negative torque/deceleration) at highway speeds at the same pedal position (in the regen on case).

One could argue that the right hand side of the curve should match with regen on vs. off when the graphs intersect, and perhaps that’s what the Mach-E and other cars with low regen and acceleration have achieved. Maybe this intersection point with the Gravity, which has very high deceleration/acelleration, is so far to the right that it occurs at speeds higher than typical highway speeds or does not intersect until 100% (imagine a straight line between 0-100% pedal position for the no-regen line). Maybe matching was attempted but resulted in an unnatural pedal feel at the matching transition point and was abandoned. We’d have to ask Lucid to get more insight.

 

[Pragmatic]

There is one additional nuance. The mapping is dynamic with regen on. The zero torque point changes based on speed. This is why the car moves forward with even a slight pedal press when stationary vs. slows down at highway speeds at the same pedal position.

Great write-up.
One question, I would think with zero regen the zero torque point does not vary with speed. It remains at zero pedal. Otherwise there would be a dead space in the pedal on the highway. I know this is a Gravity section, and I gave an Air. On the Air with regen off (which I frequently use on the highway) any movement of the pedal, when traveling at 75, results in power being added. Maybe not enough to accelerate but enough to prevent deceleration.

 

[Saksak]

Great write-up.
One question, I would think with zero regen the zero torque point does not vary with speed. It remains at zero pedal. Otherwise there would be a dead space in the pedal on the highway. I know this is a Gravity section, and I gave an Air. On the Air with regen off (which I frequently use on the highway) any movement of the pedal, when traveling at 75, results in power being added. Maybe not enough to accelerate but enough to prevent deceleration.

Yes, with regen off, the zero-torque point is always at zero pedal press and doesn’t change. It only changes with regen on. I clarified and edited my post - thanks for pointing that out!

 

[borski]

Technically, the accelerator position is for torque delivered and you are correct in that acceleration/deceleration depends on speed. Increasing speed requires more torque to maintain speed due to drag. We cannot accelerate forever at a given pedal position.

You identified the reason for the different pedal mapping with regen on vs off - the pedal range needs to cover different torque delivery profiles. Here is another way of looking at it. What is the zero torque point with regen off vs. on?

• With regen off, the zero torque point is always at zero pedal press. Any amount of pedal press applies positive torque (accelerate until drag equals power).
• With regen on, the zero torque point is at some non-zero pedal position. This is what enables one-pedal driving. The pedal is being pressed, yet the torque is zero. Releasing the pedal slightly applies negative torque, slowing you down.
• This zero torque position in the full-regen case always maps to some positive torque in the no-regen case. Also, the negative torque positions don’t exist in the no-regen case. The mappings must be different between regen and no-regen.

An illustration (curves are not real, example only):

View attachment 39904
There is one additional nuance. The mapping is dynamic with regen on. The zero torque point changes based on speed (only for regen on, not for regen off). This is why the car applies positive torque/acceleration even with a slight pedal press when stationary vs. negative torque/deceleration) at highway speeds at the same pedal position (in the regen on case).

One could argue that the right hand side of the curve should match with regen on vs. off when the graphs intersect, and perhaps that’s what the Mach-E and other cars with low regen and acceleration have achieved. Maybe this intersection point with the Gravity, which has very high deceleration/acelleration, is so far to the right that it occurs at speeds higher than typical highway speeds or does not intersect until 100% (imagine a straight line between 0-100% pedal position for the no-regen line). Maybe matching was attempted but resulted in an unnatural pedal feel at the matching transition point and was abandoned. We’d have to ask Lucid to get more insight.

This is a fantastic post. Thank you.

 

[dbsb3233]

..

Technically, the accelerator position is for torque delivered and you are correct in that acceleration/deceleration depends on speed. Increasing speed requires more torque to maintain speed due to drag. We cannot accelerate forever at a given pedal position.

