Update repo #78
Update repo #78
Conversation
|
@Ganimed do you agree with this merge? I worked mostly at the high level (matlab functions) so nothing is changed in your code |
|
Try it, then we will see. Or better make a copy of the WBM-Class branch and merge it there. |
|
@Ganimed if you prefer, I can merge |
|
|
||
| references = trajectoryGenerator(CONFIG.initForKinematics.xCoM,t,CONFIG); | ||
| CONFIG.figureCont = visualizeSimulation(t,chi,CONFIG,references); | ||
| CONFIG.consideredJoints.push_back('torso_pitch'); |
There was a problem hiding this comment.
This should be automatically extracted by the used joint lists, but at the moment that is not exposed in mex-wholebodymodel, so I see why you had to hardcode this.
| end_time = length(qj(1,:)); | ||
|
|
||
| % load the model from .urdf | ||
| mdlLdr.loadReducedModelFromFile('model/model.urdf',consideredJoints); |
There was a problem hiding this comment.
This should be exposed by the C++ to be sure that we are loading the same model.urdf of the underlying model.
There was a problem hiding this comment.
But for doing that we should first make sure that we install the meshes together with the urdf, something that at the moment we are not doing (see robotology/icub-models-generator#28 and robotology/icub-models-generator#33).
Added diagnostics messages on integration error and increased gain for integration correction on SO(3)
Added the possibility of passing to `quaternionDerivative.m` the gain for integration correction of quaternion
|
updateRepo will be merged into master so no need for this pull request anymore |
I think it is time to merge the WBM-class into master. But before doing this, it is necessary to verify the controllers are still working properly, and that's why I want to first merge this branch into WBM-class.