DARPA Launches Gremlins Program

[bobbymike]

http://www.darpa.mil/news-events/2015-08-28

For decades, U.S. military air operations have relied on increasingly capable multi-function manned aircraft to execute critical combat and non-combat missions. Adversaries’ abilities to detect and engage those aircraft from longer ranges have improved over time as well, however, driving up the costs for vehicle design, operation and replacement. An ability to send large numbers of small unmanned air systems (UAS) with coordinated, distributed capabilities could provide U.S. forces with improved operational flexibility at much lower cost than is possible with today’s expensive, all-in-one platforms—especially if those unmanned systems could be retrieved for reuse while airborne. So far, however, the technology to project volleys of low-cost, reusable systems over great distances and retrieve them in mid-air has remained out of reach.

To help make that technology a reality, DARPA has launched the Gremlins program. Named for the imaginary, mischievous imps that became the good luck charms of many British pilots during World War II, the program seeks to show the feasibility of conducting safe, reliable operations involving multiple air-launched, air-recoverable unmanned systems. The program also aims to prove that such systems, or “gremlins,” could provide significant cost advantages over expendable systems, spreading out payload and airframe costs over multiple uses instead of just one.

“Our goal is to conduct a compelling proof-of-concept flight demonstration that could employ intelligence, surveillance and reconnaissance (ISR) and other modular, non-kinetic payloads in a robust, responsive and affordable manner,” said Dan Patt, DARPA program manager.

The Gremlins program seeks to expand upon DARPA’s Request for Information (RFI) last year, which invited novel concepts for distributed airborne capabilities. It also aims to leverage DARPA’s prior success in developing automated aerial refueling capabilities, as well the Agency’s current efforts to create advanced UAS capture systems for ships.

 

[bring_it_on]

The Air-Recovery bit seems extremely interesting...

 

[Sferrin]

Hopefully it's something better than trying to snag it out of the air while it's dangling on a parachute.

"When the gremlins complete their mission, a C-130 transport aircraft would retrieve them in the air and carry them home, where ground crews would prepare them for their next use within 24 hours."

A net in the cargo hold maybe?

 

[LowObservable]

I suspect that, like tail-sitting, it's something that has become much easier with modern technology.


Hooking on to a trapeze can't be much more difficult than spearing a refueling drogue, particularly if the trapeze itself is stable and controlled.

 

[Triton]

"DARPA Wants Friendly Gremlin Drones"

Just don’t give them autonomy after midnight
By Kelsey D. Atherton Posted Yesterday at 11:15am

Source:
http://www.popsci.com/darpa-wants-friendly-gremlin-drones

It’s no secret that modern military aircraft are really, really expensive. Fighters like the F-35B cost about $134 million apiece, which makes them both attractive targets and potential liabilities in battle. To combat the high costs of modern planes, DARPA wants swarms of cheaper, useful drones, with expected lifespans of around 20 uses. They’re calling these drones “gremlins”.

DARPA says it envisions the gremlins as an intermediate space between one-use missiles and platforms like fighter jets that remain in service for decades. DARPA program manager Dan Patt explained in a statement:

“We wouldn't be discarding the entire airframe, engine, avionics and payload with every mission, as is done with missiles, but we also wouldn't have to carry the maintainability and operational cost burdens of today's reusable systems, which are meant to stay in service for decades.”

This is hardly the first time DARPA’s talked about cheap, reusable drones complementing the abilities of more expensive manned aircraft. Earlier this year DARPA released a concept video demonstrating an attack where a fighter controls drones scouting ahead of a C-130 transport plane. Once the targets are found, the C-130 releases cheap, swarming drones and cruise missiles to attack the enemy defenses, overwhelming the enemy with attackers. The Air Force has also expressed interest in cheap disposable attack drones that can be launched from other aircraft

In order for this kind of attack to work, the Air Force needs the gremlins to exist. On September 24th, DARPA is hosting a Proposer’s Day for companies to see what DARPA wants from the program. DARPA is funding program in three stages, with the third one being a full flight system demonstration. Should DARPA get the reusable yet disposable vehicle it wants, decades from now enemy aircraft may find something a little unsettling on their wing.

