神经网络及其在大地测量数据处理中的应用
|
摘要
本文主要研究了神经网络在大地测量若干领域的应用,论文的主要内容概括如下:
1.首先介绍了BP神经网络、Elman神经网络和Hopfield神经网络的基本原理,并分析三种神经网络的优缺点。
2.阐述了目前动态数据处理中常用的几种滤波方法,并介绍了关于贝叶斯估计的有关理论。
3.讨论了神经网络中最常用的学习算法——BP算法,分析了BP算法存在的缺点;并介绍了基于EKF、UKF滤波的神经网络学习算法,该方法能够提高前向神经网络收敛速度和泛化能力。
4.在EKF、UKF训练神经网络连接权的基础上,提出了自适应EKF、自适应UKF训练神经网络的思路,进一步提高了神经网络的性能。并用高程拟合算例证明了基于自适应滤波训练连接权算法的有效性。
5.对于Hopfield神经网络来说,选择合理的能量函数尤为重要,因此论述了ALR函数作为能量函数的Hopfield神经网络;另外,将抗差估计原理引入Hopfield神经网络,进一步提高了神经网络的抗差能力;分别采用高程拟合和水准网平差验证了这两种算法。
6.研究了神经网络在导航中的应用。讨论了动态数据异常检验的三种方法,并用实测数据进行了验证;采用具有反馈层的Elman神经网络处理捷联惯导精对准问题,提高了精对准的精度;采用基于自适应Kalman滤波与BP算法组合的算法处理惯导数据;将抗差自适应滤波理论应用于UKF,进一步完善了UKF理论,并用模拟算例进行了验证;针对粒子滤波的粒子退化问题,提出了自适应UKF重点方法,进一步提高粒子滤波的精度。
The applications of neural network in geodetic data processing is researched in this dissertation. The main works and contributions are summarized as follows:
1. The theories of BP, Elman and Hopfield neural networks are introduced, the merits and shortcomings of the three kinds of neural networks are analyzed.
2. Bayes estimation theory also is introduced. Several kinds of filters, which are widely applied in the dynamic data processing, are described.
3. BP algorithm, which can overcome the training of the multi-layer feed forward network, is studied and analyzed. A new algorithm that uses EKF or UKF to train the weight of the neural network is proposed, which can improve the generalization ability and divergence of the BP algorithm.
4. On the base of the new algorithms using EKF or UKF to train the weight, the adaptive EKF or UKF is introduced to compute the weight. The algorithms can further improve the generalization ability and divergence of neural network in the computation example of height fitting.
5. Energy function is important in Hopfield neural network, the augmented Lagrangian with regularization(ALR) is chosen as an energy function. The robust estimation principle is applied to improve the robustness of Hopfield neural network. It proves that the two algorithms are efficient in examples of height fitting and adjustment of leveling network.
6. The application of neural network in navigation is discussed. Three kinds of outlier detection outlier in Kalman filtering are introduced and tested in dynamic data; Elman neural network is used to improve the accuracy of refined initial alignment; the new algorithm combining the adaptive Kalman filtering and BP algorithm is applied in the data of processing INS; the theory of UKF combined with the robust adaptive filtering is applied in a simulated example; adaptive UKF is taken to improve the precision of sampling for particle filter has degeneracy.
1.首先介绍了BP神经网络、Elman神经网络和Hopfield神经网络的基本原理,并分析三种神经网络的优缺点。
2.阐述了目前动态数据处理中常用的几种滤波方法,并介绍了关于贝叶斯估计的有关理论。
3.讨论了神经网络中最常用的学习算法——BP算法,分析了BP算法存在的缺点;并介绍了基于EKF、UKF滤波的神经网络学习算法,该方法能够提高前向神经网络收敛速度和泛化能力。
4.在EKF、UKF训练神经网络连接权的基础上,提出了自适应EKF、自适应UKF训练神经网络的思路,进一步提高了神经网络的性能。并用高程拟合算例证明了基于自适应滤波训练连接权算法的有效性。
5.对于Hopfield神经网络来说,选择合理的能量函数尤为重要,因此论述了ALR函数作为能量函数的Hopfield神经网络;另外,将抗差估计原理引入Hopfield神经网络,进一步提高了神经网络的抗差能力;分别采用高程拟合和水准网平差验证了这两种算法。
6.研究了神经网络在导航中的应用。讨论了动态数据异常检验的三种方法,并用实测数据进行了验证;采用具有反馈层的Elman神经网络处理捷联惯导精对准问题,提高了精对准的精度;采用基于自适应Kalman滤波与BP算法组合的算法处理惯导数据;将抗差自适应滤波理论应用于UKF,进一步完善了UKF理论,并用模拟算例进行了验证;针对粒子滤波的粒子退化问题,提出了自适应UKF重点方法,进一步提高粒子滤波的精度。
The applications of neural network in geodetic data processing is researched in this dissertation. The main works and contributions are summarized as follows:
1. The theories of BP, Elman and Hopfield neural networks are introduced, the merits and shortcomings of the three kinds of neural networks are analyzed.
