Vismach Coordinate Systems

Hello,

I am currently trying to work on a AR2 robot simulation on Vismach. I am importing all the parts from STL assembled in a group I have 7 groups STL files. I am really struggling to understand where is the origin of the Vismach coordinate system and where are the origins of the groups compared to the Vismach origin. In fact when I try to rotate a link I get both a translation and a rotation and I cannot recover the translation since I have not understood where it is located the link compared to the origin of Vismach.
Could anybody help me please with this?
I attach two screenshots to show the behavior, the py Vismach file and the STL files I am using. The STL files should be located in a folder called AR2ForVismach to let work the py file work. The py files should be just in the directory upper to this directory.
Thanks a lot in advance.
Ugo

you need to setup the vismach file the same way you setup your controll

as of robot the 0,0,0 is most the center of the BASE
but you can make it a outside Workarea also and define the ranges you like

see the examples of the SCARA and PUMA
they are lockated under SIM/AXIS/VISMACH in your running Linuxcnc StartGui

Hello bbsr_5a,

Many thanks for your kind explanations!
I am quite new to Linux CNC and I have not got well the explanations you gave me. What do you mean with setup the vismach file as the control is setup? Is there any special keyword to be used within the vismach file?
How can I know where are the other joints with respect to the new origin?
About the Puma and Scara examples should I look to the .ini files?
Thanks in advance

drugo_

I have attached the vismach model of my Mitsubishi Robot and

To get it to work was quite a task. I basically took a working example of a robot arm, changed the stl files and worked through the code. This way I fumbled my way along. The original code I used started from the tool tip backwards to the base which seem counter intuitive but it worked.
The workflow is to create or import the first part then orient (rotate) it correctly in space then move it so that the point of rotation/connection to the next part lies in the vismach origin (this can be displayed by creating a separate visual coordinate system (mine is called "world coordinate system" and ends up being at the center of the bottom of the base.) Then you define the connection with LinuxCNC Joint using HalRotate. Then the next part is inserted -> rotated ->translated to join the rest of the arm -> connect the arm together with "Collection" -> translate to origin of the next connection -> define HalRotate for this joint -> add next part.

The reason I created different poses is the problem to create the fitting modified Denavit-Hartenberg values that then matches the genserkin kinematics to the vismach model. Thats why I ended up with the configuration in melfagui that is actually not a "natural" pose for the arm. But I have not been able to change the DH-values so that the start position is a more natural pose. Mostly because genserkins does not allow us to define theta values.

Anyway I hope you find this useful.