Development request – Pixhawk to work with a Rotormast V-22 Osprey VTOL.

1,963 views
Skip to first unread message

Ian Booth

unread,
Oct 7, 2014, 7:47:08 AM10/7/14
to drones-...@googlegroups.com

Hello,

Whilst I enjoy Quads, I would very much like to connect a Pixhawk to my scale model of a V-22 Osprey VTOL.  The model is made by a US company called Rotormast (www.rotormast.com) and I think that making it work with a Pixhawk would open up great possibilities in the world of ‘UAV’s’ for a new breed of multi-purpose vehicles capable of VTOL and transition to horizontal aircraft flight.  It would be great for duration missions, carrying of payload and especially FPV.

The Rotormast V-22 model uses 3 Helicopter gyros, one for each axis and has a proprietary hardware controller board to handle various functions including the control mixing between helicopter hover and airplane mode and apply the varying gyro gains during transition as the wingtip nacelles rotate from vertical to horizontal. 

I have spoken to the V-22 developer and owner of Rotormast, and he supports the implementation of a flight controller such as the Pixhawk into the V-22 and may be prepared to assist in providing a test model to aid the development of the code.

Whilst I am not a programmer nor do I understand the APM code but I think to use the Pixhawk in the V-22 there may be three initial functional modifications required to the existing APM helicopter code as a minimum:

·      As the roll and cyclic controls swap during the transition from hover to forward flight, the gyro rates need to vary.  The gyro gains may also need to change to maintain a consistent feel as the aircraft undergoes the conversion.  The transition position of the assigned R/C Channel controlling the nacelle angle will need to be read and the Pixhawk apply the correct gyro gains to the correct axis for that nacelle position.  (I am hoping that it is possible within the Pixhawk to program in the use of different PID values to different axis depending on the input value from a radio channel).

·      Dedicate an input to the Pixhawk for the R/C transition channel so the Pixhawk knows the position of the nacelles and therefore what PID gains to apply.   

·      The V-22 controller board provides the Cyclic,Collective,Pitch Mixing (CCPM) which is required to control the two swash plates.  (The V-22 doesn't use fixed wing flight control surfaces but the swash plates on both rotors in vertical and horizontal flight.)  The helicopter version of the Pixhawk also does this same mixing so the APM software may need to change to allow these signals to pass through the Pixhawk direct to the V-22 controller. 

·         Ongoing development could include the implementation of using the Pixhawk airspeed sensor to assist with the nacelle position and/or propeller pitch based on airspeed.  This would probably be needed for auto missions.

The V-22 developer is happy to provide whatever information is needed regarding the control of the V-22.  If you are an APM firmware programmer and up for a challenge, how about something different?

So if you're interested in helping out with the development of some APM code to fly a VTOL with a difference, I'd love to hear from you.

Cheers

Ian

Robert Lefebvre

unread,
Oct 7, 2014, 10:20:43 AM10/7/14
to drones-discuss
Hi Ian,  glad to hear from you.  I am the lead developer for the conventional helicopter controls of Arducopter.  I've been interested in doing tandems for a long time but haven't gotten around to it yet.  Both longitudinal and transverse tandems have possible performance advantages.  Obviously, transverse tandem offers the possibility of doing tilt-rotor hybrid platforms such as the V-22 model.

I would be interested in working on this, but I have to fit it around other projects which have priority.  I may have time about a month from now.

The only problem is that I am not at all interested in the concept of having the Pixhawk pass gyro values out to another processor which handles the controls mixing.  To me, this is a needless complication.  It also makes the work going into the project into a sole-supplier scenario, where the only way to fly one of these things is to buy the proprietary Rotormast controller. I'd much prefer a universal solution, where everything is handled inside the single flight controller.  And there's no reason I can see not to do that.  Also, using the single-system approach, the gain scheduling and control output can be done much more intelligently than it can with the dual-system, since the Pixhawk is much more aware of the environment than the Rotormast control system ever could be.

If we can come to an agreement on the direction to take this, I would be interested in working on it.  Otherwise, I might end up working on it anyway this winter, again depending on project priorities.

Regards,
Rob



--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

David Pawlak

unread,
Oct 7, 2014, 10:50:18 AM10/7/14
to drones-...@googlegroups.com
Nice, very interested in progress reports.

100% with Rob with respect to replacing the custom electronics. 3 Helicopter gyros would be extra cost for one, and the Pixhawk has all the power and sensors one could ask for.

I like the motor/head configuration. Compact looking.

Ian Booth

unread,
Oct 7, 2014, 6:31:18 PM10/7/14
to drones-...@googlegroups.com
Hi Robert,

Many thanks for such a quick response and I'm really excited now and looking forward to working with you to get this happening.  No problems on the time frame as we can probably use that time to scope out what can be done etc.  With regards to the proprietary Rotormast controller I fully agree with you that this would replace the 3 helicopter gyro's.  To remove the Rotormast controller entirely though we would need to add some additional functions to the Pixhawk as the existing controller also manages the angle and conversion curves for both nacelles keeping them the same, manages the conversion servo trim positions and acts as a pitch governor to both rotors.

How can I get in touch with you off line so we can discuss further.  I can then also put you in touch with the V22 developer to discuss the direction to take on this.

Cheers
Ian

Craig Elder

unread,
Oct 7, 2014, 7:37:38 PM10/7/14
to drones-discuss
We are close to having VTOL support.  You might want to look at https://github.com/diydrones/ardupilot/pull/1413

--

Robert Lefebvre

unread,
Oct 7, 2014, 8:03:32 PM10/7/14
to drones-discuss
Is that actually going into master?  It'll be an important development if it does.  However, it will still need a lot of work to fly the VTOL, as I believe the Yuneec is a multi-rotor type system, and since the V-22 is helicopter based, it will require quite different control laws, that will be more similar to Helicopter.

Robert Lefebvre

unread,
Oct 7, 2014, 8:06:39 PM10/7/14
to drones-discuss
Hi Ian, no worries about the extra functions, all stuff that needs to go into Arducopter anyway, as there will be demand from many people for this type of system once it's working, and those functions will probably be required for all anyway.

If you want, you can send me an email directly, or contact me through Skype, my ID is robert.lefebvre14.

Rob

--

Randy Mackay

unread,
Oct 7, 2014, 9:49:34 PM10/7/14
to drones-...@googlegroups.com

 

     If you are satisfied with just fixed-pitch bi-copter support (as opposed to the variable pitch of the V-22 Osprey) then I have a bi-copter library provided by Miwa-san (Japan ArduCopter group) but I just haven’t gotten around to integrating it into master yet.

 

-Randy

Tom Mast

unread,
Nov 14, 2014, 7:18:38 PM11/14/14
to drones-...@googlegroups.com
Rotormast would be more than willing to support this project with what ever resources we have available.  If a Developer has the time to work a project like this.

Craig Elder

unread,
Nov 16, 2014, 10:50:17 PM11/16/14
to drones-discuss
Tom we have several groups wanting this feature and a few groups working independently to implement it.

I'm going to see what can be done to get them collaborating and working together and we can get this done sooner rather than later. 

--

Fredrik Hedberg

unread,
Dec 14, 2014, 1:26:29 PM12/14/14
to drones-...@googlegroups.com
Any updates on this? I'm interested in working on it. Any links to the related initiatives Craig?

Craig Elder

unread,
Dec 14, 2014, 8:39:59 PM12/14/14
to drones-discuss
We had Mike and Trevor from Team Tilt rotor join us on the dev call last week and I have introduced them to the guys at Bird's Eye.  Strategically we have discussed that starting with their aircraft makes more sense and then moving to something like the V-22  I have also been helping TT with the items they need to move forward as far as parts and materials and some github assistance.

