Originally posted by truttseeker
I think the one thing that will ensure human dominance is instinct. Computers can always have faster reflexes, and reaction times. They can do more
g's and all of this stuff. But theyre only as good as theyre programed to be. They cant really learn. Once they can then it will be a problem, but
until then theyre stuck with what they have unless we upgrade them.
Instinct doesn't beat numbers.
If two threats come at you in _perfect synch_ from 2:30 and 9:30 you have to be able to turn twice as fast as they EACH do, just to stay in the
The only thing which will save manned airpower from swarming systems is the laser and then only because the first generation systems will still be so
large that 'might as well man up' doesn't make that big a difference.
Nor is the thought process inherent to air to air tactical maneuver all that sophisticated.
Around every airframe, friend or foe, is an envelope bubble which shrinks with aspect and speed and signature value for each weapons system likely to
be employed against it.
Navigating to a point where his bubble overlaps your weapons system is simply a matter of if-then-else stacking vector math subprocesses until you
find the the one which plays out soonest, with the highest threshold SSPK. If you at the same time run a /defensive/ equivalent you can further state
how long (on an optimized performance path through the sky) before YOU are in a similar position to the threat weapons systems.
If X reaches threshold .7 before Y reaches threshold .55 SSPK, go for X. At which point you call the sort on X over IFDL and all the other shooters
pick a different target UNLESS, one of them can get there before you do and then the target list morphs again.
Of course not every instance is offensive so if you or one of your robot squadron mates reaches a position where they hit a zero sum survivability
before countershot on threat X altogether a quick call can cause other elements to engage on sub-optimal 'fire for (pscyhological) effect'. And
with HOBS weapons especially, it becomes fairly simple to shoot the advantage threat off.
Indeed, I would put the creation of a synthetic replacement for all-round human vision as being 100 times more important than any element of AI
integration within the airframe tactical mission processor.
In terms of performance, what will really aid the robotic weapons platform will be two things:
1. Dual Axis Maneuver.
Even if it's only 15 and 7G, it will so vastly speed the rate of engagement as to blow manned (postive only) single-axis platforms right out of the
water as well as provide the ability to generate whacky apparent positioning versus 'intended' maneuver status. It will equally have a very large
effect on missile proportional logic as you cannot lead as far on a target which can snap back against the turn (without banking).
2. Signature Value.
A typical A2A UCAV will likely weigh less than 3,000lbs and thus be able to get away with _very_ small wings and a fuselage designed to an absolute
minimum of body height. Say 15-20ft long and 10ft in span with a 2-3ft frontal elevation. Absent conventional tails (which may fold up for landing
or be replaced by SSDs or all moving tiperons) at more than a few hundred feet, there will be no possiblity for a human to hold visual contact and
even augmented systems will have a hard time tracking the vehicle in any but planview.
With these and conventional RFLO as a given, we could design a system today (like a HiMAT/X-31 hybrid) which would so radically outperform any manned
system as to be utterly untouchable by them in close in combat.
At which point, it comes down to using the UCAV the same way a hunter uses dogs to flush game. If the threat launches early enough to beat the UCAVs,
it will face a scenario in which it will either be obvious as a discrete radar target separate from the main (friendly) raid packages. Or be utterly
handicapped in it's emissions from targeting the drones at ranges where a dense missile volley might significantly attrite the dogs. Either one
buying the threat aircraft a standoff face shot from BVR supporting leashholders.
If the threat waits to flush their QRA birds until the last moments, the drones will be in among the 'fighters' like wolves among the sheep and -at
best- you will be trading 10 million dollar UCAVs for 30-40 million dollar F-16 or JAS-39 type 'dogfighter' threats.
For the drone, successful mission accomplishment simply means closing on a threat until optical systems recognize the sky-backdrop silouhette as
discrete from only a few other platforms with similar span and heat sources (i.e. There may be 100 different trucks but only 3-4 medium fighters that
'could be an F-15', thus you can afford much more detailed models of each for point:point rejection).
At which point it either kills or bypasses said target.
Where MITL driven tactics -may- come into matters is in establishing dominant geometry position from before the merge using 'football playbook' type
tactics ala so-
Wherein, again, once you have simplified the "Fire at one of us, BVR and we will dogpile you like maddened animals from outside your sensor cone.
Try to blow through and we will convert on your large-signature and blow you away from visual range." elements of decision inherent to
microsignatures and very high degrees of on-board maneuverability, the specifics of tactical intuit are limited in terms of what a conventional threat
can do without compromising itself early or late.
As usual, the air power services /vastly/ overplay the 'accomplished warrior' degrees by which they dominate the threat as justification for
investing yet more money in systems and synergies approaches by which they secretly crutch-up the severely limited abilities of flying monkey
In this case, they have actually created a great danger to our forces however in that small numbers of highly capable shooters cannot be everywhere at
once and are themselves highly vulnerable to saturation tactics. By reestablishing the red baron air approach to fighter sweeps on the bleeding edge
(or beyond) EA protection and radar coverage, you generate a system more likely to find rogue threats and to engage them decisively before they are
_mixed in_ with other merged-plot radar tracks in a manner for which today's HOBS capable ISRM class makes nearly a coin toss for survivability.
Will we get there? No. But only because even fighter pilots are smart enough to realize that the _very instant_ you sacrifice the top of the glamour
missions to robotic supplementation, you make ALL the other missions (which are in some ways harder to do) equally subject to review.
And such would be the end of the Sky Knight conspiracy to waste funds on baby-onboard airpower that fails utterly to do its job.