摘要
针对传统的小脑模型,在保留CMAC原有增强和局部特性的基础上,结合模糊逻辑的思想,采用模糊隶属度函数作为接收域函数,提出了一种广义模糊小脑模型神经网络(GFAC)。研究了GFAC接受域函数的映射规律、隶属度函数及其参数的选取规律和学习算法。仿真结果表明GFAC具有良好的泛化能力和逼近精度。
Aiming at conventional Cerebellar Model Articulation Controller （CMAC） and combining CMAC addressing schemes with fuzzy logic idea, a general fuzzified CMAC（GFAC） was proposed, in which the fuzzy membership functions were utilized as the receptive field functions. The mapping of receptive field functions, the selection law of membership with its parameters and the learning algorithm were studied.. By using GFAC, the approximation of complex functions can be obtained which is more continuous than using conventional CMAC. The simulation results show that GFAC has good generalization, proper approximate accuracy and capacity to calculate function derivative output.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2005年第11期2708-2712,共5页
Journal of System Simulation
基金
国家自然科学基金项目(60474014)
教育部高等学校博士学科点基金项目(20040151007)
交通部应用基础项目(200432922504)
