BDDRPA: An Efficient BDD-Based Incremental Heuristic Search Algorithm for Replanning
Author(s)
Yue, Weiya
Xu, Yanyan
Su, Kaile
Griffith University Author(s)
Year published
2006
Metadata
Show full item recordAbstract
We introduce a new algorithm, BDDRPA*, which is an efficient BDD-based incremental heuristic search algorithm for replanning. BDDRPA* combines the incremental heuristic search with BDD-based search to efficiently solve replanning search problems in artificial intelligence. We do a lot of experiments and our experiment evaluation proves BDDRPA* to be a powerful incremental search algorithm. BDDRPA* outperforms breadth-first search by several orders of magnitude for huge size search problems. When the changes to the search problems are small, BDDRPA* needs less runtime by reusing previous information, and even when the changes ...
View more >We introduce a new algorithm, BDDRPA*, which is an efficient BDD-based incremental heuristic search algorithm for replanning. BDDRPA* combines the incremental heuristic search with BDD-based search to efficiently solve replanning search problems in artificial intelligence. We do a lot of experiments and our experiment evaluation proves BDDRPA* to be a powerful incremental search algorithm. BDDRPA* outperforms breadth-first search by several orders of magnitude for huge size search problems. When the changes to the search problems are small, BDDRPA* needs less runtime by reusing previous information, and even when the changes reach to 20 percent of the size of the problems, BDDRPA* still works more efficiently.
View less >
View more >We introduce a new algorithm, BDDRPA*, which is an efficient BDD-based incremental heuristic search algorithm for replanning. BDDRPA* combines the incremental heuristic search with BDD-based search to efficiently solve replanning search problems in artificial intelligence. We do a lot of experiments and our experiment evaluation proves BDDRPA* to be a powerful incremental search algorithm. BDDRPA* outperforms breadth-first search by several orders of magnitude for huge size search problems. When the changes to the search problems are small, BDDRPA* needs less runtime by reusing previous information, and even when the changes reach to 20 percent of the size of the problems, BDDRPA* still works more efficiently.
View less >
Conference Title
AI 2006: Advances in Artificial Intelligence
Copyright Statement
© 2006 Springer : Reproduced in accordance with the copyright policy of the publisher : The original publication will be available at SpringerLink (use hypertext links)