Jump to content
Sign in to follow this  
Dimitris76

Question about Dragon OSD

Recommended Posts

Hi to everyone!

It's been awhile since I last posted on this forum and going through the recent posts I couldn't help but notice the "advancements" you guys did the last couple of years!

Anyway, my set up right now is on a 60 size heli with a heading hold gyro and FMA's Copilot. This setup is working pretty good for me in stabilizing the aircraft mid-air and hopefully will do it succesfully in case of R/C signa loss (fortunatelly never had the chance to actually test it).

While browsing through Dragon's pdf manual the Autopilot feature grabbed my attention and I am wondering if it could somehow be implemented on a helicopter platform.

So here is the plan:

The Dragon's rudder output would be connected to the heli's yaw (rudder) servo and the elevator output to the collective pitch servo (while Futaba's Governor GV-1 will take care of the throttle and subsequently the engine's/main rotor's speed).

A second option (without the Futaba Governor) would be setting the collective's failsafe around center stick position and connect the Dragon's elevator output to the throttle. That way the engine hopefully will "tolerate" some gradual throttle variation during autopilot flight (without bogging down or overspeeding - like the old fixed pitch helicopters used to fly).

The front/aft cyclic failsafe will be set for a slow forward flight and left/right bank cyclic to neutral (both taken care and dampened by FMA Co-pilot).

That way I am still having the self leveling hovering that FMA offers plus I get the "return home" function in case of emergency.

Will it work or am I missing something?

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

I'm thinking if you want to use the dragon on a helicopter you're going to have to dive in and start modifying the code yourself to make it work.

Do you know C?

Share this post


Link to post
Share on other sites

Hi Mark!

No unfortunately I don't have a clue about programming... :-(

I could learn though if the hardware can do it!

Is it easy to explain why it wouldnt work and in which way the code should be modified?

Is there any chance to release a firmware version for us the "rotary modellers" in the near future?

Dimitris

Share this post


Link to post
Share on other sites

It's probably risky trying this. Doing an autopilot return-to-home feature on a heli is very complicated I would imagine. When in a hover, throttle/pitch controls altitude, but in forward flight, elevator (fore/aft pitch) acts more as the altitude control, whereas pitch controls speed. Having to deal with these dynamic circumstances is very difficult for a control system. Plus, due to the way Dragon controls the aircraft, it assumes that it is inherently stable, which we all know is NOT the case for heli's! :) I'm not saying it won't work. But it would be quite challenging most likely.

I should add... the Dragon has no direct sense of the aircraft's attitude. It does not know specifically what pitch/roll configuration the aircraft is in. It infers this based on the ROT (rate of turn) and ROC (rate of climb) information from the GPS. That is why it's important for the aircraft to be very stable. If it were to suddenly roll over, the Dragon would not know it, and would not be able to cope with the sudden change in attitude. However, Daniel is working on an IMU (Inertial Measurement Unit) to interface with Dragon that will give exact roll/pitch readings. When that comes out, I would say your chances of incorporating it into a heli are MUCH improved. In fact, I would definitely say it's possible to do then, although the autopilot algorithms may have to be re-written.

Edited by twinturbostang

Share this post


Link to post
Share on other sites

Hmm. Re-reading your first post, it may actually work. The key, which I missed the first time reading it is here:

The front/aft cyclic failsafe will be set for a slow forward flight and left/right bank cyclic to neutral (both taken care and dampened by FMA Co-pilot).

IF you set the fore/aft pitch failsafe to just the right spot, and also set the co-pilot to the right sensitivity, then it might just give a slight forward motion while still maintaining stability. The elevator output of the Dragon would hold the altitude steady, and the rudder output would turn the heli back towards home.

I would imagine if you were heading away from yourself and the system was activated, it might look a bit strange (flat turn?). And depending on the gains, it might actually end up flying backwards, if the rate of turn was set too high. That would be a bad scenario because the GPS doesn't know which way it's facing. It just knows in which direction it's traveling. In that case, it could command an incorrect rudder turn. So you can see you could possibly run into trouble. But with everything set just right, it might just work!

edit: One more thing. lol The best case would probably be to also have a magnetometer hooked up to Dragon. That way the actual heading of the heli (from the magnetometer), and the direction of travel (from the GPS) could be compared. Now you have all the information you need. You can determine if you are flying forwards or backwards, and can base your control decisions accordingly.

Edited by twinturbostang

Share this post


Link to post
Share on other sites

So in other words...

it MIGHT work but considering the slow forward speed of a heli (in comparison to a model airplane) the ROTLIMIT and ROTSTEPGAIN parameters should be set pretty low.

That way the yaw will change gradually and the heli will have adequate time to overcome it's innertia and slowly adjust it's tracking heading along the fuselage's longitudinal axis.

Of course since the GPS can sence and output only information about the track heading (and thus the autopilot can only correct only according to this) and the result will be constant yaw corrections which hopefully will result in coinciding the track with the magnetic heading.

The actual heading might be a little off (due to the wind vane phenomenon acting on the tail's aerodynamic surfaces) but that's something that applies to the airplanes too.

Am I right here or I am just trying to convince myself?

Dimitris

Share this post


Link to post
Share on other sites

Well, hopefully Daniel will chime in here also. But that sounds pretty good to me. Yes, I would think the ROT would have to be very low, which I think it is actually since he's currently relying on that to infer the roll orientation of the plane.

The thing I worry about with a heli is side slipping. You don't have that problem with an airplane because you can't fly sideways. :) But with a heli, and since you are doing "rudder turns", it's possible to get into a position where the heli is flying sidways (or even backwards as I mentioned before). Really what you want to do is to be able to roll the heli in the direction of the turn also (coordinated turn). That would require another input/output though. The Dragon does have a secondary aileron input (for planes that use dual aileron channels). It's possible you could write your own software to use that as cyclic roll control as well. Try to do somewhat of a coordinated turn by commanding rudder and aileron at specific rates. I can see it could get complicated in the algorithms. But as I said... it may be possible!! :D

BTW, this is a rather fascinating problem. I have an old .30 size heli from way back when, that I used to fly a little bit. I did not get very good at flying it, and then gave up R/C for many years. I'm back in R/C since maybe 2-3 years ago, and I have thought about getting back into heli's as well. Probably electric too, since the technology seems to have matured quite a bit. I will definitely be interested in seeing any progress on this. :)

Share this post


Link to post
Share on other sites

That's a great idea actually!

I forgot that Dragon has a 2nd aileron output. That connected to the left/right cyclic servo could introduce a "slight" roll to coordinate the turn - hopefully without any extra code writing, since the yaw output will be minimal too....

Long ago, I have seen somewhere a small circuit for sale that plugs in line with the servo and adds servo throw adjustsment.

Maybe with some trial and error the two aileron outputs could be adjusted for coordinated turns on a given forward speed.

Dimitris

Share this post


Link to post
Share on other sites
The thing I worry about with a heli is side slipping. You don't have that problem with an airplane because you can't fly sideways. :) But with a heli, and since you are doing "rudder turns", it's possible to get into a position where the heli is flying sidways (or even backwards as I mentioned before).

Well, if the system is set up like he described (copilot with pitch offset commanded by the failsafe) anf the ROT is low enough this should not even happen. The Dragon doesn't have pitch/roll info, but the copilot does. It will do the job of keeping the heli flat on the roll axis and maintaining a forward pitch which will ensure the flight direction follows the heading. Without the pitch offset then it would be like you said.

I'd be much more concerned about the collective pitch/throttle thing though. A failsafe set to "more or less hover" will be risky. Even if only a bit off the thing will go down to the ground in a matter of seconds - or even worse, get satellised, which for sure won't help about the R/C loss that threw it into failsafe.

I'd nearly be tempted to try throwing the pitch control to the alt hold of the Dragon, VERY carefully of course, with min/max limits only 1-2° away from hover on each side.

Of course all this thread must suppose you have a H1 head with independent control of all functions...

Share this post


Link to post
Share on other sites
Well, if the system is set up like he described (copilot with pitch offset commanded by the failsafe) anf the ROT is low enough this should not even happen. The Dragon doesn't have pitch/roll info, but the copilot does. It will do the job of keeping the heli flat on the roll axis and maintaining a forward pitch which will ensure the flight direction follows the heading. Without the pitch offset then it would be like you said.

As long as ROT is sufficiently low, then yes. Otherwise, not necessarily, because as Dimitris said, the heli's inertia will continue to carry the heli in the same direction even though the direction the heli is pointing will change. This causes a side slip situation where the heli is not moving in the direction it is pointing.

I'd be much more concerned about the collective pitch/throttle thing though. A failsafe set to "more or less hover" will be risky. Even if only a bit off the thing will go down to the ground in a matter of seconds - or even worse, get satellised, which for sure won't help about the R/C loss that threw it into failsafe.

I'd nearly be tempted to try throwing the pitch control to the alt hold of the Dragon, VERY carefully of course, with min/max limits only 1-2° away from hover on each side.

Dimitris mentioned this as well in his first post... sending the pitch channel to the elevator input of the Dragon, which is the altitude hold input.

Of course all this thread must suppose you have a H1 head with independent control of all functions...

Yes, good point. If you have a 120* CCPM setup, it may not be possible to do without more inputs to the Dragon and a whole bunch of mixing algorithms as well. :)

Share this post


Link to post
Share on other sites
Dimitris mentioned this as well in his first post...  sending the pitch channel to the elevator input of the Dragon, which is the altitude hold input.

Attention! You could do that (sending the pitch channel to Dragon) only if a governor takes care of the throttle in order to hold steady rotor's prm.

If you don't have a governor (or you don't want to install one) you should send the throttle channel to Dragon and set collective pitch's failsafe to an average hovering/slow forward flight value (4-5 degrees?).

All the fixed pitch helicopters (popular in the old days) did exactly this - had fixed collective (adjusted only with tools on the ground) and the pilot could vary during flight only the throttle. I am not aware of a heli that worked the other way around (fixed throttle-variable pitch).

Dimitris

Share this post


Link to post
Share on other sites

Depends on your heli. If it's electric you can just fix throttle failsafe at 80% on the TX and go with that. Glow powered will have less stable RPM indeed, but on the other hand these engines aren't made to cope with a very wide RPM range.

Share this post


Link to post
Share on other sites

I'll test it on a friend's Real Flight simulator and I hope these "real life physics" they advertise will let us know if Dragon is up to the task on a real heli...

Kilrah,

do you have a Dragon board? Did you test it on your heli yet?

Dimitris

Share this post


Link to post
Share on other sites

I have a Dragon but haven't tested the return home feature at all yet.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×