Pitch Effort mismatch due to sliding rig

[mindThomas]

I would like to verify the torque requirements of an actively held pitch joint using Isaac Sim. I have rigged the robot including the end effector before which we have the pitch joint.

The end effector includes a sliding mechanism with a parallel mechanism to hold the end level and move it down as it reaches the end.

From the masses and COM in CAD drawings that I have also added to the Isaac Sim model, I have been able to roughly calculate the expected load torque for the pitch joint to hold the end effector level (i.e. compensate for gravity). This is between 80-118 Nm.

When I start the simulation with the sliding mechanism retracted the Effort reported by the “Isaac Read Effort Node” correctly reports 80 Nm and I get the same value with get_measured_joint_forces()

However, when the sliding mechanism reaches the end and moves down, resulting in an intended collision between the sliding mechanism and the bottom plate, the pitch joint Effort jumps from 110 all the way up to 150 Nm which is incorrect. The jump just happens at the very end, exactly when some collision forces starts to come into play.

Please see a video of the mechanism here including the torque measurement in the top left corner: https://drive.google.com/file/d/1MRG4Shz7w4a_VHc3uz9qnOnA3fY4z0kV/view?usp=sharing

Why do I see this jump? I have no other joints with drivers other than the pitch joint.

I am running Isaac Sim 5.1.0 on Ubuntu 24.04 with an RTX 5090, driver Version: 580.105.08 CUDA Version: 13.0.

[michaellin6]

Hi @mindThomas, thank you for your question. Just from the video it seems like there is a load at the distal end of the sliding/prismatic joint. When it reaches the end of the travel, it seems like the load drops downward which would create a reactive force due to the impact. That force would translate to a rise in measure torque at the joint. Not sure if that is intended, but you can freeze those joints on the sliding mechanism carriage to avoid having the effect of that impact in your measurements.

[mindThomas]

I understand that it might create a reactive force between the two loads upon impact, but why does that reactive force appear on the joint further up in the prim tree hierarchy?
I would like to know the torque that the motor in this particular joint is exerting and until the end where this impact happens the value does also seem to resemble just that. But upon impact something else contributes and the value can no longer be trusted as just the motor torque.

[michaellin6]

The reaction force from the impact will be on equal an opposite on the two colliding bodies. Subsequently the impact force on the distal body will also propagate to the bodies it is connected to. This is why it shows up in the joint torque. If the impact force was exactly intersecting the revolute axis then it wouldn't show up as f x r (where f is the force vector and r is the position vector of the attachment point from the joint axis) would be 0 since r would be the zero vector.

[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 organized.

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, we recommend mentioning or linking to this original topic in your new post—this helps provide context and makes it easier for others to assist you.

Thank you for being part of the NVIDIA Isaac Sim community.

Best regards,
The NVIDIA Isaac Sim Forum Team