22 replies to this topic

[gearwolf]

I have a project that I'm almost to the point of a maiden flight. It's a cross between a heli and an autogyro. I have two modes. It has a flybarless head on it. In one flight mode it is a helicopter and in another the whole model will tilt 15 degrees upwards but still fly along the same vector as the helicopter mode (that angle may change since there is no set angle on autogyros). My question here concerns attitude stabilization of the model in both modes. If I preset on my TX to bank the model up 15 degrees in my 2nd flight mode can I still utilize the attitude stabilization of the board? Rudder control isn't an issue here since even if I were to use that from the board I know how to do a hardware bypass to achieve what I need. Also would it be a problem if during the autogyro mode I have the collective preset to a slight negative angle? I figure that I would ask this before I am able to obtain a board so I know what needs to be done during setup.

[MikeL]

Aren't both these requirements addressed by massaging the pitch/roll bias?

[gearwolf]

I was worried about that. After looking across forums on other flybarless controllers using an offset on the tx would mean a continuous climb rate. What I need is exactly what you need too. However from what it sounds like though dankers your way would be a perminent offset. I would only need that offset about half of the time and then the other half with the offset taken off. Just looking at the general scope of all the hardware, and I'm completly guessing here, if implemented I would need pipextreme and the cc board to complete this if I wanted the settings to change in flight, right? I wish I could code. I only took a class on c++ and didn't do well.

[MikeL]

I'd be happy to take a look, measure level of effort.

So what you're thinking Dankers is that flight-mode, which currently describes the modes (rate,att,axis,etc) for yaw/pitch/roll, would have an additional option which is r/p/y rotation and to adjust for little right roll you'd dial in a couple degrees of board rotation. Gyrocopter mode, say fm3 on his transmitter would dial in some pitch rotation to fool cc into maintaining that attitude. Sound about right?

[gearwolf]

MikeL: Thank you for looking into that. If that's what it will get the board to think it's level but fixed at 15 degrees positive pitch to the ground then that sounds perfect to me.

dankers: I thought this was going to be more complicated than that. It sounds like coptercontrol is like eepe to the 9x firmware, being able to execute complex mixes with a switch. I am starting to really like this already great system. And thank you, I would like to see this board in my system as well. Before I get a chance to obtain one I will probably get some lower grade gyro systems and test each component separately. Plus I also look forward to the revolution, sounds like really great stuff and possibly a way to really get a controlled test to see if I can get longer endurance with my system as opposed to either straight heli or straight autogyro.

[gearwolf]

I'm just curious MikeL but have you made any headway?

[MikeL]

I'm just curious MikeL but have you made any headway?

No, I'm afraid not, but I haven't forgotten about it either.

We're thrashing to get a release candidate ready for cc3d availability, right after that rolls out we'll have more cycles to devote to pushing the envelope a little further out.

[gearwolf]

Thanks for the update MikeL. Good luck to the release

[MikeL]

Thanks for the update MikeL. Good luck to the release

Thanks. I started a code review topic on this so we can get it in the mix, look for something soon as we can.

http://git.openpilot.org/cru/OPReview-229

[gearwolf]

That sounds excellent

[Mat Wellington]

time to go old skool

use a servo to tilt the board 15 degrees when you flick a switch on the transmitter

[gearwolf]

Lol did not think of that one. Too bad I don't have any more room on my rx for that or else I would try that.

[Mat Wellington]

Lol did not think of that one. Too bad I don't have any more room on my rx for that or else I would try that.

you could run it off channel 5 (flight mode) with a y lead

[Kenn Sebesta]

I made the comment in the review, but I'll repeat it here. It would be better not to trick the system about which way is really up. Rotation matrices can get cranky. However, it would be straightforward to mux in a static bias to an attitude set point. (After all, that's what we're talking about here. In rate mode, there's no sense of angle, there's just a sense of keeping whatever angle you happen to be at.)

So really all you need is a field that says "add bias" and have that be triggered by a switch. Of course, you can also do this in TX with the trim. (Again, in attitude mode. In rate mode, trim would be Bad™, but rate mode makes no sense for the kind of flight behavior you're requesting.)

[gearwolf]

Thank you guys for the suggestions. Kenn Sebesta, can I assume if you were talking about trim that in attitude mode the amount of stick movement = the angle the board moves and maintains? So if I added a mix in my tx that moved the pitch stick down 15 degrees the board would attempt to maintain that pitch angle, instead of a constant rate that would normally occur resulting in a loop/roll.

[Kenn Sebesta]

Yup, that's exactly it. In rate mode, you would be asking for 15 deg/s, in attitude mode you would simply be asking for 15 deg.

P.S. I see you're in Lex. Next time I'm back home, we'll have to do a mini fly-in.

[gearwolf]

Can't believe it was already built in. Well I still have to make it fly. It could hop but nothing great. And now I need to get the board. Where do you fly around? I haven't been able to find places that are for the public.

[Kenn Sebesta]

Masterson Station Park was where I used to fly planes around. If you're flying in close for hovering flight, then I'm sure any old farmers field will be fine.