Saving the U.S. Air Force

New York Post ^ | Feb. 11, 2005 | Ralph Peters

[Rokke]

"Trust me, I am acutely aware of the large fights and the small ones"

I know that. That's why your comments in your last post caught me off guard. I've enjoyed your posts in the past. They are a refreshing antidote to the usual pontification coming from the CNN-fed armchair generals that frequently post on this site.

"The AF is not investing in the proper force structure to support the fight we have."

With all due respect, if we each had a dime for everytime anyone has said that exact thing about each of the services, we could probably buy a military of our own built exactly to our specifications. You've been in 30 years. When has the Army ever been satisfied with the Air Force?

"Perhaps it was merely the certainty and triumphalism of your tone that put my antenna up."

I'll try to sound more defeatist in the future. After a few more of these threads, it won't be that hard.

"Thanks for the eggs."

Well hot damn. At least someone is thankful for something the Air Force has done.

[Rokke]

I agree that subs are an ideal cruise missile platform. They did outstanding work in OIF. But I disagree that cruise missiles aren't stealthy. While the current batch aren't technically low observable, we've already tested some that make the F-117 look like a purpose built radar reflector. Remove the man and you eliminate many of the sources of radar return.

One of the magical capabilities of the B-2 is its ability to fly an entire mission without a human touching the controls. That includes takeoff and landing. The Army is currently flying UAV's by placing a man in a simulator cockpit surrounded by a 360 degree projection of a photo-realistic terrain database of the area in which the UAV is flying. The applications of this are endless. Since the man is still the limfac for performance in a fighter, if we can figure out how to provide realtime sensory data to a remote cockpit, we can "man" aircraft with the same turning performance of an air to air missile and the endurance of a Globalhawk. I believe we are on the threshold of a huge revolution in aerial combat. And I think that is a good thing.

[RFEngineer]

"One of the magical capabilities of the B-2 is its ability to fly an entire mission without a human touching the controls"

You are thinking of the B-2 described by Arnold Swartzenegger in Terminator II, not the real one.

[ProudVet77]

Knew we were on the path to stealthy cruise missiles, but have been unable to find any particulars. The current Tomahawks are pretty bad, their IR signature is a dead give away. The real problem with cruise missiles are cost though. That is where a UAV comes in. Multiple warheads in a single stealthy package. And the dispenser comes home. Even better if it's a bomb. google SDB.
I agree we on the cusp of many changes. Even the Navy has some hot items up it's sleeve. But I've spent many years in software, with an accent on AI. It's hard to replace a pilot. The bandwidth requirements are astronomical. Plus maintaining a good connection while a UAV is doing a 15G turn is hard. The antenna won't keep up! And if the UAV is xmitting data back to a controlling force (pilot or AWACs), it can be jammed and HARMed.
Not sure if you are familiar with this, but Navy pilots rarely make their own traps now. The computer does it better. That is why an X-47 is just a few months away from carrier landings.
Sorry, you still will have to work for a living. But you may be commanding a wing of UAVs.

[ProudVet77]

Hate to tell you, but I race sailboats with a commercial airline pilot. And he can do the same thing.

[RFEngineer]

"Hate to tell you, but I race sailboats with a commercial airline pilot. And he can do the same thing."

I guess I missed that capability in all the time I spent up close and personal with B-2 avionics.

[Rokke]

"You are thinking of the B-2 described by Arnold Swartzenegger in Terminator II, not the real one."

Dammit. I always get the two confused. But all kidding aside, I've even flown my lowly Viper in a fully automatic terrain following mode that flew itself from one point to the next. But we still had to land it!

[jackbill]

Or, the $600 toilet seat/$900 hammer/$1200 coffee maker nonsense of the 70s.

That's a damnable myth. The toilet seat "cost" $600 because of the accounting procedures of the military. If a contractor provided a jet engine, a toilet seat and a wing replacement, the total cost was divided by the number of items. So the toilet seat "cost" $600, but so did the jet engine. Get real, and stop parroting the liberal line.

[RFEngineer]

"Viper in a fully automatic terrain following mode that flew itself from one point to the next. But we still had to land it!"

