Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Cruise Control improvements

This site may earn commission on affiliate links.
Regen on/off (or low, med, high) needs to be on a primary control, just like the brake or accelerator or turn signal. The shifter works great. I use it more often than I use the brake pedal.

Don't you then sometimes forget, or get confused about, which mode you are in, and the car behaves differently than intended?

Personally, I'd prefer a single specific mode which goes into muscle memory, and that I can rely on.
 
tough problem for which I've see no suggestions that personally make sense to me based on having driven the Roadster and Model S over the last 2+ years.
I'm glad you agree it's a problem; while I agree that smart drivers will learn how to manage this issue, I'm also concerned that ~90% of drivers don't pay attention to driving techniques and would benefit by some alteration in the current CC behavior.

Thinking about the input from above, my specific proposal for behavior immediately after releasing the CC (either by turning it off or pushing the stalk forward) is this:
  1. Is brake pedal depressed (at all)? If so, enable full regen and apply mechanical brake matching the brake pedal input.
  2. Otherwise, is accelerator depressed (at all)? If so, match regen/acceleration to accelerator input.
  3. Otherwise, feather in regen, starting with 0% for one second and ramping up to 100% after four seconds. [exact figures would require some testing]

This proposal would allow thoughtful drivers to have the car do exactly what they want, while providing less attentive drivers a few seconds to pick up from the CC.
 
Thinking about the input from above, my specific proposal for behavior immediately after releasing the CC (either by turning it off or pushing the stalk forward) is this:
  1. Is brake pedal depressed (at all)? If so, enable full regen and apply mechanical brake matching the brake pedal input.
  2. Otherwise, is accelerator depressed (at all)? If so, match regen/acceleration to accelerator input.
  3. Otherwise, feather in regen, starting with 0% for one second and ramping up to 100% after four seconds. [exact figures would require some testing]

I don't think you need/want your 2) and 3) to be separate: simply in either case have it transition from the cruise setting to the setting commanded by the accelerator over a couple of seconds. But exactly as you say, only a manual cancel from the stalk would give the soft transition, while the forced cancel with the brake pedal would be abrupt.

But the ultimate fix may need a hardware addition to the car. The exact same issue applies when driving a manual ICE - if you don't match the accelerator position, then you get engine braking when you cancel the cruise; however, I find that I can usually feel the relevant position in the pedal stiffness. This is just a side-effect of the way it's implemented in the ICE case (with the servo motor compressing the throttle spring in parallel with a softer spring giving the pedal return), and implementing the same feel in an EV would require an extra servo motor.

However, such a feature could be very useful to give a special variety of cruise-assist in circumstances (heavy traffic) where ordinary CC isn't useable. I find that modern cars are actually very difficult to drive at the speed limit even if you want to, especially around town. Governing the speed would be too restrictive, and devices that beep are just too annoying, but having a resistance in the pedal travel so that you can easily get the speed you want (while still having full authority) could be a useful driving aid.
 
Thinking about the input from above, my specific proposal for behavior immediately after releasing the CC (either by turning it off or pushing the stalk forward) is this:
  1. Is brake pedal depressed (at all)? If so, enable full regen and apply mechanical brake matching the brake pedal input.
  2. Otherwise, is accelerator depressed (at all)? If so, match regen/acceleration to accelerator input.
  3. Otherwise, feather in regen, starting with 0% for one second and ramping up to 100% after four seconds. [exact figures would require some testing]
Well, 1) is already there since touching the brake immediately cancels CC and car goes into full regen. 2 and 3 sound interesting. I'd like to try it out.

But the ultimate fix may need a hardware addition to the car. The exact same issue applies when driving a manual ICE - if you don't match the accelerator position, then you get engine braking when you cancel the cruise; however, I find that I can usually feel the relevant position in the pedal stiffness. This is just a side-effect of the way it's implemented in the ICE case (with the servo motor compressing the throttle spring in parallel with a softer spring giving the pedal return), and implementing the same feel in an EV would require an extra servo motor.
Many cars are fly by wire these days so you can't detect throttle cable position by pedal pressure. In the old days, yes, but not anymore. As I posted above I was able to match speed w/ my manual Corvette (fly by wire throttle) and sure, you're nearly always off by a little but it's always going to less than not pressing the pedal at all.
 
I really dislike how the cruise disengagement interacts with regen in the current software. I like Robert's proposal, except that I think the slow onset of regen should happen any time you disengage cruise, including when you disengage it by pressing the brake pedal. Rolling full regen on rapidly if you only needed a small amount of braking is likely to surprise the driver and produce an unsafe condition.

On a related topic, does anybody else find it odd that pushing the stalk away (in the direction of travel) cancels cruise, while pulling it towards you resumes? It's backwards to how my intuition suggested that the stalk should work (push forward to go faster, pull back to slow down).
 
I really dislike how the cruise disengagement interacts with regen in the current software. I like Robert's proposal, except that I think the slow onset of regen should happen any time you disengage cruise, including when you disengage it by pressing the brake pedal. Rolling full regen on rapidly if you only needed a small amount of braking is likely to surprise the driver and produce an unsafe condition.

No, absolutely not. The brake force shouldn't be any different after using CC, compared to at other times. You need to rely on the brake force which you are used to, immediately. If you want only a small amount of braking, then immediate full regen already provides that.
 
No, absolutely not. The brake force shouldn't be any different after using CC, compared to at other times. You need to rely on the brake force which you are used to, immediately. If you want only a small amount of braking, then immediate full regen already provides that.

Oh highways I sometimes drive in "pure CC mode". I use the tap up and tap down to maintain speed with the car ahead (hence my continued cry for ACC). When I disengage the CC, I want full regen - it means I want to slow down. Usually I don't need to hit the brake becuase the regen is enough and I recapture as much energy as possible.

One other point with the current cruise control software is how it handles the the tap and hold. You push the CC stalk down (not past the click point) and hold it. It lowers the target speed and seems to feather on the regen.

I really like the behaviour of the current CC and would rather it not change - until the hardware is added to allow for real ACC functionality.
 
Cruise Control for Non-Mercedes Types

My wife had a bit of fun with the cruise control today by accident (without accident). She was flipping it by reflex as the blinker as that is where BMW puts their blinker. In the process, she set the cruise at 30 mph on a 45 mph road. She drove along as normal with her foot on the accel then went to slow. The single pedal regenerative braking worked perfectly right until she hit that magic pre-set 30mph number at which point the car held the speed. She dove for the brake and that dis-engaged the cruise.

It caught her by surprise and may do others. For all us non-MB types, it is something to look out for until we get used to the stalk arrangement.
 
This is also something I have to get used to. I'm adapting and getting better at the transition. That being said, I wouldn't mind the car detecting pressure on the accelerator and cutting regen in that instance, as long as the accelerator is held in. But that may cause unintended issues, who knows?

As far as unintentional cruise activation, I always turn off the system by pushing in the lever - I've never used 'resume' in any of my cars because I like to accelerate at the speed I want it to. This avoids accidentally setting it when reaching for the turn signals.
 
I'm still screwing it up. I don't grab the wrong stalk very often anymore, but it happens occasionally. The bigger problem is the push-to-disengage and pull-to-reengage feature of the CC stalk. I still have to think about it every single time. Slows my reaction time; occasionally I've had to jab the brakes rather than think it through.
 
I'm still screwing it up. I don't grab the wrong stalk very often anymore, but it happens occasionally. The bigger problem is the push-to-disengage and pull-to-reengage feature of the CC stalk. I still have to think about it every single time. Slows my reaction time; occasionally I've had to jab the brakes rather than think it through.
In case it helps, compare the thickness of the CC and turn signal stalks in your other vehicle( s ) to the thickness on the Model S. If those comparative thicknesses are consistent, you might be able to retrain your brain to focus on that rather than location.
 
Toyota required that the Regen be in the low setting when operating the curise control. (D not B) This prevents the uncomfortable decelation when turning off the cruise.

Toyota Prius uses "creep regen". Regeneration on the accelerator is no more no less than the minimal drag one would experience with a conventional automatic transmission.

Toyota brakes are rather complex in that the initial travel produces regeneration at speeds over 7 MPH but friction braking at lesser speeds. Strangely when the friction brakes are wet they are grabby and the transition from regeneration to friction is not smooth.

- - - Updated - - -

In case it helps, compare the thickness of the CC and turn signal stalks in your other vehicle( s ) to the thickness on the Model S. If those comparative thicknesses are consistent, you might be able to retrain your brain to focus on that rather than location.

My "other car" is a 2009 Mercedes-Benz ML320 Bluetec which has essentially the exact same steering column controls as my Model S.

On the ML320 the CC stalk can be used as a "trim" to drive the car footless. Disengage if one gets boxed behind a truck, re-engage when the box opens. Click up 2 MPH to pass, back down to the closest multiple of 5 MPH for cruising. Sadly the Model S CC doesn't work that smoothly. 1 MPH trim adjustments are far too aggressive.

I agree with the original poster that disengaging the CC should temporarily suspend regenerative braking. I disagree with some theories as to when it should resume. I think not until after the accelerator has been used to propel the vehicle should regeneration be re-enabled. The Model S already has the means of temporarily restricting regeneration. Has a dashed line which appears on the "power speedometer" (the Model S equivalent of a tachometer?) to indicate reduced regeneration limits.

Others suggest its easy enough to depress the accelerator and find the current point before disengaging CC. I would be more inclined to agree if the accelerator moved with the CC as with many conventional vehicles. Doesn't really move but with throttle held open by CC only the spring under the pedal is holding it up. Press until one meets additional resistance and one has found the CC's current throttle position. As things are now one has to accelerate beyond the set point speed to gain manual control before disengaging the CC.