You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A guide on how to use the ur_desciprtion physical_parameters.yaml to calculate the inertia of specific arm model. This can help cross-checking the effort data published on the /joint_states topic.
Right now, I use effort from topic /joint_states directly to calculate the workdone for the motion or the trajectory planned. However, as I am not so familiar it, I am not sure how this effort is calculated and packed into the msg. Thus, I would like to calculate the workdone using torque which is calculated from inertia * angular acceleration which is the dv/dt. I know this might have a slight difference for the calculated result.
I am using ur5e now and here is the parameter I can find:
Thank you for your interest in this topic. I'm afraid, I cannot give you a full answer to everything, but to address some of your questions:
What is upper_arm and forearm? And why the elbow is not being called?
Inertias are calculated for the links, not the joints. The elbow isn 't a link in the kinematic chain. Upper arm and forearm are the two main tubes respectively. If you look at the model in RViz, you can easily switch the individual links on and off.
What geometry should be considered when calculating the inertia? inertia wiki Should I use a hollow cylinder or a solid cylinder?
The value is mismatch with the official DH table here
I'll add ros-industrial/universal_robot#504 as a comment to both questions. Inertia calculation is a bit more complicated than just the bodies, as motor inertia has to be taken into account, as well. We haven't had a deeper look into this, so far, so any help on this is highly welcome.
Actually, I am conducting a study (master course) on the energy consumption of a planned trajectory. I am using UR5e with ursim to conduct the research. I found out there is a topic called /joint_states which provides the effort of each joint? or link.
Would you mind guiding me on which files/nodes pack those effect data and publish it to this topic? I would like reverse engineering to find out how it has been calculated.
The reason why I would like to re-calculate the moment of inertia is to double verify whether the published effort is trustable/close-to-accurate or not.
Feature summary
A guide on how to use the ur_desciprtion
physical_parameters.yaml
to calculate the inertia of specific arm model. This can help cross-checking theeffort
data published on the /joint_states topic.Right now, I use effort from topic /joint_states directly to calculate the workdone for the motion or the trajectory planned. However, as I am not so familiar it, I am not sure how this
effort
is calculated and packed into the msg. Thus, I would like to calculate the workdone using torque which is calculated from inertia * angular acceleration which is the dv/dt. I know this might have a slight difference for the calculated result.I am using ur5e now and here is the parameter I can find:
Questions:
upper_arm
andforearm
? And why theelbow
is not being called?Related issues
N/A
Tasks
To complete this issue involves
Great thanks for any ideas!
The text was updated successfully, but these errors were encountered: