Enabling the Sense of Self in a Dual-Arm Robot

AlQallaf, A. and Aragon Camarasa, G. (2021) Enabling the Sense of Self in a Dual-Arm Robot. In: IEEE International Conference on Development and Learning (ICDL 2021), Beijing, China, 23-26 Aug 2021, ISBN 9781728162430 (doi: 10.1109/ICDL49984.2021.9515649)

[img] Text
245284.pdf - Accepted Version



While humans are aware of their body and capabilities, robots are not. To address this, we propose a first step towards a basic, minimal self-awareness in a robot. That is, we propose an experimental methodology to evaluate whether the robot can differentiate itself from the environment, and to test whether artificial self-awareness increases a robot’s self-certainty in an unseen environment. For this, we implemented a simple neural network architecture that enables a dual-arm robot to differentiate its limbs from an environment using visual and proprioception sensory inputs. The proposed experimental approach allows us to evaluate whether the robot can differentiate itself from the environment. Our results indicate that a robot can distinguish itself with an accuracy of 88.7% on average in different environmental settings and under confounding input signals.

Item Type:Conference Proceedings
Additional Information:Ali AlQallaf thanks the Kuwait Institute for Scientific Research. We also thank the support of NVIDIA Corporation for the donation of the Titan Xp GPU used in this research.
Glasgow Author(s) Enlighten ID:AlQallaf, Ali and Aragon Camarasa, Dr Gerardo
Authors: AlQallaf, A., and Aragon Camarasa, G.
College/School:College of Science and Engineering
College of Science and Engineering > School of Computing Science
Published Online:20 August 2021
Copyright Holders:Copyright © 2021 IEEE
Publisher Policy:Reproduced in accordance with the publisher copyright policy
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record