Authors:
Adham Atyabi
1
and
David M. W. Powers
2
Affiliations:
1
Flinders University, Australia
;
2
Flinders University and Beijing University of Technology, Australia
Keyword(s):
Swarm Robotics, Particle Swarm Optimization, Cooperative Learning, Transfer Learning, Knowledge Transfer.
Related
Ontology
Subjects/Areas/Topics:
Evolutionary Computation and Control
;
Hybrid Learning Systems
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Optimization Algorithms
;
Software Agents for Intelligent Control Systems
Abstract:
The study investigates the effectiveness of 2 variations of Particle Swarm Optimization (PSO) called Area Extended PSO (AEPSO) and Cooperative AEPSO (CAEPSO) in simulated robotic environments affected by a combinatorial noise. Knowledge Transfer, the use of the expertise and knowledge gained from previous
experiments, can improve the robots decision making and reduce the number of wrong decisions in such uncertain environments. This study investigates the impact of transfer learning on robots’ performance in such hostile environment. The results highlight the feasibility of CAEPSO to be used as the controller and decision maker of a swarm of robots in the simulated uncertain environment when gained expertise from past training is transferred to the robots in the testing phase.