Jump Landing Model

Main Forum
1

Hello,
I’m working on making a model of a jump landing. The model loads, but it is statically indeterminate. I’m not exactly sure what the issue is, and was hoping someone could take a look at the model. I’m hoping to compare the output from this model to the output from the new repository model (hopefully coming soon!) w/ gaitapplication2.
Thanks for your help!
Abra

2

Hi Abra

I have tried to load you model but i failed, i think you must be running on a modified repository or a older version?

Anyway i tried to review your JointandDrivers.any file and found some issues that could be related to the error you get.

The model you have contains pelvis and the right leg, so you will have to drive the following dof in the model.

six rigid body dof of the entire system
thee dof of the hip
one dof of the knee
two dof of the ankle

This adds up to 12 dof, but counting yours gives me 10? so you need to drive 2 more.

When driving the model in this way i always do a small stick figure skecth with small arrows showing the driver directions for each marker it helps understanding the model.

Please ask again if this does not help to solve the problem.

Best regards
Søren, AnyBody Support

3

Hi Soeren,
Thanks for your help so far. I see what you mean about the drivers. I had been playing with them because there were too many reactions. Unfortunately, the model is still statically indeterminate. The model probably won’t load because I put in the full filepath to the repository on my computer. I’ve attached another version w/ references to the repository changed. (I’m using version 7.1) I’ve also attached a text file of the model information in case the model still won’t load.
Thanks,
Abra

4

Hi Soeren,
I actually figured out what I was doing wrong. I was using the wrong body model. I switched to TrunkNoMusclesRightLeg and the model isn’t statically indeterminate any more. Sorry for the inconvenience and thanks again for your help!
~Abra

5

Hi Soeren,
I’m still working on the jump landing model, but am having troubles when I try to set the initial conditions. I can’t figure out what I’m doing to cause the kinematic errors. Could you maybe take a look?
Thanks,
Abra

6

Hi Abra,

Sorry for the late answer. I looked at your model, the problem is that you are not driving the appropriate selection of dof, i mean not the good directions in the MeasureOrganizer. It is not easy to know wich dof to drive and for that we very strongly recomend to draw small sketches and visualize the different dof and the effect of driving some of them.
So i made the sketches for you, showing the configuration you did and a new working configuration so that you can compare them and see the differences. In the old configuration I marked in color the redundant drivers (those are a direct cause of the kinematic failling). I hope you will understand the logic in the changes, if not please ask for it.

An other remark is tha in some drivers the Ref menber of the linear measure was not set to 0 (0 means the measure is done in the first local coordinate system). It is important that the measure for this kind of marker driver has the Ref=0.

Ialso attached a new version of the JointsAndDrivers file includeing the modifications. Notice that the feet is a bit twisted at the end, that can maybe be fixed by improving the marker placement. I let this part up to you.

By the way you will excuse my poor drawing skills, but i guess the leg and pelvis segments are recognisable.

Best regards, Sylvain.

7

Hi,
Thanks for your help Sylvain. I definitely see what you mean about the drivers. For some reason I had only thought about the model as a whole instead of thinking about the individual segments as well.
I’ve been playing around with the markers and the mannequin file, but it always seems that I get an activity envelope around 3, with a spike up to 6 at the beginning of the landing. Is this reasonable? I’ve attached the working model as well.
Thanks,
Abra

8

Hi Abra,

The activity you get is quite high. There are some reasons for for that.

  1. As i told you in the last posting the motion of the feet is not very good yet, it get twisted. This is not a natural posture and it will affect the muscle activity. One way to fix it is to modify a bit the marker position on the foot. I did that and it looks better now.

  2. The center of presure is not matching the position of the foot, especialy at the begining of the landing. This should be corrected by an offset to the center of presure location.

So i made those two modification and the activity looks better. The first pic is reduced but still there, that is normal because of the impact. Then the muscle activity is still generaly high between 2and 3. The offset I introduced may not be at the exact right location, you can try to play a bit with it and see if it makes it better. The files are attached.

I’ll see if i find something more later.

Best regards, Sylvain.

9

Hi Abra,

We hope that the help we have provided has helped you finish your modeling work.

As stated in the description of this forum we in change of spending time and effort on working with your model, ask you to submit the model back to AnyScript.org once it has been finished.

Please post the finished model with the a short description in the forum “Gallery of user models and videos” this will be to the benefit of all users in this community.

Thanks in advance,
Sebastian

10

Hello Sebastian,

I apologize for the very late reply. I have been working on the models. Abra and I were working together and she is no longer working on this project. I have taken control of her work. I am not sure how much more was done with the model after she left but you have helped me a great deal with the jump landing model. The model I was working on which is currently being used was the GaitUniMiamiTDRightLeg model which was completed with your help. If you let me know what is required when submitting the model back to AnyBody I would more than happily take care of it.

I greatly appreciate your help. Thank you.

Preet Sabharwal