Short answer: You can approximate the output of a gyro with accelerometers.

Principles:
An electronic gyro outputs a signal that is related to the rate at
which the gyro is rotating on the measured axis.

If you place two accelerometers so they are measuring the acceleration
at a tangent to the axis of rotation then you can measure the
acceleration and deceleration of the rotation of the axis.  You need
two so you can cancel out simple movement - they need to be opposite
each other (one on one side of the axis, the other on the opposite
side of the axis).  Once it's rotating at a fixed rate, though, the
accelerometers won't have any output so you'll have to integrate to
get rate.

If you also put an accelerometer some distance from the axis so it
measures perpendicular to the axis, then as long as it's turning
you'll also measure some amount of centripetal force.  This will be
related directly to the rate of turning, but is also going to have
issues with simple movements other than rotation.  If you place
another one opposite the axis then you should be able to cancel out
simple movement (as above).

So the upshot is you need 4 accelerometers (or two dual axis
accelerometers) to start measuring forces that can give you readings
similar to a gyroscope.

You'd have to define "Well performing gyroscope" before an answer for
that can be given.

-Adam

On 6/9/08, Electron <electron2k4@infinito.it> wrote:
>
> Hello,
> is there any way to make, from n accelerometers (1, 2 or 3 axis) a well performing gyroscope?
> If so, what is the best approach?
>
> Thanks!
> Mario
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>


-- 
EARTH DAY 2008
Tuesday April 22
Save Money * Save Oil * Save Lives * Save the Planet
http://www.driveslowly.org
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist