|
Does anyone know of specific algorithms to calculate smooth natural looking path distances between two points in a non-convex set? For example:
I would want a 'smooth' line distance from point X to Y as opposed to a shortest path distance with a sharp corner from X to Y. |
|
Just hand waving here, but you could sample points from your non-convex region, form a graph (each sample is a vertex), and run a shortest path algorithm like dijkstras. I think there are other methods in AI like rapidly exploring random trees that could help out with this type of search problem too. I was thinking there might be an algorithm that already exists along those lines. To add to the hand waving, the smoothness parameter might be some sort of gaussian prior between neighboring points. But since this is a relatively small part of a larger project and seems like it would be a fairly common problem in a lot of fields, I was wondering if there's something readily available.
(Dec 09 '11 at 12:18)
crdrn
|
