Build order optimization in StarCraft
作者： Michael Buro , David Churchill
DOI:
关键词: Pathfinding 、 Heuristics 、 Artificial intelligence 、 Computer science 、 Speedup 、 Scheduling (computing)
摘要: In recent years, real-time strategy (RTS) games have gained interest in the AI research community for their multitude of challenging subproblems — such as collaborative pathfinding, effective resource allocation and unit targeting, to name a few. this paper we consider build order problem RTS which need find concurrent action sequences that, constrained by dependencies availability, create certain number units structures shortest possible time span. We present abstractions heuristics that speed up search approximative solutions considerably game StarCraft, show efficacy our method comparing its performance with professional StarCraft players.
skatgame.net参考文章(6)
Alan Fern, Hei Chan, Soumya Ray, Chris Ventura, Nick Wilson, Online planning for resource production in real-time strategy games international conference on automated planning and scheduling. pp. 65- 72 ,(2007)
Martin Stolle, Doina Precup, Learning Options in Reinforcement Learning symposium on abstraction, reformulation and approximation. pp. 212- 223 ,(2002) , 10.1007/3-540-45622-8_16
Glenn A. Iba, A Heuristic Approach to the Discovery of Macro-Operators Machine Learning. ,vol. 3, pp. 285- 317 ,(1989) , 10.1023/A:1022693717366
Augusto A. B. Branquinho, Carlos R. Lopes, Planning for resource production in real-time strategy games based on partial order planning, search and learning systems, man and cybernetics. pp. 4205- 4211 ,(2010) , 10.1109/ICSMC.2010.5642498
Michael Buro, Alexander Kovarsky, Concurrent action execution with shared fluents national conference on artificial intelligence. pp. 950- 955 ,(2007)
Alan Fern, Hei Chan, Soumya Ray, Chris Ventura, Nick Wilson, Extending Online Planning for Resource Production in Real-Time Strategy Games with Search ,(2007)