机器人智能修磨方法研究
|
摘要
机器人修磨的特点是面向复杂曲面，且需精确控制磨削量。由于砂带和接触轮具有弹性，且串联工业机器人刚性不好，所以是一种非刚性加工，磨削量受多种因素的影响，其中动态因素不能测控是系统实现难题。现有系统一般基于经验公式对磨削量进行控制，根据实验数据对动态因素进行补偿，精度低，适应能力差。人工智能方法是解决这一难题的有效手段。本文围绕着应用智能方法解决机器人修磨系统磨削量的精确控制问题而展开，包括方法框架、适应建模方法和应用实现三个层面。主要工作成果和创新点如下：
1.面向动态因素的适应建模问题提出机器人智能修磨方法框架。该框架的关键点包括知识库、在位测量、适应建模与磨削参数选择。在位测量把新样本及新模型反馈至知识库，使基于知识库的建模方法具有适应动态变化的基础。
2.提出基于机器学习的稳态修磨过程建模方法。该方法通过对砂带磨损的稳态近似获得满足独立同分布条件的样本，并用于训练基于支持向量回归和回声状态网络的学习机模型，精度满足系统要求。同时，该方法解决了知识库中数据和模型的组织问题，使之可用于适应建模。
3.提出基于迁移学习的适应建模方法。该方法利用新样本去量化新旧模型的相似度，在新旧模型单调性一致的前提下，通过对相似度最高的旧模型的迁移来实现新目标函数的近似估计。局部迁移方法中，面向磨削量的归一化欧氏距离解决了相邻点对待求点迁移量的加权问题。当新样本在输入变量作用域内分布均匀时该方法具有较高精度。
4.提出融合先验知识的适应学习建模方法。该方法通过虚拟样本的形式把半经验公式中蕴含的信息融合到学习机中，解决了新样本点分布不均时，新旧模型间不能准确量化迁移量而导致的无法适应建模的问题。虚拟样本保证了学习机的稳定性，机器学习方法提高了建模精度。
5.基于粒子群算法实现修磨参数的优化。参数选择是实现智能修磨的关键步骤，多变量在线可控使参数可优化，但变量在线调节速度影响磨削量控制的准确度。提出针对不同变量进行加权的相邻点差值平方和的优化目标，并基于粒子群算法实现参数优化，有效提高了系统响应速度。
The robotic profile belt grinding (RPBG) system is often employed for machiningcomplex surface whose removal rates need to be controlled accurately. Because theabrasive belt and the contact wheel are flexible, and the industrial robot is not rigidenough, RPBG is non-rigid and its removal rate is affected by a large amount of factors.So far, empirical formulas are applied to calculate the removal rate in the RPBGsystems mostly. However, they are low accuracy and lack of adaptability. Artificialintelligence is effective for solving this problem. The central topic of this thesis is tocontrol the accuracy of the RPBG system using artificial intelligence technology. Itincludes the method framework, the methods of adaptive modeling and how to put theminto use. The contributions of this dissertation consist of the following parts:
1. To solve the problem of adapting to dynamic factors, an intelligent methodframework is presented. The keys of the framework include the knowledge database, in-situ measurement, adaptive modeling and parameters' selection. The in-situmeasurement gets some new samples for the knowledge database in each grinding. So itis possible for the model to adapt the dynamic factors.
2. In order to solve the adaptive problem, methods based on support vectorregression and echo state network are firstly applied to model the removal rate on quasi-steady state which means meeting the i.i.d. condition approximatively in this thesis.And the accuracies of these two methods meet the need. Basing on the quasi-steadylearners, samples in the database can be organized and the adaptive learning problemcan be formulized.
3. An adaptive modeling method based on transfer learning is proposed for theRPBG system. This method uses some new samples to measure the difference betweenthe old learner and the new target function. With same monotonicity, the most similarold learner is transferred to predict the target points. The normalized Euclidean distancefor removal rate is important for calculating the weights. When the new samplesdistribute evenly in the input space, this method is effective.
4. Another adaptive method is put forward by incorporating prior knowledge inmachine learning by creating virtual examples based on semi-empirical formula. The method can be applied for adaptive modeling when the new samples in input space areunevenly. In principle, the virtual samples make the model stable and the machinelearning algorithm improve the accurate.
5. It is a key step to choose the grinding parameters in the intelligent framework,which is also the purpose of the removal model. And there are more than one variablecan be adjusted online, so the grinding parameters can be optimized. The objectfunction is presented in order to reduce impact of the system response speed on thegrinding accuracy. The particle swarm optimization algorithm is applied to gain theoptimal parameters.
1.面向动态因素的适应建模问题提出机器人智能修磨方法框架。该框架的关键点包括知识库、在位测量、适应建模与磨削参数选择。在位测量把新样本及新模型反馈至知识库，使基于知识库的建模方法具有适应动态变化的基础。
2.提出基于机器学习的稳态修磨过程建模方法。该方法通过对砂带磨损的稳态近似获得满足独立同分布条件的样本，并用于训练基于支持向量回归和回声状态网络的学习机模型，精度满足系统要求。同时，该方法解决了知识库中数据和模型的组织问题，使之可用于适应建模。
3.提出基于迁移学习的适应建模方法。该方法利用新样本去量化新旧模型的相似度，在新旧模型单调性一致的前提下，通过对相似度最高的旧模型的迁移来实现新目标函数的近似估计。局部迁移方法中，面向磨削量的归一化欧氏距离解决了相邻点对待求点迁移量的加权问题。当新样本在输入变量作用域内分布均匀时该方法具有较高精度。
4.提出融合先验知识的适应学习建模方法。该方法通过虚拟样本的形式把半经验公式中蕴含的信息融合到学习机中，解决了新样本点分布不均时，新旧模型间不能准确量化迁移量而导致的无法适应建模的问题。虚拟样本保证了学习机的稳定性，机器学习方法提高了建模精度。
5.基于粒子群算法实现修磨参数的优化。参数选择是实现智能修磨的关键步骤，多变量在线可控使参数可优化，但变量在线调节速度影响磨削量控制的准确度。提出针对不同变量进行加权的相邻点差值平方和的优化目标，并基于粒子群算法实现参数优化，有效提高了系统响应速度。
The robotic profile belt grinding (RPBG) system is often employed for machiningcomplex surface whose removal rates need to be controlled accurately. Because theabrasive belt and the contact wheel are flexible, and the industrial robot is not rigidenough, RPBG is non-rigid and its removal rate is affected by a large amount of factors.So far, empirical formulas are applied to calculate the removal rate in the RPBGsystems mostly. However, they are low accuracy and lack of adaptability. Artificialintelligence is effective for solving this problem. The central topic of this thesis is tocontrol the accuracy of the RPBG system using artificial intelligence technology. Itincludes the method framework, the methods of adaptive modeling and how to put theminto use. The contributions of this dissertation consist of the following parts:
1. To solve the problem of adapting to dynamic factors, an intelligent methodframework is presented. The keys of the framework include the knowledge database, in-situ measurement, adaptive modeling and parameters' selection. The in-situmeasurement gets some new samples for the knowledge database in each grinding. So itis possible for the model to adapt the dynamic factors.
2. In order to solve the adaptive problem, methods based on support vectorregression and echo state network are firstly applied to model the removal rate on quasi-steady state which means meeting the i.i.d. condition approximatively in this thesis.And the accuracies of these two methods meet the need. Basing on the quasi-steadylearners, samples in the database can be organized and the adaptive learning problemcan be formulized.
3. An adaptive modeling method based on transfer learning is proposed for theRPBG system. This method uses some new samples to measure the difference betweenthe old learner and the new target function. With same monotonicity, the most similarold learner is transferred to predict the target points. The normalized Euclidean distancefor removal rate is important for calculating the weights. When the new samplesdistribute evenly in the input space, this method is effective.
4. Another adaptive method is put forward by incorporating prior knowledge inmachine learning by creating virtual examples based on semi-empirical formula. The method can be applied for adaptive modeling when the new samples in input space areunevenly. In principle, the virtual samples make the model stable and the machinelearning algorithm improve the accurate.
5. It is a key step to choose the grinding parameters in the intelligent framework,which is also the purpose of the removal model. And there are more than one variablecan be adjusted online, so the grinding parameters can be optimized. The objectfunction is presented in order to reduce impact of the system response speed on thegrinding accuracy. The particle swarm optimization algorithm is applied to gain theoptimal parameters.
引文
[1] Shedroff N. Experience Design. Waite Group Press,2001.
[2] Alpaydin E. Introduction to Machine Learning. Dietterich T, editor. The MIT Press,2010,Second Edition.
[3] Mitchell T M. Machine Learning. McGraw-Hill,1997.
[4]黄云,黄智.现代砂带磨削技术及工程应用.重庆大学出版社,2009.
[5] Whitton S. Adaptive robot grinding improves turbine blade repair. Industrial Robot: AnInternational Journal2003,30(4).
[6] Directorate-General III of European Commission. Summaries of projects-FourthFramework Programme (Information Technologies RTD Programme-Domain8:Integration in Manufacturing).1997. Internet: http://www.cordis.lu/esprit/src/iimhome.htm
[7] Chen X, Gong Z, Huang H. Development of Robotic System for3D Profile Grinding andPolishing. SIMTech,2000. Report Nr. AT/00/012/AMP.
[8] Huang H, Gong Z M, Chen X Q, et al. Robotic grinding and polishing for turbine-vaneoverhaul. Journal of Materials Processing Technology,2002,127(2):140-145.
[9] Huang H, Gong Z M, Chen X Q, et al. SMART robotic system for3D profile turbine vaneairfoil repair. International Journal of Advanced Manufacturing Technology,2003,21(4):275-283.
[10] Sun Y. Development of a unified flexible grinding process[Ph.D. dissertation]. Universityof Connecticut,2004.
[11] Kuhlenkoetter B, Zhang X. A Robot System for High Quality Belt Grinding and PolishingProcesses//Kordic V, Lazinica A, Merdan M. Cutting Edge Robotics. plV pro literaturVerlag Robert Mayer-Scholz,2005:755-770.
[12] Kuhlenkoetter B. Development of a Robot System for Advanced High QualityManufacturing Processes. Acta Polytechnica,2006,46(1):3-7.
[13] Zhang X, Cabaravdic M, Kneupner K, et al. Real-Time Simulation of Robot ControlledBelt Grinding Processes of Sculptured Surfaces. International Journal of Advanced RoboticSystems,2004,1(2):109-114.
[14] Hammann G. Modellierung des Abtragsverhaltens Elastischer Roboter-gefuehrterSchleifwerkzeuge[Ph.D. dissertation]. University Stuttgart,1998.
[15] Schueppstuhl T. Beitrag zum Bandschleifen komplexer Freiformgeometrien mit demIndustrieroboter[Ph.D. dissertation]. University Dortmund,2003.
[16] Cabaravdic M, Kuhlenk tter B. Bandschleifprozesse optimieren. mo Metalloberfl che,2005,59(4):44-47.
[17] Ren X, Cabaravdic M, Zhang X, et al. A local process model for simulation of robotic beltgrinding. International Journal of Machine Tools and Manufacture,2007,47(6):962-970.
[18]詹建明.机器人研磨自由曲面时的作业环境与柔顺控制研究[博士学位论文].吉林大学,2002.
[19]杨威.加工自由曲面的机器人砂带研磨系统研究[硕士学位论文].吉林大学,2007.
[20]苏运动.自由曲面砂带研磨机器人自动编程系统研究[硕士学位论文].吉林大学,2008.
[21]张海鸥,胡永凉,王桂兰, et al.面向熔射快速制模的机器人自动研磨系统的开发.中国机械工程,2005,(17):1515-1518.
[22]胡永凉.面向熔射制模的机器人研磨系统及实验研究[硕士学位论文].华中科技大学,2005.
[23]任俊.面向熔射快速制模的机器人辅助曲面自动抛光系统的研究[硕士学位论文].华中科技大学,2006.
[24]吴宝海,张莹,罗明, et al.现代数控机床的智能化发展及应用.航空制造技术,2008,(17):52-56.
[25] Jurrens K K, Soons J A, Ivester R W. Smart Machining Research at the National Instituteof Standards and Technology. Small Lots Intelligent Manufacturing Workshop. Santa Fe,New Mexico, USA;2003.
[26]石璟.面向叶片加工的数控砂带抛磨系统的研究[硕士学位论文].江南大学,2009.
[27]陈光.航空发动机结构设计分析.北京航空航天大学出版社,2006.
[28] Shibata J, Inasaki I, Yonetsu S. The relation between the wear of grain cutting edges andtheir metal removal ability in coated abrasive belt grinding. Wear,1979,55(2):331-344.
[29]龚正.砂带磨削的机理及砂带磨损的探讨.上海第二工业大学学报,1991,(01):25-32.
[30]钟华珍,阳小民.关于砂带磨损的研究.磨料磨具与磨削,1991,(05):4-9.
[31]邓朝晖,许世良,罗重常.砂带磨损机理的研究.磨床与磨削,1994,(02):41-44.
[32] Jourani A, Dursapt M, Hamdi H, et al. Effect of the belt grinding on the surface texture:Modeling of the contact and abrasive wear. Wear,2005,259(7-12):1137-1143.