 

[Sferrin]

I suspect that, like tail-sitting, it's something that has become much easier with modern technology.


Hooking on to a trapeze can't be much more difficult than spearing a refueling drogue, particularly if the trapeze itself is stable and controlled.

Wouldn't they want to bring it into the cargo bay though? (Maybe for refueling/rearming?)

 

[Dragon029]

You could just have a swinging trapeze that hangs under the rear cargo back door and then swings / retracts back into the bay once the Gremlin is hooked.

 

[Jeb]

Formation flight into the cargo bay where the idiot brave loadmaster can just reach over and hook a leash on it. ;D

 

[Sferrin]

Formation flight into the cargo bay where the idiot brave loadmaster can just reach over and hook a leash on it. ;D

Lasso that sucker. ;D

 

[AeroFranz]

Retrieving inside the cargo hold would be attractive - much more room for storage than under wings, and sferrin mentioned you could reconfigure / rearm / refuel.
The real neat trick would be to do it WITHOUT depressurizing the C-130.

 

[Sferrin]

Retrieving inside the cargo hold would be attractive - much more room for storage than under wings, and sferrin mentioned you could reconfigure / rearm / refuel.
The real neat trick would be to do it WITHOUT depressurizing the C-130.

Short of an air-lock you couldn't. But why would that matter? They deploy/drop stuff out the back all the time. ???

 

[Orionblamblam]

Seems to me it *might* be easy to string a net across the back of the payload bay, and simply fly the drone into through the lowered cargo ramp. If the drone has a few hooks, it should simply catch on the net and hang there until someone comes and gets it.

A bunch of drones could be dealt with quickly. Have a series of nets that have vertical supports on the sides, like a volleyball net; the nets, say four or five of them, are folded flush with the deck. The net on top rotates 90 degrees, so it's vertical; a drone or three gets caught on it, the net rotates another 90 degrees till it's flush with the floor the other way. The next net rotates up, snags a few drones, folds flat. Repeat as needed.

 

[Sferrin]

Seems to me it *might* be easy to string a net across the back of the payload bay, and simply fly the drone into through the lowered cargo ramp. If the drone has a few hooks, it should simply catch on the net and hang there until someone comes and gets it.

A bunch of drones could be dealt with quickly. Have a series of nets that have vertical supports on the sides, like a volleyball net; the nets, say four or five of them, are folded flush with the deck. The net on top rotates 90 degrees, so it's vertical; a drone or three gets caught on it, the net rotates another 90 degrees till it's flush with the floor the other way. The next net rotates up, snags a few drones, folds flat. Repeat as needed.

Same way they did on the BBs back in the day (and probably a million other places). That's kind of along the lines what I was thinking. The interesting bit would be relative velocity, airspeed inside the aircraft, and accuracy of the incoming UAV.

 

[Mat Parry]

Purely based on the artists concepts that (in part) started this discussion, I'd imagine the rather beefy "pitot tube" might actually be used to lock into a drogue which would then haul the gremlin into the cargo bay.

An actively stabilized drouge might make the process a little less fraught
https://youtu.be/DbHDJtyXXe0

Nice (and somewhat relevant) picture can be seen here
http://www.secretprojects.co.uk/forum/index.php/topic,3536.msg250848.html#msg250848

 

[Void]

Goodbye, Gremlin.

 

[AeroFranz]

Retrieving inside the cargo hold would be attractive - much more room for storage than under wings, and sferrin mentioned you could reconfigure / rearm / refuel.
The real neat trick would be to do it WITHOUT depressurizing the C-130.

Short of an air-lock you couldn't. But why would that matter? They deploy/drop stuff out the back all the time. ???