You identified the reason for the different pedal mapping with regen on vs off - the pedal range needs to cover different torque delivery profiles. Here is another way of looking at it. What is the zero torque point with regen off vs. on?

• With regen off, the zero torque point is always at zero pedal press. Any amount of pedal press applies positive torque (accelerate until drag equals power).
• With regen on, the zero torque point is at some non-zero pedal position. This is what enables one-pedal driving. The pedal is being pressed, yet the torque is zero. Releasing the pedal slightly applies negative torque, slowing you down.
• This zero torque position in the full-regen case always maps to some positive torque in the no-regen case. Also, the negative torque positions don’t exist in the no-regen case. The mappings must be different between regen and no-regen.

An illustration (curves are not real, example only):

View attachment 39904
There is one additional nuance. The mapping is dynamic with regen on. The zero torque point changes based on speed (only for regen on, not for regen off). This is why the car applies positive torque/acceleration even with a slight pedal press when stationary vs. negative torque/deceleration) at highway speeds at the same pedal position (in the regen on case).

One could argue that the right hand side of the curve should match with regen on vs. off when the graphs intersect, and perhaps that’s what the Mach-E and other cars with low regen and acceleration have achieved. Maybe this intersection point with the Gravity, which has very high deceleration/acelleration, is so far to the right that it occurs at speeds higher than typical highway speeds or does not intersect until 100% (imagine a straight line between 0-100% pedal position for the no-regen line). Maybe matching was attempted but resulted in an unnatural pedal feel at the matching transition point and was abandoned. We’d have to ask Lucid to get more insight.

Yes, that's what regen is - negative torque, to decelerate the car when lifting up on the accelerator pedal. Your chart is a good visualization to show the effect of regen on the car's speed.

How they choose to map it to the accelerator pedal is a different matter though. They likely *could* map them equally for the amount of positive torque being applied at any point in the curve (as is done in the Macan), but perhaps it is the stronger regen in the Gravity that drove them to use significantly disparate positive torque curves, to give them more "runway" (so to speak) for the driver to modulate a wider range of regen.

Although as you said, it's not a fixed point anyway, it's dynamic, so I suppose this all could be moot. It's not just the bottom 20% or 40% of the pedal range where regen applied when you begin to lift up on the pedal, it's the entire range of the pedal. If you're 80% of the way down and start to lift up, you get regen there too. But when you're only 20% of the way down, you still want enough room to modulate so maybe they pushed that lower end higher? No way to know what their thinking was, but it's fun to speculate. At minimum we do know that such strong regen does need a lot of pedal runway to modulate. If anything, I'd say the Gravity could use MORE runway in the pedal. A very tiny movement makes a big difference (in both positive and negative torque) so you have to be more precise than, for instance, our Macan, which has both a softer pedal and more room to modulate speed.

Our Mach-E actually made that a choice. The different drive modes affected 3 things: steering wheel ratio and tightness, regen level, and accelerator pedal range. I LOVED that I could select a broader pedal range (making it easier to be more precise with speed). As well as making the steering wheel easier to turn and less twitchy. I didn't want sporty, I wanted luxury. The Gravity has a great steering wheel feel for my tastes (and I love the squircle!), but I don't like the accelerator pedal tuning so much. It's more like the Mach-E'a tighter, sporty setting than the softer, wider range setting I liked much better.

 

[Saksak]

[…] It's not just the bottom 20% or 40% of the pedal range where regen applied when you begin to lift up on the pedal, it's the entire range of the pedal. If you're 80% of the way down and start to lift up, you get regen there too. […]

At 80% pedal and lifting up a little, you get deceleration but not regen. It may feel like regen as the car is slowing down, but the deceleration is due to less positive torque. You’ll be able to see this in the car’s power meter. The power will still be in the positive range, but less.

It’s the same as an ICE vehicle where you experience deceleration when lifting up on the accelerator slightly. The deceleration is due to less power and not due to the brakes being applied.

