November 02, 2010
Contributor: Defence IQ
Colonel Jeffrey Turcotte, USAF, Chief Air and Weapons Division, Air Force Research Laboratory, speaks at the UAV Summit, hosted by IQPC. This transcript of his presentation addresses the implications of US Department of Defense budget cuts and where UAV research falls within the spectrum of scope and funding. He also discusses the growth of military interest in unmanned capabilites, the prospect of 'one pilot - multiple UAVs' employment, and the ways in which the Laboratory focuses spending on multi role and multi service outcomes.
Chair: Jeffrey S Turcotte is the Chief of the Air Weapons Sector, Plans and Programmes Director, Air Force Research Lab at Wright-Patterson Air Force Base. He’s responsible for managing AFRL relationships with stakeholders in the Air and Weapons domains and directs a staff of 17 government employees. He also manages planning and programming for technology programmes in these domains. So the full bio-zero in the agenda so without further ado, Colonel Turcotte.
Turcotte: So, yes, the last time I agreed to speak at this kind of a summit, I came mostly by myself and I didn’t have, you know, folks with me that were the experts. And so, when it came to asking questions and asking for details, I was like, ah, ah, ah. So I thought, well, this time what I'm going to do is, yes, I'll agree to come, with support.
And so what I plan to do is just give a very brief overview and outline a little bit what you’re going to hear from the experts, and then I will also cover a few areas for which we didn’t bring experts but just to let you know, kind of, the breadth of the research that we’re doing.
That’ll just be a few, you know, spot eagle items.
The ball joint analogy
Okay, so. And then I don’t know how many of you have seen my ball joint repair slide. Bob has. Oh, I'm glad most of you haven’t. So the ball joint is this unit right here, okay. And I like to use this as an example of how we need to work together on things. In order to get the ball joint out of there, obviously you take the nut off, okay, and you start beating on this knuckle, right here, to break it loose because it’s got a tapered fitting there.
But even after you break it loose usually it doesn’t fall down, because if this bolt isn’t holding it up the rest of the control arm does hold it up and so, you need to put a big pry bar in here. And one guy stands back there and pulls the pry bar down and the other guy reaches in there and tries to get the ball joint out, right.
So, you can imagine what happens if these two guys don’t work together, right, the guy’s got his hand in between there and then oops, the pry bar slips out. So the reason I like this example, a ball joint replacement requires leverage and co-operation. That’s your lesson. If you have any questions about ball joint replacement, talk to me later. If you have any questions about UAVs or RPAs, talk to these other guys.
Okay, so again, a brief introduction of AFRL a little bit on … Well, I’ll tell you what briefings you’re going to get of course you could just read that in the agenda but… And then I’ll cover a few other programmes and let the show get on the road.
So the mission, Air Force mission, fly and fight and win in these three domains. The SNT vision for achieving that mission is based on the kill chain and all aspects of the kill chain are supported by RPAs. We pretty much do all those things with RPAs. Not saying we’re going to get rid of pilots, but we can do all those things with RPAs.
Okay and this vision is in concert with the war fighter and hopefully OSD because we don’t want to get any more of our secretaries or chiefs fired. So, in the laboratory itself our previous Commander, as this flight plan rolled out and we had discussions with him, we were able to secure a place on his short list of what is the laboratory going to focus on and here we are, UAVs is a high interest item and so it has enjoyed – it hasn’t been totally immune from cuts, but it has enjoyed support in the budget.
We also have a team, not just these four folks that join me today, but a large team across the laboratory that we use to keep the collaboration going, transfer information back and forth and so on. We do participate in the OSD, IPTs, okay, and the Air Force level IPTs.
This is one way of breaking down the work that we do, there’s lots of ways of doing it and basically this adds up to somewhere in the neighbourhood of about £100 million dollars or more per year that’s unique. Okay, there’s lots of things we do that can be used for RPAs but also for, you know, manned aircraft like sensors and things like that, but about 100 million goes into stuff that’s strictly for UAVs each year.
