SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By SimonNielsen
#57597
Hi

I am currently working on a project where i have a model glider that i need to stabilize while its landing in a gliding fashion. I have to have some kind if lateral stabilization and what i thought of was the IDG300 GYRO, and then using the readings from this to obtain the roll angle and use that to control the ailerons.

But im now realising that this might not work. i basicly want the angle and the IDG300 rate sensor could be used with some succes but due to the drift and the fact that the GYRO does not know when the glider is leveled i dont think it will work very well, especially when considering that the glider might not be leveled when i launch it.

So i was thinking if the ADXL320 accelerometer might be what i need, but i am not sure how it works exactly. I read in the ADXL320 datasheet something about obtaining the tilt from the g readings. Is this something i could use?

Hope someone can help :)


\Simon
By wiml
#57758
Accelerometer-type tilt sensors don't work when stabilizing aircraft— they can't tell the difference between straight-and-level flight and a barrel roll. (This was a big problem in aviation before the development of gyroscope-based artificial horizons, since the pilot's sense of tilt has the same weakness. Fly into a cloud or other instrument conditions and your orientation becomes random really quickly.)
By edmoore
#57786
You are treading a very, very, very well trodden route. Have a search on the forum for the 'kalman filter' and brew up a coffee and wade in. The answer to your problem is there.

Ed
By SimonNielsen
#57800
I have been looking around and i found this post:

viewtopic.php?t=12435

That guy seems to be having the same problem. I have been reading up on that Kalman filter stuff, it does seem a bit complicated though.

It seems that the solution is to use a gyro combined with an accelerometer, as stated, it also seems to be the most common solution to this kind of problem.

I just dont know if the accelerometer will give useful output in my situation with the airplane movements. But i just need to know where the ground is. The gyro should be able to do the rest i believe.
By lyndon
#57813
IIRC, the guys at Rotomotion (www.rotomotion.com) have open sourced their autopilot software and you can find it on Sourceforge along with hardware requirements. Theirs was for helicopters, but the basic principles are the same.

This isn't easy to do and you'll probably have more success if you look at how others have solved the same problem.
By sylvie369
#57815
wiml wrote:Accelerometer-type tilt sensors don't work when stabilizing aircraft— they can't tell the difference between straight-and-level flight and a barrel roll. (This was a big problem in aviation before the development of gyroscope-based artificial horizons, since the pilot's sense of tilt has the same weakness. Fly into a cloud or other instrument conditions and your orientation becomes random really quickly.)
You'll enjoy this article:

http://www.theatlantic.com/unbound/langew/turn.htm
By wiml
#57824
That's a nice article. :)

FWIW, SimonNielsen, you might want to check out the DIY Drones website too.