Gyro degradation

[Joe2003]

Hey

Im developing a system for indoor navigation and thought some of you guys might be able to help me out.

I know a gyro(s) wont work in a UAV as it needs a point of referance to maintain accuracy, which can/will be degraded through wind etc. However my application will be indoor on a smooth floor, in essence zero friction. Can a gyro be used as an accurate referance point for any calculations here? A compass bearing will not work as I will be dealing with electronic noise which seems to throw any compas modules (the RoHS CMPS03).

So once the gyro has been locked I can see how far im varying from that vector by analysising the pulse width outputed by the unit?

thanks

Joe

[zlite]

Gryo drift has nothing to do with wind. It's just a function of motion of any sort. I'd put the compass sensor on a long stalk to keep it away from noise. The alternative is very messy math in combining gyros and accelerometers. Either that or use a stationary IR beacon in the space to establish an absolute heading.

[Joe2003]

noise will result from its enviroment more than localised noise. IR beacons did cross my mind, but I cant always assume I will have line of sight with such unit.

[gael.desilles]

Dear Joe,

if you can't denoise the environment of the compass, you may then calibrate the effects of that noise on the sensor : put it on several known orientations, read the signal and tabulate as a function of the true angular position.

Another idea would be to use a small camera pointed upward and analyse the image of the ceiling. This can be very accurate with "professional" ceilings, where straight lines are often there.

Hope this helps.