UCT CS Research Document Archive

The Benefits of Adaptive Behavior and Morphology for Cooperation in Robot Teams

Hewland, Jamie and Geoff Nitschke (2015) The Benefits of Adaptive Behavior and Morphology for Cooperation in Robot Teams. In Proceedings IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2015), pages 1047-1054, Cape Town, South Africa.

Full text available as:
PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

This is a study on the role of morphological (sensor configuration) and behavioral (control system) adaptation in simulated robot teams that must accomplish cooperative tasks. The research objective was to elucidate the necessary features and computational mechanics of a method that automates the behavior-morphology design of robot teams that must accomplish cooperative tasks (tasks that cannot be optimally solved by individual robots). Results indicate that automating behavior-morphology design is beneficial as task complexity increases, compared to evolving behaviors in fixed morphology teams. However, increased task complexity does not necessarily equate to the evolution of increased morphological complexity in teams.

EPrint Type:Conference Paper
Subjects:I Computing Methodologies: I.2 ARTIFICIAL INTELLIGENCE
ID Code:1195
Deposited By:Nitschke, Geoff
Deposited On:23 November 2017