Comments on: Java, GPS Receivers, and Geocaching: Vincenty’s Formula

By: rick

rick — Fri, 07 May 2010 12:15:44 +0000

That makes perfect sense – thanks.

rick

By: Mike Gavaghan

Mike Gavaghan — Thu, 06 May 2010 19:41:49 +0000

Hi Rick,

The reason is because the shortest path between two points at the same latitude doesn’t follow the latitudinal line (unless you’re on the equator). If you’re not on the equator, the shortest path will bow out toward the nearest pole.

Try GlobalPosition(45, 100, 0.0) and GlobalPosition(45, 101, 0.0). Because these points are very close, the azimuth is 89.65 (very nearly 90).

Now, try GlobalPosition(45, 100, 0.0) and GlobalPosition(45, -75, 0.0). These two points are on nearly opposite sides. The azimuth is 3.53. In order words, the shortest path takes you right by the north pole.

Make sense?

–Mike

By: rick

rick — Thu, 06 May 2010 01:37:03 +0000

I used the Geodetic Calculator to determine the azimuth given 2 points at the same latitude using the WGS84 ellipsoid:
GlobalPosition(45, 100, 0.0);
GlobalPosition(45, 140, 0.0);

However, the azimuth returned was not 90 degrees as I expected, but about 75 degrees. As I change the longitude, I get different angles for the azimuth. Maybe I don’t understand this correctly, but the answer seems counter-intuitive. Could someone explain?

thanks

rick

By: Michael

Michael — Tue, 16 Feb 2010 20:04:25 +0000

I tried the GeodeticCalculator for use with Nasa’s open source WorldWindJava and after a quick set of tests that converted WorldWind Java’s LatLon, Position and Angle to the correct input required by the GeodeticCalculator the results were very accurate.

By: Jacky

Jacky — Fri, 02 Oct 2009 07:10:10 +0000

This API is awesome. Very useful for me. Can I include this API in my system? Does any license I need follow?

By: Peter

Peter — Thu, 20 Aug 2009 14:38:06 +0000

Can I use this code to add a certain distance to a coordinate? So calculating a coordinate that is e.g. 100miles north of a given coordinate?

By: Paul

Paul — Tue, 18 Nov 2008 14:32:51 +0000

My complete apologies, there is nothing wrong with either your or my code implementation. I’m just a brainless idiot who has spent far too long looking for a big problem when a minus sign should have been obvious. Sorry.

Paul

By: paul

paul — Mon, 17 Nov 2008 17:05:48 +0000

Hi,
Yours is the only c# implementation of Vincenty’s direct formula that I’ve come across on the net and I’ve a little problem that you might be able to help me with. Before I found this implementation I recoded the NGS downloadable fortan into c# myself and while both my and your implementations both work to give the same results, I’m finding them wrong and am wondering if you might know what the problem could be. I’m a PhD student in Ireland, Europe and if I enter the following simple values into either my or your implementation I get the same results to 10 decimal places, these are wrong though. However, if I enter these values into the NGS online calculator I get the correct results – http://www.ngs.noaa.gov/cgi-bin/Inv_Fwd/forward2.prl . Our implementations are producing an incorrect Longitude and I can’t seem to isolate the problem.
Values:
Latitude: 53.387125N
Longitude: 6.58569W
Azimuth: 218.512579
Distance: 59.352571

Any help you might be able to offer would be greatly appreciated.
Thanks in advance.

Paul