The effect of exo's mass on kinetic

[Felix (Kurt Felix)]

Hi everyone,
I'm trying to explore the effect of hip exo's weight distribution on human kinetic. I have driven the coupled human-exo model successfully based on the collected marker motion data and GRF of human walking without exo. But there are some issues that confused me.

• First, about the Reaction.Type setting of Human-exo connection. I switched the Type Off for kinetic analysis, and created CreateAnyGeneralForce6DOF for all connection. But the hip joint flexor moment was smaller with exo than without it while no external assisted joint moment was added. It's contrary to common sense.

  AnyFolder LeftThighConnection = 
   {  
     CreateAnyGeneralForce6DOF Lthigh_Forces(
     TargetFrame = Main.EnvironmentModel.exo.LeftThigh___1.thigh_connect,
     BaseFrame =Main.HumanModel.BodyModel.Left.Leg.Seg.Thigh.LeftThighConnectNode)= 
     {
       strength_Fx = 1000;
       strength_Fy = 1000;
       strength_Fz = 1000;
       strength_Mx = 1000  ;
       strength_My = 1000  ;
       strength_Mz = 1000  ;
     };
   };

With Exo:

No exo:

• Next, I'm not sure whether it's reasonable to drive the coupled model with same GRF. Is using predictedGRF with same marker motion data more suitable for this project?

Thank you in advance.

Best regards,
Felix

[Felix (Kurt Felix)]

Does anyone know anything about this?

[Bjorn (Bjørn Keller Engelund)]

Hi @Felix

So it a little tricky you question, but I don't think your results are necessarily wrong.

When you set the six strengths to 1000 it is essentially the same as using a AnyReacForce on the six DoF. Furthermore, if you do not have any other constraints on the exo, you will get the same result no matter what you set the strengths to since there will be only one solution for the solver to transfer the forces to the thigh.

Now for the drop in hip joint moment that can be correct dependent on your model and motion.
You can have a motion where the added exo mass will work opposite the GRF and thereby decrease the hip load. It is also likely that the "hand of god" aka. pelvis residual forces will increase when you apply the exo.

Lastly, there is definitely a error source in using the GRF from a trial without the exo. The error magnitude is probably dependent on the exo mass. I would the GRF prediction tool to see what that results in.

Best regards,
Bjørn