Hand force measure

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Silvia Buscaglione
Posts: 2
Joined: Fri Dec 20, 2024 1:37 pm

Hand force measure

Post by Silvia Buscaglione » Thu Feb 13, 2025 10:24 am

Hi all,

I am relatively new to OpenSim and I would like to know if I can compute the force between the hand and externa object using Opensim.

Details of the study:
I am working with a dataset of two people interacting with a robotic system and I want to determine the force between the robotic interface and a person given the body's kinematics and force plate data.

So far, I have: (1) modeled the robotic handle in OpenSim as an external body in contact with the hand, (2) performed inverse kinematics, (3) performed the inverse dynamics, (4) ran the "Joint Reaction Analysis" to obtain the joint reaction forces between the hand and the handle.

HHowever, when comparing OpenSim's computed forces with the actual forces recorded from the robot in trials where only one person interacts with it, the results do not match. Right now, I am trying to determine whether this discrepancy is due to an issue in the OpenSim setup (e.g. how the external model of the handle is implemented) or if OpenSim simply cannot provide accurate force estimations in this scenario.

I hope my explanation is clear, and I would appreciate any insights or suggestions.

Thank you in advance,
Silvia

Tags:

User avatar
Jon Stingel
Posts: 35
Joined: Fri Jan 19, 2018 5:54 pm

Re: Hand force measure

Post by Jon Stingel » Fri Feb 21, 2025 3:31 pm

Hi Silvia,

Sounds like a cool problem setup. I think a little clarification might shed light on your discrepancies. A couple of questions:
  • Could you say more about the way that you have modeled the external handle? Is it fixed to ground, and what type of joint is there with the hand?
  • What forces are included in your model, and have you included them in your analysis? Could the discrepancy be a result of the participant gripping the handle? You might imagine that without forces due to the hand gripping the handle, the force you might estimate could be much lower.
Hope this helps,
Jon

User avatar
Silvia Buscaglione
Posts: 2
Joined: Fri Dec 20, 2024 1:37 pm

Re: Hand force measure

Post by Silvia Buscaglione » Tue Feb 25, 2025 1:57 pm

Hi Jon,

Thank you for your response.
  • Regarding the handle, I created a Weld joint with the hand as the parent and the handle as the child.
  • In terms of forces, I have included the force platform data in the following way: I extracted the .mot file from the .c3d file in the CoP form (c3d export with the ForceLocation param set to 1) and inserted the .mot file as an external load in the inverse dynamics as a point force applied to the ground.
I understand your point about the gripping force. However, based on our force recording approach, the recorded forces should only reflect those that generate the system's motion and should not include any components of user gripping force.
I believe the error might be more related to the joint between the handle and the hand. Here is my thought, in case it is useful. The handle should always remain orthogonal to the ground. However, in its current setup, it moves rigidly with the hand body, causing it to deviate from this orientation.

I am not sure what you mean regarding the forces included in the model. Under the force section of my model, I have muscles and actuators, whereas contact forces and other forces are empty.

I hope this clarifies all your questions.

Looking forward to hearing from you,
Silvia

POST REPLY