Just circumstantial evidence.
C-130s have been used in the past to drop AGM-176 Griffins from a rack mounted on the cargo ramp. Simple, cheap. However they went to great lengths to subsequently design what's called a 'Derringer' door that replaces the starboard paratrooper door and mounts a couple of vertical tubes that can be loaded from the inside. Small openings/airlocks on the bottom of the door allow dropping the weapon without de-pressurizing the fuselage. Obviously this does not work if you want to recover a UAV.
Depending on the mission the UAV itself might prefer being launched or retrieved at high altitude in order to maximize time on station for the UAV, for the mothership, or some other metric - If you are recovering one UAV every hour for twelve hours, it might get annoying to fly up and down twelve times. You would probably just stay low and swallow the decrease in endurance. Or maybe you want to be above MANPADS range? Or maybe whatever UAV command and control crew is inside the fuselage working the consoles requires a pressurized fuselage at all times?
I agree it certainly makes things harder.


- Regarding flying the UAV inside the C-130 - the aero environment behind the C-130 is pretty bad. A small UAV would get tossed around quite a bit. Then there's the issue of getting a small/medium sized UAV to catch up to a C-130. I think KC-130s refuel at 105 kts or so, so that plus 10/20 knots more or less is how fast the UAV would have to go - no small feat for a small UAV that also needs to be optimized for endurance (= slow).
I think the air force would probably frown upon flying UAVs directly at a manned air vehicle.


- Regarding these things getting blown out of the sky - Fine. If they cost $200K (pulling number out of my a$$) and bad guys have to spend as much to shoot them down with MANPADS, that's still a win. There is renewed interest in the Air Force for attritable air vehicles, i think Gremlin is in that same vein.

 

[Sferrin]

Goodbye, Gremlin.

Fixed for accuracy.

 

[Orphic]

- Regarding flying the UAV inside the C-130 - the aero environment behind the C-130 is pretty bad. A small UAV would get tossed around quite a bit. Then there's the issue of getting a small/medium sized UAV to catch up to a C-130. I think KC-130s refuel at 105 kts or so, so that plus 10/20 knots more or less is how fast the UAV would have to go - no small feat for a small UAV that also needs to be optimized for endurance (= slow).
I think the air force would probably frown upon flying UAVs directly at a manned air vehicle.

These pics don't lie, perhaps its not as bad as you think?

Typhoon Strike Fighter Closing Dangerously On Hercules C130's Cargo Bay

Click to viewThis flight image sequence was allegedly taken during the filming of a Eurofighter Typhoon from a Hercules 130 cargo bay. Apparently, the director asked the pilot over the radio to come closer and closer and closer, until the pilot got fed up and whizzed right up to the entrance of...

gizmodo.com

 

[shedofdread]

Whilst the Typhoon is close to the C130, it isn't *that* close (the camera may not lie but it can mislead ) . Also, remember that the gust response of a heavily loaded thing like a Typhoon is VERY different to something like Reaper or even worse Scan Eagle.


The UAS will need to be grabbed at some distance from the 'mother ship' and brought aboard safely. How this is to be done; I'll leave others to speculate on.

 

[AeroFranz]

Precisely my thought ;D

 

[Void]

Fixed for accuracy.

And I suppose a missile which can intelligently classify any hostile jammer, distinguish true targets from decoys and automatically make the decision to attack it will still be cheaper than the command guided SAMs defending the jammer? Why, it only needs the same level of autonomy as the LRASM.

No, I lied actually. EW warfare systems which attack UAVs by sending them false commands are already on the market. So it not only needs to be able to deal with any conventional jammer, but even EW systems which have no characteristic jamming signal at all and will not be recognized by conventional ARM-type seekers.

 

[Sferrin]

Fixed for accuracy.

And I suppose a missile which can intelligently classify any hostile jammer, distinguish true targets from decoys and automatically make the decision to attack it will still be cheaper than the command guided SAMs defending the jammer? Why, it only needs the same level of autonomy as the LRASM.

No, I lied actually. EW warfare systems which attack UAVs by sending them false commands are already on the market. So it not only needs to be able to deal with any conventional jammer, but even EW systems which have no characteristic jamming signal at all and will not be recognized by conventional ARM-type seekers.

You're right, Putin's uber-badass toy will be unpossible to kill. It will bring entire air forces down all by itself and leap tall buildings in a single bound. : You guys are hilarious.

 

[AeroFranz]

Fixed for accuracy.

And I suppose a missile which can intelligently classify any hostile jammer, distinguish true targets from decoys and automatically make the decision to attack it will still be cheaper than the command guided SAMs defending the jammer? Why, it only needs the same level of autonomy as the LRASM.

Cheap multi-mode seekers have been demonstrated in the past. See LOCAAS, for example, which is pretty dated by now. Lockheed, Boeing, and Raytheon offered tri-mode seekers for JAGM, IIRC. Furthermore in a swarm you could use off-board sensing, so not all Gremlins need the EW gear. In fact distributed capabilities, which confer graceful degradation, are a tenet of swarming.
The thing about the cost battle (for lack of better term) that no one seems to mention is that in order for the US to win, you don't actually need parity. It would be nice, but as long as your expenditure is not, say, 100 times that of the enemy, if you have deep pockets and deep magazines, you can still afford it. Of course it becomes much better when that ratio comes down to something like 2 to 1 or better.
So even if a Gremlin costs twice as much as the SAM, it might still win the attrition race.

 

[Sferrin]

Fixed for accuracy.

And I suppose a missile which can intelligently classify any hostile jammer, distinguish true targets from decoys and automatically make the decision to attack it will still be cheaper than the command guided SAMs defending the jammer? Why, it only needs the same level of autonomy as the LRASM.

Cheap multi-mode seekers have been demonstrated in the past. See LOCAAS, for example, which is pretty dated by now. Lockheed, Boeing, and Raytheon offered tri-mode seekers for JAGM, IIRC. Furthermore in a swarm you could use off-board sensing, so not all Gremlins need the EW gear. In fact distributed capabilities, which confer graceful degradation, are a tenet of swarming.
The thing about the cost battle (for lack of better term) that no one seems to mention is that in order for the US to win, you don't actually need parity. It would be nice, but as long as your expenditure is not, say, 100 times that of the enemy, if you have deep pockets and deep magazines, you can still afford it. Of course it becomes much better when that ratio comes down to something like 2 to 1 or better.
So even if a Gremlin costs twice as much as the SAM, it might still win the attrition race.

IIRC they tried to make the missile in the TOR SAM system as dumb as possible, keeping the brains on the vehicle, so shooting down JDAMs wouldn't break the bank as fast.

 

[Orphic]

Whilst the Typhoon is close to the C130, it isn't *that* close (the camera may not lie but it can mislead ) . Also, remember that the gust response of a heavily loaded thing like a Typhoon is VERY different to something like Reaper or even worse Scan Eagle.

As the DARPA graphic shows something close in size to the LRASM or JAASSM, that would put it around 4.27m in length with a wingspan of 2.4m (wing extended), wing area is approx 1.326 m², weight would be approx 907 kg and you could fit 2 tiers of 6 (wings folded) in a C-130J without busting the normal payload of 15,422 kg. This would put the wing loading at approx. 684 kg/m² verses the Eurofighter which is 312 kg/m² at MTOW.

 

[AeroFranz]

I wouldn't put too much stock on a DARPA artist's impression. They're done for illustration purposes only.
Such wingloadings may be good for a cruise missile which spends its time at high speed at sea level, less so for a loitering platform at medium altitudes. I don't know the requirements yet, so it's hard to comment on optimum wingloading. Cruise missiles tend to be extreme cases, anything that flies higher or spends a good portion of its mission profile flying slower will by definition want a lower wingloading.

