Royal Canadian Air Force Sees More Sims In the Future of Fighter Pilot Training

dakohli writes "Currently, Canadian Fighter Pilots spend about 20% of their 'stick' time in Simulators. RCAF General Blondin states that this will rise to 50/50 in the future. The article goes on to state that the U.S. Army is moving in this direction, although the U.S. Air Force is a little more skeptical. Aircraft are expensive to fly, and if the fidelity of a simulator is good enough then perhaps real pilots will spend even less time actually in the air. Slashdotters, do you think that this will actually make recruiting pilots more difficult, or is it a sign of the things to come beyond Military Aviation?"

Comments

Flight Sim Tech Here

[Pikoro]

The fidelity is already there. Flight time in the sim is nearly as good as the real thing, especially considering when you are up on a motion platform.

The sims are great for procedure training since you can simulate failures which would be expensive or impossible to simulate in a real aircraft. More sim time = less cash spent on keeping the real aircraft in the air but with the same amount (or more) experience for the pilot being retained.

Re:Flight Sim Tech Here

[Anonymous Coward]

One of the interesting things that I never thought about until I experienced it, is that because flight models are typically generated based on data from prototypes during late stage development, simulated aircraft generally fly like they just rolled out of the factory. The aircraft that most pilots fly are often closer to the end of their serviceable life than the beginning. (The oldest of the tails that I currently fly has exceeded its planned service life by a factor of 3) This does make a big difference. Engines are not quite as responsive. Controls don't feel quite the same, and electronics start to do unpredictable things.

In the end though, while interesting, this is not that big of a factor. The significant limitations to simulator training are more human.

In the sim, every time something fails, it looks the same as it did last time. In the sim you never loose your weather radar halfway through penetrating a line of embedded thunderstorms. In the sim, you are never scared, the comms are always crystal clear, ATC never spontaneously forgets how to speak english, the tanker never descends to the wrong altitude and civilian traffic never busts your airspace. Chinese fighters never disguise themselves as Singaporean airliners, and fishing boats never try to blind you with lasers.

Even if we were able to integrate each of those things into the curriculum, it would not make much difference. Different weird things happen to every pilot. Almost anyone can learn to fly a plane, but gracefully and safely dealing with stuff that no one could ever anticipate is what makes someone a pilot. Real life is always more strange than anything a curriculum development committee can ever come up with, and real-world flying is currently the only way to teach pilots how to think like pilots and not just technicians.

Re:Flight Sim Tech Here

[Joce640k]

Given the number of hours things like the F22 have managed to stay airworthy I'd say simulators were the future, yes.

Why not?

[pushing-robot]

In a decade or two, most of them will be flying drones anyway.

Re:Why not?

[Hentes]

Just because a drone is controlled remotely with an interface similar to a simulator doesn't mean it behaves like a simulated aircraft. It's still flying through real-life air.

It depends ...

[perpenso]

Simulators can be very useful for pilot training. However their training value varies greatly depending on the task to be performed. Things relating to standard procedures and corrective actions for unforeseen events may be more useful, things related to air combat maneuvering (ACM) less so. Certainly ACM can be taught at an academic level in a simulator, learning the mechanics of a particular maneuver, being able to replay things from different vantage points, including your opponents. However the experience of actually feeling the g-forces during ACM is very important. Learning/practicing proper technique for maintaining consciousness, learning your personal limits, etc need actual flight time and the skills developed during this flight time are perishable. G-forces are also another input your brain learns to use. With experience a pilot can estimate how many degrees they have turned based on g-force and time, "that feels like 90 degrees", its just another thing that contributes to situational awareness and may negate the need to check a compass or external reference point. Handy if you have a more pressing thing to do.

No, it's really not.

[Anonymous Coward]

For the record, I'm a military aviator, and I've got plenty of experience in both sims and the actual aircraft.

For some platforms, yes, the sims are just fine. Less dynamic platforms (i.e. helicopters, big wing) work just fine with full motion platforms. It will never be "perfect." Many of the imperfections manifest in ways that are inherent in simplified programming, i.e. actually modeling fluid dynamics for how the jet handles with failed systems vs. just hard coding that things "will" or "wont" work at certain airspeeds.

For tactical aircraft, however, there is absolutely no comparison. Yes, basic flight operations (taking off, landing, navigating) can be done relatively decently, but tactical flying (g-force, sun blind spots, etc) cannot be replicated in anything remotely resembling our current simulators.

Not to mention that most tactical simulators dont include motion. A "full motion" sim can't replicate more than 1.0 G in any given direction, much less a sustained 5g pull. The technology simply doesn't exist.

So do simulators have their uses? Absolutely. But there is no substitute for real flight time, and until we get some Star Trek -esque technology at our disposal, there won't be.

