Creating a simple program to actuate an EOAT or turn on/off any robot-IO-controlled device and assigning them to Tool1 and/or Tool1 keys on the pendant can save lots of time.
Assign to UK[1] and [2] on the macro table
Creating a simple program to actuate an EOAT or turn on/off any robot-IO-controlled device and assigning them to Tool1 and/or Tool1 keys on the pendant can save lots of time.
Assign to UK[1] and [2] on the macro table
No problem. There are some KAREL built-ins that check reachability of a point if you wanted to add some intelligence. They only check the point is reachable, not that the path to the point is Ok.
Very interesting. Also jealous. Something I've wanted to try, but haven't had the chance to.
Fanuc supports ethernet encoders, which I believe work on ROS packets, so I would emulate one of those on the PLC side and send the data over that way.
If by ROS you are thinking the open source robot operating system, that is incorrect. Don't know what protocol the Ethernet encoder uses but I think it's probably proprietary.
You may get best response from the robot if you use a simulated encoder in the robot (set this up in the encoder screen) and just send the speed of the AGV to the robot's encoder simulation speed variable. All the encoder variables are in $ENC_STAT[x]. You can also automatically configure the encoder simulation from here too Hopefully the AGV speed is relatively constant while the robot operates on it.
There's a Rockwell module that converts machine counts from a PLC to pulses that can be picked up by the encoder board. That's another option if you're using Rockwell controls
Why do you need to know the config?
If you're trying to home the robot such that it doesn't fault out, a simple and universal way of doing that is in TP program.
Take an LPOS to a PR, then change the Z value of the PR to a known safe height, then do a linear move to that PR. It will move straight upwards. Then, do a joint move or to the home position. Wjnt on a linear move to home might work too (wjnt moves minor axes in joint and major axes as linear). Joint/wjnt will get you through most config or singularities
you should have a ROBOGUIDE Installation Support .docx file in the Support/Troubleshooting ROBOGUIDE folder of the download folder for ROBOGUIDE(not the install folder on your C, but the SFTROBOGxDVD...) folder.
That docx has some info on this issue
I ran into this before and FANUC pointed me there which is why I know about iit
What is the value of P[1]. The X value should be a reasonable positive value. The value will be relative to the position in which you taught the track frame origin
open the robot in Boot Mode (swipe in from the left side of the screen and select Boot Mode) and power cycle, then perform an INIT start from this menu.
That would reinit the robot to factory settings. This would rule out software
You will lose and programming you have done to this point on the robot
You'd have to setup a conveyor machine in HandlingPRO to accomplish the simulated iRVision+tracking. Load the robot with iRPickTool option and set it up from web interface, and set the RG conveyor and camera up to be connected to the robot in the HandlingPro invironmnet
Simulated vision wont work in iRPickPRO.
the KAREL Manual has documentation and sample code for this sort of thing.
robodec has a good suggestion. just want to add that your safety assessment should account for the robot being in non-collaborative mode for this time period.
I also wonder if the Touch Skip function could work here. there is an icon for it. you might need to experiment though but it may end up being more flexible. that being said it can still lead to excessive force errors
the only way open source products may be on pair is when functionality is limited and well defined or if the open source product is backed by a large company.
https://www.therobotreport.com/intrinsic-acqu…-robotics-corp/
check line tracking manual for M-1iA type of encoder.
scara is different, good luck with that.
why would you want to