I am teasing you, but as you say, all kidding aside, it can't fly itself as you describe, but for major portions of missions, it can bore a pilot to tears.

[Rokke]

I agree that ultimately a reusable UAV would be better than a cruise missile. Except you obviously have to double the range, increase its survivablity, and have a place to land it. It is a nice trick to sneak a sub up close, launch its missiles, and then disappear never to be seen again. And I think you are right about the bandwidth problem as well. If it can be transmitted, I'm sure it can be jammed. I've flown simulator profiles against AI adversaries that were very good, but I'm not sure what application differences there are for transfering that AI from a computer model to an actual aircraft. It seems though, that with computers and AI it is just a matter of time.

[185JHP]

I agree. Write Peters off - way too much crankiness and nuttiness to speak about the military as a whole. Let him yap about Army Specops stuff. The designer of the old A4 had a great motto: "Simplicate. And add lightness." A4s are still good dogfighters. Defense contractors need close scrutiny so they don't veer toward what's most profitable, instead of what's best.

[ProudVet77]

Just guessing it was probably not part of the B-2 package. But I can tell you Navy pilots rarely land their own planes on a carrier. The computer does a better job of it. They do it once in a while to be able to handle emergencies.
And no cracks from the AF pilots! With GPS and fly by wire you can pretty much land the plane and have it taxi up to the correct landing gate at the airport. The only thing missing is the folks at ground control can't tell your onboard computer when to give space and change landing gates when needed. But that will be a simple fix. Just integrate the two systems. Think GIG.

[Rokke]

"it can't fly itself as you describe"

I had a chance to fly the B-2 simulator. I was told the aircraft flew like a T-38. Well if the sim was any indication it flies more like a KC-135. Refueling was a nightmare. But then again, I'm more used to fly by thought. It was the B-2 pilots who told me it could fly itself from takeoff to landing. Frankly, I find it harder to believe it can't than it can. I've flown automatic landings in 727's.

[RFEngineer]

Oh, I'm not arguing that it's not possible - the B2 has the hardware capable of doing it, but going completely hands off....I don't think they've made that leap just yet, and there are just too many mission variables (and human elements) to make that a good idea in practice.

[ProudVet77]

Have you seen this?
http://www.air-attack.com/page.php?pid=28
I'd be interested in your thoughts.
I agree AI will come, but I was working with it in the 80s with a bunch of MIT types, and it was along way away then, and from my studies not much better now. The key to the problem is asking the AI the right question. Programing the solution is easy.

[RFEngineer]

I would agree that it would likely fly more like a kc-135 than a T-38. A B-1 may fly more like a T-38, but not a B-2.

The automatic landings are not so much the problem, it's the complexities of the missions that present more of a challenge. You are better off letting the computer fly on terrain following mission segments, I quite agree.

[ProudVet77]

I don't know if the B-2 is "twitchy" or something, or maybe it's just the 2Billion $$ price tag per unit, but I would think it could fly the whole mission by itself. You do avionics, so what do you need to land? Lat and Lon and height within say 10 feet. My $145 GPS does that. The final details are handled by feedback from the ground approach radar. Left/right, up/down. Should be easy. If the Navy can land planes that way on a carrier doing 30+ knots and the deck is pitching, not to mention they land at an off angle to the carrier's direction due to the angled flight deck, heck even a Marine should be able to do it.

[RFEngineer]

"I don't know if the B-2 is "twitchy" "

B-2 flying qualities are pretty decent for the most part.

[Rokke]

I think aircraft like the X-47 are the shape of the future. And they should be. I've flown alongside Predators on some very complex missions. The strangest part was talking to the Predator "pilot" via a link in the Predator. But obviously, the Predator is such a simplified version of what is yet to come that its capabilities provide only a glimpse of what is really possible. As aircraft like the X-47 and X-45 prove more concepts, the progression will be like that of the microchip.

[ProudVet77]

Sorry, wasn't meant in a derogatory fashion. Just don't know much about the plane other than the specs, and having flown, I know some planes behave differently.