Re:No, it's really not.

[CRC'99]

For the record, I'm a military aviator, and I've got plenty of experience in both sims and the actual aircraft.

For the record, I'm a commercial pilot.

Simulators have their place - but it is certainly nowhere near the experience as a real aircraft. Speaking from a commercial background, simulators are great at two things:
1) Procedure
2) Techniques

Simulators are great in showing pilots how things work. Want to know what to expect in a fogged in approach to an airport and are learning how to use the ILS etc? A simulator is *great* in this role. You can do things in this combo that are GREAT for education. Does it come anywhere close to the real thing? Hell no.

The other thing that simulators excel at is teaching things such as instrument scans - basically train you to keep an eye on all your instruments at the same time by developing an effective scan of them. No pilot flying on instruments will use a single instrument - flying is very complex and cannot be done like this. An effective instrument scan (A/H -> Airspeed -> A/H -> Altitude -> A/H -> VSI -> A/H -> DG etc) is very hard to grasp when first starting - and it is the bread and butter that keeps pilots alive when the weather is starting to deteriorate or you start to fly faster and bigger aircraft.

Your standard 737 pilot will probably spend about 15 minutes out of every flight looking out the windows. The rest is monitoring instrumentation. I cannot understate how important this skill is - and simulators are perfect at developing those skills.

So are simulators replacement for a real aircraft though? Nowhere near. Simulators should be treated as an addition to inflight training - not as a replacement for it.

Re:No, it's really not.

[CRC'99]

Oh, and I forgot to include this link in my response above...

Simulator training flaws tied to airline crashes:
http://travel.usatoday.com/flights/2010-08-31-1Acockpits31_ST_N.htm

Re:No, it's really not.

[GodfatherofSoul]

Combat flying is probably quite a bit different. I'm just an air warfare buff, so I have no firsthand knowledge but fighter pilots are very focused on what's going on outside the cockpit. They have to learn to scan for aircraft w/ or w/o a radar cue, tactics that are all about your position in relation to other aircraft, and obviously using weapon systems.

A coworker interned with the military on some secret missile tech 15 years ago (I suspect he was working with HARM trajectories from what I could glean) and he said the guys who ran the simulators were far better at evading SAMs than actual pilots because of all the sim time they had. Come to think of it, he said the actual pilots were pretty abysmal at evading them which is a tactic best rehearsed on a simulator!

Re:No, it's really not.

[SplashMyBandit]

\Not true. IEEE 1278.1-1998 DIS has been out for a long time (first version of the spec in 1993). It is designed to have massive numbers of interacting agents. Even on a Pentium II you could get a thousand clients connecting. How do I know this? I'm currently writing a multi-threaded cross-platform modern jet combat simulator (using Java 7+OpenGL/JOGL+OpenAL+JInput+OpenDIS, if you must know) and having a lot of clients and CGF (Computer Generated Forces) is well within the capabilities of one of the 8-core AMD CPUs you can get. The performance of OpenGL and modern multi-threaded Java is outstanding (compared to the existing single-threaded C++ jet combat simulations currently on the civilian market; the DCS series and Falcon BMS).

My understand is that many military aircraft these days (and certainly the simulators) have hardware support for IEEE 1278.1 so pilots can learn to operate cockpit in a massively networked environment. Hence, it appears your post is speculation rather - because the military have been able to have extensive simulated engagements for a long time - although a particularly royal air force was interested in the characteristics of the 80-player (40 a side) battles we were able to host in the Eagle Dynamics Flaming Cliffs 2 simulator.

Re:No, it's really not.

[Loki_1929]

If you're looking out the window for your enemy in a modern air combat situation, you're either about to die or lots of people screwed up in lots of ways.

Nobody has given much thought to non-BVR air combat in about 10 years and for good reason. First sight, first shot, first kill. That's the whole idea of stealth and advanced detection systems for fighters: I'm harder to see, so I see you first, so I shoot first, so I go home minus one long range missile. That's why a $140 Million F-22 makes more sense than three $40 Million Eurofighters. Once the fight is over, nobody got within 40 miles of an enemy and all you have to tally up is the cost of three planes and three trained fighter pilots versus the cost of three missiles and some gas.

Re:No, it's really not.

[mabhatter654]

But if two $40 mil planes shoot down one $140 mil plane you are ahead in the next round.

Part of the issue with Sims is that planes are SO expensive now we can't keep enough pilots ready to fly them. Back to your example, they have three ready pilots to one... The amount of things that go wrong on these expensive planes without being shot at evens the odds a bit while your pilot waits for another plane.

Also, the US military is really trained to fight Russia or China.. About the only countries that can put equivalent aircraft in the sky... So the other European countries just need planes for external threats... The occasional Middle East operation... They have no intension of going against USA, China, or Russia head on.

