[FooTheBar]

Hey!

I just started with RoboDK and want to use the python API to write a simple Picking simulation. For that, I create three Frames (for three pick-up stations) and for each frame, I create two Targets, one without an offset (grasping position) and one 100mm above the Frame (pre-grasp). Both Targets are children of the corresponding frame and now I want to move between those Targets.

After a bit of experimentation in the GUI, I found out that I need to set the Reference Frame before calling a MoveJ to the next target.  I had assumed that I can call a MoveJ on a target and the robot position will be computed by traversing the tree upwards until we find the robot base.

Why was it implemented this way? When would I use a Reference Frame that is not the parent of a Target when calling a Move-command with the Target?

[Albert]

Targets can be Cartesian and Joint targets. Cartesian targets are displayed as red icons in RoboDK (default targets) and Joint targets are displayed as green icons.

When you create a joint target it defines the position as absolute and this position is not affected by the reference frame. However, Cartesian targets need to be moved to using a defined coordinate system and tool. This is useful if you have to repeat the same sequence of movements in another reference.

You can find more information and examples about pick and place here:
https://robodk.com/doc/en/Example-Pick-and-place.html