IK problems

[Gary Lu]

Dear all:
I'm processing simulations on the arm26 model. I finished the scaling steps but then came into some unexpected mistake during IK steps.
Here are my operations so far, please tell me if i have anything go wrong:
1. To meet the sensor positions of my experiment data, I change the marker to desired positions and saved settings as markersets.
2. Generate the Setupfile for scaling. And I choose to do measurement-based scaling with the .trc file for static and moving trajectory.
3. Next, I used scaling tools to successfully generate a ideal replica since the experiment model i use is identical to the opensim one.
4. Then generate the IK setup file while other required files are generated after scaling.
5. Then I used IK tool to see if the model's movement fit the experiment movement. But it turns out to be the IK result shows that the arm is remaining still instead of swinging in the expected way.

I first thought it was because the insufficient number of markers on each body which the instructions on the website recommended to be 3.
So I tried again with 3 markers on each body segments but it came up with the same results where the error remained very big. What's more, I even used the pre to see if my .trc files work in opensim or not, fortunately it works !
I don't know where the problem is, so if anyone could help me would be very thankful.All files are attached below.

P.S. I used ansys to generate the trajectory of the arm-swinging-simulations so the marker is first arrange in the same plane.

[Dimitar Stanev]

You can try visualizing the marker data (File -> Preview Experimental Data). It always helps in pinpointing the problem.

A common mistake might be that the marker trajectories are displaced from the origin and the model. This causes problems when the model is not permitted to translate and rotate with respect to the ground frame (fixed base). If this is the case, then you can add 6DoF between the ground and the thorax body so that the model can position itself near the measured marker positions and then try to better fit the data.

[Gary Lu]

Professor Dimitar Stanev, thanks you so much for answering my question. However, I eventually found out that it was because my misplacement of the markers frame so the data couldn't fit together. Now the IK tool works and can generate the trajectory as expected. Thanks again.

[Kyler Howard]

Gary, could you explain how you both figured out and fixed what your issue was? I am having a similar problem with my IK errors, but haven't heard of or considered it being an issue with the marker frame.

Thanks in advance,
Kyler Howard