We at the El Reg Special Projects Bureau are delighted to welcome the OpenPilot initiative aboard our Low Orbit Helium Assisted Navigator (LOHAN) programme. Click here for a bigger version of the LOHAN graphic The chaps from OpenPilot got in touch to say they reckoned their CopterControl board (pictured below) could do the job …
This project is tickling parts of me that haven't been tickled since the Byker tickle outbreak of '86.
Well bowled to all involved.
Be sure to run some tests
of the plane in "glider" mode before the launch, so you can verify that the plane can land safely. Of course, this can't test high-altitude behaviour let alone rocket-powered flight, but the plane will eventually get to low altitude, and that is where it can get into trouble most.
As for plane design, I still suggest putting the rocket in front, so it pulls the plane, otherwise there is too much risk the plane will just flip over once you burn the rocket. The wings are aft separated by some fireproof beams, so the rocket won't burn them. Or you can arrange the wings in a triangle, so the rocket exhaust goes through the triangle. You can have a small control-surface wing in front with the rocket, making it a canard-style design.
Toasty tail plane batman
I suspect you may not have seen the kind of rocket motor that might be used in full effect. There isnt much that would withstand the rocket exhaust plume. Putting the rocket motor up front gives you a bunch of issues:
Assymetric thrust: as the motor has to be above or below the aircraft.
Drag as you need some kind of pod to contain the motor outside the airframe.
Thermal protection of the airframe which will add mass/complexity
A rear engined approach is probably wiser. Then you can get the thrust along the axis of the aircraft and have the motor contained within the fuselage.
The most succesful boost gliders I have seen have been rear engined (as flown by the UKs FIA R/C boost glider international gold meadlist Mike Francies and a Hybrid rocket motor powered R/C delta wing glider regularly flown by John Barber a member of the East Anglian Rocketry Society).
As to flipping, that is a risk that is inherent to Boost Gliders. Infact it is the usual flight failure mode ;). The aerodynamic requirements of a rocket and a plane are different. Rockets need neutral lift so that the fins / wings do not generate lift which would cause the rocket/plane to veer off course. Aircraft need lift so that they can fly. In boost configuration Vulture 2 has to be trimmed to give no lift. This has to be maintained until the craft reaches apogee at which point it will need to switch to plane/lift configuration so that it can fly back.
The change in the CofG of the craft needs to be addressed. Pre launch you have the full rocket motor mass at the rear fo the craft. As the motor burns it loses mass and the CofG moves forward. Providing the rocket phase of the design is good this change in CofG makes the craft more stable in rocket flight. For a rocket to be stable the Centre of Gravity needs to be infront of the Center of Pressure, CofP, (the point around which the aerodynamic foces balance). What makes for a stable rocket doesnt make for a stable plane. The plane needs to be trimmed to fly stably with a spent motor casing and fuel grain onboard not a live one. Many model rocket boost gliders eject the spent cardboard motor casing somewhere close to apogee. This shifts the CofG further forward still but probably isnt something you would want to do with a larger metal motor casing at a significant altitude
"Rockets need neutral lift so that the fins / wings do not generate lift which would cause the rocket/plane to veer off course."
"For a rocket to be stable the Centre of Gravity needs to be infront of the Center of Pressure. What makes for a stable rocket doesn't make for a stable plane."
There's a saying in radio controlled gliding (which I do a lot of..) "A nose heavy plane flies badly, a tail heavy plane flies once". So for rocket flying, you want a very forward CoG and no wings, for plane flying the CoG wants to be quite a way back from there, but not too much or it becomes uncontrollable.
That indicates to me that the wings should fold *forwards* for the rocket phase of the flight, and spring back into place when the motor burns out. That would give the no-lift, well-forward CoG to start with and a predictable CoG with good lift for the gliding phase. And could be done entirely with stored spring energy, not much added weight (as the spar could also be the spring?) and solid positioning using some sort of magnetic latch to hold them back when the spring is released...
Should have gone Ardupilot!
While CopterControl is a promising project, Ardupilot Mega has about 50,000 more hours of real time testing and a community of 17,000 members.
Nothing against Openpilot, but's Alpha compared to some of the solutions out there..