Re:No, it's really not.

[Solandri]

A previous company I worked at helped create 1278.1 (DIS). And you're correct that it's able to handle massive numbers of simulated entities interacting (easily tens of thousands; I think they managed one sim with over 100,000 entities in the mid-1990s). The crucial difference from the simulated shared environment most of us are familiar with (online games) is that the participants don't cheat. Hacking your client so it doesn't operate as programmed defeats the purpose of running a sim. Whereas in a game it's frequently in the player's best interest to cheat by hacking their client.

So in an online game, all the position, movement, actions, and collisions have to be handled by a centralized server to make it impossible to cheat. With DIS, each client calculates its own interactions and simply multicasts the consequences (e.g. movement changes) to all the other sim participants. e.g. The F-16 sim tells everyone it drops a bomb from its location with this trajectory. The M1A2 tank sim uses that to calculate that it was hit and destroyed, and it tells everyone "I'm dead now."

Since all these calculations are distributed, your computing power scales with the number of participants, unlike an online game where the server computing power is fixed. And the primary limitation on scalability is how much traffic the network can handle (the spec calls for a very minimal packet size, and a lot of work went into decreasing the frequency with which an individual sim needed to multicast updates).

Re:No, it's really not.

[serviscope_minor]

That's why a $140 Million F-22 makes more sense than three $40 Million Eurofighters.

Plus the F22 is much easier to simulate. Just put the trainee in an hypobaric chamber until they pass out.

Re:No, it's really not.

[dywolf]

"since the vietnam war"
There also hasnt been a real air war since Vietnam, and yet there's still be a few dogfight incidents.
Missiles fail, missiles can be jammed/countered/evaded. What happens then? More missles? You can only carry so many. Turn and run? Not always tactically viable, and makes you a nice big target.

Sensors can be evaded, thus negating the BVR combat space, letting him get that much closer and taking you by surprise. they can be vectored in on a blind spot (we dont -always- have AWACS radar coverage), they might appear from below you (particularly zoom climb interception profiles, aircraft that were never BVR to start with). hell, that's pretty much the whole concept behind stealth: to negate the other guys detection abilities and get him by surprise. As more and more stealth planes appear, as was bound to happen, it decreases the usefulness of the BVR-only concept, once again pointing out how focusing just on BVR will bite you in the ass.

Point is, you cannot, just CANNOT, tunnelvision on just one tactic. You must remain flexible, you must not leave a backdoor wide open. And that is why to this day we still teach dogfighting tactics, perhaps even moreso than BVR combat training (because it's more complex, and less forgiving).

You cannot build a giant nearly invulnerable death machine, and then ignore the thermal exhaust port that leads directly to the reactor core, even if it is only 2 meters wide.

Re:We almost have self driving cars

[smash]

Already exist. ACLS carrier landings have been available for over a decade now I believe, and carrier landing is probably one of the trickier things.

Re:Don't follow the Canadian example

[Dzimas]

Canada's military spending ranked 14th in the world in 2012. There are 180 nations in the world that spend less on their militaries - hardly chronically underfunded. Canadian soldiers are dedicated and extremely hard working; your attempt to slander the present day Canadian Forces because of an event that occurred 20 years ago is ridiculous. We are not proud that two Canadian soldiers beat a teenager to death in Somalia in 1993, but they don't represent the 115,000 active and reserve personnel in today's CF in any way, shape or form.

simple solution

[aapold]

Simple solution...to prevent jamming... put the simulator... inside the drone!

For the record -- why do we still need pilots?

[rocket rancher]

I'm a private pilot with a multi-engine rating. Simulators seem to be a good way to rehearse cockpit procedures, but unless they figure out a way to simulate g-forces, that's about the limit of their usefulness. Simulating a spin recovery procedure is one thing, doing it for real with a two- or three-g load from the spin is another. With that said, I don't think commercial and military pilots are going to have a viable career field for much longer. Military pilots are already being replaced by drone operators, and I think the rate of replacement is going to accelerate if the drone program keeps posting the kind of successes it has enjoyed so far. Unmanned vehicles seem to be the future of military aviation. Commercial pilots will probably last longer, because commercial airlines have to convince a skeptical public that airliners are going to be as safe with a computer at the stick as they are right now with a human. Realistically, commercial pilots have a hand on the stick only during takeoffs and landings, but all modern heavies can land and take off under autopilot, and have been able to for about thirty years. IIRC, a Douglas Skymaster made a transatlantic flight completely on autopilot, including the take-off and landing, even farther back than that (late 1940s? have to google that) so the technology is definitely out there. IMHO, pilots are still in commercial cockpits (and will be there for a while) because the paying public wants them there, not because they need to be there.