CMC dont track ankle

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Anders Sandholm
Posts: 30
Joined: Thu Feb 22, 2007 1:53 pm

CMC dont track ankle

Post by Anders Sandholm » Sat Sep 04, 2010 1:40 pm

Hi,

This is a rather strange problem with CMC. IK and RRA gives a good solution for the ankle (motion is gait), now when running CMC and then looking at the joint coordinates (xxx_Kinematics_q.mot file) the right ankle is not tracked. The left one is, but the right one has a very bad solution (its the foot which has a GRF).

So i turned up the wight for the right ankle, but nothing happens, the ankle do not track the motion. All other joints are good.

Has anyone seen this problem? (due to this I have a huge residual for the right ankle) What can i do to make the model track the IK (RRA) better? Could the kp/kv value affect the tracking of the joint?

Any suggestion is much appreciated.

cheers

User avatar
Edith Arnold
Posts: 44
Joined: Fri Apr 06, 2007 2:07 pm

RE: CMC dont track ankle

Post by Edith Arnold » Wed Sep 15, 2010 1:55 pm

Just a thought, but what's going on with the subtalar and mtp joints in your model? I lock them during CMC for walking since there aren't enough muscles to keep them under control. Problems there could affect the ankle.
-Edith

User avatar
Lode Bosmans
Posts: 12
Joined: Sat Feb 14, 2009 9:36 am

Re: CMC dont track ankle

Post by Lode Bosmans » Fri Mar 16, 2012 7:53 am

Dear all,

I seem the have the same problem. I have gait data of children with an abnormal gait pattern. In contrast with the previous post, I have problems with the tracking of the ankle_angle of the leg in swing phase.

The IK and RRA procedures give good results. When looking at the results of the cmc, the tracking of the ankle_angle of the leg in swing phase is already bad from the beginning of the CMC output …-kinematics_q file. When continuing further in the kinematics_q file, the tracking of knee_flex and hip_flex of the leg in swing phase also becomes worse, eventually leading to a total flip of the model, affecting all DOF’s.

I have checked the next things:
- Comparison between IK and RRA is good.
- None of the DOF’s reach their limits before the model starts crashing.
- None of the residual forces reach their limits.
- None of the reserve actuators reach their limits. The reserve ankle_angle actuator is still capable to produce more force. But for some strange reason, it will not do so..
- Some muscles do reach the maximal activation of 1 in the control file. But as said before, none of the reserve actuators reach the maximal force, so the reserve actuators should be able to compensate for the muscles who are not able to produce more force.
- Changing the weights of the ankle_angle in het CMC_tasks has no effect what so ever.

Is anyone familiar with this problem of does somebody know in which direction I have to search to find the cause/solution?

Thanks in advance for your answer!

Kind regards,
Lode

User avatar
Marije Goudriaan
Posts: 5
Joined: Wed Apr 11, 2012 10:40 am

Re: CMC dont track ankle

Post by Marije Goudriaan » Wed Jul 25, 2012 2:13 am

Dear all,

I am encountering the same problem as well. Has anyone found a solution to this problem?

Kind regards

Marije Goudriaan

User avatar
Ayman Habib
Posts: 2237
Joined: Fri Apr 01, 2005 12:24 pm

Re: CMC dont track ankle

Post by Ayman Habib » Thu Jul 26, 2012 2:46 pm

Hi Marije,

Please consult the documentation/troubleshooting page below on the new support website and let us know if you continue to have questions.

http://simtk-confluence.stanford.edu:80 ... leshooting

Best of luck,
-Ayman

User avatar
Marije Goudriaan
Posts: 5
Joined: Wed Apr 11, 2012 10:40 am

Re: CMC dont track ankle

Post by Marije Goudriaan » Thu Aug 09, 2012 2:00 am

Dear Ayman,

Thanx for your response!
I am interested in arm swing, so I deleted the leg muscles from the model. However, without the leg muscles and with only the leg actuators the CMC runs with large errors in the ankle angle. I did read the documentation and troubleshooting about one hundred times, but that did not help ;) . Finally I put the leg muscles back in the model, adjusted the task-, constraints- and actuator files and now it works perfectly.

POST REPLY