[faichele]

Hello!

I encountered the same problem, ending up Fixing it locally by patching the PluginManager class to take the debug suffix in the file names for libraries (“_d”) into account when searching for the .dll/.so files.

I can submit a pull request for that if that is helpful?

With best regards,
Fabian

[faichele]

Hello, Lekan, Shuo and Hugo!

I’m getting around to looking at your pull request only now, after returning from holidays and after the work mayhem before.

Could you please let me know:
– What ROS version(s) you are using to build the ROS connector plugin against?
– If you installed ROS from source, or from Debian/Ubuntu packages provided by the OSRF?

Thanks a lot!
With best regards,
Fabian

[faichele]

Hello, Vlad!

A few years back, I encountered the same problem as you described (force vector containing only zeros, externalForce vector just empty) when trying to extract contact forces from a SOFA scene.
I did manage to obtain contact forces between objects by directly tapping into the “raw” results that were produced by the SOFA solver component(s), but this naturally involved modifications to the source code of some SOFA core classes.
I can see if I could easily isolate the changes I made for this at least, if this helps.

With best regards,
Fabian

[faichele]

Hello!

As an alternative, you might want to consider using ROS to couple your RL environment with SOFA, using the SOFA ROS connector plugin. I have worked on the Neurorobotics Platform (https://neurorobotics.net/) in the past, where ROS served as middleware to integrate Spiking Neural Network simulators with robotics simulations. The ROS connector for SOFA offers a similar possibility, in that you have a convenient way to exchange data between a SOFA simulation and an external (Python-based) framework.

With best regards,
Fabian

[faichele]

Hello, Rishabh!

The MechanicalObject class has functions that apply the *initial* transform to an object: applyTranslation and applyRotation. Combined with the various getObject functions from the BaseObject class, you should be able to enumerate all MechanicalObjects in your scene and customize their initial positions before you step the simulation for the first time.

With best regards,
Fabian

[faichele]

Hello, Hugo!

Indeed, the removal of the SOFA_FLOAT/SOFA_DOUBLE CMake options and the associated impact on the Rigid<n>, Vec<n>, ExtVec<n> types was indeed the issue. Removing the specializations for 3f/3d in the template parameters for the mapping did the trick, I got the plugin code to build successfully again.

Thank you very much!
With best regards,
Fabian

[faichele]

Hello, Adam!

I have been using SOFA for robotics-related simulations for a couple of years, starting during my Ph.D. thesis.

Regarding 1., the simulation of tactile sensing: I have had a similar discussion with a roboticist in the context of the Human Brain Project in 2017; depending on your objectives, this is in principle doable with SOFA, as long as your deformable collision models are geometrically not too complex.

Regarding 3.: There is a ROS connector plugin I have recently made publically available. If you like, you can give it a try. You should however be aware that the plugin still depends on a slightly customized SOFA version; I intend to make it compatible with the default SOFA codebase in the coming weeks.

Thank you very much!
With best regards,
Fabian Aichele

[faichele]

Hello, Hugo!

I’ve only tried a test scene for the custom class derived from the ArticulatedHierarchyController with the 18.06 SOFA release so far. I’ll try digging up the original demo scene from an older SOFA version and try it with the 18.06 release.

Thank you very much!
With best regards,
Fabian

[faichele]

Hello!

You can contact us via the SOFA consortium or via one of the contact possibilities given on our homepage if you are interested in testing the ROS connector.
The first user of the plugin is the DEFROST team.

With best regards,
Fabian Aichele

[faichele]

Hello!

There is a ready-to-use SOFA ROS connector plugin, written and maintained by our team.
We have provided it to an INRIA team a few weeks ago, because the plugin is still in the “beta-testing” stage. If you are interested in testing it as well, please let me know.

With best regards,
Fabian Aichele
TruPhysics GmbH

[Author]

Posts