I assumed that since the points were very close to each other, you’d be operating over a predefined area. The linear transformation from lat-long to UTM is very fast and uses no trig.
I’m not sure what the accuracy of the Haversine formula really is.
Finally, some GPS units can output UTM directly, which will make your life much easier.
This link has a great description of the conversion to UTM (in Mathematica). http://blog.wolfram.com/2009/04/17/mapping-gps-data/
Good luck.
tronic:
Hi Scott,Thanks for you’re tips. I was unfamiliar with the UTM system, as I havent really looked into GPS before now. I’ve quickly read through the Wikipedia page on UTM and now have a better idea of what it is. So thanks for bringing that to my attention!
I am left wondering whether the technique you describe will offer me any benifit over the Haversine formula, since I have to take the additional step of converting coordinates to UTM. My original goal was to save clock cycles in the process of converting changes in coordinates to distance. And my application does not operate within a predefined region.