joint reaction force against lenght scaling

I ran the GaitLowerExtremity model (4.2). I want to see the effect of the length of thigh to the hip contact force. However, even if the right thigh was set as a lenght of 1.2 * the length of the left thigh, the hip contact force of the right joint is still the same as the model with the original parameter.
I did not ran MotionAndParameterOptimizationModel.

If the model is driven by the maker movements, why can I ran the modified model? (If the model is modified, the joint of the model and the joint in the c3d file may not be at the same point)

Please could anybody give me some detailed clue.

Thanks in advance!


I guess AnyBody performe optimization before running the inverse dynamic for the GaitLowerExtremity model, even though I do not performe MotionAndParameterOptimizationmodel.

If so, how can perform the GaitLowerExtremity model with a modified length of thigh? (small modification, perhaps 5%) ?

Hi Jerry,

When running the inverse dynamic analysis in this gait model it will read in a file containing the optimized bonelengths.

I think the name of the file is xxxx-optimizedParameters.txt, these values overrides the values the model is loaded with.

If the motion optimization has not been done this file is not changed, and you overwrite with old values.

This is why you do not see any change in the model results.

The “proper” way to do this would be to change the segment length, switch off the optimization of the segment lengths in the ModelSetup file, then run the motion optimization, it would then try to fit the subject to the motion data.

Then when running the inverse analysis it would have the new size.

Best regards


Thank you for your kind reply. I do not quite understand what is motion optimization. As you said, I should turn off the optimization of bone length, but need to run motion optimization. Sould I turn off MotionAndParameterOptimizationmodel?

Thank you.


Hi Jerry,

In the file ModelSetup file there is a variable called ThighLengthOnOff, just set it to Off like this:


Then input the length value you need in the TrialSpecificData file (here again there is a variable called ThighLength), and run the normal optimization. The thigh length will not be changed.

Best regards, Sylvain.

Thanks. But the model still tried to optimize the tight length when I turned that off and run inverse dynamic directly.

I also tried to run MotionAndParameterOptimizationModel with a modified thigh length, but the optimization stoped with an error “[SIZE=1]Kinematic analysis failed in time step 0”.

Hi Jerry,

Running the inverse dynamic directly is not the correct way. First you have to run the optimization using the “MotionAndParameterOptimizationSequence”.

Now if the optimization fails with the modified thigh length it may be because it is too different from the original size, or from the other side.

Best regards, Sylvain.