Page 1 of 1
Inverse Dynamics reference frame
Posted: Wed Nov 02, 2016 9:25 am
by aseel_ghazwan
Hi All,
I'm wondering about the reference point used in OpenSim inverse dynamics calculation? is it in the local body frame or global frame?
According to the definition of the GRF, as shown in the attached screenshot, the force is expressed in "ground". Does this mean that the moments calculated from inverse dynamics are expressed in global frame?
- GRF3.png (28.2 KiB) Viewed 762 times
Thanks in advance,
Aseel
Re: Inverse Dynamics reference frame
Posted: Thu Nov 03, 2016 6:38 am
by m.mauersberger
I think this form you showed is only applicable for the external forces you added.
But when you calculating with the "Inverse Dynamics" tool there are all the joints considered you have in your model. When you look at the "Coordinates" window in OpenSim you see all your coordinates. Thus the results from Inverse Dynamics are expressed in those. Then every force and moment is expressed each in its parent frame.
Greetings
Michael
Re: Inverse Dynamics reference frame
Posted: Thu Nov 03, 2016 6:46 am
by aseel_ghazwan
Thanks Michael for your reply.
What I understand is: the knee adduction/abduction moment, for example, is expressed in femur reference frame, is that right?
Kind regards,
Aseel
Re: Inverse Dynamics reference frame
Posted: Fri Nov 04, 2016 2:43 am
by m.mauersberger
Yes, I think that's right. When you look in the file that is created through Inverse Dynamics analysis there is (for example) a "knee_angle_l_moment" according to the "knee_angle_l" coordinate. This Coordinate is (in my example) defined by the "femur_l" reference frame to account for the angle of the "tibia_l".
Here is an excerpt of the "gait2392.xml" model:
Code: Select all
<Body name="tibia_l">
...
<!--Joint that connects this body with the parent body.-->
<Joint>
<CustomJoint name="knee_l">
<!--Name of the parent body to which this joint connects its owner body.-->
<parent_body> femur_l </parent_body>
<location_in_parent>0 0 0</location_in_parent>
<orientation_in_parent>0 0 0</orientation_in_parent>
<location>0 0 0</location>
<orientation>0 0 0</orientation>
<!--Set holding the generalized coordinates (q's) that parmeterize this joint.-->
<CoordinateSet>
<objects>
<Coordinate name="knee_angle_l">
...
</Coordinate>
</objects>
<groups />
</CoordinateSet>
...
</CustomJoint>
...
</Joint>
</Body>
Here the coordinate is defined that has been described. In the same reference the forces/moments of Inverse Dynamics are calculated.
But yes, you are right.
Best regards
Michael