Authors
Hao Pan
Publication date
2020
Conference
AIIDE Workshops
Description
We deal with the problem of inferring the opponent army size and composition in Real-Time Strategy (RTS) games with partial observability. Using StarCraft R: Brood War R, we propose a methodology to dynamically estimate the number of production facilities the opponent has at any given time in a game, then in turn reason on how many army units the opponent will produce in the near future and of which type. We conduct case studies to demonstrate the effectiveness of the proposed methodology and compare the results with those from mainstream approaches.
Scholar articles