[iCubGenova09] Possible problems on the generated model #70
Comments
|
cc @Nicogene |
|
A lof of difference are due to the fact that the order of the links in the two models are different. |
Ideally if we had a script to "normalize" a URDF, probably this would simplify checking the diff. We could also try XML-specific diff tools, that just look at the difference in matching elements, see https://extendsclass.com/xml-diff.html . |
|
@traversaro suggested that the problem has been caused by the fact that the new creo mechanism has been released in order to have iCub2.7 head on iCub3. Consequentially the In theory, we should recheck all the joints direction of the model with the robot. We should reenable the |
|
I've just checked with @S-Dafarra and @prashanthr05 the axis direction by using
I used the 20bb535 commit |
|
cc @Nicogene |
|
I can do that |
Thanks @GiulioRomualdi 👍🏻 |
On Friday we tested the walking controller on iCubGenova09. Before running the controller I updated the model to the latest commit devel 676cb8d. We noticed a strange movement of the swing foot while walking (like one of the joint axis has a wrong direction) .
So we decided to use the model stored in our fork and everything went fine.
In particular, we decided to run the controller with the iCubGazeboV3 model stored in https://github.com/GiulioRomualdi/icub-models/blob/9a1cc2d6adcab85c2f5f1c18fb9acb1b71122fda/iCub/robots/iCubGazeboV3/model.urdf
This is the result of a
gitbetween the two modelscustom_model_iCubGazeboV3.urdf: https://github.com/GiulioRomualdi/icub-models/blob/9a1cc2d6adcab85c2f5f1c18fb9acb1b71122fda/iCub/robots/iCubGazeboV3/model.urdfofficial_iCubGenova09.urdf: https://github.com/robotology/icub-models/blob/676cb8d5e53b791133987d89340253d196de20b6/iCub/robots/iCubGenova09/model.urdfThe differences are really a lot
cc @S-Dafarra @prashanthr05 @DanielePucci and @traversaro
The text was updated successfully, but these errors were encountered: