计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (26): 25-29.DOI: 10.3778/j.issn.1002-8331.2008.26.008

• 博士论坛 • 上一篇    下一篇

KMEA算法及其在多传感器融合系统中的应用

阎高伟,谢 刚,牛昱光,谢克明   

  1. 太原理工大学 信息工程学院,太原 030024
  • 收稿日期:2008-05-15 修回日期:2008-06-24 出版日期:2008-09-11 发布日期:2008-09-11
  • 通讯作者: 阎高伟

Knowledge based Mind Evolution Algorithm and its application in multi-sensor fusion system

YAN Gao-wei,XIE Gang,NIU Yu-guang,XIE Ke-ming   

  1. College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,China
  • Received:2008-05-15 Revised:2008-06-24 Online:2008-09-11 Published:2008-09-11
  • Contact: YAN Gao-wei

摘要: 将粗糙集和粒计算形成的知识获取机制融入思维进化算法,对进化过程中所产生的数据进行挖掘和知识发现,利用所发现的知识指导进化的方向,实现了知识指导下的思维进化算法,体现出人类思维活动过程中对知识的抽象和利用功能。对多传感器信息融合系统中神经网络权值优化的结果表明,该方法可降低神经网络在权值选择上的随机性缺陷,缩小搜索空间,提高网络的收敛速度和泛化能力。

关键词: 思维进化算法, 知识发现, 粗糙集, 粒计算, 信息融合, 神经网络

Abstract: The knowledge-acquisition mechanism generated from rough set and granular computing is integrated into Mind Evolution Algorithm(MEA).The evolution direction is guided by the knowledge discovered from the data produced in the evolution process.As a result,the mind evolution algorithm under the guide of knowledge is realized,which reflects the knowledge abstraction and usage during human beings’ mind activities.The result of weight-value optimization of the neural network in multi-sensor information fusion system shows that this method is able to effectively improve the study efficiency and precision for neural networks.

Key words: Mind Evolution Algorithm(MEA), knowledge discovery, rough set, granular gomputing, information fusion, neural network