Dear JJ3564,
Are you seeing this "image unstable" when teaching the 10 trays at once ? Are you teaching one tray and then trying to locate all 10 at once ?
Anyway, using the CAD file method is recommended if you are able to use it. It is the best teaching method as it works with exact object teaching.
Let me know what you think
Best regards.
Paul
jj3564
Hi Paul, We imported CAD models and used that as reference but the Robot is still placing objects slightly shifted away from the original waypoint it was set to place it at. Is this because of the angle of the camera when viewing at the objects from its snapshot position? How can we fix this so that the robot does not place objects shifted away from the designated waypoint?PaulGWF said:Hello JJ3564,
Were you able to detect your objects after all ?
Best regards.
Paul
jj3564
Yes all move functions are set to the proper sequences. Let me better explain the issue I am talking about.PaulGWF said:Helo JJ3564,
Could you please verify that you have correctly programmed the move sequence following the picking of a found object ? All positions that are relative to the position of the current object should be placed in a move sequence relative to the object position. And all other positions, I’m guessing like the drop position, should not :) Please verify this.
Best regards.
Paul
The Robotiq camera first detects the electrical boards we want to place our objects onto. Then it detects the objects to be picked up then proceeds to pick up the object and places them on the board in the designated spot. Each object is to be placed at the exact spot we set out waypoint/drop position at for each identical board.
Here is the issue, The robot places the objects onto the exact spot we want it to for the board that was set as the reference point, but for the other boards the objects are placed slightly shifted away from the designated spot. I was thinking this was due to the camera's error since the view of the boards away from the middle of the snapshot position are at an angled view.
Best regards,
J
We are running a program where the UR robot arm picks up objects and places them onto boards. We want this program to run automatically for 10 boards which means we would have to fit the 10 boards in the snapshot position for the camera to detect all boards. When trying to fit all ten boards in the snapshot, the camera is able to detect the boards but does not let us proceed with the program with an error message saying "image is unstable". We have tried many different methods to try to get the program to work on 10 boards but have not found the solution yet.
What would be the best way to approach this? Would it be better to just use a CAD file of the boards and set the program to run off of that?