Flight Tests with Arducopter 2.x

This are the pictures from my first auto fly mode wich took place today. I was happy with the results even if alt must be tuned, but I had no crash, and any flight without a crash, for me is a good flight :D

I made 2 passes of the same path and as you can see there are some alt variations, the quad looked like a frog jumping from point to point.3689426709?profile=original3689426942?profile=original3689426954?profile=original

E-mail me when people leave their comments –

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

Join diydrones


  • After a crash yesterday, wich resulted in a broken landing gear, today I got a new landing gear thanks to my friend Mr. Veloso from BeeDrones and went to test version 2.0.47 again. Let me just point to the fact that the crash had nothing to do with the hardware or software, but a loose spinner adapter was the reason. I tried stabilize, stabilize+simple, alt_hold, loiter, RTL and auto. Auto proven to work as all the times before, following the path very well with about 1mt variations, wich is nothing. Stabilize is proven to work well as before, alt_hold and loiter seemed to hold position much better (I had perfect flying conditions, 0m/s wind). RTL however did not work. The quad just loiter in place where RTL was activated and didn't come home, I engaged it 3 times if I'm not mistaken, about 30 mts from home and didn't return. Anyone having this issue ? I'm using default PIDs.

  • New flight test today with version 2.0.47. Unfortunatelly my flight was interrupted thanks to a broken propeller. My quad crashed but cleaning the dust and sand from motors, showed no great damage. A broken landing gear and 2 props to the garbage were the only damages. Luckily this was after my AUTO mode test, wich I may say was much better than 2.0.46. It was a bit windy but the quad behaved very well, I can't give you an impression on loiter or RTL but I leave the screens of my auto flight.



























    Top image is the desired path, middle image is the real path and bottom one, is the real in perspective. As you can tell, even with a light wind, the path had no great errors and when looked in perspective, the altitude was much more steady (only using barometer). I'm not having that "frog jumping" effect no more, there was a slight alt discrepancy but not too much. Thanks Jason, great code. I'll try to make some more test flights during the next days, after I repair my landing gear.

  • David, I too was nervous, but started by testing RTL and Loiter and after I got confortable with RTL decided to try a small course :) now, I'll try more complex missions, in small steps !

  • Sweet !.   looks decent so far.  I'm preparing to start testing waypoint navi soon.  I admit I'm really nervous to try it  lol.



  • Developer

    I made a slight tweak that limits the speed change to waypoints. It limits the more violent lurch to the next WP.


    Z accel code is in 46 but needs someone to have a look at optimizing it.

  • Distributor

    Nice idea to use z acc to help with jumping effect for arducopter.  And nice results Rui, looking forward to see if you manage to get any improvements with your new values.

  • ok :) I'll be doing some more tests and posting results for better analisys. Now that I'm more confortable with the commands all I wan't to do is fly. To bad not having unlimmited batteries, because the flight conditions in the last few days have been perfect. I was using Loiter_P=0.7 and alt_hold_P=1.0, I now lowered the values and I'm going for another test flight this afternoon.

  • Moderator

    Great Job Rui :) I spoke with Jason about a good upgrade on Altitude ... I think that with z acc we can solve the jumping effect :) On Arducopter32 for doing the test i disable at  the moment altitude hold on loiter and on auto mode .... emile told me that he doing some test on auto ,too

This reply was deleted.