In this paper, we investigate the application of evolutionary\udcomputation to the automatic generation of tracks for\udhigh-end racing games. The idea underlying our approach is\udthat diversity is a major source of challenge/interest for racing\udtracks and, eventually, might play a key role in contributing to the\udplayer’s fun. In particular, we focus on the diversity of a track in\udterms of its shape (i.e., the number and the assortment of turns\udand straights it contains), and in terms of driving experience it\udprovides (i.e., the range of speeds achievable while driving on the\udtrack). We define two fitness functions that capture our idea of\uddiversity as the entropy of the track’s curvature and speed profiles.\udWe apply both a single-objective and a multiobjective real-coded\udgenetic algorithm (GA) to evolve tracks involving both a wide\udvariety of turns and straights and also a large range of driving\udspeeds. The results we report show that both single-objective and\udmultiobjective approaches can successfully evolve tracks with a\udhigh degree of diversity both in terms of shape and achievable\udspeeds.
展开▼
