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

Problems with Auto Lane change in 32.12.2

This site may earn commission on affiliate links.
Thanks for the update, I have a service appointment booked for November 29th at the Costa Mesa location for the phantom proximity sensor warnings and failed lane changes, really looking forward to the findings from the engineers looking at your car.
 
My late 2018 MX was upgraded to HW3.0 a week ago. I am having the exact same problems mentioned in this thread after upgrade. I am on 2019.32.12.2 (before and after upgrade). Never had these issues before the upgrade.

Also, auto-wipers and acceleration speed after stopping behind a car got worse after the HW3.0 upgrade.
 
My late 2018 MX was upgraded to HW3.0 a week ago. I am having the exact same problems mentioned in this thread after upgrade. I am on 2019.32.12.2 (before and after upgrade). Never had these issues before the upgrade.

Also, auto-wipers and acceleration speed after stopping behind a car got worse after the HW3.0 upgrade.

May 2019 Model 3, HW3, 32.12.2.

Maybe HW3 and 32.12.2 at the same time is the problem? Hopefully the next update will fix the problem.
 
I couldn't take the car out yesterday but I did get some testing done earlier today. Placed the car back in Mild lane change mode and I did NOT have any of the hesitation / delay that I had experienced last week. Used both NoA and EAP modes. Toggled between Mild and Mad Max and honestly could not tell the difference between the two. Both made the lane change in what I thought was a "normal" period of time. A couple of times I purposely applied the turn signal just after passing slower traffic and the car recognized the vehicle (red image on the display) and my car delayed moving over until there was a clear space to do so. Never did see the "wavies" as described earlier.

From my perspective, the initial application of Mad Max mode appears to have immediately "fixed" my car. Nothing else was done, like re-booting or changing any of the other options in the Auotpilot menus. If that indeed changed / reset something else which caused my awful performance of ALC then I'm happy with the result. I'll keep the car in Mild mode (what it used to be in) and continue to monitor the lane change mode.

@Fernand: We're good. I understand your reasoning. Sorry for seeming to jump on you (and your opinion). I've been an admin at a ThinkPad forum for the last decade or so. In that time I thought I've become more inured to perceived online criticism but I guess I had a weak moment. I've been spending a lot more time here on TMC, although I have posted about my Tesla there.
 
My problems seem to have been resolved with tire pressures above 39psi. I’m the cold when they dipped I had more failed ALC. I had 0 issues the last two days.

As odd as that may be, with the swinging fall temps can y’all let me know about your tire pressures?
Grasping at straws here but.
 
Wow, THAT would be a strange one. But MY tire pressures are pretty evenly 45 cold, as per spec. They get up to 47 or even 48 hot. I had just inflated them a bit recently, either before or after the crazy stuff started. But there's something about what you're reporting that sounds like it could be "it".

@texasranger and @RayK, are you guys watching for the proximity warning "wavies" on either side of your car image on the screen? Did you have them before, and they went away? 'cause I'm pretty sure that's the sign & symptom, and it says "I can't change lanes, dude, there's something in the way". And what you really DON'T want is the converse: "I can't see nuthin' there, I'll sideswipe it for yuks".

Seeing as they've had my car for about a week, they're probably testing left field hypotheses by now like "amount of fluid in the wipers reservoir", or "candy wrappers on the floor", or "last viewed YouTube content", or "thickness of wallet pressing on seat".

I LOVE that driving into the gas station effect, wiping the windshield, adding a little air, checking out the confections racks, thanking the staff and driving off.

Another LOVE thing, that I don't do with my Nicki (when healthy) out of respect, is gonzo accel. The Model S they loaned me isn't ludicrous, but it's a D, and it's no slouch. In the absence of automation entertainment, I was testing the Hawk From Above Lane Squeeze. Get up to say 90 MPH so briefly they couldn't even think of clocking you, and perfectly safely drop into a minimum daily requirement spot between two cars one lane over. Like a knight on a chessboard. This one guy seemed to blink too slowly, 'cause bam, there was a car right in front where he could have sworn there wasn't ;-) They better get my car done before I lose my mind entirely.
.
 
Last edited:
  • Love
Reactions: HappyWino
Haha. Do you have a Model S or Model 3 @Fernand? I have the performance wheels on my Model 3 that have a recommended 42 PSI "cold". When they drop more than 4 PSI that's when I started to have issues. After filling back up and keeping them filled (large temperature fluctuations in Nevada... 30-80 degrees Fahrenheit) I stopped having issues.

