Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!gatech!newsfeed.internetmci.com!in1.uu.net!bcstec!nntp
From: Mike Fouche <mikef@hsvaic.hv.boeing.com>
Subject: Re: Stability of neural control systems
X-Nntp-Posting-Host: e677744.hv.boeing.com
Content-Type: text/plain; charset=us-ascii
Message-ID: <DMGpDF.sF@bcstec.ca.boeing.com>
Sender: nntp@bcstec.ca.boeing.com (NNTP News Access)
Content-Transfer-Encoding: 7bit
Organization: Missiles & Space Division
References: <311691F9.2781E494@elec.uq.edu.au>
Mime-Version: 1.0
Date: Thu, 8 Feb 1996 14:53:37 GMT
X-Mailer: Mozilla 1.2 (Windows; U; 16bit)
Lines: 14


We've been doing work on neural control - one of the things we looked at was
to apply root locus techniques to determine the stability of a particular network, 
which had been trained on the behavior of a linear classical controller.  Of course
a neural net is a highly nonlinear entity and root locus is geared towards 
particular setpoints of linear systems.  However, we still got good results in that
the root locus could predict which neural controllers were stable and which weren't.
As a first order "quick look" it wasn't bad.  

In terms of other work on stability I can't tell much.

Mike Fouche
Boeing Missiles & Space Division

