Learning, evolution and adaptation in racing games

Modern racing games offer a realistic driving experience and a vivid game environment. Accordingly, developing this type of games involves several challenges and requires a large amount of game contents. Computational intelligence represents a promising technology to deal effectively with such challenges and, at the same time, to reduce the cost of the development process. In this paper, we provide an overview of the most relevant applications of computational intelligence methods in the domain of racing games. In particular, we show that computational intelligence can be successfully applied (i) to develop highly competitive non-player characters,(ii) to design advanced racing behaviors such as overtaking maneuvers, and (iii) to automatically generate tracks and racing scenarios.