As defence outlays start to shrink after a decade of growth, the US Department of Defense is exploring the increased use of flight-simulation technology to cut back on fuel costs and reduce the wear and tear on its aircraft fleet.
But to what extent can simulations really replace live flying? And would that really save money?
"There is no data to accurately quantify to what extent simulators can replace live flying for Air Force fighter/bomber training," says the US Air Force's Air Combat Command (ACC), responsible for organising, training, equipping and maintaining the combat air force (CAF).
"ACC has done several studies and evaluations of live and simulator training to determine the right mix of live and virtual training that maximises combat readiness while reducing training costs in today's fiscally constrained environment," says USAF major command. While the technology has great strengths in terms of procedural training and large-force employment exercises, it also has some serious accuracy limitations. "ACC also views most simulator training as a complement versus a replacement of live fly training due to the strengths of each training medium," says USAF.
Nonetheless, simulation technology has come a long way since the start of the information age. US Marine Corps Col Arthur Tomassetti, vice-commander of the USAF's 33rd Fighter Wing, currently flies the Lockheed Martin F-35 Joint Strike Fighter. But over his long career he has flown more than 35 different aircraft types and has served as a test pilot for both the current F-35 and its X-35 concept demonstrator.
Tomassetti started initial flight training at NAS Pensacola, Florida, in 1987, in the Beechcraft T-34C Turbo Mentor basic trainer. Although there was a simulator to help student aviators practise, the device was a rudimentary procedural trainer. "While you could fly with the instrument suite to give you the impression you were flying, you didn't have any outside-the-window view," he says.
Simulators are well-suited to tactical training
Later, flying the Douglas A-4 Skyhawk
, Tomassetti experienced training on a somewhat more advanced simulator. The Skyhawk simulator offered some basic visuals and some weapons capability, as well as some level of motion capability. But, Tomassetti says, it was still limited, and used mostly to drill procedures.
It was not until Tomassetti got to the Boeing AV-8B Harrier II jump jet in 1988 that he says he experienced a simulator with decent visuals. "That was the first time you started to see the screens trying to represent the outside world with early animated graphics," he says.
But while the graphics had improved, and one could learn from the machine, the 1980s-vintage simulator did not truly present an accurate depiction of the Harrier.
"You could see some things, you could practise your procedures in there, but it wasn't the real airplane - it emulated the real airplane," Tomassetti says.
Later simulators are much better, and provide some excellent capabilities not just for procedural training but also for weapons employment. The latest US military simulators for the F-35 use the same software as the actual jet, Tomassetti says. That means that, theoretically, everything in the simulator should match what the real aircraft does.
Discussing the graphics offered by the F-35 simulator, Tomassetti notes that a student will be able to pick out the difference between different classes of armoured vehicles visually in the simulator. Visual representations have traditionally been a major limitation for flight simulators, particularly at close ranges and for terrain.
But even the state-of-the art F-35 simulator does not move, and certain factors such as flight-control response times cannot necessarily be modelled with 100% fidelity. "How the airplane flies in the sim is the best we can do without having a real airplane attached," Tomassetti says. "But even that level of effort that went into making that level of fidelity, is much more than we've had in other airplane programmes."
Even with some level of motion, one major limitation of simulators is that they cannot replicate the physiological effects on the human body. Simulators cannot train student pilots on aspects of flight such as experiencing g-loading and the associated vestibular effects; nor can they replicate physiological exhaustion. A simulator is also unable to replicate the "feel" of an aircraft, such as airframe buffet cues.
"All of this is important in dogfighting - the feel of energy bleed and increasing buffet as you trade [airspeed] for nose position," says one senior USAF officer.
A senior USAF instructor pilot says he has never flown any simulator that has been able to model the flight dynamics of a fighter accurately - despite some of the latest simulators incorporating inflating g-suits, speakers for vibration and even moving plates in the seat.
For the US Navy, the problem is even more acute. There is no substitute for some of the specialised tasks naval aviators are required to carry out - such as carrier landings - which must be performed in a real aircraft.
As Lt Cdr Ben Charles - simulation fleet project team chairman for the USN's Strike Fighter Wing Pacific and instructor pilot for the Boeing F/A-18E/F Super Hornet - notes, "there is no substitute for actual flight hours: gs, environment, flying form, taking off from the boat, landing on the boat - night and day. There is no equivalent for that, but when [student pilots] do get in a jet, they're more ready".
Carrier qualifications are an especially difficult skill to master and represent one area where live flying is essential.
There are, however, those who advocate increasing carrier qualification training in simulators to reduce naval aviators' required flight hours. "That makes us as flightcrew very nervous," Charles says.
Perhaps where the simulator is most lacking is that students are always consciously aware that there is no real danger. Another senior USAF pilot says: "You simply cannot replace the sight, sound, smell, g-forces and fear you experience when actually flying."
Charles concurs. "You can't put the fear of dying in the sim," he says. "You can't ever replicate the fact that if you get low at the boat that you will actually hit the boat."
For many younger pilots, "getting in the simulator is playing a very expensive video game", Charles says. "Now, there are folks who are very good in the sim, but do very poorly in the jet."
Charles says many young pilots are simply overwhelmed by having to operate the aircraft and its systems while simultaneously experiencing g-forces and other airborne environmental factors. "But on the other hand, there are some folks who do mediocre in the sim and get in the jet and they perform very well," he says.
