Gluonpilot

The good weather is coming so a good time to pick up the Gluonpilot.
Last time I tried to do a simple navigation i.e. CIRCLE_REL with a 30m radius. When switching from stabilized to auto mode the easy star just keep on flying away from me.

I can see from code that there is an airborne criteria that sets the home as long as we're not airborne but I cannot find the location where naviation_set_home() is invoked. Is something missing here or did I overlook something?

Hi,
i have from autonomous flight to stabilized flight then back to autonomous flight
the plane fly away see pictures blue arrow


Michael

@Jippieee: Good catch! Thanks! The current development source code is incorrect. The beginning of navigation_update should NOT be commented as it currently is. This is the result of some bench-testing which I forgot to un-comment. I'll change it ASAP. Please use the firmware version 0.5.2 from the download section.

@Sidewind: I suppose you use v0.5.2 or earlier from the download section? Could you give me your navigation commands and your datalog from the gluonconfig?


I will do some more flight tests tomorrow and put version 0.5.3 online!

Hi Tom,

I also run to same problems as Sidewind last season.

I have brought this out to you couple of month ago.
Also I have sent you kml files then.
At that point we were discussing that the problem might lie in vibrations that come from the motor that might disrupt gyroscopes.

Till then I didn't manage to retest this during flight as at the moment I have stabilization problems.

Hi Lukasz,

Yes I remember it but didn't you say you had succesfull flights after that when you managed to reduce the vibrations?

Because a lot of people are having successfull flights, and others have some problems it is not easy for me to judge whether it is a bug or a misconfiguration of the user. Therefore, please always send me your datalog files and navigation commands so I can better judge the problem.

Later after I eliminated vibrations I only had two/three flights and the problem seemed to be solved but I guess that's not a good sample as more flight would be required.

Once I solve my stabilization problems I will retest this and let you know.

Could it be that the GPS information is (sometimes) misinterpreted or accidently an offset is added? When I connected the windows configuration software the other day and from the GPS page selected "view in google maps" then sometimes the showed location is up to a few kilometers off, even with 6 satellites in sight. It is just something I noticed and did not do further testing on this as I suppossed it was a only user interface thing.

Hi Tom,

I have this problem since version 4.

Enclosed the requested RAR-file.

Regards,

Michael

@Jippieee were you inside a building when this was happening?
I also see the same but when I'm in my house, when out outside all seems to be fine.

Hello Michael,

My analysis:

The first time you flew in 'navigation mode', your plane nicely took off, climbed to the desired altitude and followed the waypoints and ended circling above the take-off position, exactly as you programmed it.

Then you did some "manual mode" flying.
Then you did some "autopilot mode" flying -> the plane started circling again
Then, for some reason, you switched to "stabilized mode" and the plane flew away from you.... The moment you switched back to "autopilot mode", the plane returned to the home position as expected. (Hint: export the datalog in "KML Classic" mode)

To me, this seems to be a perfect flight! Maybe the threshold between "Stabilized mode" and "autopilot mode" is too small on your transmitter?