Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!agate!bionet!apple!rutgers!mailrus!umich!itivax!dhw
From: dhw@itivax.UUCP (David H. West)
Newsgroups: comp.graphics
Subject: Efficient primitives for spherical-surface geometry
Keywords: sphere geometry
Message-ID: <369@itivax.UUCP>
Date: 8 Nov 88 22:58:36 GMT
Organization: Industrial Technology Institute
Lines: 20

Has anyone seen a discussion of how best to perform simple geometric 
operations on a spherical surface?  The kinds of thing I have in
mind are: calculate the intersection of two great-circle segments,
the incenter of a spherical triangle, and whether a point is inside
a spherical polygon.  All I've been able to find is how to
solve spherical triangles, which is probably not the best way to do
anything efficiently - it seems to me that there is probably a way
(e.g. using direction cosines) to consistently avoid most of the trig 
function evaluations. 

Has anyone seen a good way to partition a spherical surface to
support nearest-neighbor search? Subtriangulating a regular
polyhedron seems reasonable, except for generating neighbors of a
partition. (Somebody must have invented this wheel...)

Please email, I'll summarize if there are both interest and answers.

-David West            dhw%iti@umix.cc.umich.edu
		       {uunet,rutgers,ames}!umix!itivax!dhw
CDSL, Industrial Technology Institute, PO Box 1485, 
Ann Arbor, MI 48106