I'm creating a path-finder for navmesh. I'm wondering which method should I use after the A* has been completed, funneling or string pulling? What are the advantages/disadvantages of using them on a triangle-based navmesh?
\$\begingroup\$
\$\endgroup\$
3
-
\$\begingroup\$ I might be completely wrong here since I looked both these algorithms up the other day, but aren't they essentially one and the same thing? \$\endgroup\$Ray Dey– Ray Dey2011-01-12 16:25:54 +00:00Commented Jan 12, 2011 at 16:25
-
\$\begingroup\$ @Ray Dey, both can be used to accomplish path refining in a mesh after the path-find has been executed. \$\endgroup\$Shawn Mclean– Shawn Mclean2011-01-12 16:29:07 +00:00Commented Jan 12, 2011 at 16:29
-
\$\begingroup\$ Sorry, what I meant to say was WHEN I looked both these algorithms up, they seemed to point to the same resources (on Google anyway) \$\endgroup\$Ray Dey– Ray Dey2011-01-12 17:12:53 +00:00Commented Jan 12, 2011 at 17:12
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
If you're stopping condition for the funneling algorithm is the invalidity of the cone I think you'll get pretty much the same end result than with string pulling.
But using another stopping condition you can produce more various behaviors by setting a particular aperture threshold and/or by tweaking how the local target/velocity is chosen in the final cone.
