Unstable Surface Deformation and Sticking Artifacts in Probe–Breast Contact Simulation

[zhanglepy]

I am working on a simulation where a rigid ultrasound probe scans and compresses a soft breast phantom modeled with a tetrahedral corotational FEM force field. The setup includes collision detection between the probe and the breast surface (triangle/point collision models) and a predefined SE(3) probe trajectory updated at each time step.

During the simulation, I observe two main issues that do not match realistic ultrasound scanning behavior:

1、When the probe compresses the phantom, the surface nodes of the breast exhibit rapid, high-frequency oscillations instead of smooth quasi-static deformation.

2、The probe often appears to “stick” to the surface nodes, resulting in unrealistic adhesion during scanning.

I would be very grateful for any advice or best practices on how to stabilize the surface deformation and reduce unrealistic sticking behavior, for example by tuning contact parameters, damping, time step, or by modifying the way the probe’s motion is applied. Thank you very much in advance for your guidance.
The sofa is 23.06. thank you for your advice!

The code:
code.zip
Thank you very much for your suggestions

[zhanglepy]

[bakpaul]

Hello !

Firs of all, but you knew I would say it, v23.06 is old and actually not supported anymore. We might have fixed stability issues in the recent release so please if you can, try updating your SOFA version to the latest release (v25.06)

I will have 6 suggestion to start with, then if those doesn't help we can see other strategy:

1. Make sure the normals of your collision models are well oriented
2. educe the friction parameter of the contact. For this you need to add a parameter to your DefaultContactManager: responseParams="mu=0.3"
3. Try increasing the contact distance of your intersection component
4. I would only use point to triangle collision. And furthermore, it could be even faster and stabler if you used point for the probe and triangles for the breast (given the size of the breast triangles w/r to the probe ones)
5. Try another proximity intersector (the NewProximityIntersection for instance)
6. If none of the above worked, try reducing timestep ?

Hope that helps !

[zhanglepy]

Thank you very much for your prompt reply and suggestions. I will try to make targeted modifications according to your suggestions and hope that this problem will be improved. Thank you again.

[hugtalbot]

Thanks @bakpaul 👍 Keep us posted @zhanglepy

[github-actions[bot]]

⚠️ ⚠️ ⚠️
@zhanglepy
Feedback has been given to you by the project reviewers, however we have not received a response from you. Without further news in the coming weeks, this discussion will be automatically closed in order to keep this forum clean and fresh 🌱 Thank you for your understanding