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

1300+ miles road trip AP observation

This site may earn commission on affiliate links.
I just came back from a 1300+ miles round trip between Atlanta, GA and Ft. Lauderdale, FL. About 60% of the drive was day time and 40% of the drive was after sunset. My AP HW is 1.0 and SW is 2017.28 c528869.

Overall experience was better than a couple shorter trips I had before in an earlier software version. The Autosteering this time was smoother in general and notably when road ahead ramp up and down. In the past, the car would ping pong between the lane mark every time a small hill coming up ahead.

There were still occasional ping-ponging within lane marks that seemed very random. And sometimes during a turn it would grab the outer lane mark too close. It got scary a couple times when there was a semi in the next lane ahead of me. It did quickly corrected itself but the steering was not comfortable in such situation.

I did notice for the same road condition, Autosteering performed better (less ping-ponging) on the most left lane vs. mid-lane. I am not sure this was because the solid line on the left was easier to recognize vs. dotted line on both sides in the middle lane.

I also notice in the dark environment, Autosteering seemed more confident with high beam on vs. off. It does seem to predict the lane better when the light was projected further.

Close to the end of trip, the car started to have trouble recognizing lane mark intermittently. I disengaged AP and restarted both screens and this problem went away. I wonder if there was a computer power problem that AP2.0 and 2.5 already resolved. But man, it's annoying.

Overall, I love AP and it helped reduce my fatigue significantly. I do think the limited HW on AP1.0 made it harder to become significantly better than its current capabilities.
 
  • Informative
Reactions: vrykolas
I just came back from a 1300+ miles round trip between Atlanta, GA and Ft. Lauderdale, FL. About 60% of the drive was day time and 40% of the drive was after sunset. My AP HW is 1.0 and SW is 2017.28 c528869.

Overall experience was better than a couple shorter trips I had before in an earlier software version. The Autosteering this time was smoother in general and notably when road ahead ramp up and down. In the past, the car would ping pong between the lane mark every time a small hill coming up ahead.

There were still occasional ping-ponging within lane marks that seemed very random. And sometimes during a turn it would grab the outer lane mark too close. It got scary a couple times when there was a semi in the next lane ahead of me. It did quickly corrected itself but the steering was not comfortable in such situation.

I did notice for the same road condition, Autosteering performed better (less ping-ponging) on the most left lane vs. mid-lane. I am not sure this was because the solid line on the left was easier to recognize vs. dotted line on both sides in the middle lane.

I also notice in the dark environment, Autosteering seemed more confident with high beam on vs. off. It does seem to predict the lane better when the light was projected further.

Close to the end of trip, the car started to have trouble recognizing lane mark intermittently. I disengaged AP and restarted both screens and this problem went away. I wonder if there was a computer power problem that AP2.0 and 2.5 already resolved. But man, it's annoying.

Overall, I love AP and it helped reduce my fatigue significantly. I do think the limited HW on AP1.0 made it harder to become significantly better than its current capabilities.

I had a ~3000 mile (4800 km) from Dallas to Philadelphia and back in August 2017.
I had similar thoughts/conclusions.
I'm "guessing" I had about 60% AP2, 20% TACC, and 20% manual drive.

Basically yes, when there are 3 lanes, the middle lane seemed to work best. Otherwise, in 2-lanes left lane was better than right lane. Right lane was not comfortable to use AP2 because it sometimes wasn't sure to exit or not even though navigation was on to stay on the highway.
 
I had a ~3000 mile (4800 km) from Dallas to Philadelphia and back in August 2017.
I had similar thoughts/conclusions.
I'm "guessing" I had about 60% AP2, 20% TACC, and 20% manual drive.

Basically yes, when there are 3 lanes, the middle lane seemed to work best. Otherwise, in 2-lanes left lane was better than right lane. Right lane was not comfortable to use AP2 because it sometimes wasn't sure to exit or not even though navigation was on to stay on the highway.

It does seem AP2 software works quite differently. On my AP1, the left lane with one solid line on the left worked better than two dotted line in the middle lane. My guess is that the limited computer power on AP1 does not project dotted lane mark into a reliable path for the car well. In some areas when the dotted strokes seem longer, it worked better. It also worked better during the night where there are reflective raised pavement markers.
 
  • Informative
Reactions: vrykolas