From: sneftel@gmail.com   
      
   I have a bidirectional 2D euclidean graph, which I'd like to find many   
   shortest paths on. Is there a way to use A*, or a variant of A*, or a   
   completely different algorithm like LPA*, to perturb the source and   
   goal nodes slightly without having to fully re-plan each time? That   
   is, given that I've already gone to the trouble of finding a shortest   
   path from position (15,35) to position (80, 5), a way to reuse most of   
   the computation to find a shortest path from (10,35) to (80,5), or a   
   path from (15,35) to (85,5)? This would be repeated over time, with   
   the source and goal tending to wander around.   
      
   If only the goal is perturbed, one approach would be to reuse the open-   
   list, recalculating all nodes on it for the new heuristic. I'm not   
   sure if that's the most efficient approach, though, and I have no idea   
   how to move the source without completely recalculating.   
      
   Ben   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)