How to solve ground reaction force through using weld constraint?

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Simon Jeng
Posts: 87
Joined: Fri Sep 07, 2018 8:26 pm

How to solve ground reaction force through using weld constraint?

Post by Simon Jeng » Sat Jul 04, 2020 8:58 am

Hi, helpers

I'm working on calculating the ground reaction force and moment during squat lifting according to a prescribed motion. I built a weld constraint between each foot and ground in OpenSim4.0. I prescribed a state file(including coordinate value and speed) and inputted it into the 'Analyze tool' to calculate the constraint force. This constraint force represents the ground reaction force.
I found that the resulting constraint force is calculated about one fixed point(i.e. the location of the weld I had specified in the .osim file). However, in reality, the center of pressure is changing during squat lifting so my resulting constraint force is incorrect. How to solve this problem?
Thanks for your help :D

Best wishes

Tags:

User avatar
Carmichael Ong
Posts: 401
Joined: Fri Feb 24, 2012 11:50 am

Re: How to solve ground reaction force through using weld constraint?

Post by Carmichael Ong » Wed Jul 08, 2020 10:50 am

A WeldConstraint shouldn't allow for the two bodies to move relative to each other both in translation and rotation, but it does mean that your calculated resultant force and moment are based on the point you chose for the WeldConstraint.

The resultant force output calculated should not change based on the point chosen. However, the resultant moment (or torque) calculated will change based on the point chosen. You could use the resultant moment and force to calculate where the center of pressure is located.

While the resultant force/moment from your constraints are exactly "correct" for your data, it is likely that they will look "incorrect". One is that the GRF estimated will be based on your kinematics, which tend to be much noisier than force plate data. Also, this motion seems to involve both feet being on the ground, which can lead to problems in calculating this values because it is difficult to know how to split the forces between both constraints.

POST REPLY