Thanks Andy for the reply.
So I think the second option might be the most convenient, there would be no problem having robot B sit and send out pings while robot A moves around and tries to pick up those pings. How exactly does the pinging work? I've read a bit in the api about capture and continuous.
EDIT: Actually I'm having a bit of a crisis with my project, I'm not 100% sure on how I will implement things. My main areas of concern are:
Location, obstacle detection and avoidance.
I was thinking maybe I could draw out a big area with black tape, robot A can use it's colour sensor to move along the tape, if there's obstacles, it moves in another direction (Essentially very similar to the JavaOne).
So that is at one end of the area and at the other will be robot B, waiting for A to find it.
Does that sound doable?