Tuning the Ardupilot

Hi

I do have some trouble. I've flown the 2.5 code a few times with a EM406 and it seems to work, sort of.

My problem is that i do not know there to start with the tuning.

In stabilize everything works perfect. No wing rocking or pitch oscillations. Really does work well, recovers fast after a dive. Banking is less than in manual but it does bank enough for gentle flying.

To me this means that the roll_P and Pitch_P should have the correct values.

If I try RTL mode it doesn't seen to work all that nice anymore. If I set the Heading_P value to the easystar default i get a lot of wing rocking(Left to right), it comes close to home but never really circles around it. It does semi-eliptical figures and ugly figure-8's ect(80m Radius) . If i decrease the heading_P it doesn't really return to home ever...

Altitude hold does not seem to work at all.(Note- I can't get a value for MAX AIRSPEED in the config tool. Only 1 as the value---- Altitude is correct though)

I have seen some of the users kml files and their circles seem to be near perfect on top of each other. Even in wind.

So to the people with the know how. Where do i start? Do i go back to the roll_p and increase it till it wobbles in stabilize or what.

Lets get a nice HOW-TO going to put in the wiki.
eg. Increase roll_P until oscillating, decrease by 80% and set.
     Increase pitch_P until oscillating, decrease by 80% and set.

So, please do comment, but if you have advice, please give it as a step by step. It doesn't help anyone if the reply is only "Your gains are wrong, adjust it". Please explain it

Thanks :)
   

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

Join diydrones

Email me when people reply –

Replies

  • Ok... So had my first successful flight yesterday evening... Yippy! It works,. sort of.

    On the gains Issue: Tried adjusting it... No luck... Then i thought well, last try- I'll set the rudder link to the minimum throw...
    And... Viola! The circles aren't that tight but it seems to do a lot better.. No more crazy rocking... So now the question..

    Would setting the Rudder max degrees have the same effect? In manual i would still like to keep the large throws, can't turn on a dime anymore...
  • Developer
    The control loops from the current 2.5.02 version are really from the older version of 2.4. They have some limitations and can be harder to tune. 2.5.1 uses some new approaches that I think you'll find much easier to deal with.
    One such approach is using groundspeed to dampen the roll of the plane. What we've found is that the plane flies one way downwind and another upwind. It overshoots leading to a seeking motion.

    Doug had a great post about tuning recently and he was spot on. I'd look for that. I'll be posting a new version of 2.5.03 in a day or so with the improvements from last weekends flight testing and tuning session.

    One thing I did notice is that the rudder size on the easystar can greatly effect the tuning of that plane. the header file from a perfectly tuned plane can be wrong for another plane with different rudder size or weight.
  • Developer
    Your approach on adjusting P gain is spot on. Just keep at it and for most things leave I gain to zero. If stab mode is working great than you got it and leave it alone.
    Alt hold: as written is currently not working for anyone from what I understand.
    Heading or RTL controller should be working but the gains are really sensitive with changes in ground speed. You have to use the deweighting software Doug uses in 2.5.1 (not sure if it is in 2.5?) and even then you really have tweak the gains because it is dynamic controller that is constantly changing. For the constant over shoot you can also try and add some derivative gain but keep it a lot lower than P.

    good luck
This reply was deleted.

Activity

DIY Robocars via Twitter
22 hours ago
DIY Robocars via Twitter
RT @SahikaGenc: AWS DeepRacer & Hot Wheels Track https://youtu.be/4H0Ei07RdR4 via @YouTube
Sep 14
DIY Robocars via Twitter
Sep 8
DIY Robocars via Twitter
RT @davsca1: We are releasing the code of our Fisher Information Field, the first dedicated map for perception-aware planning that is >10x…
Sep 8
DIY Robocars via Twitter
RT @SmallpixelCar: How this works: 1)object detection to find cones in single camera image, 30 frames/sec on @NVIDIAEmbedded Xavier. 2)comp…
Sep 8
DIY Robocars via Twitter
RT @SmallpixelCar: Use two color cones to guide the robocar. No map needed, on onsite training needed. Just place the cones and it will fol…
Sep 7
DIY Robocars via Twitter
Sep 7
DIY Robocars via Twitter
RT @roboton_io: Great to see http://roboton.io running at 60fps on the cheapest #chromebook we could find! #edtech #robotics #educat…
Sep 3
DIY Robocars via Twitter
RT @openmvcam: Crazy in-depth article about using the OpenMV Cam for Astrophotography: https://github.com/frank26080115/OpemMV-Astrophotography-Gear https://t.co/BPoK9QDEwS
Sep 3
DIY Robocars via Twitter
RT @openmvcam: Hi folks, it's finally here! Our first draft of our Arduino Interface Library is out! It works over SoftwareSerial, Hardware…
Sep 3
DIY Robocars via Twitter
RT @chr1sa: Please let them have an open API. This would be perfect for @DIYRobocars races https://twitter.com/NintendoAmerica/status/1301513099707658246
Sep 3
DIY Robocars via Twitter
RT @SmallpixelCar: Lanenet pretty much used all my GPU power on @NVIDIAEmbedded Xavier since I optimized with tensorRT. I need to run anoth…
Sep 3
xemone liked Max Gilson's profile
Aug 31
DIY Robocars via Twitter
RT @LyftLevel5: Our @kaggle competition on Motion Prediction for Autonomous Vehicles is now live! Experiment with the largest-ever self-dri…
Aug 24
DIY Robocars via Twitter
RT @chr1sa: Our next @DIYRobocars virtual AI car race will be on Sept 26th. Sign up here https://www.meetup.com/DIYRobocars/events/272786977/ https://t.co/UENKGSOWO8
Aug 24
DIY Robocars via Twitter
New ready-to-run @NVIDIAEmbedded JetRacer car from Waveshare. Perfect for the next @diyrobocars race as soon as we… https://twitter.com/i/web/status/1297960223013867520
Aug 24
More…