Unmanned (But Armed) Aircraft Experiments In 2001

John Warden, architect of the Gulf war air campaign, believes that by 2025 90% of combat aircraft will be unmanned. Next spring, the first armed aircraft without pilot, the X-45A UCAV will make its maiden flight. Replacing the pilot by a ground controller cuts the price of each unit by two-thirds, and makes it easier to transport. The Economist has more, and states 'the decision to fire weapons should be made by a human, to reduce the risk of "friendly fire."' This is not logical: Since the planes can be networked and thus know each other's relative positions, preventing friendly fire is a much simpler problem than the visual recognition required to determine what to shoot at, unless you don't mind hitting non-military targets. I wonder what Asimov would think.

Comments

Re:Incorrect assumption

[Sanchi]

Please, learn something before you spew your nonsence. First some background on me. I work on the computer program that runs on AWACS. I have seen what the system can do. The F-22 will not run any active systems when on an attack run. All of the detection is left up to us (awacs). and the radar system on it sucks, its only an plainer two pass (hard to explain).

And IFF is used by every single airplane in the air, not just NATO.

Sanchi

Incorrect assumption

[SuiteSisterMary]

This is not logical: Since the planes can be networked and thus know each other's relative positions, preventing friendly fire is a much simpler problem than the visual recognition required to determine what to shoot at, unless you don't mind hitting non-military targets. I wonder what Asimov would think.

Ok, so they won't shoot each other. But what about other friendly forces? Sure, put a location beacon on them, too. Then the enemy either a) tracks in on the frequency and shoots them, or b) jams them and watches chaos ensue. Humans will always point the trigger, if only so that the brass knows who to point the finger at later.

Re:Incorrect assumption

[Sinical]

As someone who works in the defense industry,
let me just say that weapons systems are *hard*.
You do not have the luxury of going "dang, a bug"
when your missile just decided to blow up
friendlies by mistake

Now, I work entirely on missiles, which have a
fairly small operational scope (kill *that*),
and I know how many hours (read, YEARS) missiles
spend in development, how much testing is done,
how many simulation runs are made, and the idea
of trying to build algorithms that try and decide
whether a *human* *being* should DIE is not
something I would relish or encourage.

IFF sytems break, they are destroyed in combat,
and maybe they are jammed. Allied systems aren't
compatible, or a wire gets loose, or whatever.

In my very not humble opinion, only PEOPLE get
to decide when people die. Remember, KISS,
and AI fire systems are most definitely not simple.

Re:Incorrect assumption

[Iron Monkey]

I agree. One place that computers still haven't surpassed humans is dealing with complex scenarios with insufficient information. Putting IFF on everything in sight may seem reasonable at first, but then someone might jam them. Put in swanky image recognition software, and the enemy repaints their planes to look like your allies... and so on.

Basically, any method you use to try and ensure no screw-ups occur can be broken by the enemy.. Find me a computer that can deal with situations as complex as identifying friend vs. foe in a heated battle situation - with very little time, and when the enemy is actively trying to decieve it, and I'll show you a human brain.

Perhaps a computer like this lies somewhere in the future.. I hope so, quite frankly. But I firmly believe that until then, humans are the best thing we have - by a long shot.

One other thing, regarding the ethical situation. Sure, there are losses to friendly fire in every war - these are likely inevitable. But risking large amounts of human life based on some new program or machine is potentially very stupid, given just how well tested the human being is in combat.. a very safe bet over the latest technological development.

Re:Incorrect assumption

[vheissu]

For some reason, this discussion seems to have focused on the idea that we have only two choices: A) Machines picking targets with little or no human intervention, but with the powers of radar, IFF, radiation counters, video cameras, etc. or B) People controlling the machines without any sort of electronic back up. This is ridiculous. People and machines have different strengths and weaknesses. Even if an IFF works 100% of the time, it still needs a human to determine whether an enemy should be targeted, ignored, or avoided. And that is one of the easier problems for a machine. On the other hand, people have problems too--we're relativly fragile, get tired, need heavy life support, and can't detect radio signals. What this system does is exactly what makes sense--it allows the people to control the machine remotely and make the hard decisions, while the machine gets up close and personal.

