corrected dimensions and positions of inertias
I'm by far not an expert in working with gazebo or inertias, but it seemed wrong to me: - The upper arm inertia of the ur10 is not centered in the visual arm segment - CoM in the wrist links don't sit inside the correct links. E.g. wrist1 has its CoM inside the end of the forearm for all robots. - Because of the second point the inertia's geometry of wrist3 is matching wrist2 instead of the actual moving part of wrist3. - Wrist dimensions of ur5 were completely off. - On the ur5e the arm inertias weren't centered in the visuals.