If air-to-air missiles learned a few tricks from cruise missiles...

I've been trying to think of ways to improve air-to-air missile capabilities vs. current limitations, and I think there's some interesting improvements that could be made if some ideas were borrowed from the "book of cruise missiles:"

1] Propulsion:

AMRAAM is limited by the energy stored in it's solid-rocket motor. A target with sufficient head start can out-distance the missile, or get far enough away that the missile won't have enough energy left to manoeuvre and hit.

Proposal: Switch to ramjet power. It's more expensive but will considerably improve range and speed, particularly at the extremes where a comparable solid-rocket motors tend to fall short

2] Target ID:

"Rules of engagement" often limit the range of radar-guided long range missiles, for fear of targeting non-hostile aircraft.

Proposal: Use GPS/INS to fly the missile to it's target (AMRAAM/Tomahawk style), but use IR-imaging + electro-optical sensors for terminal guidance, rather than radar. Cruise missiles use IR/Optical "digital scene matching" systems to pick out their target once they get close; an AMRAAM with this capability could fly out to it's target, take a picture, and compare it to a list of likely targets before the final decision to attack is made.

3] AIM => AGM option:

AMRAAMs are currently anti-aircraft missiles only. If you're expecting enemy fighters and find yourself targeted by SAMs, your AMRAAMs aren't going to do you much good.

Proposal: If you stick to what I've described above, you've go a weapon with GPS guidance and IR homing abilities. Sparrows/Standard Missiles carry 86/250lb warheads, enough to do plenty of damage to ground targets (i.e. a SAM battery) even if the warhead was originally optimized for airborne targets. Add SAMs to the list of target images in memory, and you can use it as an "anti-air defence" air-to-air/air-to-ground weapon.


So that's quite a long and rambling post I just wrote. I'll leave it up for discussion from here!

Originally posted by RedMatt

1] Propulsion:

AMRAAM is limited by the energy stored in it's solid-rocket motor. A target with sufficient head start can out-distance the missile, or get far enough away that the missile won't have enough energy left to manoeuvre and hit.

Proposal: Switch to ramjet power. It's more expensive but will considerably improve range and speed, particularly at the extremes where a comparable solid-rocket motors tend to fall short

I think the european countries i think it Britain is already doing this. It's called the Meteor
www.fas.org...

Some good ideas.

The ramjet power would be well suited to many missiles - especially a 'revised' Phoenix missile - possibly even a scram jet could power such a missile. The Phoenix works, literally, like a miniature ICBM - complete with detatchable re-entry vehicle (it never left the atmosphere - but the warhead detatches from the rest of the missile and is guided to the target by active radar like a homing meteor).

A Scram jet could potentially yield far greater ranges (although I have no idea from how far away you're trying to hit your victim.... Intercontinental just doesn't seem fair ).

Many of the issues currently existing with weaponry are easily solved on an engineering front - but on a monetary front - lose dramatically. The amount of processing power needed in a missile to allow for a comparison of radar and IR signatures to ID an aircraft as friend or foe is not cheap. Add into it that people are not naturally trusting of the deductive reasoning capabilities of a silicon wafer..... and you have a no-deal. It's a nice idea - and I'd incorporate it into my own private airforce (if I had one) - in fact, I'd be lightyears ahead of declassified systems integration (hell, that isn't hard - there's no competition).

The only one I see as impractical is the multi-role missile. This is due to the nature of the two different warheads. Intercept missiles are designed with fragmentary warheads designed to spew shrapnel out in a specific pattern in order to maximize the lethality and likelyhood of a hit. They detonate based on given data about the distance, movement, and velocity of the target - trying to aim their shrapnel into the path of the aircraft.

Although this would be like a claymore mine strapped to a rocket for infantry - it wouldn't be much of a menace to even a lightly armored vehicle (except for requiring the crew to change their pants).

The only truely effective solution - which isn't yet available (to my knowledge) would be a warhead that is preserved in an alternate form - suppose, a gel or paste, that is moved between two chambers - the forward chamber creating a shaped charge to peirce armor - and the rear chamber being for destroying enemy aircraft with shrapnel encasing the forward chamber or inbetween the forward and rear chambers. It then solidifies when exposed to an electrical current (within a very short time frame) - possibly forming two different compositions based on frequency or a voltage threshold. Then detonates when exposed to precussion.

But I'm no chemist (yet), so I can't help much with that shape-shifting warhead.

This is pretty much the same system that ch1466 has been proposing for... well.... ages.

Only I think his idea is more akin to UCAVs, being totally independant of manned/piloted [usually referred to as "the waste of space up front" or something similar in his posts ] fighter/interceptors.

Originally posted by Aim64C
Some good ideas.

Thanks!

A Scram jet could potentially yield far greater ranges (although I have no idea from how far away you're trying to hit your victim.... Intercontinental just doesn't seem fair

).

