Electron (Mario) wrote:
> is there any way to make, from n accelerometers (1, 2 or 3 axis) a well
> performing gyroscope?

To be blunt, no.

You can get some idea of "vertical" using accelerometers, assumuing you're
motionless in a gravity field, but for any real idea of attitude in a
dynamic situation, you're out of luck. You'll need additional input from
some other type of sensor in order to disambiguate the data.

> If so, what is the best approach?

Take the accelerometers back and get gyroscopic sensors (e.g., MEMS rate
gyros).

What is it that you're really trying to accomplish?

I have some experience building inertial navigation and attitude systems
that combine accelerometers, rate gyros, magnetometers and multiple GPS
receivers for one of my clients. I can give you a good idea of what you
can accomplish with various subsets of these sensors, if I have some idea
of where you're headed with this (sorry for the pun :-).

-- Dave Tweed
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist