search

ros-industrial / motoman

ROS-Industrial Motoman support (http://wiki.ros.org/motoman)
146 stars 195 forks source link

Xacro mh12 correction #370

Closed leitoxve closed 3 years ago

leitoxve commented 3 years ago

in https://github.com/ros-industrial/motoman/blob/kinetic-devel/motoman_mh12_support/urdf/mh12_macro.xacro

inverse the flench translate values in xacro line 144 with 149 ORIGINAL 144-origin xyz="0 0 0" rpy="0 0 0" 149-origin xyz="0.100 0 0" rpy="3.1415926 -1.570796 0" to MODIFICATION 144-origin xyz="0.10 0 0" rpy="0 0 0" 149-origin xyz="0.000 0 0" rpy="3.1415926 -1.570796 0"

gavanderhoorn commented 3 years ago

It would be good if you could provide a little more context and explanation here.

Do you have a particular reason you want this change applied?

EricMarcil commented 3 years ago

The standard is that Tool0 represent the flange and it should not be edited. If you are adding a tool to the robot then you should create a new joint with Tool0 as the parent. Technically, the change that you are proposing shouldn't change the result.
The way it is currently done, matches the way Yaskawa does its modeling in our simulator (MotoSim) except that we call Flange what Ros calls Tool0.

gavanderhoorn commented 3 years ago

I'll try not to confuse everyone by mentioning the flange frame (I'll mark my comment as off-topic), but we do have a flange frame. It just wouldn't be aligned with any TCP, but with REP-103 standard orientation of frames. This would be for connection of EEF geometry to xacro macros from robot support packages.

See REP-199 for a WIP proposal (I really need to finish that). Ignore the Vendor Nomenclature Mapping, that will be removed.

For additional tool frames, they should not be made children of tool0, but siblings, and then have flange as their parent.

tool0 must not be changed, and the difference between tool0 and flange would be the orientation only on most robots.

gavanderhoorn commented 3 years ago

I'm going to close this, as I believe @EricMarcil's response should clear this up.

@leitoxve: please feel free to keep commenting on the issue.

If it turns out we should actually change or fix something here, we'll re-open.

leitoxve commented 3 years ago

what I mean is that in real life the robot does not have a washer in the last joint in the middle of nowhere Screenshot from 2020-11-27 15-45-33

EricMarcil commented 3 years ago

Oops, you are correct that shouldn't be there. Sorry, for misunderstanding your initial message. Now that the issue is clear. You suggestion would fix it but to remain consistent with other models, it's probably the origin of the model that needs to be fixed. I'll double check and make a PR to correct the model.

EricMarcil commented 3 years ago

Submitted PR #372 to fix this issue.

gavanderhoorn commented 3 years ago

Thanks for reporting @leitoxve, this was indeed a mistake in the model.

Should be fixed now with the merge of #372.