Hello,
We teach our robot a waypoint, when going to this waypoint a few moments later it is shifted a few millimeters.
Lencon2.urp (22.9 KB)
default.installation (2.5 KB)
Here is our code
Maybe someone knows what is wrong?
Thank you in advance!
To be clear, my understanding is that the waypoints dont actually change.
The coordinates stay the same, but the robot moves to a different position, therefor it seems like a loss of calibration. Is my interpretation correct?
Yes I will change my question
Is the calibration only lost after the robot is (re)initializated, just at random or after some specific event?
not anymore lloyd, we thought at first it only happened after emergency stops or resets but now it is also happening “runtime”
It sounds like a hardware issue, but i like to rule out some things first:
-The problem is that the robot wil move to the some waypoint twice, but end up in a different physical position, despite the readout showing it is at the same coordinates?
-How are you reading the position of the robot? Are you sure you are using the right view?
-Is the robot able to hold its position while running? (if not the end of arm tool weight is set wrong)
-Is nothing loose? There should be no play on anything.
-How are you measuring the robot position? Are you sure the robot is wrong and not your end of arm tool?
-Does the issue still occur if you make the robot move between 2 hardcoded joint (position defined bij angles of each joint) waypoints? (To make sure the problem isn’t in your script)
-If yes, does it also occur on each joint individually?
These tests should enable us to narrow down the suspects.
-The problem is that the robot wil move to the some waypoint twice, but end up in a different physical position, despite the readout showing it is at the same coordinates? Yes
-How are you reading the position of the robot? Are you sure you are using the right view? We are making sure we read everything from the base-view, we read everything from the UR Controller Interface
-Is the robot able to hold its position while running? (if not the end of arm tool weight is set wrong) Weight is set to the right amount
-Is nothing loose? There should be no play on anything. No nothing is loose, everything including the robot is bolted to the table
-How are you measuring the robot position? Are you sure the robot is wrong and not your end of arm tool? we thought this too but we only make a tcp offset in the Z-axis and the difference occurs in the X and Y
-Does the issue still occur if you make the robot move between 2 hardcoded joint (position defined bij angles of each joint) waypoints? (To make sure the problem isn’t in your script) Every waypoint we make is done via the UR Controller Interface, polyscope, not via a script
-If yes, does it also occur on each joint individually? We have not tested it for every joint individually