Dear all,

I would like to seek the experts' help, or who ever can help, to analyze my flight data in order to enhance the navigation performance in AUTO mode. 

I am using a bit older version of ArduPlane (2.4) and APM2.0, but it's doing well so far. However, my colleagues and I would like to enhance the controller performance in order to have better landing. So the main issue is in the landing. Once the plane comes to the landing waypoint, we can observe that it's not aligned with the landing waypoint. 

Is this due to GPS accuracy (we are using MTK), or it's a control issue?!

Do I need to keep a fairly long straight path before the landing waypoint to get better alignment?

I also observe some roll shaking once it's landing on the road! Does that mean I need to decrease the Integral part of the nav/servo controller to reduce the oscillations ?!

BTW, we are using a big airframe, it's C17 Globemaster III 

http://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=17784

I am also attaching the parameters.

So, can you please help me to further tune the controller of this big frame?

Thanks in advance.

2013-02-21 10-23 1.log

flight_param.param

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Thanks Andrew. 

    So you are suggesting to use the latest release (Arduplane .273)?

    If so, what are the suggested gains of the controllers that will work with this airframe (C17).

    Thanks.

  • Developer

    Hi Mohamed,

    Unfortunately I can't help you with that log. It is from a very old version of ArduPlane, which had quite poor logging. It is also much easier for me to analyze telemetry logs than flash logs.

    If you get a telemetry radio and the current version of ArduPlane then I'd be better able to help. The navigation control in newer versions of ArduPlane is also a lot better.

    Finally, for autoland Paul and I are working on a new altitude/throttle controller which we hope will make automatic landing a lot better. That will be in either 2.74 or 2.75.

    Cheers, Tridge

  • These are also the used PID gains. Are they Ok??

    PID_gains.JPG

    https://storage.ning.com/topology/rest/1.0/file/get/3692736964?profile=original
This reply was deleted.

Activity