[33] Bigerelle M, Hagege B, El Mansori M. Mechanical modelling of micro-scale abrasion insuperfinish belt grinding. Tribology International,2008,41(11):992-1001.
[34] Mezghani S, El Mansori M, Sura E. Wear mechanism maps for the belt finishing of steeland cast iron. Wear,2009,267(1-4):86-91.
[35]傅水根.机械制造工艺基础.清华大学出版社,2004.
[36]罗兵,李成群,贠超, et al.三自由度砂带磨床的设计与分析.机械设计与制造,2007,(02).
[37]王留呆,贠超,彭伟, et al.一维力控制机器人砂带磨削机设计.机械设计,2008,(06).
[38]李成群,付永领,贠超, et al.三自由度砂带磨床的运动学及其灵巧度分析.机械设计与制造,2010,(11):122-124.
[39]杨扬.面向机器人修型磨削的智能建模方法研究[硕士学位论文].清华大学,2010.
[40]李成群,付永领,贠超, et al. PLC在一维力解耦砂带磨削机床中的应用.制造业自动化,2010,(09):103-105.
[41]张国雄.三坐标测量机的发展趋势.中国机械工程,2000,11(1-2):231-235+9.
[42]谢金.自由曲面非接触式测量方法研究及系统研制[硕士学位论文].浙江大学,2003.
[43]刘祚时,倪潇娟.三坐标测量机(CMM)的现状和发展趋势.机械制造,2004,(08):32-34.
[44]黄智,黄云.砂带磨削原理及其应用.金属加工(冷加工),2008,(24):28-30.
[45] Mezghani S, El Mansori M. Abrasiveness properties assessment of coated abrasives forprecision belt grinding. Surface and Coatings Technology,2008,203(5-7):786-789.
[46] Abebe M, Appl F C. Theoretical analysis of the basic mechanics of abrasive processes:Part I: General model. Wear,1988,126(3):251-266.
[47] Abebe M, Appl F C. Theoretical analysis of the basic mechanics of abrasive processes:Part II: Model of the ploughing process. Wear,1988,126(3):267-283.
[48] Marinescu I D, Hitchiner M, Uhlmann E, et al. Handbook of Machining with GrindingWheels. Boothroyd G, editor. CRC Press,2007.
[49] Malkin S, Guo C. Grinding Technology: Theory and Applications of Machining withAbrasives. Industrial Press Inc.,2008,2nd.
[50] Wang J, Sun Y, Gan Z, et al. Process Modeling of Flexible Robotic Grinding.2003International Conference on Control, Automation, and Systems (ICCAS2003). Gyeongju,Korea;2003.
[51] Brinksmeier E, T nshoff H K, Czenkusch C, et al. Modelling and optimization of grindingprocesses. Journal of Intelligent Manufacturing,1998,9(4):303-314.
[52] Brinksmeier E, Aurich J C, Govekar E, et al. Advances in Modeling and Simulation ofGrinding Processes. CIRP Annals-Manufacturing Technology,2006,55(2):667-696.
[53]李鑫,黄云,黄智, et al.镁合金砂带磨削表面质量分析及其预测.机械设计与制造,2009,(11):186-187.
[54] Zhang X, Kuhlenk tter B, Kneupner K. A Learning-Based Contact Model in Belt-GrindingProcesses. Proceeding of the Advances in Neural Networks-ISNN2004. Dalian, China.,2004:761-766.
[55] Zhang X, Kneupner K, Kuhlenkotter B. A new force distribution calculation model forhigh-quality production processes. The International Journal of Advanced ManufacturingTechnology,2006,27(7-8):726-732.
[56] Weinert K, Blum H, Kuhlenk tter B, et al. New methods for calculating the forcedistribution within belt grinding processes. Production Engineering,2007,1(3):285-289.
[57]张学工.模式识别.清华大学出版社,2010,第三版.
[58] Vapnik V N. The Nature of Statistical Learning Theory. Springer,2000,2nd.
[59] Vapnik V N. Statistical Learning Theory. John Wiley&Sons, Inc.,1998.
[60] Funahashi K. On the approximate realization of continuous mappings by neural networks.Neural Networks,1989,2(3):183-192.
[61] Funahashi K-i, Nakamura Y. Approximation of dynamical systems by continuous timerecurrent neural networks. Neural Networks,1993,6(6):801-806.
[62]周会娜.基于BP神经网络和遗传算法对工程陶瓷磨削力建模的研究[硕士学位论文].天津大学,2007.
[63]刘红星,魏强,屈梁生.基于人工神经网络建模的磨削烧伤预测.磨床与磨削,1996,(03):28-30.
[64] Wang Y T, Jan Y J. Grinding force models in finishing processes. Proceeding of the2001IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM): Vol.2.Como, Italy: Institute of Electrical and Electronics Engineers Inc.,2001:822-827.
[65]宋洪涛,宾鸿赞.把神经网络应用于丝杠磨削过程的建模与控制.光学精密工程,2001,(04):364-367.
[66] Lukosevicius M, Jaeger H. Reservoir Computing Approaches to Recurrent Neural NetworkTraining. Computer Science Review,2009,3(3):127-149
[67] Jaeger H. The "echo state" approach to analysing and training recurrent neural networks.German National Research Institute for Computer Science,2001. Report Nr.148.
[68] Jaeger H, Haas H. Harnessing Nonlinearity: Predicting Chaotic Systems and SavingEnergy in Wireless Communication. Science,2004,304(5667):78-80.
[69] Vapnik V. Estimation of Dependences Based on Empirical Data. Springer,2006,2nd.
[70]边肇祺,张学工.模式识别.清华大学出版社,2000,第二版.
[71] Scholkopf B, Smola A J. Learning with Kernels-Support Vector Machines, Regularization,Optimization, and Beyond. Dietterich T, editor. The MIT Press,2002.
[72] Vapnik V N. The Nature of Statistical Learning Theory. Springer,1995,1st Edition.
[73] Cortes C, Vapnik V. Support-Vector Networks. Machine Learning,1995,20(3):273-297.
[74] Jaeger H. Tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF andthe "echo state network" approach. German National Research Institute for ComputerScience,2002. Report Nr.159.
[75] Chang C-C, Lin C-J. LIBSVM: a Library for Support Vector Machines.2001.
[76] Wang W. Determination of the spread parameter in the Gaussian kernel for classificationand regression. Neurocomputing,2003,55(3-4):643-663.
[77] Cherkassky V, Ma Y. Practical selection of SVM parameters and noise estimation for SVMregression. Neural Networks,2004,17(1):113-126.
[78] Mohri M,. A R I. Stability bounds for non-iid processes. Proceeding of the Advances inNeural Information Processing Systems,2007.