2. Bayes estimation theory also is introduced. Several kinds of filters, which are widely applied in the dynamic data processing, are described.
3. BP algorithm, which can overcome the training of the multi-layer feed forward network, is studied and analyzed. A new algorithm that uses EKF or UKF to train the weight of the neural network is proposed, which can improve the generalization ability and divergence of the BP algorithm.
4. On the base of the new algorithms using EKF or UKF to train the weight, the adaptive EKF or UKF is introduced to compute the weight. The algorithms can further improve the generalization ability and divergence of neural network in the computation example of height fitting.
5. Energy function is important in Hopfield neural network, the augmented Lagrangian with regularization(ALR) is chosen as an energy function. The robust estimation principle is applied to improve the robustness of Hopfield neural network. It proves that the two algorithms are efficient in examples of height fitting and adjustment of leveling network.
6. The application of neural network in navigation is discussed. Three kinds of outlier detection outlier in Kalman filtering are introduced and tested in dynamic data; Elman neural network is used to improve the accuracy of refined initial alignment; the new algorithm combining the adaptive Kalman filtering and BP algorithm is applied in the data of processing INS; the theory of UKF combined with the robust adaptive filtering is applied in a simulated example; adaptive UKF is taken to improve the precision of sampling for particle filter has degeneracy.
引文
[1]陈哲.捷联惯导系统原理[M].北京:宇航出版社,1986
[2]董长虹.神经网络与应用[M].国防工业出版社,2005
[3]付梦印,邓志红,张继伟.Kalman滤波理论及其在导航系统中的应用[M].北京:科学出版社,2003
[4]高为广,杨元喜,张婷.神经网络辅助的GPS/INS组合导航自适应滤波算法[J].测绘学报,2007,36(1)21-30
[5]郭鹏.Hopfield网络及其在电厂中的应用[D].华北电力大学博士论文,2004
[6]何海波,杨元喜.GPS动态测量连续周跳检验[J].测绘学报,1999,28(3):199-204
[7]何芳,陈收.基于扩展Kalman滤波的神经网络学习算法在股票预测中的应用[J].系统工程,2003,21(6):75-79
[8]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策,2005,20(4):361-365
[9]胡伍生.GPS精密高程测量理论与方法及其应用研究[D].河海大学博士论文,2001
[10]胡伍生.神经网络理论及其工程应用[M].测绘出版社,2006
[11]李朝奎,徐望国,曾卓乔,等.非线性函数空间平差方程的解法及其特征[J].测绘学院学报,2001,18(1):8-11
[12]李磊磊,陈家斌,谢玲,等.粒子滤波方法在GPS/DR组合导航系统中的应用[J].微电子学与计算机,2004,21(10):97-99
[13]李涛.非线性滤波方法在导航系统中的应用研究[D].国防科技大学博士论文,2003
[14]廖晓峰,刘光远,虞厥邦.基于几种误差准则的稳健BP算法[J].信号处理,1997,13(3):235-240
[15]刘光远,廖晓峰,邱玉辉.基于稳健误差估计器的快速BP算法[J].计算机科学,1997,24(2):66-68
[16]刘光远,廖晓峰,虞厥邦,等.一种神经网络稳健估计方法的推广性研究[J].电子科学学刊,1999,21(1):128-131
