HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Apr 30, 2010, 11:41 PM
Registered User
Joined Aug 2005
4,093 Posts
mark,

Did you have your aileron/rudder, and elevator hooked up through the OSD?

Cool - glad to know the elevon mixing works. You are one smart guy to figure out how to activate it before I wrote it up, haha.

About the stabilization, it should work even without autopilot active, btw. As long as the autostabilizergain is more than 0.0, it should kick in. I should warn that I am not sure if I am correcting in the right direction so that it might actually be de-stabilizing the plane more than stabilizing it. If you have a slider or knob free, you can assign it to the GAINCHAN. The knob/slider can then be used to vary the gain from 0 to whatever value you have set in AUTOSTABILIZERGAIN. In this way you can put in a large value and experiment with it to see the impact.

The code is up to 6.2BETA4 now. Once I verify that stabilization is kinda working, I will promote it to release version.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Last edited by Daniel Wee; Apr 30, 2010 at 11:49 PM.
Reply With Quote
Sign up now
to remove ads between posts
Old Apr 30, 2010, 11:54 PM
Registered Loser :)
mark hitchman's Avatar
New Zealand
Joined Sep 2009
1,928 Posts
Quote:
Originally Posted by Daniel Wee View Post
mark,

Did you have your aileron/rudder, and elevator hooked up through the OSD?

Cool - glad to know the elevon mixing works. You are one smart guy to figure out how to activate it before I wrote it up, haha.

About the stabilization, it should work even without autopilot active, btw. As long as the autostabilizergain is more than 0.0, it should kick in. I should warn that I am not sure if I am correcting in the right direction so that it might actually be de-stabilizing the plane more than stabilizing it. If you have a slider or knob free, you can assign it to the GAINCHAN. The knob/slider can then be used to vary the gain from 0 to whatever value you have set in AUTOSTABILIZERGAIN. In this way you can put in a large value and experiment with it to see the impact.

Daniel
Yes, I had aileron and elevator through the osd, when I enabled it in the shed the elevons were twitching so I thought I would give it a go I will check through the settings and see where things are at. Was blowing a bit here when I flew, 30 km into wind and 100 km with it so not the ideal day to test things.
wont be long and you will see it for yourself. At one stage the ahi was 90 degrees out of sync but I think the wind was making it slide sideways

Mark
mark hitchman is offline Find More Posts by mark hitchman
Reply With Quote
Old Apr 30, 2010, 11:58 PM
Registered User
Joined Aug 2005
4,093 Posts
mark,

Make sure you have the servo outputs assigned to EVN1 and EVN2 if you have not done so already. I suspect you have already done it. You can also change the mixing behaviour with the Elevon mix parameter, which supports two modes at the moment.

Note that violent turns will saturate (max-out) the gyros which will cause a temporary loss of orientation. The implication here is that the IMU works best when flown with less than 300-degrees/s rolls and pitches. Sudden rolls and pitches don't go down well with the IMU.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Reply With Quote
Old May 01, 2010, 12:04 AM
Registered Loser :)
mark hitchman's Avatar
New Zealand
Joined Sep 2009
1,928 Posts
Yeah I tried the elevon mix on 1 and it reversed the direction of control which is very easy to set up. Yeah i did evn1 and 2, had them set opposite for a start and it took me a while to figure out how to correct it. I tried reversing servos in the transmitter but that just made down elevator for up input. All is good now.

thanks

mark
mark hitchman is offline Find More Posts by mark hitchman
Reply With Quote
Old May 01, 2010, 12:06 AM
Registered Loser :)
mark hitchman's Avatar
New Zealand
Joined Sep 2009
1,928 Posts
DOSD+ 6.2Beta with elevon mixing in dosd. Transmitter set to normal plane.

DOSD+ 6.2beta fun jet (8 min 3 sec)
mark hitchman is offline Find More Posts by mark hitchman
Last edited by mark hitchman; May 01, 2010 at 12:42 AM.
Reply With Quote
Old May 01, 2010, 01:22 AM
Registered User
Joined Aug 2005
4,093 Posts
mark,

Fantastic video! It looks like the IMU might have been mounted slightly slanted? Other than that, it seems to be working quite well. I can see where it went off towards the 3/4 mark in the video. It almost looks as if there was increased vibrations on the platform at that point but it's hard to tell. The vibrations could be affecting the IMU.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Reply With Quote
Old May 01, 2010, 01:43 AM
Just clumsy. Oh, forgetful too
MikeTheCrash's Avatar
undefined
Joined Jun 2008
3,252 Posts
115200 did it thanks
MikeTheCrash is offline Find More Posts by MikeTheCrash
Reply With Quote
Old May 01, 2010, 01:52 AM
Registered Loser :)
mark hitchman's Avatar
New Zealand
Joined Sep 2009
1,928 Posts
I see what you mean about the vibrations, I will have to check that out and the level of the imu. It's the first flight with goggles on dosd+, in the morning I will go through the menu in flight to increase gain for the autopilot if I cant get the digital slider to work properly. Looks like its working pretty sweet though. If you have time can you explain some of the autopilot settings and whether autopilot switched on returns home or stabilisers the aircraft. I just re-read your previous post, with the gain turned up flight stabilisation works all the time?

