A question about RRA

[Simon Jeng]

Hi,

I am researching squatting with OpenSim. I input measured kinematic data and ground reaction forces into scaled model to run RRA. After that I find that the model presents unexpected motion.

Its motion is squatting but also sliding on the ground. Also, it swings in the coronal plane. What causes this?

I think that a video could make you have a better understanding of my question, but I do not know how to upload it.

Best wishes,
Simon

[Dimitar Stanev]

Hi Simon,

Have you tried increasing the tracking task weights of RRA? Also, note that RRA may not be appropriate for squatting tasks. RRA is typically used for walking and running. It is not a necessity in the analysis pipeline.

[Simon Jeng]

Hi Simon,

Have you tried increasing the tracking task weights of RRA? Also, note that RRA may not be appropriate for squatting tasks. RRA is typically used for walking and running. It is not a necessity in the analysis pipeline.

Hi, Stanev

Thanks a lot. I tried to increase some tracking task weights and the unexpected motions disappeared. But I am not clear about why RRA is not appropriate for squatting tasks. Could you tell me?

[Dimitar Stanev]

In principle you could use RRA to reduce the residuals when the subject is in touch with the force plates. In the user documentation:

https://simtk-confluence.stanford.edu/d ... +RRA+Works

it is mentioned that RRA is primarily intended for walking and running. I am not familiar with the internal implementation of RRA to argue about this, but I would be a bit skeptical of the algorithm's generalization to other movements.

Maybe people that have used RRA can share their opinion on whether it is good practice to use it for movements other than walking and running.

Thanks!

[Quinten Humphrey]

In principle you could use RRA to reduce the residuals when the subject is in touch with the force plates. In the user documentation:

https://simtk-confluence.stanford.edu/d ... +RRA+Works

it is mentioned that RRA is primarily intended for walking and running. I am not familiar with the internal implementation of RRA to argue about this, but I would be a bit skeptical of the algorithm's generalization to other movements.

Maybe people that have used RRA can share their opinion on whether it is good practice to use it for movements other than walking and running.

Thanks!

I am trying to execute RRA and the subject stands on the force plates during pick and place tasks. Does that cause complications since the subject never steps on the force plates?