Title: Improving Search Capabilities of Reduced Real Swarm of Robots by Virtual Particle Swarm
Authors: Otávio Santos, Sergio Oliveira
Abstract: This paper introduces a new approach to improve the search capabilities of a reduced real swarm of robots based on the classical Particle Swarm Optimization (PSO) by employing a larger virtual swarm. The central idea is to add more particles in the virtual swarm using them to aid the real search. This approach aims to reduce costs once that increasing the quantity of robots in the real swarm implies higher costs related to robots acquisition and its maintenance. To work around this issue, we insert particles only in the virtual swarm. These particles are able to interact with the real robots and help them. The description functions of virtual and real environment are slightly different to represent dynamic changes on real data environment. Preliminary results indicates that the virtual particles aid the real ones improving its search mechanisms. The virtual aided the few number of real robots obtained equivalent solutions than the larger number of robots in the same search space.
Key-words: Swarm Intelligence; Particle Swarm Optimization; Swarm of Robots
DOI code: 10.21528/CBIC2019-83
PDF file: CBIC2019-83.pdf
BibTeX file: CBIC2019-83.bib