|
|
Subject: Re: GPS Accuracy Now.
From: Lu Abel (lu.abel@XXX.XXX)
Date: Fri May 05 2000 - 00:32:37 EDT
Sorry, I have to disagree with some of Bill Trayfor's comments (see below).
While I agree there's no idiot like a computer-assisted idiot (and I've
been one!), erroneous use of GPS is the navigator's fault, not GPS's fault!
At 07:17 PM 5/4/00 +0100, Bill Trayfor wrote:
>It's indeed a wonderful thing that DoD has been instructed to "turn off"
>selective availability, and that civilians now will have access to the same
>GPS signals the military has enjoyed. This action is long overdue.
>
>However, one predictable consequence of enhanced GPS system "accuracy" is
>that some of our fellow mariners, and maybe even an aviator or two, are
>gonna have GPS-assisted groundings! This will happen because word will get
>around that GPS is now an extremely accurate positioning system and some
>people will use it blindly. There are several reasons to think twice about
>how GPS is used:
So far, so good. Know the limitations of your navigation system!
>1. Even if GPS were 100% accurate 100% of the time, there are a lot of
>charts in common use which are referenced to a different datum than WGS-84;
>differences can be LARGE, e.g., some charts of the Virgin Islands have
>islands "misplaced" by 1/4 mile or more. I have personal knowledge of a
>custom 70' sloop which was lost on Neckar Island after successfully
>completing a round-the-world voyage with a professional crew; this yacht
>ran aground just 1/4 mile north of the indicated GPS position and, guess
>what....the datum error at that location is 1/4 mile North/South!
Most GPS systems provide for translation to other geoids. Again, only
fools do 0/0 approaches to coral reefs without understanding the accuracy
of the chart data and what datum it is drawn on. Ditto for using L/Lo's
from charts drawn in the 1800s (which many of the Pacific Island charts
allegedly are).
>2. GPS isn't available 100% of the time. If system availability and
>accuracy are, say, 99% reliable that means that during the year it will not
>be available or will not be accurate during some 5,256 minutes or 87.6
hours!!
Here I have to disagree. The GPS system specs call for at least 99.95%
uptime (it may be even more, I can't recall of the top of my head), which
increases reliability a hundred-fold. More important, one has to remember
these are statistical estimates. There's no guarantee that the system WILL
be down for x number of minutes a year (nor that it will not be down for
way more than that, either). It just says that the system contractors had
to prove, via statistical analyses, that the system will have that
probability of being up at any moment of time.
Looking beyond statistics, it's highly unlikely that more than one
satellite will fail and modern receivers can give pretty good info even
with one bad satellite. An informed analysis of GPS reliability would show
that receiver failure is a much, much more likely cause of being able to
get navigational info than failure of all the other parts of the system put
together.
>3. There are numerous other sources of GPS error -- both technical and
>user related -- which can render GPS readings misleading for the mariner.
For example?? I suspect one can come up with a similar list for almost any
other soure of navigational data. Far, far more collisions and groundings
have, for example, occurred when radar was being used as the primary source
of navigational data than GPS, yet no one blames these on the use of radar.
Ditto for celestial. Could there be some strange masochism among
navigators which says "the harder the information is to get and interpret,
the better it must be?" I'll answer my own question by noting that many
traditional methods (using bearings is an excellent example) let a good
navigator see the information coming together and detect problems with it
(mis-recorded bearings, poor crossing angles) whilst the GPS box just gives
an answer with little ability on the part of the navigator to determine its
quality.
>For most practical navigation uses, I believe the turning off of SA
>will/should have little impact (except, perhaps, GPS-indicated speed will
>be more accurate).
When I lived in New England I used Loran to find harbor entrances in some
pretty thick fogs. I had pretty strong confidence it could put me where I
wanted to be +/- 100 ft. I never would have done that with GPS as long as
SA was turned on. Now that it's turned off, I personally would use GPS
with similar confidence. Others are free to disagree.
>I will continue to believe that GPS readings should be
>treated as just one source of positioning information to be compared with
>all other available sources at all times. Yes, GPS is the best thing since
>sliced bread. But, unless you use it intelligently it can cut your finger
off!
Total, absolute agreement. A prudent navigator uses ALL possible sources
of navigational information and depends on NO single source...
Lu Abel
|