Benning Teams with University to Study Free-Flying UAS

Benning Teams with University to Study Free-Flying UAS

Infantry officials at Fort Benning, Ga., want squads and platoons to one day get their battlefield intelligence from formations of unmanned aerial systems that fly with complete autonomy.

The Maneuver Center of Excellence at Benning and the Georgia Institute of Technology recently conducted an experiment aimed at testing hardware and algorithms designed to help multiple UAS fly without human involvement.

Georgia Tech scientists have been studying “how these vehicles can autonomously cooperate with one another in multiple groups,” said Charles Pippin, senior research scientist at Georgia Tech.


Benning officials are interested in the research to help shape the procedures guiding how infantry units will operate with UAS in the future.

“We have to pull soldiers from the squad and deplete our fighting force to control these systems,” said Harry Lubin, chief of the experimentation branch at Benning’s Maneuver Battle Lab.

“If we can get to a point where we can basically upload mission sets into unmanned aircraft and have them perform these reconnaissance missions autonomously through cooperating with each other, what we are then doing is taking all the soldiers you might have needed to control X number of UAS and putting them back into the squad.”

The effort is still in the early research phase, but Georgia Tech officials spent the first week of June testing the concept using quarter-scale Piper Cubs equipped a special payload that transmits algorithms to an off-the-shelf auto-pilot devices.

During the experiment, officials flew UASs in leader and follower roles, using the university’s four UAS. Georgia Tech hopes to use the findings to determine what types of UASs are suited for specific tasks, Pippin said.

Georgia Tech’s research, which is internally funded,  is designed to ensure that the algorithms and hardware can be loaded onto other UAS platforms, Pippin said.

One of the program’s future goals is to make the technology more robust for a combat environment, Pippin said.

Join the Conversation

Great concept. Worked it several years ago for naval assured access systems. Logistics is certainly an issue with air autonomous vehicles. Despite all of the wonderful DoD work on alternate fuels, energy density remains the principal limiting factor for unmanned vehicles. Propulsion and data processing are principal energy draws. Small air vehicles such as these apparently are will moving processing offboard and requires sufficient bandwidth, which in a difficult EW environment is very precious. Onboard processing is key to true autonomy and limiting bandwitdh for video which may be principal need for infantry in the field (I do not know). Individual vehicle autonomy is realtively easy, multiple (pack, fur ball) vehicle autonomy is more of a challenge, but can permit dissimilar vehicles (and/or dissimilar hosted sensors). in the end manpower to manage vehicle logistics cannot be glossed over as easily as the article indicates.

Really — you want to have no mission control in flight ? For a small unit level UAV ? How Fail is that ?

VeP: control is an illusion. object is to have sufficient air vehicles to cover an area (with revisit if sufficient sensors are not available to maintain coverage give a particular target speed vector). end result is not controlling a specific vehicle, but defining the area the vehicles are in and the spacing of vehicles. think in terms of a school of fish, only less dense. Need for clear area for manned aircraft or artillery needs to be part of the area definition. if you have to control aircraft individually, the concept fails due to lack of manpwer. Object is to provide additional awareness, not add burden to the people you have.

On the other hand, lots of vehicles means lots of data means lots of people to decide the significance of that data, and lots of bandwidth. I’m not against UAV’s but they are clearly being oversold right now as the “do everything” platform, and costs be dammed because everyone knows UAVs do everything better and cheaper, no matter what data there might be to the contrary.

This is just counterlogical. You are (one presumes) up against a thinking enemy, and that enemy will learn the flight patterns and defeat your reconnaissance assets easily by taking simple passive countermeasures. It takes time and tactics to do good tactical reconnaissance. This kitchen sink approach is nonsensical hogwash. What is the use of collecting data with sensors if you have no one on the receiving end paying any attention to it. What really frosts me about this is how penny-wise and pound foolish the Fort Benning argument really is — we need men in the trenches, can’t spare even one to man the scout UAV. Really.

Oh and I should point out the span of control principles would enable a single person to control 3–5 UAVs, given appropriate tools. Not teleoperation per se, but that is certain a possibility in a tight spot. But you couldn’t control the whole bunch that way. There has got to be a happy medium between a Robotics NCO in every platoon and what these people are talking about. Wake me up when the Maneuver “centers of excellence” start doing good combat development work again.

CapeE: As a former UAV’er and force developer w/ the US Army Intelligence Center and School, I can only say Bravo. It’s about time someone identify that there are/will be unforeseen manpower challenges caused by broadening a concept to include a wider user base. Identifying the “bill payers” (in terms of MOS mixes, pay grades, etc.) ALWAYS seems to be left in the ashes and ALWAYS comes back to bite us.

Sorry — this is the price of transformation. What were once specialized capabilities that required specialized training spread out through the force as a system matures. Dumbing down the problem (and keeping the force dumbed down) is not an option. We have to master the systems we build and deploy. Are there limits ? Sure. I’m not sure that the organic Class I UAV held at platoon level was all that hot an idea — in some ways it could be as much an encumbrance as a capability. But then, FCS threw out the Class II and III (company and battalion level) UAVs, keeping the brigade and platoon level platforms — before killing them all. And yet — when driving down the road and doing movement to contacts up in my head — I can see how one would use these things — not just in the advance guard, but to the flanks and rear. Screw the special interests and the branch loyalties. We need to get on with advancing the art and science of war, and perish all who get in the way. Speed. And Power.

The challenge in my mind is defining limited objectives for the UAV “swarm.” What is it we must know, would like to know, and are uninterested in. I am not a ground pounder and am not an expert. If it is done right, autonomous behaviors will have the vehicles managing their flight profiles (individually random) within group area of interest. Sufficient vehicles to assure randomness with programming to ensure it, is important. Have the vehicles in air with the capability to detect/monitor what you are interested in within an area of interest (you can bet the bean counters will not provide sufficient air assets) then the commander should have the information to make a difference. Lots of unknowns, but knowledge (not data) is only power when associated with the means to achieve objectives.

The basic flawed assumption behind this research is EVERY mission goes as planned so no field personnel needs to be diverted to manage battlefield intel. In the real world, we only need to ask ourselves how often does any mission goes exactly as planned? To be able to maintain control of your real-time area of operations intelligence should be a primary importance and should negate the flawed assumptions behind this obvious ploy for faux research funding and continued job security.

As to the issue of field personnel being diverted for intelligence management and thus leaving insufficient numbers to carry out the mission , the obvious answer is to provide the appropriate complement of personnel to handle the mission requirements AND the intelligence management aspect in the first place.

Other than PC and armchair mission commanders who have never set a foot outside of the wire, who else would be so foolhardy to fail to recognize that missions rarely go exactly as planned?

Boeing did the same thing in 2011 http://​www​.gizmag​.com/​u​a​v​-​s​w​a​r​m​-​t​e​c​h​n​o​l​o​g​y​/​1​9​5​81/

*required

NOTE: Comments are limited to 2500 characters and spaces.

By commenting on this topic you agree to the terms and conditions of our User Agreement

AdChoices | Like us on , follow us on and join us on Google+
© 2014 Military Advantage
A Monster Company.