[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Paparazzi-devel] ahrs_float_dcm
From: |
Christophe De Wagter |
Subject: |
Re: [Paparazzi-devel] ahrs_float_dcm |
Date: |
Tue, 10 Jul 2012 10:07:10 +0200 |
If gyro's are not perfectly perpendicular, (which is more common in manually assembled separate chip per axis situations than in modern single chip 3axis) then you want to compensate for it.
e.g.
When rotating perfectly around the yaw axis, you expect roll and pitch to measure nothing. Or better, if you mount your imu in a perfect cube and align it until you measure no roll and pitch when yawing, then when you turn your cube and do rolling, you expect to measure no yawing nor pitch. In reality there will always be a residual.
Checking this however is not so easy as you need a decent calibration setup. And again, with modern chips containing 3 gyros on one piece of sillicium, the alignment is guaranteed to be within 1 degree, so you can probably not calibrate better than that anyway.