No more "wavy" lines in the back unless I am next to a curb like it should be. I did not wash the car -- but today is day 3 without having ALC hesitation.
 
@texasranger, so you DID have the wavies, and they went away? Wow. My tires were down 6 PSI last month, so I brought them up to 45 as per the door sticker. Around the time of 12.1 or 12.2 and the start of ALC problems, I topped up one that was down to 44. So, like the switching to MadMax issue, it's not a universal cause and effect.

I have a Model S loaner. After I documented the lane change issues, my Model 3 has been under Service Center observation for a week now.

The Model S loaner is a very basic Enterprise rental, with version 2019.32.12.2 and no automated anything. I'm so used to NOA, and depend on it, that it's really a downgrade. No wavies, but then again what would it know of lane changes, as it lacks even the basic AP.
.
 
@Fernand Any luck at the SC with info? Been 3 or 4 days right?

Side Note: I seem to get fewer aborts when I inflate my tires to the recommended 42psi. When it gets colder and they drop to 37-38 I seem to get more aborts. I’m not sure if there is a correlation there but I’ll ask others to try if able.

I can't explain why but changing my tire pressures to 42psi when cold (46psi in the afternoon when warm) has completely fixed my proximity warnings problem and the car has been changing lanes perfectly since Monday.

I will continue to monitor things and report back in a few days.
 
When driving, you are surrounded by people you don't know, won't meet, and will never interact with. Does it matter what they think?

To an extent, no. But it will make other drivers wary of you, and perhaps react in ways that cause undesirable situations. Plus I personally feel that being a good driver also means being a polite driver to others. If you never care about what strangers think, they will always BE strangers.
 
Ever since I downloaded my model 3 wont auto change lanes anymore. Fully aborts.
Contacted Tesla service & they had me try a couple of things that didn’t work.
Now saying bug in 32.12.2 and I have to wait till 36.1 to have fixed. Wow !!
 
Glad to hear others are also experiencing this. I’ve resorted to turning off Navigate on Autopilot for now. The lane changes aborts are horrible. Hopefully they start pushing a new version soon.
 
  • Disagree
Reactions: kpeng
Yes, my experience is that it's taken a step back. If you are in a rural area with low density traffic then it's okay. But a 3 or 4 lane highway in a city with heavy traffic then people think you're fresh out of driver's ed. Another issue that has regressed is exiting. Before if anything it was a little cautious about passing close to your exit. Now it wants to pass a semi that is going 2 mph faster than you with 1/2 mile to go before the exit. It gets about halfway through that procedure before you can see the gears in its head going "oh jeez, that was a bad idea". Frankly, I just have turned it to manual for all but rural driving trips. It's not a bother to me at all and I'll likely keep it that way until things improve.
hmmm, odd that I have not experienced any of these issues at all in our MX100
 
So, saying that I was not having any problems did really screw the pooch. Of a dozen lane changes on the way home this evening, one of them was smooth and assertive. In all the rest of the cases, the car waited past nice long open stretches to signal and attempted to change lanes just as a car was approaching. Of course, the lane changes aborted, sometimes several times.

Then most notable was when I tried to change lanes next to a truck. The car refused to believe that there was an empty lane between me and the truck. The display kept flashing between a red dotted line and the blue target.

Sometimes I want version 9 back.
 
I can't explain why but changing my tire pressures to 42psi when cold (46psi in the afternoon when warm) has completely fixed my proximity warnings problem and the car has been changing lanes perfectly since Monday.

I will continue to monitor things and report back in a few days.
I have a theory on the tire inflation suspicion. I too have had the issue of my car not making left lane changes under NOA or Autopilot. I also had my tire pressures drop significantly with the onset of Fall cool temps. I have had my tires inflated properly and will test my theory today. Could the sound emitted by the under inflated tires cause the Ultrasonic sensors to pick up a signal?

It is just a guess but..
 
I have a theory on the tire inflation suspicion. I too have had the issue of my car not making left lane changes under NOA or Autopilot. I also had my tire pressures drop significantly with the onset of Fall cool temps. I have had my tires inflated properly and will test my theory today. Could the sound emitted by the under inflated tires cause the Ultrasonic sensors to pick up a signal?

It is just a guess but..
Maybe the TPS transmitters are sending an error message that they don't send when the tire is properly inflated? Maybe that signal is picked up by the Ultrasonic sensors? Maybe the frequencies caused bu the contact patch on an under inflated tire generates a masking frequency? I am an audio guy, so my ears are my finest honed of my senses.