Newsgroups: comp.robotics
Path: brunix!sgiblab!swrinde!cs.utexas.edu!convex!news.duke.edu!godot.cc.duq.edu!nntp.club.cc.cmu.edu!honeydew.srv.cs.cmu.edu!localhost!nivek
From: nivek@cmu.edu (Kevin Dowling)
Subject: GPS info
Message-ID: <NIVEK.94Apr25235137@scythe.cmu.edu>
Reply-To: nivek@cmu.edu (Kevin Dowling)
Sender: news@cs.cmu.edu (Usenet News System)
Nntp-Posting-Host: scythe.frc.ri.cmu.edu
Organization: Robotics Institute, Carnegie Mellon
Date: Tue, 26 Apr 1994 03:51:37 GMT
Lines: 63



There is a newsgroup devoted to GPS topics: sci.geo.satellite-nav
It is fairly recent, however there is a moderated mailing list that is
posted to that group. 

In any case - to answer a few general questions that have appeared here:

GPS techniques allow accuracies to mm range using kinematic
techniques. Basically you need time to give you greater accuracies.

Differential GPS utilizes fixed stations to broadcast the errors in
the position to nearby (within several 10's of km) GPS receivers. Thus the
pseudorandom noise can be eliminated through this technique. The Coast
Guard is established DGPS stations along the coasts and private
companies are establishing DGPS stations utlizing local FM station
sidebands to transmit the error codes. This will allow those
subscribing to such services to receive these codes.

The arguments regarding the 'crippling' of the transmitted positions
are moot. The DoD is loathe to give everyone cm accuracies that can be
misused by others. GPS is now under a civilian agency (NOAA, I
believe) and this may change. Doesn't matter, you can get the
accuracies you need now.

Carrier-phase reconstruction ignores most of the information on the
P-code signals and looks only at the carrier wave. After having
established the wave count and wave fraction to the satellite you can
monitor the carrier and, with multiple antenna, give position and
orientation to sub-cm accuracies in real-time. These systems aren't
cheap and are not widely available but they are getting there.
There are some newer C-P techniques that do not require the long
reconstruction time either.

Indoor GPS - Psuedolite (Psuedo Satellites) have been around since the
early 80's for testing purposes. Recently, these have been combined
with the C-P techniques to allow rapid and accurate 6DOF measurements
for indoor environments. There are some exciting developments in this
area but little in the commercial arena - yet...

Cost - New units from Motorola, Trimble, Magellan and others are in
the below $300 OEM costs in small quantity. Check the Moto Oracle and
Sixgun units - I think they are below $200 now.

In our work we've seen the costs for our first GPS robot systems in
the mid-80's go from $100K to a few $K. The CMU Fastnav system utlized
GPS combined with INS and odometry for path tracking. Obstacle
avoidance via laser scanner was also integrated and the system
eventually became a commercial product for Caterpillar for guiding
large mining vehicles (Cat 785's) A few of these complete systems have
been sold for autonomous haulage in surface mines. GPS has a million
applications not yet thought of...

					nivek

aka:    Kevin Dowling         Project Scientist
tel:    412.268.8830          Carnegie Mellon University
fax:    412.682.1793          The Robotics Institute
net:    <nivek@cmu.edu>       5000 Forbes Avenue
                              Pittsburgh, PA 15213



