Unexpected wheel "jump" when calling straight() after turn()

[TeamGomeNeve]

High precision issue when using turn() followed by straight() and vice versa

I'm working with my robotics class to achieve high precision in FLL.
When using drive_base.turn() or drive_base.straight(), the robot works perfectly.
However, when I call the straight() function immediately after a turn(), the robot makes a small "jump" with only the outer wheel before it starts moving straight.

The only solution I’ve found is to use the reset_angle() function on both the right and left motors.

1. Does this affect the accuracy of turning and driving?

After solving this issue, I also noticed that sometimes—though not always—when the robot finishes a straight movement after a turn, it makes a small forward "jump" when stopping, influencing precision.
2. How can I solve this and enhance precision?

[BertLindeman]

Hello,

Just a thought:
You want the movement to go on smoothly.
Did you try to do the straight() and turn() with Stop.NONE?

Something like:

drive_base.use_gyro(True)
drive_base.reset()
wait(250)  # stay still for a moment to allow the gyro to get ready
for _ in range(8):
    drive_base.straight(120, Stop.NONE)
    drive_base.curve(100, 45, Stop.NONE)

Bert

[TeamGomeNeve]

Well, it's mainly about precision, without resetting the motor turn degrees every run I get different results, but by resetting the motors the precision increases, even though the straight 'jump' still creates problems

[BertLindeman]

My knowledge about the background of drivebase is limited.

1. I assumed you use the gyro.
2. The specified measurements of the wheel diameter and axle track are important
3. Your robot does not spin out of control, so the left / right motor definitions are OK.
4. The Stop definition on the drivebase movements are of importance.

This is about all I can help you with.

[TeamGomeNeve]

1. I use the gyro with drive_base.use_gyro(true)
2. My measurements are more calibration accurate than measure accurate, as listed in the wiki
3. The motors seems defined okay
4. To stop I use the hold type, even though i tried different stopping methods and none of them worked

I can provide a video of the problem if this could help understand my problem.
Thanks you very much Bert for your help

[zoligyenge]

I think the "jump" issue is due purely to mechanics, namely motor play.

When you turn, one of the motors is engaged clockwise, the other one counterclockwise. Hence, when you drive straight after a turn, one of the wheels locks immediately, the other motor needs to turn by the amount of play (say 0.5°) before it engages.

[TeamGomeNeve]

I think this plays a role in the problem even though i don't think is the key to solve it, any suggested solution?