[gps-talkusers] Re: GPS mathematics
- From: "John Mattioli" <jmattioli@xxxxxxxxxxxxxxx>
- To: <gps-talkusers@xxxxxxxxxxxxx>
- Date: Thu, 20 May 2004 12:41:17 -0400
Greetings,
Interesting you should ask this question. I did some distance
traveling with the BN GPS last week and found some interesting things,
some of which may very well be related to the mathematics of GPS.
First I should say that I don't have an answer to your question.
What I do know is that you can't find the distance in an X y coordinate
system for a latitude and longitude. This is because the earth is not
flat. Hence, if you are at the equator you'll find that longitude lines
are a particular distance apart (I don't know the specific distance, but
if you know the circumference of the earth and you divide accordingly,
you'll figure it out. Anyway, as you travel north from the equator the
longitude lines will get closer and closer together. Eventually, they
all meet at the north pole.
So, in order to compute the distance between two points you need
to account for the curve of the earth. For short distances,
particularly in computations near the equator, any error introduced by
not doing this may be fairly minor. For longer distances, particularly
as your distance from the equator increases, these errors will be
dramatic.
So, as to the interesting thing I saw last week.
I was in a car. We traveled from Arlington Massachusetts to
Stamford Connecticut. On the trip down, the BN GPS measured the
distance at about 200 miles. I felt this was not accurate. I'd always
believed the distance to be more like 180 miles. On the return trip,
taking the same roads, the car clocked the distance at 177 miles. My BN
GPS crashed about three times on that trip so I couldn't clock the trip
again.
At one point though, we were driving east on I-90 (the Mass
pike) when the driver saw a sign saying 26 miles to I-95. From that
sign, to the I-95 ramp, the BN GPS clocked nearly 36 miles. This is a
pretty substantial error. Clearly such a substantial error cannot have
existed throughout the entire 177 mile trip otherwise BN GPS would have
reported a distance that was 68 miles in error. So I'm wondering if BN
GPS is more inaccurate when traveling east/west then north/south because
the earth's curviture is not accurately being accounted for.
Since the Mass Pike is a bit north of Longitude 42N it is nearly
half way to the pole. Is it possible that BN GPS isn't accurately
dealing with this fact?
John
-----Original Message-----
From: gps-talkusers-bounce@xxxxxxxxxxxxx
[mailto:gps-talkusers-bounce@xxxxxxxxxxxxx] On Behalf Of Matthias
Weingart
Sent: Thursday, May 20, 2004 10:47 AM
To: gps-talkusers@xxxxxxxxxxxxx
Subject: [gps-talkusers] GPS mathematics
Hi,
maybe I am asking in the wrong group (this is a group for GPS users, and
not developers?), but I would like to know something of the GPS basics.
In fact I would like to calculate the distance and angle between 2 GPS
coordinates (or position in a x-y diagram, where one coordinate is
(0,0)). (starting from the NMEA-string, result should be meters).
I would be very happy, if somebody here could point me to some good
ressources.
Matthias
Other related posts:
