Small inertia values from URDF set to 0

[user56551]

I am trying to track down some issues with a robot simulation, so I was wondering if these updates were implemented (and the issue is not isaacsim), or it’s ongoing.

Isaac Sim Inertia Treatment

Dear Isaac Sim Team, I recently ported a MuJoCo model to Isaac Sim using the available importer. I noticed that the off-diagonal elements were simply ignored, so I added the information about the off-diagonal elements in the USD file via the physics:principalAxes element of the PhysicsMassAPI. Given that I was unsure whether I needed to present the quaternion in (w,x,y,z) or (x,y,z,w) convention, I simply tried both and looked at the simulation results. To my surprise, the results were the sam…

The link's inertial properties do not match the ones in URDF

Hi, I wrote a standalone Python script to import a robot from URDDF. In general, it works well. However, on closer inspection, I realized if I query robot link inertia properties from the ArticulationView, it only partially matches the ones in the URDF file. There are two sources of differences. (1) when accessing the body CoMs through ArticulationView::get_body_coms(), the CoM frame rotations are different from URDF. (2) when accessing the body CoMs through ArticulationView::get_body_inerti…

I am having issues with small inertia values in meters. When they are in centimers in the urdf it works, but in meters it seems to be saying they are 0.

[rgasoto]

The UI rounds to zero when the value is small, but they are still authored appropriately. You can use the Mass Distribution Manipulator to check the Inertia tensor and verify that it’s correct.

[user56551]

Thanks, I’ll take a look through that. The issue wasn’t just the UI, its also in the USD attr, and when running the physics it has a warning that inertia cannot be 0.