-
Hello, I am testing the state transition data of a quadruped Go1 obtained from the Raisim simulation in Pinocchio C++. In Raisim, I stored the generalized coordinate, generalized velocity, generalized acceleration, and joint torques and confirmed that the Lagrangian equation of motion From my investigation, I found the following differences between Raisim and Pinocchio:
Pinocchio convention (local = joint frame)
Lastly, I normalized the generalized coordinates. However, despite these efforts, there are still issues. After several attempts, I found that while the mass matrix and generalized gravity are computed the same way, the nonlinearity or nonlinear effect parts seem to be completely different. Due to my limited knowledge, I am not sure where this discrepancy arises. Below is the code and the results I used.
Results
Does anyone have insights into the possible cause of this issue? I would greatly appreciate any assistance in solving this mystery. Thank you. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
Hello, Here's the corrected conversion function:
This resolved the remaining errors in my project. I hope this helps anyone facing the same issue. |
Beta Was this translation helpful? Give feedback.
Hello,
I found that adding the derivative of the rotation matrix term resolved the errors.
Here's the corrected conversion function: