The altitude hold function on the apm2.5 seems to be very unstable and I have tried all the fixes/suggestions I can find on the forums to no avail. It seems I can get the quad to hover if I am climbing relatively fast when I go to the alt hld mode. If I hover for 10 seconds before engaging the altitude hold the quad will slowly descend to the ground and at times will oscillate( tried oscillation fixes). Barometer covered etc. It would seem that throttle calculations are off by quite a bit. Has anyone got this mode to work in trustworthy manner? I certainly wouldn't trust the altitude function to be used in flying waypoints etc, its just to unstable. The quad flies very stable and handles great in stable mode.
You need to be a member of diydrones to add comments!
Replies
Althold def. worked better on 2.7.3 than in 2.8.0/1. Seems like fixing the gps issue in 2.7.3 did something evil to althold.
Hopefully the current software does not use gps altitude information at all. BTW i coded an althold with baro/acc fusion for mwii wich gives me naza like results without throttlepid stuff or vario pid etc. Just 2 parameters for althold and one to compensate for fast movement. I know it's mathematically wrong, but it works. It can even be altered to reach a specific hight at desired speed.
I include the code here again, this time i stripped it down to the core parts for an easy overview. I hope that a DEV can make something out of it.
So long
Kraut Rob
MwiiBaroAccSkeleton.zip
I have been playing with alt hold as I noticed a few issues. There are a few recent threads that you should review.
I have a few changes to make before I test again.
What do you have? Sonar?
Your altitude when you hover? More info is helpful. Do you have a log available?
Randy