While simulators cannot replicate the physiological and psychological aspects of air combat training, they are suitable for tactical training. At the unit level, pilots can rehearse tactical scenarios in simulators before going up in the air to fly the same scenario. For example, one senior USAF officer describes rapidly running multiple simulator scenarios with a flight of four F-15C Eagles using night-vision goggles and Joint Helmet Mounted Cueing Systems (JHMCSs), and subsequently digitally debriefing those same scenarios.
The digital debriefing system offers a God's eye, three-dimensional display as well as reproductions of all four cockpits and their displays. "This is outstanding for pre-flight upgrades, warming up, knocking the rust off switchology, practising 3-1 comms and the like," the senior USAF officer says. "Nowadays, you don't go fly a four-versus-four upgrade ride without briefing, e-flying, debriefing and passing the same scenario in the simulator-student upgrades and IP [instructor pilot] upgrades."
Charles says that the USN uses a similar system for its unit-level simulators for the Super Hornet. But the USN wants eventually to link disparate units from across the globe virtually so they can train together in a scheme called fleet synthetic training.
However, the USN system will not just be limited to aircraft, Charles says. Aegis air and missile defence warships such as the Ticonderoga-class cruisers and Arleigh Burke-class destroyers will eventually be linked. While initially the system will primarily be shore-based, it may eventually be deployed on board carriers. When the system becomes operational, the USN will be able to run large-force exercises with forces distributed around the globe completely virtually.
Simulator graphics have come a long way
"That's one of the things we're leaning forward on," Charles says. "Obviously, we don't have that full capability yet."
The USAF is already implementing such a system under its distributed mission operations (DMO) construct. Already a number of bases are fully operational with the system in place.
Col Mustafa Koprucu is commander of the USAF's 505th Command and Control Wing at Hurlburt Field, Florida, and responsible for leading the air force's DMO and live, virtual, and constructive (LVC) capabilities.
The nexus of the DMO network is the Distributed Mission Operations Center (DMOC) at Kirtland AFB, New Mexico. From there, the various units scattered around the globe can be linked together for exercises. The system can link together disparate units for day-to-day training, but the real value comes from assembling virtual air campaigns involving hundreds of aircraft, US army ground assets and even USN warships.
"What I can do out there at the DMOC is to put together a synthetic battlespace, where I can tie all those simulators together, where they can see a common enemy, a common air picture," Koprucu says. "We can tie them in and do basically what is like a Red Flag."
The simulated exercise is called Virtual Flag, with the "battlespace" incorporating "Red Air" enemy fighters - sometimes controlled by human operators - and advanced surface-to-air threats. Unlike in real-world exercises, there are no unrealistic limits on missile fly-out distances. There are realistic probabilities of kill, and if one is engaged by a hostile surface-to-air missile system, there is a realistic kinematic and visual representation of that event. Additionally, the full range of electronic warfare and intelligence-gathering capabilities can be simulated.
But an exercise like Virtual Flag goes further than that. It can afford the joint forces air component commander (JFACC) the opportunity to control large numbers of aircraft at the same time, Koprucu says. The goal, Koprucu says, is one day to be able to rehearse an entire air war in the virtual space.
As the DMOC grows in capability, US air forces - including not only the USAF, but also the USN, USMC and US Army - should in theory be able to rehearse an entire campaign in simulators. "Really, our objective is to give that JFACC confidence that the plan is executable," Koprucu says.
The USAF and USN are using the virtual battlespace to refine the Pentagon's AirSea battle construct, which will eventually be used to gain entry into anti-access/area-denial environments.
The USAF is continuing to expand its incorporation of live, virtual and constructive capabilities, blending live aircraft flying training missions with pilots flying simulators in the virtual world, with constructive, computer-generated aircraft and weapons systems, where everyone sees each other in a single integrated battlespace.
Additionally, Koprucu says that the USAF must seamlessly integrate its fifth-generation Lockheed Martin F-22 Raptors and F-35s into the DMO network. The major hurdle in the case of the Raptor and F-35 is that those aircraft require multilevel security built into the network to prevent unauthorised eyes from viewing their full capabilities, he says.
"I want to bring in, for example, our F-22s and other assets now, so I'm dependent on finding the technical solutions to do that," Koprucu says. "It's going to take a while to get there."
Although the simulation capabilities of the DMO offer outstanding tactical training, they do not address the mundane tasks of building airmanship skills, one senior USAF instructor pilot says. Those lessons can only be learned through actual flying in a real aircraft, he adds.
During a real Red Flag exercise, for example, a flight lead for a four-ship of fighters might have to fly though bad weather and navigate to where he needs to go in suboptimal conditions. The pilot may also have to deal with snags with air traffic control and or administrative tasks, which happen as a matter of course during a regular flying day.
It may be necessary to deconflict from other airborne traffic or manage safety rules, or deal with issues such as an aircraft malfunction or bad radios. Charles cites situations where a USN flight lead might have to substitute a different pilot or aircraft for a certain task mid-flight as a result of unforeseen problems.
"There is friction in just getting to the fight that is not represented in the sim," the senior USAF pilot says. "There are pluses to the sim, but I still think that you need to have the majority of your training in the air."
Ultimately, what may happen is that basic aircraft training is mostly flown live while high-end large-force exercises are mostly flown virtually.
"I wouldn't be surprised if in the future, we end up doing our high-end training in the DMO and we do our basic flying manoeuvre training live," Koprucu says. "The capabilities we have in the DMO world are just increasing, and that might be a natural transition."