DocumentCode :
797981
Title :
A best-first search algorithm guided by a set-valued heuristic
Author :
Lark, James W., III ; White, Chelsea C., III ; Syverson, Kirsten
Author_Institution :
Dept. of Syst. Eng., Virginia Univ., Charlottesville, VA, USA
Volume :
25
Issue :
7
fYear :
1995
fDate :
7/1/1995 12:00:00 AM
Firstpage :
1097
Lastpage :
1101
Abstract :
Presents an algorithm, called AG, for finding the least-cost path from start node to goal node set in an OR-graph, where arc costs are scalar-valued and the cost of each path is the sum of the concomitant arc costs. Search is guided by a set, H, of real-valued functions on the node set. Such a heuristic set can be a useful representation of knowledge acquired from human knowledge sources. If H={h: l⩽h} for given function l, then AG essentially becomes A*. If H is bounded, then successors of the newly expanded node may not be placed on OPEN, the set of all generated nodes that are candidates for expansion. The authors address issues of termination, completeness, admissibility, and heuristic comparison. The authors then consider a less ambitious, and less numerically taxing, objective: find the first arc of a least-cost path from start node to goal node set. This objective is achieved by the algorithm 1AG, a modification of AG
Keywords :
artificial intelligence; graph theory; search problems; set theory; 1AG algorithm; A* algorithm; AG algorithm; OR-graph; admissibility; best-first search algorithm; completeness; concomitant arc costs; goal node; heuristic comparison; human knowledge sources; least-cost path; real-valued functions; scalar-valued costs; set-valued heuristic; start node; termination; Argon; Costs; Heuristic algorithms; Humans; Industrial relations; Systems engineering and theory; Terminology; US Department of Transportation;
fLanguage :
English
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9472
Type :
jour
DOI :
10.1109/21.391289
Filename :
391289
Link To Document :
بازگشت