Re: Magnetometer

[Peter Hollands]

Tareen,

I have answered your questions inline below. I suppose the subject o this post should really be "No Magnetometer".
I have taken the liberty of posting this reply  on uavdevboard, as sometimes thses answers are useful to other people as well.

On Tue, Dec 13, 2011 at 1:47 PM, Tareen Dawood <tar...@gmail.com> wrote:

HI Peter,

 

I hope you are well...

I have a small question.. with no magnetometer we are advised to power up with the plane facing the north direction... This allows the board to find its reference quicker? Yes, that is absolutely right. When the Direction Cosine starts up, it assumes it is pointing North.

 

If you dont do this it takes longer to know its orientation, right? Yes, That is correct.

 

And if you dont how does it find its orientation..  only with GPS? Yes, the plane find's its orientation  when it is moving by using the heading from the GPS data. The GPS can only has a heading when it is moving. We assume that the plane always flies and points in the direction of movement.

 

The DCM correction is not immediate. Only a small proportion of the error between the DCM yaw orientation and the GPS heading is corrected each second. It takes about 20 seconds for a 180 degree error to be corrected. So if you point your plane south at startup, it will take about 20 seconds of flight until the DCM is correctly orientated. You can see this effect when processing SERIAL_UDB_EXTRA data for a flight in Google Earth. Often you wll see the plane take off backwards in Google Earth, and then turn around to face the correct way during the first 20 seconds of flight.

 
 In fact, MatrixPilot is even smarter than the above. If there is a strong wind, then the GPS heading is not actually the orientation of the plane. The plane will be pointing one way, and the GPS heading (the track over the ground ) will be different.
So MatrixPilot estimates the wind (some clever maths that Bill dreamt up) and then corrects the GPS heading into a Wind Corrected GPS derived orientation. It is  that Wind Corrected GPS derived orientation that is really used to correct the Direction cosine matrix. 

Best wishes, Pete

 

 

Thank you!!!