From: Laserlight <laserlight@q...>
Date: Thu, 5 Jul 2001 11:05:11 -0400
Subject: AI Fighters link
http://www.cosc.wustl.edu/doc3.html Maneuver Prediction in Air Combat via Artificial Neural Networks
From: Laserlight <laserlight@q...>
Date: Thu, 5 Jul 2001 11:05:11 -0400
Subject: AI Fighters link
http://www.cosc.wustl.edu/doc3.html Maneuver Prediction in Air Combat via Artificial Neural Networks
From: David Griffin <carbon_dragon@y...>
Date: Thu, 5 Jul 2001 08:17:20 -0700 (PDT)
Subject: Re: AI Fighters link
Very interesting. Of course as an aid to a pilot, by the time it tells you what you should do and you do it, it might be too late. Perhaps some kind of visual cue (symbols on the hud) could make this work. As much as a fighter pilot tries to do the unexpected, he probably does the expected fairly often. Good pilots also fly the plane well (executing maneuvers well) and can take more g's. Of course a computer plane would probably do both of those near flawlessly. I'm not sure I want to be a fighter pilot in the 21st or 22nd centuries when this all comes together. > --- Chris DeBoe <LASERLIGHT@QUIXNET.NET> wrote:
From: Roger Books <books@m...>
Date: Thu, 5 Jul 2001 11:29:22 -0400 (EDT)
Subject: Re: AI Fighters link
> On 5-Jul-01 at 11:18, David Griffin (carbon_dragon@yahoo.com) wrote: Note the accuracy, 55%-95%. I'm not sure I'd want to trust my life to a system that at times dropped as low as 55%. I'm not even sure much is being done now (the article was 1988-1989) with Neural Nets, they seem to have reached a bit of a dead end. Anyone here know about any recent NN work?