[79] Amini M-R, Habrard A, Ralaivola L, et al. Learning from non IID Data:Theory,Algorithms and Practice. Proceeding of the ECML PKDD2009. Bled, Slovenia,2009.
[80] Zhou Z-H, Sun Y-Y, Li Y-F. Multi-instance learning by treating instances as non-I.I.D.samples. Proceeding of the26th Annual International Conference on Machine Learning.Montreal, Quebec, Canada: ACM,2009:1249-1256.
[81]潘志伟.统计学习算法：多分类及非独立同分布抽样下的回归[博士学位论文].中国科学技术大学,2009.
[82] Pan S J, Yang Q. A Survey on Transfer Learning. IEEE Transactions On Knowledge AndData Engineering,2010,22(10):1345-1359.
[83] Dai W, Yang Q, Xue G-R, et al. Boosting for transfer learning. Proceeding of the24thinternational conference on Machine learning. Corvalis, Oregon: ACM,2007:193-200.
[84] Wu P, Dietterich T G. Improving SVM accuracy by training on auxiliary data sources.Proceeding of the twenty-first international conference on Machine learning. Banff,Alberta, Canada: ACM,2004:871-878.
[85] Caruana R. Multitask Learning. Machine Learning,1997,28(1):41-75.
[86] Bakker B, Heskes T. Task clustering and gating for bayesian multitask learning. Journal ofMachine Learning Research,2003,4:83-99.
[87] DaumeIII H. Bayesian multitask learning with latent hierarchies. Proceeding of theTwenty-Fifth Conference on Uncertainty in Artificial Intelligence. Montreal, Quebec,Canada: AUAI Press,2009:135-142.
[88] Wu W, Li H, Hu Y, et al. Multi-Task Learning: Multiple Kernels for Multiple Tasks.Microsoft,2010. Report Nr. MSR-TR-2010-87.
[89] Babcock B, Babu S, Datar M, et al. Models and issues in data stream systems. Proceedingof the twenty-first ACM SIGMOD-SIGACT-SIGART symposium on Principles ofdatabase systems. Madison, Wisconsin: ACM,2002:1-16.
[90] Gaber M M, Zaslavsky A, Krishnaswamy S. Mining data streams: a review. SIGMODRecord,2005,34(2):18-26.
[91] Bifet A. Adaptive Learning and Mining for Data Streams and Frequent Patterns[Ph.D.dissertation]. Universitat Politecnica de Catalunya,2009.
[92] Raina R, Battle A, Lee H, et al. Self-taught learning: transfer learning from unlabeled data.Proceeding of the24th international conference on Machine learning. Corvalis, Oregon:ACM,2007:759-766.
[93] Jiang J, Zhai C. Instance weighting for domain adaptation in NLP. Proceeding of the45thAnnual Meeting of the Association for Computational Linguistics,2007:264-271.
[94]曹杰.统计机器翻译领域自适应问题研究[硕士学位论文].中国科学院,2010.
[95] Zadrozny B. Learning and evaluating classifiers under sample selection bias. Proceeding ofthe twenty-first international conference on Machine learning. Banff, Alberta, Canada:ACM,2004:114.
[96] Fan W, Davidson I, Zadrozny B, et al. An Improved Categorization of Classifier'sSensitivity on Sample Selection Bias. Proceeding of the Fifth IEEE InternationalConference on Data Mining,2005.
[97] Huang J, Smola A, Gretton A, et al. Correcting Sample Selection Bias by Unlabeled Data.Proceeding of the Advances in neural information processing systems19,2007:601-608.
[98] Shimodaira H. Improving predictive inference under covariate shift by weighting the log-likelihood function. Journal of Statistical Planning and Inference,2000,90(2):227-244.
[99] Sugiyama M, Suzuki T, Nakajima S, et al. Direct importance estimation for covariate shiftadaptation. Annals of the Institute of Statistical Mathematics,2008,60(4):699-746.
[100] Arnold A, Nallapati R, Cohen W W. A Comparative Study of Methods for TransductiveTransfer Learning. Proceeding of the Seventh IEEE International Conference on DataMining Workshops. Omaha, NE: IEEE Computer Society,2007:77-82.
[101] Dai W, Yang Q, Xue G-R, et al. Self-taught clustering. Proceeding of the25th internationalconference on Machine learning. Helsinki, Finland: ACM,2008:200-207.
[102] Jebara T. Multi-task feature and kernel selection for SVMs. Proceeding of the twenty-firstinternational conference on Machine learning. Banff, Alberta, Canada: ACM,2004:55.
[103] Argyriou A, Evgeniou T, Pontil M. Multi-task feature learning. Proceeding of theAdvances in Neural Information Processing Systems19,2007.
[104] Dai W, Xue G, Yang Q, et al. Co-clustering based classification for out-of-domaindocuments. Proceeding of the13th ACM SIGKDD international conference on Knowledgediscovery and data mining. San Jose, California, USA: ACM,2007:210-219.
[105] Lee S-I, Chatalbashev V, Vickrey D, et al. Learning a meta-level prior for feature relevancefrom multiple related tasks. Proceeding of the24th international conference on Machinelearning. Corvalis, Oregon: ACM,2007:489-496.
[106] Argyriou A, Micchelli C A, Pontil M, et al. A Spectral Regularization Framework forMulti-Task Structure Learning. Proceeding of the20th Annual Conference on NeuralInformation Processing Systems,2008:25-32.
[107] Liao X, Xue Y, Carin L. Logistic regression with an auxiliary data source. Proceeding ofthe22nd international conference on Machine learning. Bonn, Germany: ACM,2005:505-512.
[108] Quinonero-Candela J, Sugiyama M, Schwaighofer A, et al. Dataset Shift in MachineLearning. The MIT Press,2009.
[109] Evgeniou T, Pontil M. Regularized multi-task learning. Proceeding of the tenth ACMSIGKDD international conference on Knowledge discovery and data mining. Seattle, WA,USA: ACM,2004:109-117.
[110] Lawrence N D, Platt J C. Learning to learn with the informative vector machine.Proceeding of the twenty-first international conference on Machine learning. Banff,Alberta, Canada: ACM,2004:65.
[111] Schwaighofer A, Tresp V, K. Yu. Learning Gaussian Process Kernels via HierarchicalBayes. Proceeding of the17th Annual Conference on Neural Information ProcessingSystems,2005: pp.1209-1216.
[112] Bonilla E, Chai K M, Williams C. Multi-Task Gaussian Process Prediction. Proceeding ofthe20th Annual Conference on Neural Information Processing Systems,2008:153-160.
[113] Gao J, Fan W, Jiang J, et al. Knowledge transfer via multiple model local structuremapping. Proceeding of the14th ACM SIGKDD international conference on Knowledgediscovery and data mining. Las Vegas, Nevada, USA: ACM,2008:283-291.
