Real-time challenge balance in an RTS game using rtNEAT

Abstract

This paper explores using the NEAT and rtNEAT neuro-evolution methodologies to generate intelligent opponents in real-time strategy (RTS) games. The main objective is to adapt the challenge generated by the game opponents to match the skill of a player in real-time, ultimately leading to a higher entertainment value perceived by a human player of the game. Results indicate the effectiveness of NEAT and rtNEAT but demonstrate their limitations for use in real-time strategy games.