Loaded Walking Model RRA problem

[Ali Khalilianmotamed Bonab]

Hi,
I am recently working on LOADEDWALKING model, I faced a problem in RRA that I could not fix it. https://simtk.org/frs/?group_id=1082

Based on the model of loaded walking, there are two different coordinates named "knee_angle_r_beta" and two constraints on them which I do not understand why it is added to this model and I would appreciate if someone can explain them.

Furthermore, When I am performing, first, scaling tool and then RRA tool on this model without any modifications on the model setting, the model cannot track the kinematics and gives a high error, the only error that RRA gives to me is mentioned below and the model's behavior has been attached as a video. I would be grateful if someone can help me to fix this problem.

opensim_video.zip

(749.94 KiB) Downloaded 25 times

Additionally, when I am disabling the constraints, this error is not appearing and kinematics tracking error is decreasing, However, the error is still remarkable.

Code: Select all

AssemblySolver::track() attempt Failed: SimTK Exception thrown at assembler.cpp:973:
  Method Assembler::track() failed because:
Unable to achieve required assembly error tolerance.
Assembly error tolerance achieved: 8.4324322211032546e-09 required: 1e-10.
Model unable to assemble: AssemblySolver::assemble() Failed: SimTK Exception thrown at assembler.cpp:897:
  Method Assembler::assemble() failed because:
Unable to achieve required assembly error tolerance.
Assembly error tolerance achieved: 8.4324322208105915e-09 required: 1e-10.
Model relaxing constraints and trying again.

[Ali Khalilianmotamed Bonab]

I solved this issue by performing the following tasks:
1) Disabling the constraints,
2) Locking the knee_angle_r_beta and knee_angle_l_beta,
3) Changing the weight of lumbar coordinates weight in RRA task,
4) Increasing the optimal force of coordinate actuators of lumbar coordinates.

However, I still do not understand why the LoadedWalking group introduced the knee_angle_r/l_beta coordinates to the system while even they are not defining actuators on them on RRA and CMC actuators. This does not make any sense here.

Secondly, now I changed some properties on the system and I do not know is it a correct process of fixing this problem or not, And also, I do not know how they will affect on the results of the Joint Moments.

I am still waiting to get a reasonable reply from the OpenSim and especially, the Loaded Walking project/paper authors.

Sincerely,
Ali.

[jimmy d]

However, I still do not understand why the LoadedWalking group introduced the knee_angle_r/l_beta coordinates to the system while even they are not defining actuators on them on RRA and CMC actuators. This does not make any sense here.

The model used for that project is the [urlhttps://simtk.org/projects/full_body/]Rajagopal gait model[/url]. There is a description of how the knee was modeled in the paper

Secondly, now I changed some properties on the system and I do not know is it a correct process of fixing this problem or not, And also, I do not know how they will affect on the results of the Joint Moments.

Neither do I. Each model is different and each decision you make depends on the research question you are trying to answer. The only way to figure that out is to perform an analysis of the model and the modifications to examine the any differences in variables that you feel are important. This paper may be useful to you

[Ali Khalilianmotamed Bonab]

Thanks Dr. James Dunne,
I will go over the papers,
Thanks for your attention.
Sincerely,
Ali.