[114] Mihalkova L, Huynh T, Mooney R. Mapping and revising markov logic networks fortransfer learning. Proceeding of the22nd National Conference on Artificial Intelligence(AAAI-07),2007:608-614.
[115] Davis J, Domingos P. Deep transfer via second-order Markov logic. Proceeding of the26thAnnual International Conference on Machine Learning. Montreal, Quebec, Canada: ACM,2009:217-224.
[116] Mihalkova L, Mooney R J. Transfer learning from minimal target data by mapping acrossrelational domains. Proceeding of the21st international jont conference on Artificalintelligence. Pasadena, California, USA: Morgan Kaufmann Publishers Inc.,2009:1163-1168.
[117]戴文渊.基于实例和特征的迁移学习算法研究[硕士学位论文].上海交通大学,2009.
[118] Vapnik V N.统计学习理论.徐建华,张学工, translator.电子工业出版社,2009.
[119]孙喆.融合先验知识的机器学习研究[博士学位论文].清华大学,2010.
[120] Lauer F, Bloch G. Incorporating prior knowledge in support vector machines forclassification: A review. Neurocomputing,2008,71(7-9):1578-1594.
[121] Lauer F, Bloch G. Incorporating prior knowledge in support vector regression. MachineLearning,2008,70(1):89-118.
[122] Epshteyn A, DeJong G. Rotational prior knowledge for SVMs. Proceeding of the SixteenthEuropean Conference on Machine Learning: Vol.3720. Porto, Portugal,2005:108-119.
[123] Ratsaby J, Maiorov V. On the value of partial information for learning from examples.Journal Of Complexity,1997,13(4):509-544.
[124] Han F, Huang D-S. Improved extreme learning machine for function approximation byencoding a priori information. Neurocomputing,2006,69(16-18):2369-2373.
[125] Joerding W H, Meador J L. Encoding a priori information in feedforward networks. NeuralNetworks,1991,4(6):847-856.
[126] Mangasarian O L, Shavlik J W, Wild E W. Knowledge-Based Kernel Approximation.Journal of Machine Learning Research,2004,5:1127-1141.
[127] Pelckmans K, Espinoza M, Brabanter J, et al. Primal-Dual Monotone Kernel Regression.Neural Processing Letters,2005,22(2):171-182.
[128] Poggio T, Vetter T. Recognition and Structure from one2D Model View: Observations onPrototypes, Object Classes and Symmetries. Massachusetts Institute of Technology,1992.Report Nr. AIM-1347.
[129] Abu-Mostafa Y S. Hints. Neural Computation,1995,7(4):639-671.
[130] Sch lkopf B, Burges C, Vapnik V. Incorporating invariances in support vector learningmachines. Proceeding of the1996International Conference on Artificial Neural Networks.Bochum, Germany: Springer-Verlag,1996:47-52.
[131] Niyogi P, Poggio T, Girosi F. Incorporating Prior Information in Machine Learning byCreating Virtual Examples. Proceedings on the IEEE,1998,86(11):2196-2209.
[132] Johnson K L. Contact Mechanics.徐秉业, translator. Cambridge University Press,1985.
[133] Lv H, Song Y, Jia P, et al. An Adaptive Modeling Approach Based on ESN for RoboticBelt Grinding. Proceeding of the2010IEEE International Conference on Information andAutomation. Harbin, Heilongjiang, China: IEEE,2010:787-792.
[134]高理富,宋宁,葛运建, et al.航天机器人用六维腕力传感器动态特性研究.机器人,2002,(04):319-323.
[135] Liang W, Song Y, Lv H, et al. A novel control method for robotic belt grinding based onSVM and PSO algorithm. Proceeding of the2010International Conference on IntelligentComputation Technology and Automation: Vol.1. Changsha, China: IEEE ComputerSociety,2010:258-261.
[136] Liang W, Song Y, Lv H, et al. An effective trajectory optimization method for robotic beltgrinding based on intelligent algorithm. Proceeding of the2010IEEE InternationalConference on Robotics and Biomimetics. Tianjin, China: IEEE,2010.
[137] Kennedy J, Eberhart R. Particle swarm optimization. Proceeding of the IEEE InternationalConference on Neural Networks: Vol.4. Perth, WA, Australia: IEEE,1995:1942-1948.
[2] Alpaydin E. Introduction to Machine Learning. Dietterich T, editor. The MIT Press,2010,Second Edition.
[3] Mitchell T M. Machine Learning. McGraw-Hill,1997.
[4]黄云,黄智.现代砂带磨削技术及工程应用.重庆大学出版社,2009.
[5] Whitton S. Adaptive robot grinding improves turbine blade repair. Industrial Robot: AnInternational Journal2003,30(4).
[6] Directorate-General III of European Commission. Summaries of projects-FourthFramework Programme (Information Technologies RTD Programme-Domain8:Integration in Manufacturing).1997. Internet: http://www.cordis.lu/esprit/src/iimhome.htm
[7] Chen X, Gong Z, Huang H. Development of Robotic System for3D Profile Grinding andPolishing. SIMTech,2000. Report Nr. AT/00/012/AMP.
[8] Huang H, Gong Z M, Chen X Q, et al. Robotic grinding and polishing for turbine-vaneoverhaul. Journal of Materials Processing Technology,2002,127(2):140-145.
[9] Huang H, Gong Z M, Chen X Q, et al. SMART robotic system for3D profile turbine vaneairfoil repair. International Journal of Advanced Manufacturing Technology,2003,21(4):275-283.
[10] Sun Y. Development of a unified flexible grinding process[Ph.D. dissertation]. Universityof Connecticut,2004.
[11] Kuhlenkoetter B, Zhang X. A Robot System for High Quality Belt Grinding and PolishingProcesses//Kordic V, Lazinica A, Merdan M. Cutting Edge Robotics. plV pro literaturVerlag Robert Mayer-Scholz,2005:755-770.
[12] Kuhlenkoetter B. Development of a Robot System for Advanced High QualityManufacturing Processes. Acta Polytechnica,2006,46(1):3-7.
[13] Zhang X, Cabaravdic M, Kneupner K, et al. Real-Time Simulation of Robot ControlledBelt Grinding Processes of Sculptured Surfaces. International Journal of Advanced RoboticSystems,2004,1(2):109-114.
[14] Hammann G. Modellierung des Abtragsverhaltens Elastischer Roboter-gefuehrterSchleifwerkzeuge[Ph.D. dissertation]. University Stuttgart,1998.
[15] Schueppstuhl T. Beitrag zum Bandschleifen komplexer Freiformgeometrien mit demIndustrieroboter[Ph.D. dissertation]. University Dortmund,2003.
[16] Cabaravdic M, Kuhlenk tter B. Bandschleifprozesse optimieren. mo Metalloberfl che,2005,59(4):44-47.
[17] Ren X, Cabaravdic M, Zhang X, et al. A local process model for simulation of robotic beltgrinding. International Journal of Machine Tools and Manufacture,2007,47(6):962-970.
[18]詹建明.机器人研磨自由曲面时的作业环境与柔顺控制研究[博士学位论文].吉林大学,2002.
[19]杨威.加工自由曲面的机器人砂带研磨系统研究[硕士学位论文].吉林大学,2007.
[20]苏运动.自由曲面砂带研磨机器人自动编程系统研究[硕士学位论文].吉林大学,2008.
[21]张海鸥,胡永凉,王桂兰, et al.面向熔射快速制模的机器人自动研磨系统的开发.中国机械工程,2005,(17):1515-1518.
[22]胡永凉.面向熔射制模的机器人研磨系统及实验研究[硕士学位论文].华中科技大学,2005.
[23]任俊.面向熔射快速制模的机器人辅助曲面自动抛光系统的研究[硕士学位论文].华中科技大学,2006.
[24]吴宝海,张莹,罗明, et al.现代数控机床的智能化发展及应用.航空制造技术,2008,(17):52-56.
[25] Jurrens K K, Soons J A, Ivester R W. Smart Machining Research at the National Instituteof Standards and Technology. Small Lots Intelligent Manufacturing Workshop. Santa Fe,New Mexico, USA;2003.
[26]石璟.面向叶片加工的数控砂带抛磨系统的研究[硕士学位论文].江南大学,2009.
[27]陈光.航空发动机结构设计分析.北京航空航天大学出版社,2006.
[28] Shibata J, Inasaki I, Yonetsu S. The relation between the wear of grain cutting edges andtheir metal removal ability in coated abrasive belt grinding. Wear,1979,55(2):331-344.
[29]龚正.砂带磨削的机理及砂带磨损的探讨.上海第二工业大学学报,1991,(01):25-32.
[30]钟华珍,阳小民.关于砂带磨损的研究.磨料磨具与磨削,1991,(05):4-9.
[31]邓朝晖,许世良,罗重常.砂带磨损机理的研究.磨床与磨削,1994,(02):41-44.
[32] Jourani A, Dursapt M, Hamdi H, et al. Effect of the belt grinding on the surface texture:Modeling of the contact and abrasive wear. Wear,2005,259(7-12):1137-1143.
[33] Bigerelle M, Hagege B, El Mansori M. Mechanical modelling of micro-scale abrasion insuperfinish belt grinding. Tribology International,2008,41(11):992-1001.
[34] Mezghani S, El Mansori M, Sura E. Wear mechanism maps for the belt finishing of steeland cast iron. Wear,2009,267(1-4):86-91.
[35]傅水根.机械制造工艺基础.清华大学出版社,2004.
[36]罗兵,李成群,贠超, et al.三自由度砂带磨床的设计与分析.机械设计与制造,2007,(02).
[37]王留呆,贠超,彭伟, et al.一维力控制机器人砂带磨削机设计.机械设计,2008,(06).
[38]李成群,付永领,贠超, et al.三自由度砂带磨床的运动学及其灵巧度分析.机械设计与制造,2010,(11):122-124.
[39]杨扬.面向机器人修型磨削的智能建模方法研究[硕士学位论文].清华大学,2010.
[40]李成群,付永领,贠超, et al. PLC在一维力解耦砂带磨削机床中的应用.制造业自动化,2010,(09):103-105.
[41]张国雄.三坐标测量机的发展趋势.中国机械工程,2000,11(1-2):231-235+9.
[42]谢金.自由曲面非接触式测量方法研究及系统研制[硕士学位论文].浙江大学,2003.
[43]刘祚时,倪潇娟.三坐标测量机(CMM)的现状和发展趋势.机械制造,2004,(08):32-34.
[44]黄智,黄云.砂带磨削原理及其应用.金属加工(冷加工),2008,(24):28-30.
[45] Mezghani S, El Mansori M. Abrasiveness properties assessment of coated abrasives forprecision belt grinding. Surface and Coatings Technology,2008,203(5-7):786-789.
[46] Abebe M, Appl F C. Theoretical analysis of the basic mechanics of abrasive processes:Part I: General model. Wear,1988,126(3):251-266.
[47] Abebe M, Appl F C. Theoretical analysis of the basic mechanics of abrasive processes:Part II: Model of the ploughing process. Wear,1988,126(3):267-283.
[48] Marinescu I D, Hitchiner M, Uhlmann E, et al. Handbook of Machining with GrindingWheels. Boothroyd G, editor. CRC Press,2007.
[49] Malkin S, Guo C. Grinding Technology: Theory and Applications of Machining withAbrasives. Industrial Press Inc.,2008,2nd.
[50] Wang J, Sun Y, Gan Z, et al. Process Modeling of Flexible Robotic Grinding.2003International Conference on Control, Automation, and Systems (ICCAS2003). Gyeongju,Korea;2003.
[51] Brinksmeier E, T nshoff H K, Czenkusch C, et al. Modelling and optimization of grindingprocesses. Journal of Intelligent Manufacturing,1998,9(4):303-314.
[52] Brinksmeier E, Aurich J C, Govekar E, et al. Advances in Modeling and Simulation ofGrinding Processes. CIRP Annals-Manufacturing Technology,2006,55(2):667-696.
[53]李鑫,黄云,黄智, et al.镁合金砂带磨削表面质量分析及其预测.机械设计与制造,2009,(11):186-187.
[54] Zhang X, Kuhlenk tter B, Kneupner K. A Learning-Based Contact Model in Belt-GrindingProcesses. Proceeding of the Advances in Neural Networks-ISNN2004. Dalian, China.,2004:761-766.
[55] Zhang X, Kneupner K, Kuhlenkotter B. A new force distribution calculation model forhigh-quality production processes. The International Journal of Advanced ManufacturingTechnology,2006,27(7-8):726-732.
[56] Weinert K, Blum H, Kuhlenk tter B, et al. New methods for calculating the forcedistribution within belt grinding processes. Production Engineering,2007,1(3):285-289.
[57]张学工.模式识别.清华大学出版社,2010,第三版.
[58] Vapnik V N. The Nature of Statistical Learning Theory. Springer,2000,2nd.
[59] Vapnik V N. Statistical Learning Theory. John Wiley&Sons, Inc.,1998.
[60] Funahashi K. On the approximate realization of continuous mappings by neural networks.Neural Networks,1989,2(3):183-192.
[61] Funahashi K-i, Nakamura Y. Approximation of dynamical systems by continuous timerecurrent neural networks. Neural Networks,1993,6(6):801-806.
