Expected behavior: Publishing consistent angular velocities in 3 dimensions, even when rotating in gimbal lock rotations (such as pitch = 90deg)
Actual behavior: Angular velocity becomes a very high value when the vehicle rotates to a gimbal lock rotation
Resolution
The angular velocities were calculated using the Euler angles, which can lead to singularities. I was able to resolve the issue by utilizing the quaternion rotation representation of the model. I created a pull request with the fix (linking this issue).
Environment
Description
Resolution
The angular velocities were calculated using the Euler angles, which can lead to singularities. I was able to resolve the issue by utilizing the quaternion rotation representation of the model. I created a pull request with the fix (linking this issue).