Page 1 of 1

Mag calibration by samples instead of time?

Posted: Thu Oct 02, 2014 10:59 pm
by Arakon
I'm no coder at all, so take this as just a theory and I have no idea how difficult this would be.
Wouldn't it make more sense to change the compass calibration to run until a certain amount of samples of all directions is collected? Rather than 30 seconds and praying that you got all directions properly.
Similar to APM.. beep when you got a good "picture" of the magnetic field.
Just thinking, since I had a few times when I did the calibration procedure same as always, yet it was way off.. i.e. the compass display drifting back to a certain direction after rotating.

Re: Mag calibration by samples instead of time?

Posted: Sat Oct 11, 2014 9:18 am
by haydent
i had thought of this myself, and i think its a good idea if possible

Re: Mag calibration by samples instead of time?

Posted: Sat Oct 11, 2014 2:55 pm
by copterrichie
I have watched guys with DJI products calibrate mag before flight, Interesting to say the least and have wondered the same myself.

Re: Mag calibration by samples instead of time?

Posted: Tue Nov 04, 2014 9:06 am
by brewski
copterrichie wrote:I have watched guys with DJI products calibrate mag before flight, Interesting to say the least and have wondered the same myself.


This is exactly how APM 2.5 8 bit & Pixhawk 32bit Mag cal works. You can examine the code using arduino IDE. Check out How to Build Ardupilot with Arduino in APM WiKi.