 

[shedofdread]

AeroFranz - You beat me to it


Further expanding on that point, considering the sort of platform one typically wants for persistence, bringing something with that sort of span aboard may present a challenge or two.... (V/G???)

 

[zaphd]

It's impossible to tell size based on just that artist's impression. The drones might as well be MALD-sized, which would put their wing loading significantly lower. Considering Raytheon has been studying launching swarms of MALDs off of C-130s, having them return would be the logical next step.

 

[jsport]

It's impossible to tell size based on just that artist's impression. The drones might as well be MALD-sized, which would put their wing loading significantly lower. Considering Raytheon has been studying launching swarms of MALDs off of C-130s, having them return would be the logical next step.

More 'tube and wing' drag queens. UAVs include the word Aerial. So how about flying a little bit.

 

[AeroFranz]

It's impossible to tell size based on just that artist's impression. The drones might as well be MALD-sized, which would put their wing loading significantly lower. Considering Raytheon has been studying launching swarms of MALDs off of C-130s, having them return would be the logical next step.

More 'tube and wing' drag queens. UAVs include the word Aerial. So how about flying a little bit.

It is a given this uav will have to fold its wings and empennage. That right there makes it harder to maintain a clean outer mold line. Not impossible, but it's certainly a more constrained design problem. Also, it has to be cheap.

 

[jsport]

It's impossible to tell size based on just that artist's impression. The drones might as well be MALD-sized, which would put their wing loading significantly lower. Considering Raytheon has been studying launching swarms of MALDs off of C-130s, having them return would be the logical next step.

More 'tube and wing' drag queens. UAVs include the word Aerial. So how about flying a little bit.

It is a given this uav will have to fold its wings and empennage. That right there makes it harder to maintain a clean outer mold line. Not impossible, but it's certainly a more constrained design problem. Also, it has to be cheap.

please explain why folding wings. It is expensive to have low endurance.
.. even more of a disaster to have wings folding and unfolding and cutting things and breaking during recovery etc.

 

[AeroFranz]

I kind of assumed they would. How are you going to fit a swarm of UAVs within the mothership unless they have folding wings? It's either that or they're really small, or they are recovered and stored externally, but that's awkward.
It would be convenient to drop these (but not recover them) from a number of platforms. As long as they're < 2,200 lbs, just about any tactical fighter will be able to carry at least a couple - but you still need to comply with ordnance carriage requirements, which kind of implies folding of surfaces.
It would also be nice to carry these internally in the CSRL of a B-1 and B-52, which also requires folding.


So you probably wouldn't want to fold the wings from an aero/structures/cost/reliability point of view if you didn't have to; it's just that there are a lot of other good reasons to do it.

 

[bobbymike]

Between MALD development and this is sure looks like future air warfare against an adversary with A2AD systems will be bombarded with hundreds maybe thousands of these UAVs in order for next generation sensor craft like the F-35, LRS-B and others to pinpoint and counter these systems.

 

[Matej]

Lockheed Minion(s) came to my mind. It's purpose was a bit different but the platform can be overall the same.

 

[jsport]

I kind of assumed they would. How are you going to fit a swarm of UAVs within the mothership unless they have folding wings? It's either that or they're really small, or they are recovered and stored externally, but that's awkward.
It would be convenient to drop these (but not recover them) from a number of platforms. As long as they're < 2,200 lbs, just about any tactical fighter will be able to carry at least a couple - but you still need to comply with ordnance carriage requirements, which kind of implies folding of surfaces.
It would also be nice to carry these internally in the CSRL of a B-1 and B-52, which also requires folding.


So you probably wouldn't want to fold the wings from an aero/structures/cost/reliability point of view if you didn't have to; it's just that there are a lot of other good reasons to do it.

