人工内分泌系统调节人工神经网络的控制模型  

A dynamic control model for modulating using artificial neural networks using the artificial endocrine system

作　　者：林广栋[1] 王煦法[1] 

机构地区：[1]中国科学技术大学计算机科学与技术学院,安徽合肥230027

出　　处：《中国科学技术大学学报》2012年第2期148-153,160,共7页JUSTC

基　　金：国家自然科学基金(60875027)资助

摘　　要：人体的内分泌系统是一个复杂的控制系统,与神经系统、免疫系统一起维持机体的内平衡.受内分泌系统与神经系统的相互作用机制启发,提出一种内分泌系统调节神经网络系统的EMNCS(endocrine modulated neural control system)控制模型.EMNCS模型中,内分泌系统可以根据环境变化动态调节神经网络系统,达到动态控制的目的.把该模型应用于动态环境下的机器人控制系统,实验表明,EMNCS模型能有效提高机器人在动态环境下的适应能力.The endocrine system in the human body is a complex control system and cooperates with the neural and immune systems to maintain homeostasis.Inspired by the mechanism by which the endocrine system interacts with neural system,an EMNCS（endocrine modulated neural control system） model was proposed in which the artificial endocrine system modulated and artificial neural system.In EMNCS,the endocrine system could modulate the parameters and structures of the neural system in order to achieve dynamic control.The algorithm was applied in the robot system and experimental results show that EMNCS model can improve performance and enhance its adaptability.

关 键 词：人工内分泌系统 人工神经网络 机器人 人工智能 计算智能 动态控制 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]