Abstract
We demonstrate how Dijkstra's algorithm for shortest path queries can be accelerated by using precomputed shortest path distances. Our approach allows a completely flexible tradeoff between query time and space consumption for precomputed distances. In particular, sublinear space is sufficient to give the search a strong "sense of direction". We evaluate our approach experimentally using large, real-world road networks. © Springer-Verlag Berlin Heidelberg 2006.
Cite
CITATION STYLE
Maue, J., Sanders, P., & Matijevic, D. (2006). Goal directed shortest path queries using precomputed cluster distances. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4007 LNCS, pp. 316–327). Springer Verlag. https://doi.org/10.1007/11764298_29
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
