Abstract
Parallel Genetic Algorithms (PGAs), implemented on the APE100/Quadrics SIMD architecture, were applied to automatic design of membership functions and fuzzy rules for robotic control. They run multiple simultaneous searches, differently balancing exploration of the solution space and fine tune of the best solutions available at each generation. Migration spreads the best individuals of each population in local neighborhoods. The approach reduces the time required for fitness evaluation (each population has less individuals, decreases the generations required for acceptable solutions and increases the probability of identifying optimal solutions.
References
P.S.Paolucci R.D'Autilia G.M.Todesco S.Cabasino. The tao language. Report I.N.F.N., March 1995.
V.Maniezzo. Genetic evolution of the topology and weight distribution of neural networks. In IEEE Transactions on Neural Networks, pages 39–53, January 1994.
F.Herrera and M.Lozano. Gradual distributed genetic algorithms. Technical Report 97103, DECSAI, 1997.
G.Mondelli G.Castellano G.Attolico A.Distante. Fuzzy controller design by parallel genetic algorithms. In Control system designs and simulation models, SPIE Aerosense, Orlando, April 1998.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mondelli, G., Castellano, G., Attolico, G., Distante, C. (1998). Parallel genetic evolution of membership functions and rules for a fuzzy controller. In: Sloot, P., Bubak, M., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1998. Lecture Notes in Computer Science, vol 1401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0037234
Download citation
DOI: https://doi.org/10.1007/BFb0037234
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64443-9
Online ISBN: 978-3-540-69783-1
eBook Packages: Springer Book Archive