I am worried.

[shren]

The US Government has avoided or gotten pressured out of a lot of wars because American Soldiers were dying. Each technology designed to fight a battle without putting men on the field or in the sky will help move a political impediment to war.

Most people would consider this a bad thing.

Re:what asimov would think

[andyh1978]

Perhaps the Asimov reference is an ironic one referring to the story 'The Feeling of Power'.

In this (slightly heavy handed) story, the superpowers' computers battle each other, with fully automated weapons. Humans have become reliant on computers to do simple maths; why bother learning it when everyone has a computer?

But the military want a way to beat the enemy's computer weapons; it's too costly to put larger and larger computers in the weapons. So, they re-invent the idea of doing maths on paper (a shocking concept to those assembled, and they name it 'graphitics'), and with it the 'manned missile'.

The general drove on. "At the present time, our chief bottleneck is the fact that missiles are limited in intelligence. The computer controlling them can only be so large, and for that reason they can meet the changing nature of antimissile defenses in a unsatisfactory way. Few missiles, if any, accomplish their goal, and missile warfare is coming to a dead end; for the enemy , fortunately as well as for ourselves.

"On the other hand, a missile with a man or two within, controlling flight by graphitics, would be lighter, more mobile, more intelligent. It would give us a lead that might well mean the margin of victory. Besides which, gentlemen, the exigencies of war compel us to remember one thing. A man is much more dispensable than a computer. Manned missiles could be launched in numbers and under circumstances that no good general would care to undertake as far as computer-directed missiles are concerned-"

Full story can be found at this site.

PIO, G and other little problems...

[costas]

There are better reasons than cost to create UCAVs:

G-forces. At this day and age, the true limit of a fighter's performance isn't engine power or structural integrity: it's how many Gs the pilot can stand. Even with the best pressure suits, a UCAV has an obvious advantage.

PIO (Pilot Induced Oscillation): if you're gonna pull any tricky aerodynamics like the X-45 does (inverted swept wing, stealth profile) you need dynamically unstable aircraft. The problem with unstable designs in fighters is usually that the pilot overcompensates flight corrections --i.e., the resolution of the human is much lower than the resolution that the flight corrections must be made; in essence, the pilot is correcting the aircraft at a lag. Modern control systems of course correct for this already --by trying to determine what the pilot *wants* to do, rather than what he's putting in the stick-- but with higher Gs (and thus higher speeds) the human is the weak link.

Weight, of course. If I remember my Design courses correctly, the extra systems for the pilot account for about 20%-25% of a fighter's Take Off Gross Weight: armor plating, cockpit controls, air conditioning, etc. Weight is an aircraft's Number 1 limiting design factor.

OTOH there is one huge disadvantage to a UCAV: in a dog-fight, or whenever human perception is needed to reduce the decision tree to something manageable, they will always (well, for the next few decades anyway) be outmanned. Pun intended.

Disturbing Trend

[vergil]

I've noticed a disturbing trend when it comes to modern weaponry, war and the public's perception of both.

Recall the "conflict" (it wasn't formally a "war") in the Persian Gulf and the lavish media coverage fawning over the tricked-out American arsenal of depleted uranium, ship-launched cruise missiles and so-called "smart bombs."

I was in high school at the time, and remember well the glossy graphics in the corporate press extolling the efficiency of "fire-and-forget" rockets.

Later came a few insightful (but quickly forgotten) editorials criticizing America's "video game mentality" of combat.

Perhaps automated weapon systems are more efficient than those manned by humans. Maybe they'll even cut down on "friendly" casualties, and, in the long run, shave some dollars off of our bloated defense budget.

What really concerns me ain't efficiency, or cost savings. It's accountability. I think many fail to realize that war -- whether conducted with knives or napalm, whether hand-to-hand or computerized -- is about killing. Smart bombs and fire-and-forget missiles abstract killing to a small blip on a phosphorescent screen far removed from the actual event.

Unmanned flying gunships, I'm afraid, are a step in the wrong direction.

Sincerely,
Vergil