Animal Flocks as Natural and Dynamic Spatial Clues in Adventure Video-games

摘要

This paper proposes the use of animal flocks to naturally provide spatial clues in adventure video-games. This approach complements the so far proposed mechanisms to support navigation in virtual worlds, addressing particular conditions that the previous approaches do not cover. The proposed model is natural and peripheral. Moreover, it does notlimit the player freedom, it is adequate for dynamic and large environments, and it does not create false expectations. Our model is composed of two complementary algorithms. The first one is a flocking algorithm (based on an extension of Reynolds' boids algorithm) which creates an emergent group behaviour fromsimple individual rules. The second one is a control algorithm which manages the list of places, decides the particular place tosuggest at each time and parameterises the flocking behaviour.