Abstract—In this paper, chaos based a new arithmetic crossover operator on the genetic algorithm has been proposed. The most frequent issue for the optimization algorithms is stuck on problem's defined local minimum points and it needs excessive amount of time to escape from them; therefore, these algorithms may never find global minimum points. To avoid and escape from local minimums, a chaotic arithmetic crossover operator has been employed on a genetic algorithm. Unimodal and multi modal benchmark functions have been used for comparing and test procedures. The genetic algorithm with this new arithmetic crossover operator has yielded better results than original arithmetic crossover operator does. With this new chaos based arithmetic crossover operator diversity and uniqueness of the genetic algorithm’s late population are increased. Therefore, even in the late stages of the optimization process, the genetic algorithm tried to search the entire search space and improved the best solution.
Index Terms—genetic algorithm, chaos, crossover operator
Cite: Hüseyin Demirci, Ahmet Turan Özcerit, Hüseyin Ekiz, and Akif Kutlu, "Knowledge-Based System Framework for Training Long Jump Athletes Using Action Recognition," Vol. 6, No. 4, pp. 217-220, November, 2015. doi: 10.12720/jait.6.4.217-220
Copyright © 2013-2020. JAIT. All Rights Reserved
This work is licensed under the Creative Commons Attribution License (CC BY-NC-ND 4.0)