The two groups are going to meet after Christmas and I'm hoping that Bird's Eye will donate a couple of aircraft to the project.

In the mean time we are discussing how best to structure the location of the code in the repository.




Team Tiltrotor

unread,
Dec 14, 2014, 9:35:44 PM12/14/14
to drones-...@googlegroups.com
Fredrik, Good to meet you. Just to add to what Craig Said, We think we are getting close to having a really good software that would be perfect for the Rotormast. The only thing keeping us from taking on the rotor mast right now is incorporating the traditional heli code to properly control the swashplates, and probably some other traditional heli code that we are less familiar with. I think once we flush out the AUTO Mode on the Fire Fly, the Rotor mast will be a good airframe add on starting with to the Fire Fly code.

Just out of curiosity, have you been working on any VTOL code? Just wondering if you have worked out some of the difficulties with the exchange/mixing of control output for roll/yaw/pitch during the thrust axis changes?
Are you flying a VTOL now?

Thanks
Trevor

Robert Lefebvre

unread,
Dec 15, 2014, 7:51:32 AM12/15/14
to drones-discuss
I had actually been planning on starting work with the Tilt Rotor guys shortly, as I now have much more time on my hands for something like this.  I was going to take the first step of getting basic hovering flight working a dual swashplate design, based only on standard Arducopter code.  I was daunted by the forward flight portion, however.  Now that Trevor and Mike are involved, I'm not sure if I should still pursue this, or leave it to them?

Rob

Team Tiltrotor

unread,
Dec 15, 2014, 8:08:57 AM12/15/14
to drones-...@googlegroups.com
Rob, good to meet you. We've also been looking forward to discussing some things with you. We are still trying to get familiar with all the developer tools that we did not know existed. When it comes to the rotor mast we will need a lot of help from your RC helicopter knowledge. No doubt we will be working together on that project. We'll also be getting with you on some Fire Fly 6 implementation.

Thanks!

Fredrik Hedberg

unread,
Dec 16, 2014, 7:09:58 PM12/16/14
to drones-...@googlegroups.com
My idea was basically the same as Robert, to fork AP_MotorsHeli into a tandem version as a first step - haven't given transition or forward flight a thought yet.

How have you guys structured the VTOL code? Is it publicly available?

Meier Lorenz

unread,
Dec 17, 2014, 3:28:16 AM12/17/14
to drones-...@googlegroups.com
Without wanting to steal developers, I thought I you might be interested in this video from the PX4 flight stack running on Pixhawk:
One particular recommendation for fast development I have is to pick initially a small airframe that is “human safe”. We got a FireFly Y6 in the lab (kindly provided by BirdsEyeView) which will be the next step.

The main aspect that allows us to move fast is that the PX4 codebase is a set of individual apps and there is no “copter” or “plane” specific main loop, which makes VTOL support rather an addition on the control side than to having to think about a whole new category.

It comes with the side effect of making our infrastructure also easier to reuse, which in turn helps other projects (including APM). And it does help us a lot, as e.g. Tridge is one of the main contributors to PX4 infrastructure and really one of the key factors getting it where it is now.

Cheers,
Lorenz


Robert Lefebvre

unread,
Dec 17, 2014, 8:35:37 AM12/17/14
to drones-discuss
Hi Trevor,

I actually just wanted to point out quickly something I'm not sure you've seen.  On the Dev call, I mentioned the idea of forgoing integration of the L1 waypoint controller in favor of the Copter Spline Nav controller.  At least in the short term.  This would probably greatly decrease the workload to integrate VTOL into the copter codebase.  So, I just wanted to show you and example of how Spline nav works for high speed flight.  This is on a helicopter, but I think you can see that it would generate a flight path which would be perfectly acceptable for airplane mode.