[17]刘光远,邱玉辉,廖晓峰,等.一种新颖的神经网络稳健估计方法[J].计算机研究与发展,1999,36(5):567-571
[18]鲁铁定,周世健,臧得彦.关于BP神经网络转换GPS高程的若干问题[J].测绘通报,2003,8:7-10
[19]马儒宁.神经网络与支持向量机相关问题研究[D].复旦大学博士论文,2003
[20]马野,王孝通,戴耀.基于UKF的神经网络自适应全局信息融合方法[J].电子学报,2005,33(10):1914-4916
[21]缪玲娟.小波分析在光纤陀螺信号中的应用研究[J].宇航学报,2000,21(1):43-46
[22]聂莉娟.捷联惯导系统初始对准滤波技术研究[D].硕士论文,哈尔滨工程大学,2004
[23]覃光华,丁晶,刘国东.自适应BP算法及其在河道洪水预报上的应用[J].水科学进展,2002,13(1):37-41
[24]覃光华.人工神经网络技术及其应用[D].博士论文,四川大学,2003
[25]秦永元.卡尔曼滤波与组合导航原理[M].西安:西北工业大学出版社,1998
[26]沈慧峰.基于粒子滤波的多目标跟踪算法研究[D].浙江大学硕士论文,2006
[27]史玉峰,曹俊如.基于遗传BP神经网络的变形数据分析处理[J].矿业研究与开发,2004,24(4):8-10
[28]唐涛,张涯辉,黄永梅.粒子在被动定位跟踪中的应用[J].仪器仪表学报,2005,26(8):211-214
[29]陶本藻.卡尔曼滤波模型误差的识别[J].地壳形变与地震,1999,19,15-20
[30]田晓宇,李明干,刘沛:基于Kalman滤波的神经网络学习算法及应用[J].计算机与数字工程,2005,33(2):40-42
[31]万德钧,房建成.惯性导航初始对准[M].南京:东南大学出版社,1998
[32]王国富,朱建军.含有粗差观测值的自适应滤波[J].测绘通报,2004,4:19-21
[33]王甫红,刘基余.星载GPS载波相位测量的周跳探测方法研究[J].武汉大学学报:信息科学版,2004,29(9):772-774
[34]王新洲.非线性模型线性近似的容许曲率[J].武汉测绘科技大学学报,1997(2):119-121
[35]王新洲.非线性模型参数估计的直接解法[J].武汉测绘科技大学学报,1999a(1):64-67
[36]王新洲.非线性模型能否线性化的实用判据[J].武汉测绘科技大学学报,1999b(2):145-148
[37]王新洲.非线性模型平差中单位权方差的估计[J].武汉测绘科技大学学报,2000(4):358-361
[38]王新洲,邓新升.大坝变形预报的模糊神经网络模型[J].武汉大学学报,信息科学 版,2005,30(7):588-591
[39]王耀南.基于Kalman滤波的神经网络快速学习算法及应用[J].系统工程与电子技术,1997,(3):44-47
[40]王志刚,芮国瑞,胡昊.UPF算法及其在目标跟踪中的应用[J].雷达科学与技术,2005,3(1):20-24
[41]魏海坤.神经网络结构设计的理论与方法.国防工业出版社,2005
[42]吴玲,卢发兴,刘忠.UKF算法及其在目标被动跟踪中的应用[J].系统工程与电子技术,2005,27(1):49-51
[43]夏克文,李昌彪,沈钧毅.前向神经网络隐含层节点数的一种优化算法[J].计算机科学,2005,32(10):143-145
[44]熊伟,何友,张晶炜.多传感器多目标粒子滤波算法[J].光电工程,2005,32(4):1-4
[45]徐培亮.非线性函数的协方差传播公式[J].武汉测绘科技大学学报,1986(2):92-99
[46]许丽佳,陈阳丹,崔平远.基于神经网络的组合导航状态估计[J].中国惯性技术学报,2004,12(2):40-46
[47]杨建刚.人工神经网络使用教程[M].浙江大学出版社,2001
[48]杨旭,程杨,曹喜滨,等.粒子滤波在卫星轨道确定中的应用[J].控制理论与应用,2005,22(4):573-577
[49]杨元喜.参数M估计及验后精度综述[J].测绘通报,1991(6):6-10
[50]杨元喜.相关观测抗差估计[A].抗差估计论文集[C],测绘出版社,1992c,14-21
[51]杨元喜.抗差估计理论及其应用[M].北京:八一出版社,1993
[52]杨元喜.抗差估计的概念及其任务[J].测绘通报,1994(4):36-39
[53]杨元喜.异常影响诊断与抗差估计[J].测绘通报,1994(5):34-36
[54]杨元喜.等价权原理-参数平差模型的抗差最小二乘解[J].测绘通报,1994(6):33-35
[55]杨元喜.动态系统的抗差Kalman滤波[J].郑州测绘学院学报,1997,14(2):79-84
[56]杨元喜.Kalman滤波异常检验[J].测绘科学与工程,2005,25(4):1-4
[57]杨元喜.自适应动态导航定位[M].测绘出版社,2006
[58]杨元喜,何海波,徐天河.论动态自适应滤波[J].测绘学报,2001,30(4):293-298
[59]杨元喜,高为广.基于方差分量估计的自适应融合导航[J].测绘学报,2004a,33(1): 22-26
[60]杨元喜、高为广.基于多传感器观测信息抗差估计的自适应融合导航.武大学报信息科学版,2004b,29(10),885-888
[61]杨元喜,文援兰.卫星精密轨道综合自适应抗差滤波技术[J].中国科学(D辑),2003,32(11):1112-1119
[62]杨元喜,徐天河.基于移动开窗法协方差估计和方差分量估计的自适应滤波[J].武汉大学学报,信息科学版,2003,28(6):714-718
[63]杨元喜,张双成,高为广.GPS导航解算中的几种非线性Kalman滤波的理论分析与比较[J].测绘工程,2005,14(3):4-7
[64]姚剑敏.粒子滤波跟踪方法研究[D].中科院博士论文,2004
[65]俞济祥.卡尔曼滤波及其在惯性导航中的应用[M].西安:航空专业教材编审组,1984
