[Yotamish]

Hi,
我想to verify in advance, if there exists a linear path in the gripper space, from one joints configuration to another.

I've looked at your API and found MoveL_Test - my only issue with it is that I can only enter the end position of the gripper (and not the end position of the joints configuration).

Is there a possibility to add this feature, or maybe there is another function that does what I need?

thanks in advance!

Yotam

[mnurpeiissov]

Hi Yotam,

I think you could just solve the Forward Kinematics and get the POSE with desired joint configurations with the help of robot.SolveFK(joints) (If your robot is in the library of RoboDK, otherwise you could manually construct transformation matrices using DH table). Afterwards, put the pose into MoleL_test. I believe this function needs not only the end position but rotation matrix as well. MoveL_Test(j1, POSE, minstep_mm=-1).

Hope it helps


Best Regards,

mnurpeiissov

[Jeremy]

I agree with mnurpeiissov,

You just need to transform the end joint configuration into a pose, then add the transformation between the flange and the TCP and you should be able to do what you are looking for.

Considering that you are looking for a MoveL, your configuration should not vary that much.

Jeremy

[Yotamish]

Himnurpeiissov&Jeremy, thanks for the replies!
The issue I wanted to solve was that using the roboDK MoveL, I sometimes received an error that no linear movement was possible from start pos to end pos, even though there was one. I assumed it's because MoveL doesn't look at all possible IK solutions.

In addition, since I want to plan, in advance, motion between several via points, I want to be able to construct the entire trajectory in advance. The way I see it, they may very well be more than one joint configurations that allow going from point A to point B, but when I add other points, I might need to select the joint configurations explicitly to go through the entire trajectory.

I would love to hear your opinion,
Thanks,
Yotam