mike regish writes:

There was a story a while back (don't know if it's true or not, but sounded
legit) that some guy was demonstrating his latest, greatest GPS by using it
to taxi into his hangar. It wasn't quite that accurate and the repair bill
wasn't cheap.

One problem with GPS is that accuracy can be rapidly and significantly
degraded by the presence of buildings or mountains or other obstacles
that reflect or block signals. This is why GPS isn't likely to be
very accurate in the streets of Manhattan. The system itself provides
good accuracy, but in order to obtain that accuracy, you have to be
able to receive the signals without interference. On the ocean, in
the countryside, or in the open sky, you can receive signals very well
indeed, but once you are on the ground, the situation changes.

Another problem, not actually part of GPS per se, is moving maps.
Your GPS position may be accurate, but that doesn't guarantee that the
map is accurate. If the mountain on the map is in the wrong place in
relation to its real-world position, having high accuracy from GPS
won't help you. Very often map errors are more of a problem than
errors in the GPS itself.

Note that WAAS and LAAS will _not_ compensate for either of the above
types of error. Differential GPS systems like this work best when you
are at exactly the spot used as a reference for the corrections. If
you are anywhere else, the corrections may not be right for your
position. The further away you are from the surveyed reference
position used to generate the corrections, the more likely it is that
your position will be incorrect.

Some of these systems also correct for atmospheric and other effects,
but here again, the corrections are most useful when you are in the
exact position for which they are generated. If the reference point
is in Cheyenne and you are in Denver, the corrections may be well off
the mark.

--
Transpose mxsmanic and gmail to reach me by e-mail.