[62]周会娜.基于BP神经网络和遗传算法对工程陶瓷磨削力建模的研究[硕士学位论文].天津大学,2007.
[63]刘红星,魏强,屈梁生.基于人工神经网络建模的磨削烧伤预测.磨床与磨削,1996,(03):28-30.
[64] Wang Y T, Jan Y J. Grinding force models in finishing processes. Proceeding of the2001IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM): Vol.2.Como, Italy: Institute of Electrical and Electronics Engineers Inc.,2001:822-827.
[65]宋洪涛,宾鸿赞.把神经网络应用于丝杠磨削过程的建模与控制.光学精密工程,2001,(04):364-367.
[66] Lukosevicius M, Jaeger H. Reservoir Computing Approaches to Recurrent Neural NetworkTraining. Computer Science Review,2009,3(3):127-149
[67] Jaeger H. The "echo state" approach to analysing and training recurrent neural networks.German National Research Institute for Computer Science,2001. Report Nr.148.
[68] Jaeger H, Haas H. Harnessing Nonlinearity: Predicting Chaotic Systems and SavingEnergy in Wireless Communication. Science,2004,304(5667):78-80.
[69] Vapnik V. Estimation of Dependences Based on Empirical Data. Springer,2006,2nd.
[70]边肇祺,张学工.模式识别.清华大学出版社,2000,第二版.
[71] Scholkopf B, Smola A J. Learning with Kernels-Support Vector Machines, Regularization,Optimization, and Beyond. Dietterich T, editor. The MIT Press,2002.
[72] Vapnik V N. The Nature of Statistical Learning Theory. Springer,1995,1st Edition.
[73] Cortes C, Vapnik V. Support-Vector Networks. Machine Learning,1995,20(3):273-297.
[74] Jaeger H. Tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF andthe "echo state network" approach. German National Research Institute for ComputerScience,2002. Report Nr.159.
[75] Chang C-C, Lin C-J. LIBSVM: a Library for Support Vector Machines.2001.
[76] Wang W. Determination of the spread parameter in the Gaussian kernel for classificationand regression. Neurocomputing,2003,55(3-4):643-663.
[77] Cherkassky V, Ma Y. Practical selection of SVM parameters and noise estimation for SVMregression. Neural Networks,2004,17(1):113-126.
[78] Mohri M,. A R I. Stability bounds for non-iid processes. Proceeding of the Advances inNeural Information Processing Systems,2007.
[79] Amini M-R, Habrard A, Ralaivola L, et al. Learning from non IID Data:Theory,Algorithms and Practice. Proceeding of the ECML PKDD2009. Bled, Slovenia,2009.
[80] Zhou Z-H, Sun Y-Y, Li Y-F. Multi-instance learning by treating instances as non-I.I.D.samples. Proceeding of the26th Annual International Conference on Machine Learning.Montreal, Quebec, Canada: ACM,2009:1249-1256.
[81]潘志伟.统计学习算法：多分类及非独立同分布抽样下的回归[博士学位论文].中国科学技术大学,2009.
[82] Pan S J, Yang Q. A Survey on Transfer Learning. IEEE Transactions On Knowledge AndData Engineering,2010,22(10):1345-1359.
[83] Dai W, Yang Q, Xue G-R, et al. Boosting for transfer learning. Proceeding of the24thinternational conference on Machine learning. Corvalis, Oregon: ACM,2007:193-200.
[84] Wu P, Dietterich T G. Improving SVM accuracy by training on auxiliary data sources.Proceeding of the twenty-first international conference on Machine learning. Banff,Alberta, Canada: ACM,2004:871-878.
[85] Caruana R. Multitask Learning. Machine Learning,1997,28(1):41-75.
[86] Bakker B, Heskes T. Task clustering and gating for bayesian multitask learning. Journal ofMachine Learning Research,2003,4:83-99.
[87] DaumeIII H. Bayesian multitask learning with latent hierarchies. Proceeding of theTwenty-Fifth Conference on Uncertainty in Artificial Intelligence. Montreal, Quebec,Canada: AUAI Press,2009:135-142.
[88] Wu W, Li H, Hu Y, et al. Multi-Task Learning: Multiple Kernels for Multiple Tasks.Microsoft,2010. Report Nr. MSR-TR-2010-87.
[89] Babcock B, Babu S, Datar M, et al. Models and issues in data stream systems. Proceedingof the twenty-first ACM SIGMOD-SIGACT-SIGART symposium on Principles ofdatabase systems. Madison, Wisconsin: ACM,2002:1-16.
[90] Gaber M M, Zaslavsky A, Krishnaswamy S. Mining data streams: a review. SIGMODRecord,2005,34(2):18-26.
[91] Bifet A. Adaptive Learning and Mining for Data Streams and Frequent Patterns[Ph.D.dissertation]. Universitat Politecnica de Catalunya,2009.
[92] Raina R, Battle A, Lee H, et al. Self-taught learning: transfer learning from unlabeled data.Proceeding of the24th international conference on Machine learning. Corvalis, Oregon:ACM,2007:759-766.
[93] Jiang J, Zhai C. Instance weighting for domain adaptation in NLP. Proceeding of the45thAnnual Meeting of the Association for Computational Linguistics,2007:264-271.
[94]曹杰.统计机器翻译领域自适应问题研究[硕士学位论文].中国科学院,2010.
[95] Zadrozny B. Learning and evaluating classifiers under sample selection bias. Proceeding ofthe twenty-first international conference on Machine learning. Banff, Alberta, Canada:ACM,2004:114.
[96] Fan W, Davidson I, Zadrozny B, et al. An Improved Categorization of Classifier'sSensitivity on Sample Selection Bias. Proceeding of the Fifth IEEE InternationalConference on Data Mining,2005.
[97] Huang J, Smola A, Gretton A, et al. Correcting Sample Selection Bias by Unlabeled Data.Proceeding of the Advances in neural information processing systems19,2007:601-608.
[98] Shimodaira H. Improving predictive inference under covariate shift by weighting the log-likelihood function. Journal of Statistical Planning and Inference,2000,90(2):227-244.
[99] Sugiyama M, Suzuki T, Nakajima S, et al. Direct importance estimation for covariate shiftadaptation. Annals of the Institute of Statistical Mathematics,2008,60(4):699-746.
[100] Arnold A, Nallapati R, Cohen W W. A Comparative Study of Methods for TransductiveTransfer Learning. Proceeding of the Seventh IEEE International Conference on DataMining Workshops. Omaha, NE: IEEE Computer Society,2007:77-82.
[101] Dai W, Yang Q, Xue G-R, et al. Self-taught clustering. Proceeding of the25th internationalconference on Machine learning. Helsinki, Finland: ACM,2008:200-207.
