Augmenting Scalable Communication-Based Role Allocation for a Three-Role Task
Open Access
- 31 December 2019
- journal article
- research article
- Published by IBERAMIA: Sociedad Iberoamericana de Inteligencia Artificial in INTELIGENCIA ARTIFICIAL
- Vol. 22 (64), 152-165
- https://doi.org/10.4114/intartif.vol22iss64pp152-165
Abstract
In evolutionary robotics role allocation studies, it is common that the role assumed by each robot is strongly associated with specific local conditions, which may compromise scalability and robustness because of the dependency on those conditions. To increase scalability, communication has been proposed as a means for robots to exchange signals that represent roles. This idea was successfully applied to evolve communication-based role allocation for a two-role task, with one communication channel. However, it was necessary to reward signal differentiation in the fitness function, which is a serious limitation as it does not generalize to tasks where the number of roles is unknown a priori. We show that rewarding signal differentiation is not necessary to evolve communication-based role allocation strategies for the referred two-role task, and we improve reported scalability, while requiring less a priori knowledge. We extend the previous work to a three-role task and we propose and compare two cognitive architectures, to increase the number of communication channels, and several fitness functions to evolve scalable controllers. Our results suggest that communication might be useful to evolve role allocation strategies for increasingly complex tasks.Keywords
This publication has 10 references indexed in Scilit:
- Using Communication for the Evolution of Scalable Role Allocation in Collective RoboticsPublished by Springer Science and Business Media LLC ,2018
- Task Allocation in Evolved Communicating Homogeneous Robots: The Importance of Being DifferentAdvances in Intelligent Systems and Computing, 2016
- The Effect of Fitness Function Design on Performance in Evolutionary RoboticsPublished by Association for Computing Machinery (ACM) ,2015
- Communication based dynamic role allocation in a group of homogeneous robotsNatural Computing, 2014
- Beyond black-box optimization: a review of selective pressures for evolutionary roboticsEvolutionary Intelligence, 2014
- Fitness functions in evolutionary robotics: A survey and analysisRobotics and Autonomous Systems, 2009
- Evolution of Signaling in a Multi-Robot System: Categorization and CommunicationAdaptive Behavior, 2008
- Evolving controllers for a homogeneous system of physical robots: structured cooperation with minimal sensorsPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2003
- Evolving Neural Networks through Augmenting TopologiesEvolutionary Computation, 2002
- Evolutionary robots with on-line self-organization and behavioral fitnessNeural Networks, 2000