Real-Time Heuristic Search with a Priority Queue
- Chris Rayner, Department of Computing Science, University of Alberta
- Katherine Davison
- Vadim Bulitko
- Kenneth Anderson
- Jieshan Lu, University of Alberta
Learning real-time search, which interleaves planning and acting, allows agents to learn from multiple trials and respond quickly. Such algorithms require no prior knowledge of the environment and can be deployed without pre-processing. We introduce Prioritized-LRTA* (P-LRTA*), a learning real-time search algorithm based on Prioritized Sweeping. P-LRTA* focuses learning on important areas of the search space, where the importance of a state is determined by the magnitude of the updates made to neighboring states. Empirical tests on path-planning in commercial game maps show a substantial learning speed-up over state-of-the-art real-time search algorithms.
Citation
C. Rayner, K. Davison, V. Bulitko, K. Anderson, J. Lu. "Real-Time Heuristic Search with a Priority Queue". International Joint Conference on Artificial Intelligence (IJCAI), Hyderabad, India, pp 2372 - 2377, January 2007.| Keywords: | learning, heuristic search, real-time, learning real-time search, prioritized, path-planning, path-finding, video games |
| Category: | In Conference |
| Web Links: | First author's research page |
| IRCL publication db |
BibTeX
@incollection{Rayner+al:IJCAI07,
author = {Chris Rayner and Katherine Davison and Vadim Bulitko and Kenneth
Anderson and Jieshan Lu},
title = {Real-Time Heuristic Search with a Priority Queue},
Pages = {2372 - 2377},
booktitle = {International Joint Conference on Artificial Intelligence
(IJCAI)},
year = 2007,
}Last Updated: August 17, 2009Submitted by Chris Rayner
