Hi all,
I have two LBR4+ robots and two KRC controller boxes. I'm trying to configure them the use the Fast Research Interface so that I can research nonlinear control laws; for this purpose, I really don't want to use most of the overhead associated with the KRC. What I want to do instead is have the robot start up and begin listening for motion commands on an ethernet UDP port, preferably without requiring manual human intervention to enable or configure it first.
I know that there are ways of wiring an X11 connector and reconfiguring the I/O software jumpers to do most of this (e.g. not requiring an attached teach pendant, not requiring manually placing the robot in AE mode, not manually starting FRI, etc) - I can find mentions of this in several places online. But I can't find any actual documentation for the X11 wiring or the right settings for the I/O jumpers, much less an explanation for how or why these things work.
Stanford has what appears to be a very nice software package to do what I want with ethernet commanding (http://cs.stanford.edu/people/tkr/fri/html/pages.html). But this website seems to assume the robot is already hardware configured appropriately, and when I follow the steps listed on the website, which only require running a KRL program that starts the interface, the KRC complains that I'm not allowed to issue the commands in the program (mostly it gives "Active commands inhibited" warning, but depending on whether I'm in T1/T2 or AE more these change).
So it looks like I don't have the hardware configuration right. Does anyone know of a resource or documentation spelling out how to do this?