The decision-making process of computer-con- trolled opponents in video games is called game AI. Adaptive game AI can improve the entertainment value of games by allowing computer-controlled opponents to fix weaknesses automatically in the game AI and to respond to changes in human-play- er tactics. Dynamic scripting is a reinforcement learning approach to adaptive game AI that learns, during gameplay, which game tactics an opponent should select to play effectively. In previous work, the tactics used by dynamic scripting were de- signed manually. We introduce the evolutionary state-based tactics generator (ESTG), which uses an evolutionary algorithm to generate tactics auto- matically. Experimental results show that ESTG im- proves dynamic scripting’s performance in a real- time strategy game. We conclude that high-quality domain knowledge can be automatically generated for strong adaptive game AI opponents. Game de- velopers can benefit from applying ESTG, as it con- siderably reduces the time and effort needed to cre- ate adaptive game AI.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below