Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《China Mechanical Engineering(中国机械工程)》 2008-17
Add to Favorite Get Latest Update

Experimental Study of Lateral Position Based on Algebra Neural Networks Information Fusion

Ma Binliang1,2 Huang Yumei1 Shi Enxiu1 Cai Tao11.Xi'an University of Technology,Xi'an,7100482.Dushanzi Petrochemical Company,Kelamayi,Xinjiang,833600  
To improve positioning precision of a mobile robot,the effect of measuring distance d and incident angle α of ultrasonic sensor was analyzed on measuring accuracy by experiments.Lateral positioning model was designed using algebra neural networks fusion algorithm because the fusion algorithm can precisely map in sample space and has better nonlinear approximation ability.By comparison experiments,the positioning fusion model has higher precision,location error is decreased to less than 0.9957mm and angle error is decreased to less than 0.2966°.Finally,the positioning fusion model was applied to a self-developed mobile robot.The experimental results show the robot has location error of ±2.5mm and angle error of ±0.42°,and meets the positioning needs.
【Fund】: 国家自然科学基金资助项目(50075069);; 陕西省教育厅产业化培育计划资助项目(03JC17)
【CateGory Index】: TP242
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved