You'd need long range rfid readers and they cost a small fortune, and they aren't very long range, I think the max is like 8ft using active tag technology .. passive max'es out at like, 8 inches lol.
It's every developers fantasy to be able to track people in real time as they move from room to room because of the number of home automation possibilities that capability presents.
There's a way to do it using 802.11 but the tags are pretty big and need batteries (active), and those systems aren't cheap either. Plus you need multiple AP's to provide a radar like functionality where locations are "approximated" by signal strength between 3 different reference AP's.
It's all pretty crustry at the moment. I've seen improvised systems but they all have impracticalities like active tags, multiple APs, dozens of readers or users needing to check in at locations to tell the computer they've moved location.
I saw one the other day and milk bottles have tags and when they are put in the garbage the computer orders more milk. Like, how impractical is that? you get a replacement bottle of milk a day "after" you've run out .. really lame.
Anyway. Part of the fun of developing is having a great idea and then spending 2 years realizing it's not doo'able, or, you need a piece of technology that doesn't presently exist lol.
I can think of a thousand cool things to do with micros but unless you're an expert in every field of technology, it can take a while to discover all the issues of taking a concept to finished product. The thing I like best about this community is its a hive of ideas and new innovation. People here try the craziest things so if you've got an idea throw it around and people might be able to point you in the right direction = )