[66]袁信,俞济祥,陈哲.导航系统[M].北京:航空工业出版社,1992
[67]张鹏,王金城.自适应滤波算法的神经网络实现.电气传动自动化,2003,25(5):28-29
[68]张红梅,邓正隆,高玉凯.UKF在基于修正罗德里格参数的飞行器姿态确定中的应用[J].宇航学报,2005,26(2):164-167
[69]张勤.非线性最小二乘理论及其在GPS定位中应用研究[D].武汉大学博士论文,2002
[70]张友民,戴冠中,张洪才.一种前馈神经网络的卡尔曼滤波学习方法[J].信息与控制,1994,23(2):113-118
[71]赵启林,卓家寿.BP网络的最大误差学习算法[J].河海大学学报,2000,28(1):113-115
[72]郑高峰.带动量项的BP神经网络收敛性分析[D].大连理工大学硕士论文,2005
[73]周志华,曹存根.神经网络及其应用[M].北京:清华大学出版社,2004
[74]朱利锋,鲍其莲,张炎华.捷联惯导系统初始对准的H_∞滤波及卡尔曼滤波比较研究[J].中国惯性技术学报,2005,13(3):4-9
[75]Athanasios Dermanis and Fernando Sanso.Nonlinear Estimation Problems for Nonlinear Models[J].Manuscripa Geodatica,1995(20):110-122
[76]Blank T B,Brown S D.Adaptive,global,extended Kalman filters for training feed-forward neural networks[J].Journal of Chemometrics,1994,(8):391-407
[77]Bucy R S,Renne K D.Digital synthesis of nonlinear filter[J].Automatic,1971,7(3): 287-289
[78]Butes. D. M. and Watte. D.G . Nonlinear Regression Analysis and Its Applications[J]. John Wiley & Sons, Inc, 1988
[79]Cichocki A and Unbehauen R. Switched-capacitor artificial neural networks for nonlinear optimization with constraints [J], Proc. IS-CAS-90 Symp., May, 1990, pp. 2809-2812
[80]Cichocki A and Unbehauen R. Neural networks for solving systems of linear equations and related problems[J]. IEEE Trans. Circuits and Systems-1: Fundamental Theory and Applications, 1992, 39(2): 124-138
[81]Dah Jing Jwo. GPS navigation solutions by analogue neural network least-squares processors. The Journal of navigation, 2005, 58:105-118
[82]E A. Wan, R van der Merwe.. The unscented Kalman filter for nonlinear estimation[J]. in Proc. of IEEE Symposium 2000, Lake Louise Alberta, Canada, Oct. 2000
[83]Hopfield J J. Neural networks and physical systems with emergent collective computational abilities[J], Proceedings of the National Academy of Sciences, 1982, 79: 2554-2558
[84]Hopfield J J. Neurons with graded response have collective computational properties like those of two-state neurons[J], Proc. Nat. Acad. Sci U.S.A., 1984 81: 3088-3092
[85]Kazufumi Ito Kaiqi Xiong. Gaussian filters for nonlinear filtering problems[J]. IEEE Transactions on Automatic Control, 2000, 45(5): 910-927
[86]Koch K. R. Einfuehrung in Die Bayes-Statistik [M]. Springer, Berlin Heidelberg, 2000, New York
[87]Koch K R, Yang Y . Robust Kalman Filter for Rank Deficient Observation Model[J]. Journal of Geodesy, 1998, 72(8): 436-441
[88]Lary D J., Mussa H Y.. Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations [J]. Atmos. Chem. Phys. Discuss., 2004,4: 3653-3667
