Investigate if it makes sense to port matlab-multi-body-sim in wbc

[gabrielenava]

I would like to port matlab-multi-body-sim in the wbc, but before I need to understand the ratio between effort and benefits

[gabrielenava]

It makes sense to me in general, but there are two blocking points:

1) the best solution would actually be to include Gazebo as a Simulink block to simulate the system's dynamics, or maybe just implement the forward dynamics integration inside Simulink. In this way controllers are not duplicated in Matlab.

2) the iDyntree visualizer requires to have a modified set of meshes in order to be able to see the robot visualization in Matlab. Also the way to automatically point the urdf models stored in the icub-models repository has to be designed, as at the moment I just have a copy of these models in the matlab-multi-body-sim repo.

So to me it is a NO, as:

• it requires a lot of work;
• if I have to spend a lot of energy, I would prefer to do it for having forward dynamics integration/Gazebo included in the Simulink blocks, as it is something other people in the lab are interested in (cc @fjandrad, @diegoferigo )

therefore, closing the issue.