Newsgroups: comp.lang.smalltalk,comp.lang.c++,comp.databases.object
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!news.kei.com!nntp.coast.net!torn!nott!emr1!news
From: Mike Craymer <craymer@emr.ca>
Subject: Re: Question re: Global Positioning (GPS) technology in a railroad app
X-Nntp-Posting-Host: craymer.dialup.emr.ca
Content-Type: text/plain; charset=us-ascii
Message-ID: <DJvB5v.3L@emr1.emr.ca>
To: perform@iu.net
Sender: news@emr1.emr.ca
Content-Transfer-Encoding: 7bit
Organization: Geodetic Survey of Canada
References: <perform.64.00231B88@iu.net>
Mime-Version: 1.0
Date: Wed, 20 Dec 1995 04:29:55 GMT
X-Mailer: Mozilla 1.1N (Macintosh; I; 68K)
X-Url: news:perform.64.00231B88@iu.net
Lines: 60
Xref: glinda.oz.cs.cmu.edu comp.lang.smalltalk:32414 comp.lang.c++:165784 comp.databases.object:8619

perform@iu.net (PERFORMANCE Software Services, Inc.) wrote:
>I'm interested in having a discussion with persons who
>has had experience with developing software applications
>using GPS (Global Positioning) technology ... preferably
>in a object-oriented development framework.
>
>We are working with a company who intends to use GPS to 
>identify "positioning" within a tolerance of about 10 ft.
>There present view of this GPS application is that it is 
>a real-time application.  I am interested in learning what
>experience others have had in this arena.  How real-time 
>is this?  Does the application demand a specific computing
>architecture and language selections?

For more informative GPS feedback and discussions, post to the 
<news:sci.geo.satellite-nav> newsgroup. On the Web, check out the Navstar 
GPS Internet Connections at
<gopher://unbmvs1.csd.unb.ca:70/hPUB.CANSPACE.GPS.INTERNET.SERVICES.HTML> 
and the GPS FAQ at <ftp://sundae.triumf.ca/pub/peter/index.html>. For 
applications of GPS technology, including railways, see back issues of GPS 
World <http://www.advanstar.com/GEO/GPS/> (also available online at 
<http://www.dc.enews.com/magazines/gps/>).

Basically, most receivers should give you standard position updates as 
often as every second. The problem is accuracy is only about 100 m 
(2drms). For 3 m accuracy you'll need to use DGPS (differential GPS) or 
WADGPS (wide area differential GPS). The idea is to reduce the errors in 
the system by differencing and applying corrections. The GPS FAQ and the 
numerous GPS resources on the Web can explain these techniques much better 
than I.

Most GPS receivers use the National Marine Electronics Association's 
(NMEA) standard for data communication with a computer or other 
instruments. For the technical details see the NMEA FAQ, also at 
<ftp://sundae.triumf.ca/pub/peter/index.html>. There are even some free 
(or shareware) PC and Mac GPS interface software there, some with source 
code to get you started.

The only criteria for your computer and software will be that you'll have 
to finish your processing of a single position fix before you get the next 
one. The position update frequency is usually selectable, so you can 
probably tailor it to the speed of your application.

gw@hplb.hpl.hp.com (Gunther Walther) wrote:
>In the real world, you will not always see enough satellites, and imho,
>a system that goes away and comes back all the time, is
>unlikely to be hard-real time. But then, this depends on whether
>you back GPS somehow up with a different technology.

Nonsense. For well over a year now, GPS has been fully operational, which 
means there are at least four satellites (usually six) in view all the 
time anywhere in the world (thus the Global in GPS). Why else would the 
FAA be planning to use GPS?
--
Mike Craymer
Geodetic Survey of Canada
<mailto:craymer@emr.ca>
<http://www.geod.emr.ca/~craymer/>


