What are the differences between "Spatial Contact Force" block in Simulink/Simscape Multibody Library and "Simscape Multibody Contact Forces Library", if any?

Hello,

Is there any difference between "spatial contact force block" in Simulink/Simscape Multibody Library and "Simscape Multibody Contact Forces Library"?

• Spatial Contact Force Block: https://www.mathworks.com/help/physmod/sm/ref/spatialcontactforce.html
• Simscape Multibody Contact Forces Library: https://www.mathworks.com/matlabcentral/fileexchange/47417-simscape-multibody-contact-forces-library

As far as I see, both of them works in the same manner however Spatial Contact Force block is a plug-and-play block whereas Simscape Multibody Contact Forces Library requires to determine the possible contact types in advance and requires to use related contact-type blocks accordingly.

Are there any advantages or disadvantages of one over the other?

Let me kindly mention you please @Steve Miller.

Thanks in advance.

Best regards,

Hakan

Answer 1

[Steve Miller]

Both are intended to help users model contact forces in Simscape Multibody models. If the Spatial Contact Force meets your needs, I highly recommend you use that block.

The Simscape Multibody Contact Forces Library was created long before the Spatial Contact Force block was added to Simscape Multibody. We created that library to examine use cases and prioritize features for the Spatial Contact Force block. We are maintaining that library on the File Exchange because it is an example of how you could create your own custom contact force law and custom friction law. If you are stuck in a release prior to when the feature you need was added to the Spatial Contact Force block, you could use that library.

Users of the Simscape Multibody Contact Forces Library should be able to (relatively) easy upgrade their models to the Spatial Contact Force block. There are a few contact types that are not (yet) direcly supported by the Spatial Contact Force block (inner surfaces of cylinders, cones, and spheres) but there are new options being added reach release so keep checking.

--Steve

[Hakan Ertugrul]

Dear Steve,

Thank you very much indeed for your prompt and detailed answer.

I am trying to model a peg-in-hole task with robotic manipulator assembly and I think spatial contact force block meets my needs. I am modeling the hole with 4 brick blocks for each side (because of convex hull) and place spatial contact force blocks between each of these blocks and the peg. The picture of the hole is below. I hope it will work. If you have any suggestions, I will be more than happy to get.

Best regards,

Hakan

[Steve Miller]

This looks good to me! Depending on the degrees of freedom involved (if the peg, for example, cannot rotate), you might even be able to get away with just Joint Limits along some degrees of freedom. The option you have chosen is the most general and should work fine.

[Hakan Ertugrul]

Perfect. Thanks for your encouragement :-)

I will start with no rotation degree of freedom to be simple as you mentioned but I plan to generalize with 6-DOF.

I think I should actuate the peg with force, leave the automatic calculation for motion, otherwise peg can penetrate the solids of holes, right?

I also plan to measure force/torque values at the tip during contact. Can I do it with 6-DOF Joint or Bushing Joint because there is obviously no force/torque measurement sensor like "Transform Sensor"? Or do you have any suggestion for measuring the reaction force/torques in vectoral form?