We investigate two methods of using limited search to improve admissible heuristics for planning, similar to pattern databases and pattern searches. Since one method involves searching in an AND/OR graph, we also develop a new algorithm for searching such graphs. Results indicate that these methods are cost-effective for some kinds of hard planning problems.
Keywords: Planning, Heuristic search
Citation: Patrik Haslum: Improving Heuristics through Search. In R.López de Mántaras and L.Saitta (eds.): ECAI2004, Proceedings of the 16th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2004, pp.1031-1032.