Intercontinental AMRAAM, hmmmmm... Actually, I originaly pictured this as something carried by an AWACS or their "escorts" which could be used at the extremes of their radar coverage (250-400 miles, I think).

Many of the issues currently existing with weaponry are easily solved on an engineering front - but on a monetary front - lose dramatically. The amount of processing power needed in a missile to allow for a comparison of radar and IR signatures to ID an aircraft as friend or foe is not cheap. Add into it that people are not naturally trusting of the deductive reasoning capabilities of a silicon wafer..... and you have a no-deal.

I wasn't planning for the missile to independantly ID it's target. The goal would be to track the target using manned radar (i.e. AWACS platform) and try to predict what was being fired on (we *think* they're either Mig 29s or SU-27s...). The missile would then try to match what it saw to the predicted target (... that doesn't look like a fighter jet...)... and could abort/radio back a picture and swing around for another pass while the operators decided what to do (... whoops, thats a passenger plane, abort!).

I'm one of those people who wouldn't want the "silicon wafer" acting entirely on it's own... especially when it's so easy to have a "man-in-the-loop" providing a failsafe where programing comes up short.

The only one I see as impractical is the multi-role missile. This is due to the nature of the two different warheads. Intercept missiles are designed with fragmentary warheads designed to spew shrapnel out in a specific pattern in order to maximize the lethality and likelyhood of a hit. They detonate based on given data about the distance, movement, and velocity of the target - trying to aim their shrapnel into the path of the aircraft.

Although this would be like a claymore mine strapped to a rocket for infantry - it wouldn't be much of a menace to even a lightly armored vehicle (except for requiring the crew to change their pants).

Actually I don't see that being as much of a challenge. First, because I'm not trying to punch through anything substantial (radar dishes/mobile missiles would be the main target) and also because warheads with different effects are possible. LOCAAS uses a warhead that can detonate and produce fragments (good for aircraft), a shapped-charge effect, or an explosiveley formed penetrator (EFP) - depending on which detonator is used.


As far as cost goes, I'd claim "multi-roll" overcoming specialization: one missile replaceing three (short/long range air-air missiles, anti-radar missiles) - saving on development costs and numbers of missiles needed (you only need to carry 1 vs. 3 for the same capability).

[edit on 7-8-2006 by RedMatt]

Ch1466 agree on some concepts - but I'll be damned if a chip takes away my future wings. I don't care if I do get shot at - I gladly return the favor in a more effective manner (they are dead - not me), wilst being able to adapt to a situation rather than being programmed with a textbook.

Flying is an art - and the closest thing to art that a computer can do is 'fractal' or 'non linear' equations. I'm sure you could adapt this math into the AI of a drone - but I don't give it much merit.

Now - a drone can fly routine patrols and support missions while I go knock on their front doors with an F/B-23 - I could live with that. I'd still like a 'manual override' on those things where a person can take control and play armchair maverick via satelite link.

Systems integration is one of my key developments in 'my' fighters. All modified fighters of 'my' creation have the designation F/A-X (for fighters) - which I just pulled out of my rectum for a programming project at school (and decided to stick with it for the book I'm developing...... and it's going to be a while before it's ready....).

Anyway - 'my' para-military group operates from a small number of underground air bases in Nevada, Australia (have yet to nail down a location, there), Syberia, and a bizzare looking submarine that makes bath toys out of aircraft carriers (permitting I can work the engineering aspect into it - it may just be too big no matter how many 'super materials' I make up). Anyway - all of these bases are linked via a central computerized network and digital assistant AI. At any given moment - any computer in the base can recieve radar data and target telemetry from any aircraft within the group.

Beyond this - all weapons are programmable to a very flexible degree - and assisted by on-board AI computer systems which update via radar data every few microseconds and cross-reference previous maneuvers from the aircraft (even when not engaged) to calculate a missile flight-path that will give the best kill ratio available.

Weapons are also group-oriented - automatically checking against data from other aircraft to ensure no aircraft is unknowingly double-targetted. The weapons, themselves, are also linked into the comm network - allowing them to re-target in-flight, or acquire a more recent and priority target. This also makes recovery from a missed intercept a greater reality (although still slim).

Systems within the aircraft are almost so deep I can't think of where to start - and whether or not I can put half of it into a post here.... but pretty much immagine that every role on the aircraft can be performed from every console on the network (including AI assistance - even acting as your RIO) - and a 3d 'hud' which acts like a computer desktop - allowing various displays to be moved, magnified, expanded, etc - everything from a technical read-out of a target to emulated games for long flights where the computer can effectively take control. Why not - the organization has, literally, unlimited physical resources and is the most advanced combat force on the face of the Earth.

AIM64C,

I've actually "lurked" here awhile and I think I know what you mean about ch1466. I'll simply say that I agree with him on some points, disagree on others.


First off, you don't want to use fractals for programming a UCAV pilot. You're better off with something simpler: statistics. The idea is to assess the situation (location of friendly and enemy forces, objectives, and capabilities of the craft) and then select the response most likely to accomplish the objective. The best example of this was Deep Blue (chess playing computer), which simulate some 200,000,000 different "next moves"/second, analyze each, and select the one that was most likely to result in checkmate for it's opponent.