[89]Lu G, Lachapelle G. Reliability analysis for kinematic GPS position and velocity estimation[J]. Proceedings of IAG Symposium 107 on Kinematic Systems in Geodesy, Surveying and Remote sensing. New York :Springer Verlag, 1990
[90]Miima J. B., Niemeier W. Kraus B. Aneural network approach to modeling geodetic deformation[C]. First International Symposium on Robust Statistics and Fuzzy Techniques in Geodesy and GIS, 2001
[91]R van der Merwe, Wan E A. Efficient derivative-free Kalman filters for online learning. In European Symposium on Artificial Neural Networks, Bruges, Belgium, 2001
[92]Rudolph van der Merwe, Nando de Freitas, Arnaud Doucet, etc. The unscented particle filter[J]. Adv. Neural Inform, Process. Syst.,Dec. 2000
[93]Salzmann M. Real-time adaptation for model errors in dynamic systems[J]. Bulletin Geodesigue, 1995,69:81-89
[94]S J. Julier, J K. Uhlmann. A general method for approximating nonlinear transformation of probability distributions[J] . ( 1994 Aug ) [online] . available: http://www.robots.ox.ac.uk/～siju
[95]S J. Julier, J K. Uhlmann. A new approach for filtering nonlinear system[C]. in Proc. Am. Conf., Seattle, WA, 1995
[96]S J. Julier, J K. Uhlmann. A new approach for nonlinear transformation of means and covariance in filters and estimators[J]. IEEE Trans. Automatic Control, 2000, 45: 477-482
[97]S J. Julier, Simon J. H. F. Durrant-Whyte. Navigation and parameter estimation of high speed road vehicles[J]. In Robotics and Automation Conference, 1995, 101-105
[98]Sunahara Y. An approximate method of state estimation for nonlinear dynamical systems[J]. Joint Automatic Control Conf, Univ. of Colorado, 1969
[99]Tank D W and Hopfield J J. Simple 'neural' optimization networks: an A/D converter, signal decision circuits, and a linear programming circuit[J]. IEEE Trans. Circuits Syst., 1988, CAS-33: 533-562
[100]Teunissen P J G. Quality control in navigation systems[J]. IEEE Aerospace and Electronic Systems Magazine, 1990, 5(7): 35-41
[101]Teunissen P J G, Salzmann M A. A recursive slippage test for use in state-space filtering[J]. Manuscripta Geodaetica, 1989,14(6): 383-390
[102]T. Saito. The Nonlinear Least Squares of Condition Equations[J]. Bull, Geod, 1973
[103]Yang Y. Robust estimation for dependent observations[J]. Mans. Geod. 1994, 19: 10-17
[104] Yang Y, Gao W. Comparison of adaptive factors on navigation results[J]. The Journal of Navigation, 2005, 58(3): 471-478
[105]Yang Y,Gao W.An optimal adaptive Kalman filter[J].J Geod,2006,80:177-183.
