机器人修磨中融合先验知识的适应学习建模方法

摘要
图/表
参考文献
相关文章 (8)

全文: PDF (1927 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
吕洪波, 宋亦旭, 贾培发. 机器人修磨中融合先验知识的适应学习建模方法[J]. 机器人, 2011, 33(6): 641-648.. Lü Hongbo, SONG Yixu, JIA Peifa. Incorporation of Prior Knowledge in Adaptive Learning for Modeling the Robotic Profile Grinding. ROBOT, 2011, 33(6): 641-648..

摘要

针对机器人修磨磨削量建模中处理突变因素的难题，本文首先从机器学习建模方法的角度指出该问题与统计学习的不同点，并把问题形式化，然后在此基础上提出了融合先验知识的适应学习建模方法．该方法基于半经验公式生成虚拟样本，不但弥补了适应学习建模中新样本不足的问题，而且把半经验公式中的信息更充分地融合到学习机模型中．实验结果证明，该方法使适应建模具有更快的速度和更高的精度，在实际应用中可提高加工效率，降低由于动态因素变化带来的废品率．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	吕洪波
	宋亦旭
	贾培发

关键词 ：机器人修磨, 机器学习, 适应建模, 融合先验知识, 经验公式, 虚拟样本

Abstract：

When modeling the removal rates of the robotic belt grinding system,it is hard to deal with the affecting factors which changes suddenly.In order to solve this problem,the difference between the belt grinding modeling approach based on machine learning and the statistical learning theory is pointed out firstly.And this difference is formalized.After that, an adaptive learning method incorporating prior knowledge is put forward,which produces virtual examples from semi-empirical formula to remedy the lack of new examples in adaptive learning modeling,and extracts the information from the semi-empirical formula which is imported into the learner then.It is proved by the experiments that this method makes the model adapt to the variances more quickly and with higher precision.Moreover,this method will help to improve the processing efficiency and reduce the reject rate caused by the dynamical factors in the practical use.

Key words： robotic profile grinding machine learning adaptive modeling incorporation of prior knowledge empirical formula virtual example

收稿日期: 2011-01-04 录用日期: 2011-08-25

TP242.2

基金资助:

国家自然科学基金资助项目（50875141）

通讯作者: 吕洪波 E-mail: LvHongbo@gmail.com

作者简介: 吕洪波（1976- ），男，博士生．研究领域：智能机器人．
宋亦旭（1972- ），男，博士后，讲师．研究领域：机器人智能控制．
贾培发（1946- ），男，硕士，教授．研究领域：机器人技术，计算机信息系统．

链接本文:

https://robot.sia.cn/CN/ 或 https://robot.sia.cn/CN/Y2011/V33/I6/641

[1] 黄云,黄智.现代砂带磨削技术及工程应用[M]. 重庆:重庆大学出版社,2009.Huang Y,Huang Z.Belt grinding's modern technologies and applications[M]. Chongqing:Chongqing University Press,2009.
[2] 杨扬.面向机器人修型磨削的智能建模方法研究[D]. 北京:清华大学,2010.Yang Y.Research on intelligent modeling method for robotic profile grinding[D]. Beijing:Tsinghua University,2010.
[3] Sun Y G.Development of a unified flexible grinding process[D]. Storrs,CT,USA:University of Connecticut,2004.
[4] Kordic V,Lazinica A,Merdan M.Cutting edge robotics[M]. Mammendorf,Germany:Pro Literatur Verlag,2005.
[5] Hammann G.Modellierung des abtragsverbaltens elastischer roboter-gefuehrter schleifwerkzeuge[D]. Stuttgart,Germany:University Stuttgart,1998.
[6] Schuppstuhl T.Beitrag zum bandschleifen komplexer freiformgeometrien mit dem industrieroboter[D]. Dortmund,Germany:University Dortmund,2003.
[7] Huang H,Zhou L,Chen X Q,et al.SMART robotic system for 3D profile turbine vane airfoil repair[J]. International Journal of Advanced Manufacturing Technology,2003,21(4):275-283.
[8] Lü H B,Song Y X,Jia P F,et al.An adaptive modeling approach based on ESN for robotic belt grinding[C]//IEEE International Conference on Information and Automation.Piscataway,N J,USA:IEEE,2010:787-792.
[9] Vapnik V N.The nature of statistical learning theory[M]. 2nd ed.New York,NY,USA:Springer,2000.
[10] Vapnik V N.Statistical learning theory[M]. Hoboken,NJ,USA:John Wiley & Sons Inc.,1998.
[11] Class F,Kaltenmeier A,Regel P,et al.Fast speaker adaptation for speech recognition systems[C]//Intemational Conference on Acoustics,Speech,and Signal Processing ICASSP.Piscataway,NJ,USA:IEEE,1990:133-136.
[12] Takahashi J I,Sagayama S.Vector-field-smoothed Bayesian learning for fast and incremental speaker/telephone-channel adaptation[J]. Computer Speech & Language,1997,11(2):127-146.
[13] Yang Y,Song Y,Wang J,et al.An adaptive SVR modeling method based on VFS for robotic grinding[C]//IEEE International Conference on Computer Science and Information Technology.Piscataway,NJ,USA:IEEE,2010:438-442.
[14] 孙喆.融合先验知识的机器学习研究[D]. 北京:清华大学,2010.Sun Z.Research on incorporation of prior knowledge in machine learning[D]. Beijing:Tsinghua University,2010.
[15] Lauer F,Bloch G.Incorporating prior knowledge in support vector machines for classification:A review[J]. Neurocomputing,2008,71(7/8/9):1578-1594.
[16] Lauer F,Bloch G.Incorporating prior knowledge in support vector regression[J]. Machine Learning,2008,70(1):89-118.
[17] Poggio T,Vetter T.Recognition and structure from one 2D model view:Observations on prototypes,object classes and symmetries[R]. Cambridge,MA,USA:MIT,1992.
[18] Abu-Mostafa Y S.Hints[J]. Neural Computation,1995,7(4):639-671.
[19] Schnkopf B,Burges C,Vapnik V.Incorporating invariances in support vector learning machines[M]//Lecture Notes in Computer Science:Vol.1112.Berlin,Germany:Springer-Verlag,1996:47-52.
[20] Niyogi P,Girosi F,Poggio T.Incorporating prior information in machine learning by creating virtual examples[J]. Proceedings of the IEEE,1998,86(11):2196-2209.
[21] Milanic S,Strrncnik S,Sel D,et al.Incorporating prior knowledge into artificial neural networks-An industrial case study[J], Neurocomputing,2004,62(1-4):131-151.
[22] Jaeger H.The "echo state" approach to analysing and training recurrent neural networks[R]. St.Augustin,Germany:German National Research Institute for Computer Science,2001.
[23] Jaeger H,Haas H.Harnessing nonlinearity:Predicting chaotic systems and saving energy in wireless communication[J]. Science,2004,304(5667):78-80.
[24] Lukosevicius M,Jaeger H.Reservoir computing approaches to recurrent neural network training[J]. Computer Science Review,2009,3(3):127-149.
[25] Verstraeten D.Reservoir computing:Computation with dynamical systems[D]. Gent,Belgium:Ghent University,2009.
[26] Jaeger H.Tutorial on training recurrent neural networks,coveting BPPT,RTRL,EKF and the "echo state network" approach[R]. St.Augustin,Germany:German National Research Center for Information Technology,2002.
[27] Ren X Y,Kuhlenkotter B.Real-time simulation and visualization of robotic belt grinding processes[J]. International Journal of Advanced Manufacturing Technology,2008,35(11):1090-1099.
[28] Johnson K L.Contact mechanics[M]. Cambridge,UK:Cambridge University Press,1987.