Hello Pros,

My colleague and I are trying to program our UR5 relative positioning movements. Still working out the kinks but after we started messing around with the scripts we started to run into a problem. our UR would do the motion of the program and then all of the joints would lock up and we'd get the error "No Controller Found." We are consistently getting this error every time we run the program. Has anyone ever experienced this?

Here is the program

Hello @DuongD, I would need your installation file and all the subroutines that are called within your program please. Also, what is the version of your controller (CB3 or CB3.1), Polyscope, and any Robotiq URCaps installed?

I have the same issues, I have e-series 5.9 polyscope. After 2 times cycling the program it showed me on the polyscope "No controller Found". Focus, what cause this issue and how i can solve it ? 

We have historically seen this issue around two scenarios.  First is excessive load on the controller caused by things such as intensive math computations, spamming the controller logs with textmsg calls, etc.  The second cause was XMLRPC related calls where the robot program was stopped while a call was being processed and the return value back from the server caused the controller to crash.  That appears to have been addressed by UR with some of the later software updates, I do not recall the version numbers.

The first issue is more of an issue on e-series controllers than CB3 as there was more CPU overhead on the CB3 due to the lower frequency the control loop was running at.

Without more details such as the program files, series of robot, version of polyscope, installed caps and their versions, its really hard to explain exactly why you are seeing issues with a program.  The only thing I can say is that usually when the robot is not performing correctly, unless there is a bug that we have found in the actual UR program most issues can be attributed back to something that we are doing incorrectly in the program

Thanks Matt, I have more than 3k line of program, with thread, call, gripper urcap software, venation, software so, As you said Matt may be the load of the program causes

Hard to say, we have many robots running in the field with over 5k lines of script code on them.  We are just very diligent about cleaning up threads when they are not needed, breaking up computationally heavy processes so that they are performed on multiple clock cycles of the robot controller, etc.

what do you recommended me to fix the problem?