asking help of tuning gain

[dzy666fly]

Hello, everyone. I want ur help about using gain tuner.
When I try to use gain tuner on the 6-DOF robot arm UR10 (and UR10e), I find that setting up joint limit seems to be important.
If every joint is limited to [-6.0, 6.0], the visual curve shows well.
However, if I set up the joint limit as the real world robot arm in our lab, it is weird that the second joint shoulder_lift_joint moves unexpected as follows. Well, other joints works well. How can I solve this problem?

The natural frequency is set to 25.0 and damping ratio to 1.0.
Every joint limit is set to [50,80], [-80,-40], [80,120], [240,300], [135,150], [120,150].
Thank u all so much!

[michaellin6]

Hi @dzy666fly thank you for posting your question. Can you clarify what is unexpected on the generated charts? I see that in the position chart, the sinusoidal command is offset and swinging from -80 to -40 as expected due to the joint limits. Then you see that the joint positions track the command to certain accuracy. The notches are due to the robot reseting its state for each of the other joint sinusoidal wave experiments. The arm resets to position 0 and then moved to the closest position within the joint limits (at -40 deg).

[michaellin6]

Hello!

We noticed that this topic hasn’t received any recent responses, so we are closing it for now to help keep the Github Issues.

If you’re still experiencing this issue or have additional questions, please feel free to create a new issue with updated details. When doing so, referencing or linking to this original discussion would be helpful for context.

Thank you for contributing to the NVIDIA Isaac Sim community!

Best regards,
The NVIDIA Isaac Sim Team