Addressing autonomy and reliability
Okay, so these are the four briefings, there’s a break in between and I think there’s another briefing by another person in between here somewhere but… So these are, kind of, the areas that you’re going to get a detailed briefing on and then some other areas that I just want to touch on just so that you can see that we’re not ignoring these other things.
Okay and you’ve seen this chart, I just use this to introduce the idea that we understand, we get it, okay, that we need autonomy and more autonomy but it’s a fairly slow process as you’ve seen, it’s not a simple matter to convince... I mean, it’s not necessarily technology limited, right, it’s there’s a cultural limitation here and the policy is slow to change and that’s just a fact of life.
So we have to work hard to make sure the reliability is going to be there, to make sure that we don’t have accidents along the way here, along the road, okay. If we rush this and start having accidents, it’s going to end up slowing us down instead, so one of the big pushes is Mac up, right, that’s what Scam used to say, Mac up. Everybody knows Scam, right? Does anyone know what he’s going to do now that he’s retired? Somebody does, you’ll see him pop up somewhere, like a whack a mole; not that I want to whack him on the head or anything.
Okay, so, obviously the idea here is one pilot, one control station controls more than one UAV, okay, as long as you’re not trying to run four joy sticks at the same time, we can do that, right? You’re just entering way points or somebody mentioned camera guided or you know, there’s lots of different ways to do that. But basically you’ve got to take some decisions out of the hand of the controller and turn them over to automation, that’s not hard, right? It’s more of a policy issue again.
Optimising the interface and interoperability
Beyond that, okay, you don’t only want to control say several MQ1s, you may want to control a couple of MQ9s, maybe some other assets, maybe a Shadow or who knows what could happen in the future if you build your architecture, as was mentioned this morning, to control many different platforms.
And of course you’ve got to have the inner operability built in because, you know, the airplane is the airplane, and unless you’re going to modify the airplane to accept a different wave form you’re going to have to produce the wave form that it understands, and get it up to that platform to control it and to control all the sensors or whatever payload happens to be on board.
So Heart is the DARPA programme we manage for them and it’s pretty well concluding now. I think they’re saying IOC2010 and I don’t know if that’s still accurate but it shouldn’t be too far off. You can see they talk about things like decouple, you know, the sensor operator and the exploiter and the Intel team and so forth from the controlling of it, and that’s not hard either, right, just use a different wave form for the sensor and take the button off of the cockpit, right, or make two separate cockpits so…
But we are though starting to get into some complexities now that we haven’t dealt with before and also the proprietary issues of having different wave forms, different protocols, different, you know, file formats and so on for the data flows that we’ve got to tackle that, okay?
And it was mentioned this morning, the interoperability, that’s something that’s worried me for quite a while now and we got to the point where we – I don’t feel that it’s the LAV’s mission to set standards but we’re willing to do it if somebody comes up with the money. First I wasn’t even willing to do it but then somebody volunteered, I said, okay, you volunteer, okay, I will go look for the money
I looked for the money and I couldn’t find it, so this is a problem now because, you know, everybody wants to be in our operability, everybody wants to be modular, everybody wants to have, you know, a standard interface here. Whatever the sensor is, you meet my standards, it’ll plug right in.
Okay, just like your phone jack plugs in to the wall, right? It doesn’t matter what phone you buy, they all plug into the same jack. Somebody’s got to pay the bill to do that, I think it should be OSD but… Do we have any OSD folks in here? See that, they don’t show up, they know you’re going to ask for money, they don’t show up. Okay, so that’s where we are with that.
Okay, here’s where we get real complicated now. Like was mentioned this morning, you know, it’s bad enough if you’re trying to control a swarm of UAVs, right, and you want them to act like the Borg, like everyone knows what everyone else is doing and they’re all under one big control, right? Except, we don’t know how to do the Borg, right?
That’s science fiction. I’d just like teleportation, right? If you can do the transporter, you can do the Borg. I think the Borg is the hardest thing we’re up against here. I mean, it’s bad enough to try to do it without any contested air space, but now you have people come in shooting your guys down, flying between your guys. And I was talking to some gentlemen this morning about software that could help us with understanding systems of systems. This, I think, is really the most difficult technical challenge we have, is as the systems get more complex, the reliability decreases unless you do something in terms of redundancy.