 

[dbsb3233]

At 80% pedal and lifting up a little, you get deceleration but not regen. It may feel like regen as the car is slowing down, but the deceleration is due to less positive torque. You’ll be able to see this in the car’s power meter. The power will still be in the positive range, but less.

It’s the same as an ICE vehicle where you experience deceleration when lifting up on the accelerator slightly. The deceleration is due to less power and not due to the brakes being applied.

Well I will admit I don't think I've even cruised at 80% power before to test it (that has to be what, 120 MPH in the Gravity? ).

It was really just an illustration to make the point. I don't know where regen tops out (if it does at all), but I know it's at least 85 MPH (>50%) because I've pulled up on the accelerator pedal at that speed multiple times when highway cruising and it definitely applied regen (as you said, you can tell by the power meter, as well as simply feeling the huge difference between hard regen and loss of coasting momentum).

 

[knewman]

Technically, the accelerator position is for torque delivered and you are correct in that acceleration/deceleration depends on speed. Increasing speed requires more torque to maintain speed due to drag. We cannot accelerate forever at a given pedal position.

You identified the reason for the different pedal mapping with regen on vs off - the pedal range needs to cover different torque delivery profiles. Here is another way of looking at it. What is the zero torque point with regen off vs. on?

• With regen off, the zero torque point is always at zero pedal press. Any amount of pedal press applies positive torque (accelerate until drag equals power).
• With regen on, the zero torque point is at some non-zero pedal position. This is what enables one-pedal driving. The pedal is being pressed, yet the torque is zero. Releasing the pedal slightly applies negative torque, slowing you down.
• This zero torque position in the full-regen case always maps to some positive torque in the no-regen case. Also, the negative torque positions don’t exist in the no-regen case. The mappings must be different between regen and no-regen.

An illustration (curves are not real, example only):

View attachment 39904
There is one additional nuance. The mapping is dynamic with regen on. The zero torque point changes based on speed (only for regen on, not for regen off). This is why the car applies positive torque/acceleration even with a slight pedal press when stationary vs. negative torque/deceleration) at highway speeds at the same pedal position (in the regen on case).

One could argue that the right hand side of the curve should match with regen on vs. off when the graphs intersect, and perhaps that’s what the Mach-E and other cars with low regen and acceleration have achieved. Maybe this intersection point with the Gravity, which has very high deceleration/acelleration, is so far to the right that it occurs at speeds higher than typical highway speeds or does not intersect until 100% (imagine a straight line between 0-100% pedal position for the no-regen line). Maybe matching was attempted but resulted in an unnatural pedal feel at the matching transition point and was abandoned. We’d have to ask Lucid to get more insight.

Really nice post! I've never given much thought to the differences in throttle mapping wrt regen enabled/disabled. The example visualization helped a lot.

I wish Lucid would expose more of this sort of data in the UI, perhaps when in sprint or swift mode.

 

[hwy17er]

I tend to think about the way Regen kicks in when you lift the pedal on Gravity to be a lot like downshifting in an iCE stick shift, where you are indeed using the motor to slow the vehicle, e.g. like when entering a curve going downhill. Thus the pedal tuning in my GGT seems absolutely ideal to me. I've never driven a car with such precise speed control on curvy mountain roads without a stick shift (and in such a heavy vehicle no less!!).

We used to have a convertible hard top Miata 6speed stick and you could downshift into curves and control the speed quite well. But then when some kook on his cellphone didn't see traffic had come to a stop around the corner and "accordioned" it (4car pile-up; totaled such a fun car!), my wife found a nearly identical car (same year, same color) but when I first walked up to it and looked inside I was horrified that between the seats the gear shift said "PRNDL"! Ugggh!
I never could get those paddle shifters in that CVT transmission down well enough to drive the car like that so after a "sporty" Audi Q5 (yawn) and even driving my wife's Taycan I've finally got that precise control of the vehicle back with the GGT!

Some may be seeking a more "luxury" experience with softer accel/decel but I fully see what Lucid engineers were trying to create here and it's awesome!!