Rob

On 15 December 2014 at 08:08, Team Tiltrotor <teamti...@gmail.com> wrote:
Rob, good to meet you. We've also been looking forward to discussing some things with you. We are still trying to get familiar with all the developer tools that we did not know existed. When it comes to the rotor mast we will need a lot of help from your RC helicopter knowledge. No doubt we will be working together on that project. We'll also be getting with you on some Fire Fly 6 implementation.

Thanks!

Team Tiltrotor

unread,
Dec 17, 2014, 10:41:49 AM12/17/14
to drones-...@googlegroups.com
@Lorenz- Awesome, I had no idea this was in work. I'll have to take a look at the app structure and see how it's works.

@Rob. That's very impressive! I think your right on spine making things simpler after watching that. We'll be doing some homework and get some planning done over the holidays on navigation.

Samuel Dudley

unread,
Dec 18, 2014, 5:57:30 AM12/18/14
to drones-...@googlegroups.com
Hi all,
I'd like to take part in this dev if at all possible... I have two JR Forza 450's, a pixhawk and a small workshop at my disposal. I would be more interested working with the Ardupilot code base.

Cheers,
Sam

Craig Elder

unread,
Dec 18, 2014, 1:16:44 PM12/18/14
to drones-discuss
Sam I'd like to discuss your interest here.  I will PM you.

Fredrik Hedberg

unread,
Dec 18, 2014, 5:02:32 PM12/18/14
to drones-...@googlegroups.com
I duplicated and wrenched AP_MotorsHeli into a AP_MotorsHeliDual today by removing the tail and adding a second swash. Seems to work logically ok in STABILIZE on the bench using two 450s but obviously a long long way to go...

Is this a good approach Rob or are you already on this?

Team Tiltrotor

unread,
Dec 18, 2014, 6:18:13 PM12/18/14
to drones-...@googlegroups.com
That's awesome Fredrik, didn't take you long!. Mike and I are in the process of moving our VTOL code over to a more current version of copter, and we also need to make some changes since we are using pixhawk for the first time. I'm guessing we have a few weeks of work on this since software is not our forte. But now that we have the help of the developers I think we will have less obstacles. We will probably need to fly this software on our own tiltrotor or on a fire fly 6 before we attempt to convert a rotor mast. I can't give you a time line on that. But if you have the everything to hover the rotor mast that's a great start! I can't remember, does the rotormast have lateral and longitudinal cyclic or just longitudinal on each head?

I don't know if each of us ( Fredrik rob and us ) can each get a rotormast or not, but that would be a good collaboration. We'll standby until we get pixhawk up and running on a VTOL 
You received this message because you are subscribed to a topic in the Google Groups "drones-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/drones-discuss/HC0b1L5MIuE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to drones-discus...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.


--
-Team Tiltrotor

Randy Mackay

unread,
Dec 18, 2014, 7:39:13 PM12/18/14
to drones-...@googlegroups.com

Fredrik,

 

      That’s really cool.  I’d love to see that flying some time.  There’s an item on the to-do list to support tandem helicopters so once Rob has a look I suspect we’d like your patch in master!

                https://github.com/diydrones/ardupilot/issues/134

 

-Randy

Robert Lefebvre

unread,
Dec 18, 2014, 8:09:39 PM12/18/14
to drones-discuss
It's something I wanted to do, but never got around to it.  I was planning on contacting Rotormast soon to get a machine but I'm not sure if there's any point now.

I do think we need to sit down and plan this stuff out before rushing headlong into it.  It's the only way it'll end up in master.

If we want this stuff to get into Copter, instead of becoming it's own vehicle, we'll need to create an layer that fits between the desired acceleration, and the copter frame.  Basically it would handle any thrust redirection (tilting rotors).  Logically, the first step would be to create a quadcopter with 4 tilting motors, where to fly forward, you would rotate the motors forward, instead of tilting the frame forward.  So any forward acceleration demand, would result in motor tilt, not frame tilt.  The frame would always remain level.  Similar in concept, this would be able to allow a helicopter arranged like the Sikorsky X3.