High endurance & High dynamic maneuver, durability, reliability, simplicity, reuseability are only going to be afforded via flat stacking. Didn't see that these things would compete w/ MALD as external ordnance requiring carriage. Mald is still essentially an expendable munition.

 

[bobbymike]

I don't know about the 'recovery' part but a B-2 could probably drop a few dozen like the 500 lbs bombs in this video posted on the bomber thread a couple of days ago

https://www.youtube.com/watch?v=KdzJWciha4A

 

[AeroFranz]

High endurance & High dynamic maneuver, durability, reliability, simplicity, reuseability are only going to be afforded via flat stacking.

That's a bit...categorical ??? . I agree that folding wings carry aerodynamic, structural, complexity penalties, but i don't know better ways of reducing the dimensions envelope for storage. We must be envisioning vehicles of very different size and capability. For smaller UAVs, folding may not be necessary, but i don't see the place for small UAVs in the Asia-Pacific scenario, where range and speed dictate a minimum size.

 

[jsport]

High endurance & High dynamic maneuver, durability, reliability, simplicity, reuseability are only going to be afforded via flat stacking.

That's a bit...categorical ??? . I agree that folding wings carry aerodynamic, structural, complexity penalties, but i don't know better ways of reducing the dimensions envelope for storage. We must be envisioning vehicles of very different size and capability. For smaller UAVs, folding may not be necessary, but i don't see the place for small UAVs in the Asia-Pacific scenario, where range and speed dictate a minimum size.

span is endurance ..is everything and your launching and recovering from a prop when your supporting jet ranges. can't even agree w/ your projected scenario.. Flat stacks from ships make sense in Asia. I non folding wing tube & wing sounds dud lie ???

 

[AeroFranz]

Flat stacks from ships make sense in Asia. I non folding wing tube & wing sounds dud lie

Sorry - are you typing this on a cellphone? I am not sure i understand.


The DARPA Gremlin's proposers'day is later this month. The only things we know so far are from the RFI earlier this year. They're talking about payloads of "100 pounds or less", and that they want "air launch and recover with minimal modifications of existing large existing airplane types".


By the time you add 100 lbs of payload, speed in excess of 110 kts (to catch up to a C-130), a typical 30% structural weight fraction, and enough fuel to be a useful loitering platform at standoff ranges, you have a large UAV. By the way, it's a given that it will have a thirsty small turbojet because it has to burn heavy fuel and there are no real small turboprops available. Getting piston engines to reliably start at altitude after being stored for years is a PITA, while there is plenty of experience with engines like the F107.


We will know more in a few months when DARPA gets feedback from Industry, but i'd be surprised if the vehicle was small enough to permit storage with fixed wings within the cabin of a C-130.

 

[jsport]

Flat stacks from ships make sense in Asia. I non folding wing tube & wing sounds dud lie

Sorry - are you typing this on a cellphone? I am not sure i understand.


The DARPA Gremlin's proposers'day is later this month. The only things we know so far are from the RFI earlier this year. They're talking about payloads of "100 pounds or less", and that they want "air launch and recover with minimal modifications of existing large existing airplane types".


By the time you add 100 lbs of payload, speed in excess of 110 kts (to catch up to a C-130), a typical 30% structural weight fraction, and enough fuel to be a useful loitering platform at standoff ranges, you have a large UAV. By the way, it's a given that it will have a thirsty small turbojet because it has to burn heavy fuel and there are no real small turboprops available. Getting piston engines to reliably start at altitude after being stored for years is a PITA, while there is plenty of experience with engines like the F107.


We will know more in a few months when DARPA gets feedback from Industry, but i'd be surprised if the vehicle was small enough to permit storage with fixed wings within the cabin of a C-130.

see nothing dispelled. Just would add that 100lb payload craft w/ any endurance would be indeed impractical to launch internally from cargo planes rendering the whole program a pipe dream. Likewise the necessary, cost effective, fuel efficient small propulsors need development first even if some sort of conformal launch (a different program) evolves and don't see it.