[102] Jebara T. Multi-task feature and kernel selection for SVMs. Proceeding of the twenty-firstinternational conference on Machine learning. Banff, Alberta, Canada: ACM,2004:55.
[103] Argyriou A, Evgeniou T, Pontil M. Multi-task feature learning. Proceeding of theAdvances in Neural Information Processing Systems19,2007.
[104] Dai W, Xue G, Yang Q, et al. Co-clustering based classification for out-of-domaindocuments. Proceeding of the13th ACM SIGKDD international conference on Knowledgediscovery and data mining. San Jose, California, USA: ACM,2007:210-219.
[105] Lee S-I, Chatalbashev V, Vickrey D, et al. Learning a meta-level prior for feature relevancefrom multiple related tasks. Proceeding of the24th international conference on Machinelearning. Corvalis, Oregon: ACM,2007:489-496.
[106] Argyriou A, Micchelli C A, Pontil M, et al. A Spectral Regularization Framework forMulti-Task Structure Learning. Proceeding of the20th Annual Conference on NeuralInformation Processing Systems,2008:25-32.
[107] Liao X, Xue Y, Carin L. Logistic regression with an auxiliary data source. Proceeding ofthe22nd international conference on Machine learning. Bonn, Germany: ACM,2005:505-512.
[108] Quinonero-Candela J, Sugiyama M, Schwaighofer A, et al. Dataset Shift in MachineLearning. The MIT Press,2009.
[109] Evgeniou T, Pontil M. Regularized multi-task learning. Proceeding of the tenth ACMSIGKDD international conference on Knowledge discovery and data mining. Seattle, WA,USA: ACM,2004:109-117.
[110] Lawrence N D, Platt J C. Learning to learn with the informative vector machine.Proceeding of the twenty-first international conference on Machine learning. Banff,Alberta, Canada: ACM,2004:65.
[111] Schwaighofer A, Tresp V, K. Yu. Learning Gaussian Process Kernels via HierarchicalBayes. Proceeding of the17th Annual Conference on Neural Information ProcessingSystems,2005: pp.1209-1216.
[112] Bonilla E, Chai K M, Williams C. Multi-Task Gaussian Process Prediction. Proceeding ofthe20th Annual Conference on Neural Information Processing Systems,2008:153-160.
[113] Gao J, Fan W, Jiang J, et al. Knowledge transfer via multiple model local structuremapping. Proceeding of the14th ACM SIGKDD international conference on Knowledgediscovery and data mining. Las Vegas, Nevada, USA: ACM,2008:283-291.
[114] Mihalkova L, Huynh T, Mooney R. Mapping and revising markov logic networks fortransfer learning. Proceeding of the22nd National Conference on Artificial Intelligence(AAAI-07),2007:608-614.
[115] Davis J, Domingos P. Deep transfer via second-order Markov logic. Proceeding of the26thAnnual International Conference on Machine Learning. Montreal, Quebec, Canada: ACM,2009:217-224.
[116] Mihalkova L, Mooney R J. Transfer learning from minimal target data by mapping acrossrelational domains. Proceeding of the21st international jont conference on Artificalintelligence. Pasadena, California, USA: Morgan Kaufmann Publishers Inc.,2009:1163-1168.
[117]戴文渊.基于实例和特征的迁移学习算法研究[硕士学位论文].上海交通大学,2009.
[118] Vapnik V N.统计学习理论.徐建华,张学工, translator.电子工业出版社,2009.
[119]孙喆.融合先验知识的机器学习研究[博士学位论文].清华大学,2010.
[120] Lauer F, Bloch G. Incorporating prior knowledge in support vector machines forclassification: A review. Neurocomputing,2008,71(7-9):1578-1594.
[121] Lauer F, Bloch G. Incorporating prior knowledge in support vector regression. MachineLearning,2008,70(1):89-118.
[122] Epshteyn A, DeJong G. Rotational prior knowledge for SVMs. Proceeding of the SixteenthEuropean Conference on Machine Learning: Vol.3720. Porto, Portugal,2005:108-119.
[123] Ratsaby J, Maiorov V. On the value of partial information for learning from examples.Journal Of Complexity,1997,13(4):509-544.
[124] Han F, Huang D-S. Improved extreme learning machine for function approximation byencoding a priori information. Neurocomputing,2006,69(16-18):2369-2373.
[125] Joerding W H, Meador J L. Encoding a priori information in feedforward networks. NeuralNetworks,1991,4(6):847-856.
[126] Mangasarian O L, Shavlik J W, Wild E W. Knowledge-Based Kernel Approximation.Journal of Machine Learning Research,2004,5:1127-1141.
[127] Pelckmans K, Espinoza M, Brabanter J, et al. Primal-Dual Monotone Kernel Regression.Neural Processing Letters,2005,22(2):171-182.
[128] Poggio T, Vetter T. Recognition and Structure from one2D Model View: Observations onPrototypes, Object Classes and Symmetries. Massachusetts Institute of Technology,1992.Report Nr. AIM-1347.
[129] Abu-Mostafa Y S. Hints. Neural Computation,1995,7(4):639-671.
[130] Sch lkopf B, Burges C, Vapnik V. Incorporating invariances in support vector learningmachines. Proceeding of the1996International Conference on Artificial Neural Networks.Bochum, Germany: Springer-Verlag,1996:47-52.
[131] Niyogi P, Poggio T, Girosi F. Incorporating Prior Information in Machine Learning byCreating Virtual Examples. Proceedings on the IEEE,1998,86(11):2196-2209.
[132] Johnson K L. Contact Mechanics.徐秉业, translator. Cambridge University Press,1985.
[133] Lv H, Song Y, Jia P, et al. An Adaptive Modeling Approach Based on ESN for RoboticBelt Grinding. Proceeding of the2010IEEE International Conference on Information andAutomation. Harbin, Heilongjiang, China: IEEE,2010:787-792.
[134]高理富,宋宁,葛运建, et al.航天机器人用六维腕力传感器动态特性研究.机器人,2002,(04):319-323.
[135] Liang W, Song Y, Lv H, et al. A novel control method for robotic belt grinding based onSVM and PSO algorithm. Proceeding of the2010International Conference on IntelligentComputation Technology and Automation: Vol.1. Changsha, China: IEEE ComputerSociety,2010:258-261.
[136] Liang W, Song Y, Lv H, et al. An effective trajectory optimization method for robotic beltgrinding based on intelligent algorithm. Proceeding of the2010IEEE InternationalConference on Robotics and Biomimetics. Tianjin, China: IEEE,2010.
[137] Kennedy J, Eberhart R. Particle swarm optimization. Proceeding of the IEEE InternationalConference on Neural Networks: Vol.4. Perth, WA, Australia: IEEE,1995:1942-1948.