Evolutionary Robotics: a new scientific tool for studying cognition
journal contributionposted on 2023-06-07, 19:06 authored by Inman HarveyInman Harvey, Ezequiel Di Paolo, Rachel Wood, Matt Quinn, Elio Tuci
We survey developments in Artificial Neural Networks, in Behaviour-based Robotics and Evolutionary Algorithms that set the stage for Evolutionary Robotics in the 1990s. We examine the motivations for using ER as a scientific tool for studying minimal models of cognition, with the advantage of being capable of generating integrated sensorimotor systems with minimal (or controllable) prejudices. These systems must act as a whole in close coupling with their environments which is an essential aspect of real cognition that is often either bypassed or modelled poorly in other disciplines. We demonstrate with three example studies: homeostasis under visual inversion; the origins of learning; and the ontogenetic acquisition of entrainment.
- Published version
Department affiliated with
- Informatics Publications
NotesOriginality: The field of Evolutionary Robotics was pioneered here at Sussex in the early 1990s, and now is an established field. This brings the field up to date summarising the distinctive approaches coming out of our Sussex group. Rigour: It critically reviews work done in both software and hardware, and explains the common themes. Significance: Review of a major body of work that now has much influence around the world. Impact: Eg the Indiana Univ Cognitive Sciences Program is now building up the US version of the Sussex group. Citations: Google Scholar 14 citations Web of Knowledge 8 citations
Full text available