Michael Görner

Thank you for tracking this down @exo-core ! It works. @christian-rauch It's rather annoying that this is not the default behavior upstream yet. Can't you just squash-commit and maybe add...

> I guess we are a little spoiled from Gitlab, which has a "squash commits" option in the merge menu The same is true for github. Also, unless the pr...

Thanks for the feedback! I played around a bit more with the longest_valid_segment_fraction (which really should not be called "resolution" in the GUI, as a higher "resolution" actually makes checks...

> > `kinematics_solver_search_resolution` doesn't seem to have any effect > > In `bio_ik`, the stop criterion can be configured using the parameters `drot`, `dpos` and `dtwist` (default: `1e-5`). As far...

It might make sense in some cases to have the additional tolerance in rad though you would need some other objective to get a gradient in this area if you...

I can't reproduce this on a custom non-debian build of ROS either. > Load a urdf, e.g. roslaunch pr2-description upload_pr2.launch This is missing `/robot_description_semantic` et. al. to load a MoveIt...

> It is also not a problem with my setup, I can reproduce the issue in a docker container. > I attached an archive containing the reproducible example: [minimal_example.tar.gz](https://github.com/TAMS-Group/bio_ik/files/6021725/minimal_example.tar.gz) Thanks...

> :+1: I also have this problem when using python bindings that use bio_ik Feel free to look into this. I planned to investigate it, but did not find the...

Thank you for finding the cause of the issue. > ``` > -static-libgcc > -static-libstdc++ > ``` > > Why link to these statically? Probably the same reasons why you...

That is true. We worked around this before by adding (secondary or a second primary) target goal to avoid joint displacement. This works best with a custom `computeCartesianPath` loop though....