Thanks

Mark
mark hitchman is offline Find More Posts by mark hitchman
Reply With Quote
Old May 01, 2010, 01:58 AM
Registered User
Joined Aug 2005
4,093 Posts
mark,

Not the gain for the autopilot, but the Auto-stabilize gain. Autopilot doesn't handle the stabilization. Also, if you don't assign a channel for GAINCHAN (ie. leave it at 15), the stabilizer will just use whatever value Auto-stabilize gain is set at. GAINCHAN is an additional way to control the gain. As long as Auto-stabilize gain is not 0.0, it will operate all the time - the degree of which depends on the actual gain value (and if enabled, the GAINCHAN position). Zero-ing the gain in effect turns stabilization off.

The autopilot and stabilizer are two separate functions. I think it might be best to just test the stabilizer first.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Reply With Quote
Old May 01, 2010, 02:03 AM
Registered Loser :)
mark hitchman's Avatar
New Zealand
Joined Sep 2009
1,928 Posts
So stabilizer will try to fly level wings with no climb or dive? If I launch with the gain too high it will oscillate and I wont be able to turn it off?

Mark
mark hitchman is offline Find More Posts by mark hitchman
Reply With Quote
Old May 01, 2010, 03:03 AM
Registered User
Joined Aug 2005
4,093 Posts
Well, you will still retain control but to be honest, I'm not sure how it will perform as yet.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Reply With Quote
Old May 01, 2010, 03:30 AM
Registered User
Joined Aug 2009
1,530 Posts
normally it should keep the plane level with no stick input, stick input are a difference added to this level flight, so it won't force the plane level and straight while you fly, only correcting non-stick input (wind etc f.ex)
at least its what i expect from it, and what others do
jimbob00 is offline Find More Posts by jimbob00
Reply With Quote
Old May 01, 2010, 03:40 AM
Registered User
Joined Aug 2005
4,093 Posts
jimbob00,

That's about right. The amount of correction decreases as it approaches level. You will still be able to control the plane. This is assuming, of course, that everything works as planned. At this point it remains to be seen how it works in real life. As I mentioned before, I'm not sure I have the correction direction correct.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Reply With Quote
Old May 01, 2010, 04:00 AM
Registered User
Joined Aug 2009
1,530 Posts
yeah, however that's just software so i'm pretty confident you will work it out.
i'm a bit more concerned about the imu "going off" like in the video

of course, when the imu is wrong, it's going to correct and crash the plane ^ ^

i hope it's not the hardware giving up too easily or something like that
jimbob00 is offline Find More Posts by jimbob00
Reply With Quote
Old May 01, 2010, 04:11 AM
Registered User
Joined Aug 2005
4,093 Posts
jimbob00,

There are some limitations to the hardware and the IMU doesn't like vibrations as it messes up detection of gravity. This is the reason why I am still treating this as a kind of test phase for the ArduIMU. I have another more expensive IMU that I am about to test as well.

On the other hand, the IMU does seem to perform reasonably in most cases so that's a good thing.

mark,

It just occurred to me that you can actually test the stabilizer on the ground. Just put a very large gain value and then tilt the plane to see if the control surfaces act correctly.

Daniel
Daniel Wee is offline Find More Posts by Daniel Wee
Last edited by Daniel Wee; May 01, 2010 at 04:58 AM.
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Dragon OSD Level Flight FPV Talk 22 Dec 31, 2013 08:08 PM
For Sale New Intelligent Flight Dragon OSD bfischer Aircraft - General - Miscellaneous (FS/W) 7 Jul 19, 2010 11:56 AM
Discussion FS Ready to fly Intelligent Flight Dragon OSD bfischer FPV Talk 0 Mar 03, 2010 05:57 AM
Discussion RV OSD or Dragon OSD djdaveq FPV Talk 1 Feb 06, 2009 12:18 AM
Discussion RV OSD and Dragon OSD - so near to being Really good! magnetman FPV Talk 8 Nov 11, 2008 05:13 AM