Well done chaps, very impressive. I've flown a lot of the AutoPilots around from the Open Source ones to the high end commercial variants, OpenPilot is by far the most promising and the most serious in the Open Source world, that is the right choice without doubt.
@Anonymous Coward: Ardunio is a fun stuff, but is not even close to the same league, I sold my Ardupilot long ago as it is poor and over priced. Time to retire all that 8 bit stuff and get with the times.
To the people at Openpilot and the LOHAN team
Guys, you rock!!!
Trifling legal issues
Don't you need a licence to operate a UAV?
Also, the whole system needs compliance with standards... which does not exist yet.
There's a draft version of STANAG 4703 (Light UAS Systems Airworthiness Requirements) around, and the constraints are quite hefty.
Depends on jurisdiction
PARIS lofted from sunny Spain, where they might be more relaxed about ballocketing about with this stuff. Besides, it's not Unmanned at all: remember the Playmonaut!
I give you the "Council Regulation (EEC) No 3922/91 of 16 December 1991 on the harmonization of technical requirements and administrative procedures in the field of civil aviation "
Unless the Playmonaut is sentient, I''m pretty sure its covered by the UAV regs.
p.s. I use to be a rocket scientist
p.p.s Maybe LOHAN project meetings shouldn't be scheduled for Friday lunchtimes? (icon)
In more paranoid countries this can be classed as OpenWarhead.
Not for non commercial purposes...
That OpenPilot stuff looks way cool.
And just remembered that I've got a cheapy Maplins microlight model sat in a box somewhere, which I never used much bcos it only had binary (on/off) controls for power and turn. Always meant to do something with that to make it properly controllable, but never quite got round to it.
Only if they catch you. :)
Thing is, think about where there could be problems. Going up, the weather balloon gets you past the flight envelope of every plane in existence (even including the U2 and SR71). So what's LOHAN going to hit when it separates from the balloon? No safety worries there.
Then there's the coming down phase. Is it more dangerous to have LOHAN gliding under electronic control than to have it simply falling like a brick (as PARIS did) or gliding in a random direction with an inherently-stable aircraft (as PARIS was supposed to but didn't)? Simple answer: not really.
And if you're launching a weather balloon, there's already going to be a NOTAM saying "don't go near this place today, otherwise you might hit something". So people will know to stay away. The problem with UAVs is when they stooge around in a crowded airspace - as happened with the army around Salisbury plain.
Ah, I see that someone with practical experience in building aircraft is noting the same problem I spotted and remarked on in previous LOHAN discussions: the thin air in the upper atmosphere is insufficient to provide enough lift, or control.
I think I know a time-tested way to work around this: drogues. Even Mars landers have used drogue chutes during the initial descent through the upper atmosphere, and Mars doesn't have much atmospheric pressure even at ground level. You wouldn't really even need a drogue chute for LOHAN, just something like a long streamer. I thought about it, you could put two drogue streamers, one from each wingtip, this would keep the nose pointed down and also prevent spinning. Then when the aircraft reached lower altitudes with enough air to actually fly, the drogues could be released, and it can level off and fly in a controlled path.
Drogues work pretty well, I used to build light model rockets that used drogues rather than parachutes. That worked perfectly in high winds, when you didn't want a long descent time to allow the winds to blow the rocket into the next county.
I was browsing RC sites...
And found this gem: http://www.twitt.org/1partdrib.html
I don't know how practical it would be, but it would be a blast if LOHAN's Vulture II actually looked like a vulture!
Wish I could join in the fun!
Very thin air
Exciting times, this in no way is simple, I speak to the CAA regularly about airspace and UAS so not too afraid about that. If they need it I am a CPL and a couple of chums are ATPLs so a suitably qualified commander could be found.
Obviously the playmonaut will have the last say.
The rocket part is the hard bit, gliding from alt has been done many times
- YARR! Pirates walk the plank: DMCA magnets sink in Google results
- Pics Whisper tracks its users. So we tracked down its LA office. This is what happened next
- OnePlus One cut-price Android phone on sale to all... for 1 HOUR
- UNIX greybeards threaten Debian fork over systemd plan
- Ex-US Navy fighter pilot MIT prof: Drones beat humans - I should know