In my mind, this is the first step towards integrating transitioning VTOL's into copter.  Let's not worry about wing lift and aerodynamics yet, and just get motor tilt figured out.

My understanding is that the Rotormast has no control surfaces on wings, and relies completely on rotor cyclic even in forward flight? This would also make it an ideal logical step in the path.

Another step in the path, is some kind of TECS equivalent, where the auto controllers knows to use frame pitch instead of thrust, in order to control altitude.  This would be used in conventional helicopters as well.

Fredrik Hedberg

unread,
Dec 19, 2014, 3:24:04 AM12/19/14
to drones-...@googlegroups.com
Just a little expectations management; the swashes seems to be moving in the right directions when i move the frame by hand but wouldn't pop the bubbly just yet ;) Really just a copy-paste version of you guys' excellent work from AP_MotorsHeli with yaw included as a component in the swash calculation. Have no idea how it would fly, and I won't have too much time over the holidays but I'll hopefully pick it up after NYE again.

Rob's VTOL proposal sounds very sensible, something way over my head however.

Also, just wanted to say thanks for the awesome job Rob and Randy - long time ArduCopter fan, first time hacking on it.


Fredrik

Fredrik Hedberg

unread,
Dec 21, 2014, 9:40:26 AM12/21/14
to drones-...@googlegroups.com
Robert, Randy, Ian, Tom, Sam,

I've submitted my first stab at dual swashes here: https://github.com/diydrones/ardupilot/pull/1699

Again, big disclaimer; this is only really bench tested and for practical reasons I won't be able to take it to real hover test anytime soon unfortunately :( 

Would love some input on the general structure (as a first time ArduCopter hacker), as well as help with testing this out and taking it further if there's any adventurous minds with a tandem helicopter and a Pixhawk out there.

An important thing that remains to be added is the pre-compensation of yaw when applying DCP.

Best,

Fredrik

Robert Lefebvre

unread,
Dec 21, 2014, 10:27:43 AM12/21/14
to drones-discuss
I had a quick look and left you some comments.  So far I think it's a great first attempt. 

The roll-yaw coupling is an important point.  We already have collective-yaw coupling compensation.  There's also a case to be made for collective-pitch compensation on conventional helis.  Maybe we should have a think if there's a better way to handle this?  Maybe there isn't.

If Randy did want to move ahead with this PR, I would probably request that you simply strip out everything out of the tandem class that has to deal with the motor control stuff.  As I mentioned, I want to pull that out of AP_MotorsHeli, and put it in it's own class.  I wanted to do this anyway.  (though this will highlight the problem with the naming of AP_Motors, because the new class would be something like AP_MotorsHeliMotor ?)  :)

I personally like the way you have made a new class.  In the past however, when I have talked about supporting a 4-servo swash,IIRC Randy had said he wanted to simply use the single 3-servo class, and just expand it to accept 4 servos, while leaving 1 of them "dead" when being used on 3-servo mode.  I didn't like this approach as it wastes a channel.  But, if the same logic were to be applied, the tandem heli class is really the same as a 3-servo heli, just with 6-servos and slightly different mixing.  So the same argument could be used that we just allow up to 6 servos on a heli, and leave 2 of them dead on a 3-servo machine.

So, we'll see what Randy would like to do.  I prefer different classes for these machines, or maybe add another layer of child/parent.  It's just as relevant for helis with different servo counts and mixing as it is for multi-rotors.


jolyboy

unread,
Dec 22, 2014, 2:43:55 AM12/22/14
to drones-...@googlegroups.com
I built a "twinn rexx" ages ago, based on trex 450 hardware. 6 servos total, with rotors coupled via a 5mm driveshaft. Mixing was achieved with a TH-2 mixer board, which allows for programmable mixing (DCP included) for the 6 servo channels. It required double the maintenance of a traditional heli so I dismantled it.

