Recent Research on AI in Games

Abstract

Games tend to have the properties of vast state space and high complexity, making them excellent benchmarks for evaluating various techniques, including AI ones. Techniques utilized in games capable of making them more attractive, immersive, smarter etc. can all be considered to be certain forms of game AI. Considering there are few reviews on the more recent work in the game AI field from the perspective of essential applications, in this paper, we make a systematic review of typical research from 2018 on three application fields of game AI: believable agents in non-player characters research, game level generation in procedural content generation, and player profiling in player modeling. We also provide a timeline of game AI history to give the readers a clearer picture of the game AI field. Moreover, general game AI and hybrid intelligence for games are discussed.