Lorenzo Rapetti

This is not tested because the tests are commented out. (See https://github.com/robotology/idyntree/blob/master/src/estimation/tests/BerdyHelperUnitTest.cpp#L337-L341)

> In general take everything you find in the existing inverse kinematics class with a grain of salt, see #282 If it's the case, I will try to port some...

Regarding this kind of performance analysis, I think it would be to worth to have cpp interface for reading the [Statistics](https://www.coin-or.org/Doxygen/Ipopt/classIpopt_1_1SolveStatistics.html) of the IK? For my understanding, at the moment...

> It seems `iCubGazeboV2_5` does not have `l_sole` link. I took the model inside my `robotology-superbuild/install`folder. @lrapetti do you have the same error? No, I don't see that error, I...

> > It seems `iCubGazeboV2_5` does not have `l_sole` link. I took the model inside my `robotology-superbuild/install`folder. @lrapetti do you have the same error? > > No, I don't see...

> Can't we use the link names instead? Those are much less ambiguous. See also #805 . Yes, it can be done using the option `linksToModify` in place of `linkIndices`....

I agree with @nunoguedelha that removing link indices from whole iDynTree might be limiting, but I think the discussion here was strictly related to the interface in the matlabWrapper visualizer....

> Open Robotics [currently supports macOS](https://ignitionrobotics.org/docs/fortress/install) even though I cannot guarantee that `scenario` will work ootb. Following the ignition gazebo installation, I was able to install both `scenario` and `gym-ignition`...

Still, I am experiencing the command line tool mentioned in https://github.com/robotology/gym-ignition/issues/100#issuecomment-552233722. It looks like the problem is still present in the ignition version I have installed. I was able to...

> Can you please report if the following command executed from a normal terminal succeeds, and if it does, its output? > > ``` > ign gazebo --versions > ```...