hi,
If you're using LabVIEW, why not use the NI vision code instead of using GRIP? Look at the vision processing sample code. It'll get you most of the way there. The camera will be on the robot and all images will go to roboRIO (they can of course also be routed to the console so the user can see them) for processing.
If you insist on using GRIP, be sure to read the "New for 2017" section on screenstepslive
https://wpilib.screenstepslive.com/s...8-new-for-2017
Regarding your other questions, I'm not sure what you're referring to when you say "computer" - is this the driver station or some other embedded processor?
With regards to using multiple cameras in LabVIEW, see some of the other threads on the forum.
Good luck
-lou