robotiq_s_model_articulated_msgs vs robotiq_s_model_control.msgs

[shenlirobot]

What are the differences between robotiq_s_model_articulated_msgs vs robotiq_s_model_control.msgs? Is the articulated one a different gripper hardware from the regular one?

I am trying to simulate the gripper in Gazebo by sending robotiq_s_model_articulated_msgs to /SMRobotOutput. However, when dealing with the real robot, I have to send robotiq_s_model_control.msgs to /SMRobotOutput. Is there a gazebo simulator for the non-articulated gripper?

Sorry to ask this stupid question here. I could not find any relavant information elsewhere

[shaun-edwards]

@Shentheman, thanks for asking. Documentation is certainly lacking. Unfortunately, I'm not familiar with using Robotiq with Gazebo, so I can't answer your question. Hopefully another user will be able to answer this.

[wxmerkt]

The messages have the same content and thus the same md5 hash. You should be able to use them interchangeably. I definitely agree that this collection of packages requites some TLC and happy to put some time in if we can then move it towards being released on the build farm/as debians - e.g. unify the S-Model messages for control and articulated as robotiq_s_model_msgs and release them (they are e.g. required on clients that do not require the drivers and the whole meta package to be built).

$ rosmsg md5 robotiq_s_model_articulated_msgs/SModelRobotInput
4d0701156e580a420c48833f57bc83f3
$ rosmsg md5 robotiq_s_model_control/SModel_robot_input
4d0701156e580a420c48833f57bc83f3

[shaun-edwards]

@wxmerkt, thanks for helping answer. I totally agree with the improvements that you suggest. The lack of progress is due to a lack of time on my part. I'm happy to accept PRs submitted by the community.

[wxmerkt]

On another note, I just noticed that the folks at Clearpath noticed that a field in the SModelOutput was missing and added it (incompatible with this repository/upstream):

https://github.com/DualUR5Husky/robotiq/commit/73e5219cfab7d9d19f3c1a6b76548c1288ff9b76

[christian-rauch]

I just encountered the same issue and mixed up the robotiq_3f_gripper_articulated_msgs and robotiq_3f_gripper_control. While the "input" messages have indeed the same hash, the "output" messages differ:

$ rosmsg md5 robotiq_3f_gripper_articulated_msgs/Robotiq3FGripperRobotOutput 
d0c643ca7dd9fdb2d935922bf6659387
$ rosmsg md5 robotiq_3f_gripper_control/Robotiq3FGripper_robot_output
31ba91390a569c669af204c3d006a806

since the rGLV is missing in robotiq_3f_gripper_articulated_msgs as @wxmerkt pointed out.

I think it would make sense to go for a single set of messages for the very same purpose. I don't see the point in having twice the same messages. I am in favour of using the robotiq_3f_gripper_articulated_msgs since they already contain documentation about message content.

Would you accept a PR that replaces the robotiq_3f_gripper_control messages by robotiq_3f_gripper_articulated_msgs (including rGLV)?