I know my way around the GRIP interface pretty well, but I don’t know how it works during deployment to the roboRIO. My understanding is that you create a little algorithm/flowchart in GRIP on your desktop depending on the object you want to track. You then deploy from GRIP to the roboRIO. GRIP automatically sends a version of GRIP compiled for the roboRIO and the project. Finally, in the code you just run GRIP from the command line and read in the values send to the NetworkTables. For the code, I just followed the example on the wiki since we are currently coding in Java. Since I’m also a newbie to vision tracking, I am confused on how to base the movement of the robot upon the values published in NetworkTables.
This tutorial should help you. It does a pretty good job of explaining how to use NetworkTable values in code, and you can apply that to having the robot respond to data sent by GRIP.