If the computer has enough information and processing power, it will start behaving like a human would - or at least the humans that programmed it. Which is also the limitation: Deep Blue was programmed (in part) by chess plays. Any viable UCAV pilot is going to need "training" by humans. And as tactics/combat evolves, you'll continue to need humans to update the program's response to new threats.

Humans evolve + Computers don't = you'll always need a human to program the computer.

However, I think we've reached the point where we'd better leave the pilot on the ground. This can be proved with a little math:

Missile Speed: .8 miles/second (~mach 4 at sea level)
Jet Cruising speed: .2 miles/second (~Mach .89)
Engagement range: 15 miles.

Situation: you're flying to your target when a mobile SAM flicks on, targets your plane, and launches a missile. A little easy math says you've got ~15 seconds before the missile reaches you. I hear lots about pilot improvisation, but on that timescale you really don't have time to think ("well that didn't work, so lets try plan B"). Either you go the right way first, eject, or die.

US pilots do as well as they do because we simulate every possible engagement/outcome hundres of times before we encounter it on the battlefield. Our pilots make the right moves because it's been trained into them over and over. CNN actually reported it perfectly when the interviewed a pilot: "I was a bit nervous when the AAA started, but all the training took over..." And at that point you're better off programming that scenario into a UCAV, which can pick the right option based on many more variables and options than a pilot will have a chance to think about.

And worse case scenario, your UCAV screws up, picks the wrong evasive action, and gets blown out of the sky by a weapon/technique you hadn't seen coming. At which point you can go back and update the software (reprograming your jets in flight) to adapt to the new situation.


That doesn't mean you shouldn't fly. I plan on flying.... by getting a pilot's licence and my own plane at some point. But I think that future combat jets will be more effective if we don't put people in them...


Sorry for such a long rambling post, it's late, I'm bored... and I'm enjoying the conversation!

Hmm....

I don't know... for me - half of the thrill of flying would be the high chance of dying - those high stakes. I work better under high pressure - when others are freaking out - I'm the guy that stands up and gets everyone doing SOMETHING - even if it isn't the truely best course of action - any group coordination is better than chaos.

If I didn't have my life dangling in front of me - it just wouldn't have the same flare - both good and bad. I need the pressure to work. Stick me in front of a computer screen with a remote control aircraft - and it becomes nothing more than a toy to me - no matter how much otherwise is drilled into my head. My life is not directly dependent upon the survival of that aircraft.

Chances are I wouldn't be willing to take any risk in a UCAV than I would a normal aircraft. My goal is accomplishing my mission - and smashing what stands in my way. If I'd take a straffing run at a radar installation threatening me, I'd do it in a UCAV or in 'standard' combat. I'd just have more heart in it if it were ME there.

Maybe I'm crazy and have a death wish..... Maybe I'd change my mind after one real combat sortie. But I doubt it.

The concerns I have with UCAVs isn't their adaptability - but that the very means of adaptability will become the target of the enemy. You can severely limit ranges of UCAVs with strikes on satelite relay stations - or satelites, themselves. You are then mostly limited to visual-range control.

I know that seems impractical - but, it's the stuff you don't see coming that gets you. And it'll go down as a REALLY big tactical blunder to replace most of our air forces with UCAVs and then be unable to give them instructions beyond visual range because a coordinated 'terrorist' strike (special ops for all that matters - they sneak in and blow it up) takes out a number of satelite relay stations - both military and civilian.

You can also jam communications - or attempt to - keeping those aircraft from recieving human assistance - possibly even target-confirmation - keeping them from ever firing a weapon.

Then you have the predictability of UCAVs. Given the odds - they will react the same way in similar instances repeatedly. Perhaps that isn't a fatal flaw anymore... but that sort of screams 'cannon fodder' when it comes to ACM.

For what it's worth, these are all good ideas and I think to some extent have mostly been pursued, or at least thought about.

I can't talk much about the Target ID part, because I honestly don't know enough about radars to really give any good input. However....

Propulsion- The ramjet idea is a good one, but probably best reserved for the large missiles. Because the solid-state fuels are cheaply made and offer fair range/performance I don't think that we really need to revamp this system. A really-long-range (RAAMRAM I screwed up the acronym-I don't care. It's late. cut me some slack) missile shouldn't be counted out of importance, though. It would be a compliment to the BVR capabilities of any aircraft. The only thing is that when the missile is launched the aircraft has lots of time to do something about it.

AIM-AGM- This is a great idea that I have been pondering for a while. It would be so very, very nice to have a missile that I could just switch radar modes with, and explode some SAMS. This is the very definition of multirole. The F-15, the F-16, they would have haydays with this kind of thing. Ditto helicopters. This is an idea I would like to see pursued until completion, although I don't expect immediate action to do so.