How to correct DH parameters

[cxj666]

When I used the DH parameters to model the robotic arm and perform the positive kinematics analysis, I found that different DH parameters may cause different positive kinematics results, and I would like to ask how to obtain the DH parameters of UR5 in coppeliasim.

[coppelia]

Hello,

you can use the Denavit Hartenberg extractor add-on

Cheers

[cxj666]

Maybe you don't understand what I mean, I use the DH parameter extractor that comes with coppeliasim to extract the DH parameters of UR5, as well as the DH parameters I calculated by myself from the pose of the coordinate system of the UR5 joint in coppeliasim. But when I applied these two DH parameters to the positive kinematics, I found that the end attitude of the robotic arm returned is inconsistent with the end attitude of the robotic arm in coppeliasim, and I have adjusted the initial attitude of the robotic arm, but the results I get are still not right, I would like to ask what may be the reason for this?

[coppelia]

indeed, something is broken again...

At the same time: DH parameters do not encode the current joint angles, or then should only be considered with all joint angles at 0. In a similar way, the reference frame of a joint can be randomly selected (even if e.g. one agrees to always align the rotation/translation axis with the ref. frame's Z axis, as in CoppeliaSim): a rev. joint A behaves the same with a rev. joint B, even if B is rotated by a random angle around its Z axis.

Cheers