A common problem with the tandem configuration is rotor vortex interference. Coming out of forward flight, the rear rotor enters the wash of the front rotor, causing sudden loss of lift in the rear rotor and extreme loss of altitude as the attitude controller compensates. Users were able to predict it, so it wasn't too big of a deal unless they forgot to be ready to juice the collective.

Yaw control became an issue when I converted to FBL rotors, because the feathering was too rigid. There wasn't sufficient deflection of the disk to allow for adequate yaw control. Softer dampers would solve the issue.

I am really confident you guys will find a viable solution with the tandem frame. The issues I had aside from increased maintenance can be overcome with some smart programming. Really excited to see where this goes!!

Ron Graham

unread,
Mar 17, 2015, 2:04:59 AM3/17/15
to drones-...@googlegroups.com
I am interested in this topic, but note the last input was December last year.Is there now no interest, or has the discussion moved?
I am in the process of obtaining an Osprey A-22 kit with the intention of (hopefully):
a) building it and getting it flying with the supplied controller.
b) add an APM, GPS, telemetry RX/TX, battery voltage and current sensing board in order to have that telemetry available via MAV link .. leaving the original controller to fly the machine.
c) Progress and swap the APM to a Pixhawk with the Pixhawk controlling everything. Hence the interest in this discussion
I may be contacted by PM at ron@rambling.info

Fredrik Hedberg

unread,
Mar 17, 2015, 4:34:00 AM3/17/15
to drones-...@googlegroups.com
Hi,

I've put together dual-swash rotor support for both tandem and transverse helicopters. Mike and Trevor are working on the hard parts, i.e. transitioning to plane mode, but using the FireFly6 hexacopter.

The dual-swash code has still seen limited full flight-testing, but is available at https://github.com/diydrones/ardupilot/pull/1860 if you want to try it out. It should support the Rotomast setup if you configure the positions of servos 1, 2, 4 and 5 correctly to 0, 180, 0, 180 degrees respectively, leaving out servo 3 and 6, as differential-collective-pitch will be used for roll.

Tom from Rotormast ordered a Pixhawk for some testing but don't think he has gotten around to it yet. I'm not too familiar with the supplied controller, but I think you will find it hard to fly in anything but manual mode if you add ArduCopter on top of it.


Best,

Fredrik


On Tuesday, March 17, 2015 at 7:04:59 AM UTC+1, Ron Graham wrote:
I am interested in this topic, but note the last input was December last year.Is there now no interest, or has the discussion moved?
I am in the process of obtaining an Osprey A-22 kit with the intention of (hopefully):
a) building it and getting it flying with the supplied controller.
b) add an APM, GPS, telemetry RX/TX, battery voltage and current sensing board in order to have that telemetry available via MAV link .. leaving the original controller to fly the machine.
c) Progress and swap the APM to a Pixhawk with the Pixhawk controlling everything. Hence the interest in this discussion
I may be contacted by PM at r...@rambling.info

Team Tiltrotor

unread,
Mar 17, 2015, 6:57:44 PM3/17/15
to drones-...@googlegroups.com
Hi Ron, We are very much hard at work on a tiltrotor code.

If you did not see our progress last year read this

Since we posted this we have been hard at work updating the code and integrating with "master". We also moved from APM 2.5 to Pixhawk.
In order to expedite a working tiltrotor code, we have acquired a Fire Fly 6. We plan on working on this aircraft until we have a tiltrotor code that can do a "take off- convert- cruise -convert- landing" autonomous mission.
After that is accomplished we would love to go back to tiltrotor like the Rotor Mast. Enabling the Rotor Mast will not be too much work after we get done on the Fire Fly 6, Thanks to Fredrik's tandem swashplate code.
 
It's hard to predict a timeline, but I imagine this summer will yield lots of progress in tiltrotor code.

