Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!lard.ftp.com!bloom-beacon.mit.edu!world!pbrennan
From: pbrennan@world.std.com (Patrick M Brennan)
Subject: Re: Indoors positioning problem
Message-ID: <D8sDBs.JEJ@world.std.com>
Organization: The World Public Access UNIX, Brookline, MA
X-Newsreader: TIN [version 1.2 PL2]
References: <3p9ont$31e@ceres.bibsys.no>
Date: Thu, 18 May 1995 18:14:16 GMT
Lines: 28

Roar :

You have put your finger on a set of extremely difficult and
complex problems which are still on the active & unsolved list
as far as most roboticists today are concerned.

: The interesting part is positioning systems. I figured
: using a device (light or sound) that could measure a distance in
: a narrow beam...

This is a bad assumption.  I mean, it'll work in the simulator,
of course, but there is no device which will work exactly the
way you want it to.  This is the reason there is so much interest
in so-called "sensor fusion"; so that different sensors can 
combine to eliminate their individual defects.  Think more seriously
about modelling your sensors with some accuracy.

: ...I have assumed that it is possible
: to measure the distances within say 0.5% margin, and that it be possible
: to narrow the beam down to a degree, giving 360 separate measurements
: round the circle. Do in fact such hardware exist at a reasonable
: price? 

There are so-called Laser Radars, and they have very interesting
properties, but neither of us can afford one.  I think your 0.5% is
far too optimistic, not to mention your one degree beamwidth.

Patrick
