Geodetic distance on WGS84 earth ellipsoid

Rapidly computes the geodetic distance between coordinates on the ellipsoidal earth.

Description

In 1975, Vincenty published a rapidly converging algorithm for computing the distance between points on an ellipsoidal earth. The algorithm is precise to within a few millimeters. Since then, his algorithm has since seen significant implementation in geodesy and engineering. After adjusting the algorithm to converge in all cases (the original suffers from convergence failure in a few outlying cases), I have provided it here in MATLAB form. The function itself does not require the Mapping Toolbox, but I have included a (commented-out) code section following the body of the function, which you can use if you have the Mapping Toolbox to compare the accuracy of this algorithm to spherical earth distances. Note that in that toolbox, the Mathworks uses a fast but somewhat less precise method for computing geodetic distances on an ellipsoid. This function will provide rapid, extremely precise results. Please see code comments for references.

• MATLAB release: MATLAB 6.5 (R13)

[Fahad Al Mahmood]

I have verified the output of this function with my Garmin MapSource software and I got the answers to be close enough. I found this function very useful.

[Ashish Misra]

I have verified the output of this function with my Garmin MapSource software and I got the answers to be close enough. I found this function very useful

[Gabor Vali]

Very helpful. Much more transparent than the equaazim.m routine.

[Gerald Dalley]

Very nice. It would be convenient to allow the input coordinates to be vectors/matrices.

[Michael Kleder]

See the following URL (Matlab File Exchange item #8607) for vectorized inputs and forward / reverse azimuth outputs, in addition to the Vincenty distance:
http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=8607&objectType=FILE

[Tro Han]

Returns imaginary result for lon1=0 lon2=0.
Fix by preventing asin of # > 1 in alpha equation i.e.
alpha = asin(max(min(cos(U1)*cos(U2)*sin(lambda)/sin(sigma),1),-1));

[The Author]

Thank you for your helpful comment. The problem you reported is actually addressed in the function which has superceded this one. The new function is vectorized, computes azimuth as well as distance, and features improved handling of anomalous cases. Please see
http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=8607&objectType=FILE

[Frederic B.]

Thanks!
The file has been really useful!

[Chuck B]

Works great.

[Michael Draugelis]

Works like a charm. I love it. Thanks for sharing!

[xzn dasds]

 

[Goran Nystrom]

 

[catur anindhito]

thanx for sharing. That is very good