Trevor

Ron Graham

unread,
Mar 24, 2015, 5:50:52 PM3/24/15
to drones-...@googlegroups.com
Thanks to Fredrik & Team Tiltrotor for their response and that they are still progressing with this challange. I have only just started to assemble my Osprey A-22, so not sure how long it will take. Plus, of course, getting it to fly with the original controller. Hopefully, by then, some others will be changing over to the Pixhawk controller .. note someone did say that Tom has ordered a Pixhawk! 
 
Ron.

Ian Booth

unread,
Apr 29, 2015, 10:13:54 PM4/29/15
to drones-...@googlegroups.com
Good day everyone,

Looks like there has been some significant inroads into this since my original post.  How are you getting on with your V-22 build Ron?  If I can offer any assistance or if you have any queries please let me know and as before I'm happy to conduct testing on my V-22.

I noticed with interest that S Korea have just unveiled their first unmanned tiltrotor aircraft prototype, the TR-60 UAV.

http://www.airrecognition.com/index.php/archive-world-worldwide-news-air-force-aviation-aerospace-air-military-defence-industry/global-news-2015/april/1684-south-korea-unveils-new-tr-60-unmanned-aerial-vehicle-prototype.html

Cheers
Ian

Ron Graham

unread,
May 1, 2015, 4:20:16 AM5/1/15
to drones-...@googlegroups.com
Hi Ian and other Group Members,

My V-22 build is progressing .. have essentially finished the
mechanical side of things, plus most of the wiring. In the process of
verifying / debugging the electronics and note some possible hurdles ahead.
Awaiting the arrival of the USB programming interface for the Castle ESC's.


Ian, my email is r...@ramblin.info .. would you mind just dropping
me an email so that I will have your address. Having someone available for
offering any possible assistance would be great. Possibly save me quite a
bit of time as well.

Regards, Ron.


At 07:10 PM 4/29/2015 -0700, you wrote:
>Good day everyone,
>
>Looks like there has been some significant inroads into this since my
>original post. How are you getting on with your V-22 build Ron? If I can
>offer any assistance or if you have any queries please let me know and as
>before I'm happy to conduct testing on my V-22.
>
>I noticed with interest that S Korea have just unveiled their first
>unmanned tiltrotor aircraft prototype, the TR-60 UAV.
>
>
>
>Cheers
>Ian
>
>
>On Tuesday, October 7, 2014 at 9:47:08 PM UTC+10, Ian Booth wrote:
>Hello,
>
>Whilst I enjoy Quads, I would very much like to connect a Pixhawk to my
>scale model of a V-22 Osprey VTOL. The model is made by a US company
>called Rotormast (<http://www.rotormast.com/>www.rotormast.com) and I
>think that making it work with a Pixhawk would open up great possibilities
>in the world of ‘UAV’s’ for a new breed of multi-purpose vehicles
>capable of VTOL and transition to horizontal aircraft flight. It would be
>great for duration missions, carrying of payload and especially FPV.
>
>The Rotormast V-22 model uses 3 Helicopter gyros, one for each axis and
>has a proprietary hardware controller board to handle various functions
>including the control mixing between helicopter hover and airplane mode
>and apply the varying gyro gains during transition as the wingtip nacelles
>rotate from vertical to horizontal.
>
>I have spoken to the V-22 developer and owner of Rotormast, and he
>supports the implementation of a flight controller such as the Pixhawk
>into the V-22 and may be prepared to assist in providing a test model to
>aid the development of the code.
>
>Whilst I am not a programmer nor do I understand the APM code but I think
>to use the Pixhawk in the V-22 there may be three initial functional
>modifications required to the existing APM helicopter code as a minimum:
>
>· As the roll and cyclic controls swap during the transition from
>hover to forward flight, the gyro rates need to vary. The gyro gains may
>also need to change to maintain a consistent feel as the aircraft
>undergoes the conversion. The transition position of the assigned R/C
>Channel controlling the nacelle angle will need to be read and the Pixhawk
>apply the correct gyro gains to the correct axis for that nacelle
>position. (I am hoping that it is possible within the Pixhawk to program
>in the use of different PID values to different axis depending on the
>input value from a radio channel).
>
>· Dedicate an input to the Pixhawk for the R/C transition channel so
>the Pixhawk knows the position of the nacelles and therefore what PID
>gains to apply.
>
>· The V-22 controller board provides the Cyclic,Collective,Pitch
>Mixing (CCPM) which is required to control the two swash plates. (The
>V-22 doesn't use fixed wing flight control surfaces but the swash plates
>on both rotors in vertical and horizontal flight.) The helicopter version
>of the Pixhawk also does this same mixing so the APM software may need to
>change to allow these signals to pass through the Pixhawk direct to the
>V-22 controller.
>
>· Ongoing development could include the implementation of using
>the Pixhawk airspeed sensor to assist with the nacelle position and/or
>propeller pitch based on airspeed. This would probably be needed for auto
>missions.
>
>The V-22 developer is happy to provide whatever information is needed
>regarding the control of the V-22. If you are an APM firmware programmer
>and up for a challenge, how about something different?
>
>So if you're interested in helping out with the development of some APM
>code to fly a VTOL with a difference, I'd love to hear from you.
>
>Cheers
>
>Ian
>
>--
>You received this message because you are subscribed to a topic in the
>Google Groups "drones-discuss" group.
>To unsubscribe from this topic, visit
><https://groups.google.com/d/topic/drones-discuss/HC0b1L5MIuE/unsubscribe>https://groups.google.com/d/topic/drones-discuss/HC0b1L5MIuE/unsubscribe.
>To unsubscribe from this group and all its topics, send an email to
><mailto:drones-discus...@googlegroups.com>drones-discus...@googlegroups.com.
>For more options, visit
><https://groups.google.com/d/optout>https://groups.google.com/d/optout.

