An improved shuffled frog leaping algorithm with cognitive behavior

• Published in: 2008 7th World Congress on Intelligent Control and Automation pp. 6197 - 6202
• Main Authors: Xuncai Zhang, Xuemei Hu, Guangzhao Cui, Yanfeng Wang, Ying Niu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2008
• Subjects: Algorithm design and analysis; Cognition; Cognitive behavior; Evolution (biology); Evolutionary computation; Heuristic algorithms; Memetic algorithms; Optimization; Particle swarm optimization; Shuffled frog leaping; Space exploration
• ISBN: 1424421136; 9781424421138
• DOI: 10.1109/WCICA.2008.4592798

Shuffled frog leaping (SFL) is a population based, cooperative search metaphor inspired by natural memetics. Its ability of adapting to dynamic environment makes SFL become one of the most important memetic algorithms. In order to improve the algorithmpsilas stability and the ability to search the global optimum, a novel dasiacognition componentpsila is introduced to enhance the effectiveness of the SFL, namely frog not only adjust its position according to the best individual within the memeplex or the global best of population but also according to thinking of the frog itself. To validate the improved SFL (ISFL) method, numerous simulations were conducted to compare SFL and ISFL using six benchmark problems for continuous and discrete optimization. According to the simulation results, adding the cognitive behavior to SFL significantly enhances the performance of SFL in solving the optimization problems, and the improvements are more evident with the scale of the problem increasing.