Approximation Models of Combat in StarCraft 2

Real-time strategy (RTS) games make heavy use of artificial intelligence (AI), especially in the design of computerized opponents. Because of the computational complexity involved in managing all aspects of these games, many AI opponents are designed to optimize only a few areas of playing style. In games like StarCraft 2, a very popular and recently released RTS, most AI strategies revolve around economic and building efficiency: AI opponents try to gather and spend all resources as quickly and effectively as possible while ensuring that no units are idle. The aim of this work was to help address the need for AI combat strategies that are not computationally intensive. Our goal was to produce a computationally efficient model that is accurate at predicting the results of complex battles between diverse armies, including which army will win and how many units will remain. Our results suggest it may be possible to develop a relatively simple approximation model of combat that can accurately predict many battles that do not involve micromanagement. Future designs of AI opponents may be able to incorporate such an approximation model into their decision and planning systems to provide a challenge that is strategically balanced across all aspects of play. Background and Motivation StarCraft 2 is a real-time strategy (RTS) computer game created by Blizzard Entertainment. This game is a sequel to the original StarCraft and, like its predecessor, has proven very popular. Released in July 2010, StarCraft 2 sold over 3 million copies worldwide after only one month on the market, making it one of the fastest-selling RTS games of all time.1 Many professional computer gaming leagues that previously ran StarCraft tournaments have recently converted to running StarCraft 2 tournaments, including the GOMTV Global StarCraft 2 League (GSL), which has awarded approximately $2.5 million USD in prize money since its inception in 2010.2 The rising popularity of "e-sports" and the potential of winning substantial prize money in tournaments have contributed to an increased interest in strategies and technical analyses of many competitive computer games, especially StarCraft 2.3 StarCraft 2, like many other titles in the RTS genre, is a game for two or more players. Each player chooses from one of three different races (Protoss, Zerg, or Human) and builds an army of units to defeat his or her enemies on a specific map: a 2-dimensional grid of finite size with