On Design of an Effective AI Agent for

Tutaj przychodzi karta pracy dyplomowej; oryginał wstawiamy do wersji dla archiwum PP, w pozostałych kopiach wstawiamy ksero. This thesis focuses on the field of real-time strategy games artificial intelligence. We introduce this area of research by comparing it to the widely known traditional games like chess to show the unique challenges and problems related to this field. To prove that the field is relevant for research, we show the resemblance between the real-life problems and those encountered in the real-time strategy games. Our research focuses on the game StarCraft by Blizzard Entertainment, a representa-tive of the real-time strategy games and one of the best selling computer titles in history. We chose this game primarily because it has recently become a test-bed for testing AI methods and many researchers evaluate their solutions in this game’s environment. We implement a fully functional artificial intelligence for StarCraft. We base our solution on the UAlbertaBot, which we extend and improve to play Terrans. The tw