Evolving Virtual Ecology

Abstract

Within the field of artificial life it has been possible to create numerous virtual models that have allowed the study of the behaviour of living organisms and their interactions within artificially created ecosystems. Whilst the methods employed in this field have been mostly explored by various researchers in their projects, they had not been broadly applied to the entertainment and art fields. This paper focuses on a system (digital toy) which contains artificial life agents. These agents learn to interpret external audio commands and adapt to their environment using evolutionary computation and machine learning.