I am working on an application where we have an I shaped vacuum EOAT picking boards from a stack. There is a decent amount of adhesion forces making picking a complicated process. I have been working on a zoned picking cycle where I start one of the suction cups, which pulls the TCP down on that side, then correcting said motion to somewhat “peel” the board off in the corner. That works fine, but when I engage the other suction cups, I am running into a “Path deviates” prot stop on Wrist 2. I currently have the payload being set with a transition time of 10s. The issue seems to be caused by the wrist being pulled down in one direction by the suction cups. Any advice is welcome, and I can post pictures or videos if needed.
So, while researching solutions, I also found UR Log Viewer and started looking through the data. My issue is that I have no idea what I am looking at. The error claims a Wrist 2 devaition, but I am not seeing anything super obvious off the cuff. Can someone help my understand this better?
So I don’t think you can get out of this by setting the transition time, even though I do not have that much experience with it.
If I were you, I would try to make some programming to solve the issue. The Force node is more lenient towards “unexpected” outside forces.
So you could, for example, move the robot down to pick up and enable the one corner suction cup, and then make the robot move upwards with a Force node set to Frame (which allows you to set the force and speed).
That makes sense. I am using a force node to engage the suction cup, but haven’t tried doing so when retracting. I have seen some things about using force mode at low settings, and the boards + payload near the limit of the UR20’s capacity (at max thickness). Do you recommend a high force setting or will the force I enter be on top of the force required to move the payload?
If you enable the force control while the robot is not lifting the board at all, then it will require a high force. The Force node resets the FT sensor before it starts.
