Newsgroups: sci.op-research,comp.constraints
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!newsxfer.itd.umich.edu!uunet!newsflash.concordia.ca!CC.UMontreal.CA!IRO.UMontreal.CA!iros1!pesant
From: pesant@nil.IRO.UMontreal.CA (Gilles Pesant)
Subject: Re: Optimization of Geometric Constraints
In-Reply-To: caseau@dmi.ens.fr's message of 5 Sep 1994 09:53:46 GMT
Message-ID: <PESANT.94Sep8005732@nil.IRO.UMontreal.CA>
Sender: news@IRO.UMontreal.CA
Organization: Universite de Montreal -- Laboratoire Incognito
References: <34epra$788@nef.ens.fr>
Date: Thu, 8 Sep 1994 04:57:32 GMT
Lines: 44
Xref: glinda.oz.cs.cmu.edu sci.op-research:1742 comp.constraints:278

In article <34epra$788@nef.ens.fr> caseau@dmi.ens.fr (Yves Caseau) writes:

> From: caseau@dmi.ens.fr (Yves Caseau)
> Newsgroups: sci.op-research,comp.constraints
> Date: 5 Sep 1994 09:53:46 GMT
> Organization: Ecole Normale Superieure, PARIS, France
> Lines: 17
> NNTP-Posting-Host: dmi.ens.fr
> 
> Hello,
> 
> I am looking at some optimization problems where the constraints can
> be seen as "geometric": relative position, distances between points, lines
> and planes (all in 3D), angles ...
> All domains are bounded (floats).
> 
> What are the relevant techniques that I should know ? Are there systems that
> would solve such problems ? Can I try some generic CSP techniques 
> (and products) or should I exploit the geometric nature of the problem with
> some specific algorithms ?
> 
> 
> Thanks in advance,
> 
> 
> -- Yves Caseau

The experimental CLP language QUAD-CLP(R) introduces a partial solver
for quadratic constraints, the main motivation being "geometric"
constraints. The idea is to reduce or approximate each original
constraint using a Boolean combination of linear constraints. There is no
direct support (yet) for optimisation, though.

A paper outlining the algorithm and presenting some applications
appeared in the proceedings of PPCP 94 which will be published soon in
Springer Verlag's LNCS series. In the meantime, it can be retrieved by
ftp from june.cs.washington.edu in directory pub/constraints/ppcp94. 
You can also contact me for more details.

I hope this is of some help to you.

Gilles Pesant
University of Montreal
pesant@iro.umontreal.ca