Al B

unread,
May 11, 2015, 12:22:13 PM5/11/15
to drones-...@googlegroups.com
Here is a similar hybrid design like in the V-22 Osprey, but with tiltwings rather than tiltrotor.

Nigel Cartwright

unread,
Mar 16, 2016, 9:08:26 AM3/16/16
to drones-discuss


Hi all

Has there been any progress with this project?? I have a Rotormast V22 and I'd like to use a Pixhawk...

Robert Lefebvre

unread,
Mar 16, 2016, 10:38:15 AM3/16/16
to drones-discuss
No, I don't think there has been.  I'd be happy to take on the project if it's something that somebody would be willing to fund development.

Rob

On 16 March 2016 at 05:10, Nigel Cartwright <ni...@modeltek.com> wrote:


Hi all

Has there been any progress with this project?? I have a Rotormast V22 and I'd like to use a Pixhawk...

--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Craig Elder

unread,
Mar 16, 2016, 11:56:21 AM3/16/16
to drones-discuss
It should be quite a bit easier to do now that we have the quadplane http://plane.ardupilot.com/wiki/quadplane-support/ available

Robert Lefebvre

unread,
Mar 16, 2016, 12:23:56 PM3/16/16
to drones-discuss
Yeah, I was thinking about that if it should be based on plane code and copter code.  That's a large topic, for another day...

Randy Mackay

unread,
Mar 16, 2016, 7:58:54 PM3/16/16
to drones-...@googlegroups.com

 

     I have a severely bit-rotted bi-copter motor library from a person here in Japan that I could dig out for someone if they take on the project.

 

-Randy

Fredrik Hedberg

unread,
Mar 17, 2016, 12:45:20 PM3/17/16
to drones-discuss
I've flown something similar to the Rotormast V-22 with this branch: https://github.com/fhedberg/ardupilot/commits/copter-dual-heli-tr

Does only work in STABILIZE and ACRO, and is really weird to fly when transitioning to forward flight. But should be pretty straight forward to integrate with Tridge's VTOL-plane code.
Reply all
Reply to author
Forward
0 new messages