Well, veganode, you're attacking an absolutely formidable problem, one that took decades of work by thousands of people to accomplish.
1) The thrust phase of a model rocket motor is only a couple of seconds long, and you'll only have active control during that time. You could conceivably add some movable control surfaces to the rocket, so you could control it during its descent, but control surfaces small enough to not interfere with the ascent won't really be able to have much effect.
2) The gimbals needed to adjust the thrust angle on a rocket motor are formidable. You will need quite good machining skills to make such devices.
3) You won't need just one gyroscope; you'll need three, one for each axis. You'll also need three linear accelerometers, plus the electronics to decode the sensor values and solve a complex set of differential equations to give you the position of the vehicle. You're looking at microprocessor-level electronics, probably a StrongARM or 286 or better. It took tens of thousands of man-hours of work to make an inertial guidance system that could fit in a cubic foot; you simply will not be able to accomplish it by yourself. Here's a document on the history of inertial guidance systems:
http://www.imar-navigation.de/downlo...troduction.pdf
4) You can purchase a variety of off-the-shelf radio control equipment from a hobby shop. You can buy the transmitter and receiver units, plus servos and other equipment.
My honest advice? Scrap the idea of building an inertial guidance system. Scrap the idea of gimballing a rocket motor. Go to a hobby shop, buy a pre-made model airplane body and radio equipment, and start small. Once you're an expert on radio controlled airplanes, then you can start strapping rate gyros and microcontrollers in them to start learning guidance. It's a very, very long road ahead.
- Warren