Just... You know, in order to put systems together and evaluate them you have to model each sub-system, right, and then somehow you have to put those models together and then excite them to see how they behave, right? You put inputs and you’re trying to predict the outputs; what are the outputs going to be, are they predictable and are they reliably predictable? Are they repeatable and, you know, how many sneak circuits are you going to have? How many little bits of code are going to cause this thing to go haywire? Has anybody seen The Terminator movie? This is the one the FAA says, if you don’t destroy all copies of The Terminator movie, we will never get, you know, Sense and Avoid through the FAA.
Okay, I don’t believe Skynet became self aware. I don’t worry about that, I don’t worry that the computer’s going to get smart and try to kill the humans, but I worry that we’re not smart enough to design the computer that anticipates everything that could go wrong. And again, you know, it’s a systems of systems thing that we need to somehow be able to interrogate the final design, to know that it’s reliable and that’s a problem I’m stuck on. So think about that one. If you have an answer, give me your business card later.
Communications and sensor challenges
Okay, layered sensing. Sensing’s very important, the only thing I’d like to say about this is these guys, what they really want to do is the one guy, the one commander says, I want — he’s looking at the satellite output and he says, I want to get a closer look at that. Well, sir, it’s zoomed in all the way.
Okay, cue that sensor down there, right, and then that sensor looks around and I want to get a closer look over there. Sir, it’s zoomed in all the way. Well, cue that sensor down there. That’s, kind of, what we’re talking about here is going down and back up again, being able to control all these different sensors somehow through the gig, right, and then all this data comes in and then you’ve got this PED issue, right. You’ve got this PED issue, I want to talk a minute about that but let me go through the rest of these slides.
As far as data links are concerned, this is just an example of making them smaller, right. Okay, we got Sense And Avoid, we have one that fits on the RQ fleet, we don’t have it small enough yet for the MQ fleet, so we had to make everything smaller, okay, and smaller then again and then when Doctor Perkins gets up and talks about the little fly-sized things, imagine how small your data link has to be to fit on that fly.
And interoperability and connecting the boots on the ground to all the assets that are involved in the fight, okay, and having them all speak the same language and communicate with each other, that’s obviously prime on our list of things to do.
And just a little bit about — you guys listened to the pilots this morning, this is a new career field for the Air Force, right, okay, and the requirements for this work force is a little bit different than for pilots that actually fly on the jets, right, and so we have to think about how we’re going to do that and we have to do research into, okay, what are the requirements? What are the limitations? There are some, right, you know, the guy that — what the guys say at Mississippi State, Bobby said, honey, I crashed my plane, I’ll be home at four, right, because he wasn’t in the plane so he had to come home early that day. Anyway, so obviously we’ve got to think about that and I know there’s some people here that are working in this area and we appreciate that, we need that kind of work done.
The final word
Okay, and before the conclusion, let me just say one thing about PED. So we have a lot of work going on in PED and automating PED, I have some slides on that that are not releasable at the public level. If you’re interested in that and folks at the flight plan do have those slides, if you’re interested in that and you think you’re eligible to receive…
These are unclassified but they’re, like, distribution C, so you’d have to be a DOD contractor or employee with a need to know. And I assume if you’re in this room you have a need to know. So if you’re interested in that…
The second plan is, as part of the... You know, there’s a UAS flight plan, there’s also an ISR flight plan. And within that flight plan really is where the requirements for automating PED are embedded, and we are answering that flight plan probably sometime in the summer; the implementation plan for that will come out or at least be in draft form and then you can see, within that, what AFRL is doing to answer the requirement for automating PED.
Okay, and finally just conclusions. Basically, you know, we are working with Air Force A2 to answer the flight plan. Most of the things in there actually are already within the scope at or above TRL6, but there are some things that I’ve shown you here today and that you’re also going to hear that we need to continue to work on to, you know, until we’re at 2047 if we’ve met all those requirements. But I appreciate your time today, I appreciate everyone coming. Thank you.
Posted by Naxal Watch at 7:19 AM