ess (phil hunt) wrote:

:On Fri, 19 Dec 2003 06:17:02 GMT, Kevin Brooks wrote:
:
:"phil hunt" wrote in message
g...
: Even
: LCCM's are fairly high technology, and 'dead reckoning' isn't as easy
: as it sounds.
:
: Why not?
:
:Accumulated error, for one thing; you can't count on GPS for positional
:updates.
:
:Say the error is 1%. Then it'd be 1 km off on a 100 km journey.
:That's close enough for terminal homing to

No, it isn't. Which way is the target and how far is it from where
you actually are? What's the FOV of your sensor. How long can you
hang around and survey looking for your target? Will you even
recognize it when you see it?

I'd also suggest you look at the specs for real IMUs that fit in your
price tag before you start making assumptions about how much error
you're going to get.

:Your LORAN idea fell flatter than a pancake.
:
:No it didn't.

Well, the stations to support it fell flatter than pancakes the minute
you got invaded. Now you have no guidance. Now what?

: So you are now left
:with trying to cobble together an inertial nav system--more weight and
:complexity, more R&D required, and in the end it is not going to give you
:the kind of accuracy you need over the distances you will have to negotiate.
:
:Are you an expert on inertial nav systems? If so, how much
:weight/cost? If not...

I sit in a room with a bunch of GNC types. You don't have a clue.

: and if you start going for IR systems, you've just stopped
: being "cheap".
:
: That's mostly true, IR cameras cost around $5000. Probably it'd be
: best to have plug-in sensors so ther operator could choose to add IR
: when it's necessary for the job.
:
:Now you need a whole new set of target data--more R&D again.
:
:I'm not sure it would be that much more. For the main application of
:spotting moving vehicles you could probably use essentially the same
:software. Also, the shape of objects under IR is the same as under
:visual light.

Not even. There are algorithms for converting from one to the other,
but you have to have those in hand.

: Also, computer's and programs that can pick out
: targets against ground clutter are somewhat more difficult-- note the
: fact that even now the U.S. still prefers laser guided missiles, and I
: don't believe we have any autonomous weapons like this in stock
: (although some are being made ready). The problems are tremendous.
:
: sarcasmWell, obviously, if the USA can't do it, no-one else
: can./sarcasm
:
:The hell with your sarcasm, the fact is that it is a hell of a lot harder
:nut to crack than you seem to comprehend. If you think otherwise, you need
:to go into business for yourself and offer us this wonderful, cheap, easily
:produced autonomous attack system to ther DoD.
:
:My understanding -- and I've heard this from multiple sources -- is
:that in defence procurement it's not how good your product is, it's
:who you know.

And I'd bet none of those sources had contracts, either. Sour grapes
always tells a good story.

: Swarm co-ordination is a software problem. To solve it, you need a
: few clever postgrad students, properly managed.
:
:ROFLOL! Gee, I guess you also consider AI to be something you can acheive
:over next weekend, right?
:
:Er, no, I didn't say that. And in any case, swarm co-ordination is
bviously not AI-complete, as you would know if you knew anything
:about it at all.

It's also very difficult, as you would know if you knew anything about
it at all.

: Your habit of taking every serious problem with
:your pet theory here and writing it off as a "software problem which is easy
:to take care of" is getting a bit monotonous.
:
o you know anything about software? I've been a programmer all my
rofessional life, and I like to think that I do have some
:understanding of the field.

You may know everything there is to know about PROGRAMMING. You're
still clueless as to how hard some of the problems are that you're
just hand waving away.

: The idea that Italy couldn't make a cruise missile is silly IMO.
:
:Sure they could--but they can't make the autonomous uberweapon you have
:posited. Nothing to be ashamed of--right now neither can we. But you can,
:because all of the problems are mere exercises in writing a few lines of new
:code, right?
:
:A few hundred thousand lines, more like.

Keep counting.

--
"Millions for defense, but not one cent for tribute."
-- Charles Pinckney