A study of fuzzy membership functions for dependence decision-making in security robot system. Issue 1 (January 2017)

• Record Type: Journal Article
• Authors: Kim, Suntae; Lee, Malrey; Lee, Jongho
• Is Part Of: Neural computing & applications. Volume 28:Issue 1(2017)
• Page Start: 155
• Page End: 164
• Publication Date: 2017-01
• Subjects: Robots -- Fuzzy decision-making -- Fuzzy membership function -- Artificial intelligence; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-2044-3
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract This paper proposes comparison of fuzzy membership functions for decision-making in security robot system. Robot's decision-making speed depends on fuzzy membership functions. The results of the study indicate that sigmoidal membership function coincide the best, two-sided Gaussian, Gaussian, generalized bell, and zero on both extremes with a rise in the middle member ship function coincide well. However, sigmoidal, with a mirror image membership function that opens to the right, and asymmetrical polynomial membership functions do not coincide well with the trapezoidal membership function. The result of the study indicates that trapezoidal MF gives the best performance, and triangular MF response is very close to that of trapezoidal MF.