Mixing Population-Based Metaheuristics: An Approach Based on a Distributed-Queue for the Optimal Design of Fuzzy Controllers

Mancilla A., Castillo O., Garcia Valdez M.

4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Turkey, 19 - 21 July 2022, vol.504, pp.839-846 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 504
  • Doi Number: 10.1007/978-3-031-09173-5_96
  • City: Bornova
  • Country: Turkey
  • Page Numbers: pp.839-846
  • Keywords: Asynchronous algorithms, Fuzzy control, Population-based metaheuristics, SYSTEMS
  • Istanbul Technical University Affiliated: No


In this work, we present a distributed platform to execute multi-population metaheuristics. As proof of concept, we present an implementation using two metaheuristics: Genetic Algorithms, and Particle Swarm Optimization. We execute these multi-population algorithms asynchronously using a queue-based architecture. We optimize the parameters defining the membership functions of a rear-wheel fuzzy controller. We compare the results with a non-distributed sequential alternative and show the benefits of mixing the algorithms' populations and integrating a migration process between them.