[106]Yang Y,He H,Xu G.Adaptively robust filtering for kinematic geodetic positioning[J].Journal of Geodesy,2001,75(2/3):109-116
[107]Yang Y,Xu T,He H.On adaptively kinematic filtering[J].Selected Paper for English of Acta Geodetica Cartographica Sinica,2001b,25-32
[108]Yang Y.,Xu T.and Song L.Robust estimation of variance components with application in GPS network adjustment[J].Journal of Surveying Engineering,2005,131(4):107-112
[109]Zhang S and Constantinides A G.Langrange Programming Neural Networks[J],IEEE TRANSACTIONS ONCIRCUITS AND SYSTEMS- Ⅱ:ANALOG AND SIGNAL PROCESSING 1992,39(7):441-452
[2]董长虹.神经网络与应用[M].国防工业出版社,2005
[3]付梦印,邓志红,张继伟.Kalman滤波理论及其在导航系统中的应用[M].北京:科学出版社,2003
[4]高为广,杨元喜,张婷.神经网络辅助的GPS/INS组合导航自适应滤波算法[J].测绘学报,2007,36(1)21-30
[5]郭鹏.Hopfield网络及其在电厂中的应用[D].华北电力大学博士论文,2004
[6]何海波,杨元喜.GPS动态测量连续周跳检验[J].测绘学报,1999,28(3):199-204
[7]何芳,陈收.基于扩展Kalman滤波的神经网络学习算法在股票预测中的应用[J].系统工程,2003,21(6):75-79
[8]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策,2005,20(4):361-365
[9]胡伍生.GPS精密高程测量理论与方法及其应用研究[D].河海大学博士论文,2001
[10]胡伍生.神经网络理论及其工程应用[M].测绘出版社,2006
[11]李朝奎,徐望国,曾卓乔,等.非线性函数空间平差方程的解法及其特征[J].测绘学院学报,2001,18(1):8-11
[12]李磊磊,陈家斌,谢玲,等.粒子滤波方法在GPS/DR组合导航系统中的应用[J].微电子学与计算机,2004,21(10):97-99
[13]李涛.非线性滤波方法在导航系统中的应用研究[D].国防科技大学博士论文,2003
[14]廖晓峰,刘光远,虞厥邦.基于几种误差准则的稳健BP算法[J].信号处理,1997,13(3):235-240
[15]刘光远,廖晓峰,邱玉辉.基于稳健误差估计器的快速BP算法[J].计算机科学,1997,24(2):66-68
[16]刘光远,廖晓峰,虞厥邦,等.一种神经网络稳健估计方法的推广性研究[J].电子科学学刊,1999,21(1):128-131
[17]刘光远,邱玉辉,廖晓峰,等.一种新颖的神经网络稳健估计方法[J].计算机研究与发展,1999,36(5):567-571
[18]鲁铁定,周世健,臧得彦.关于BP神经网络转换GPS高程的若干问题[J].测绘通报,2003,8:7-10
[19]马儒宁.神经网络与支持向量机相关问题研究[D].复旦大学博士论文,2003
[20]马野,王孝通,戴耀.基于UKF的神经网络自适应全局信息融合方法[J].电子学报,2005,33(10):1914-4916
[21]缪玲娟.小波分析在光纤陀螺信号中的应用研究[J].宇航学报,2000,21(1):43-46
[22]聂莉娟.捷联惯导系统初始对准滤波技术研究[D].硕士论文,哈尔滨工程大学,2004
[23]覃光华,丁晶,刘国东.自适应BP算法及其在河道洪水预报上的应用[J].水科学进展,2002,13(1):37-41
[24]覃光华.人工神经网络技术及其应用[D].博士论文,四川大学,2003
[25]秦永元.卡尔曼滤波与组合导航原理[M].西安:西北工业大学出版社,1998
[26]沈慧峰.基于粒子滤波的多目标跟踪算法研究[D].浙江大学硕士论文,2006
[27]史玉峰,曹俊如.基于遗传BP神经网络的变形数据分析处理[J].矿业研究与开发,2004,24(4):8-10
[28]唐涛,张涯辉,黄永梅.粒子在被动定位跟踪中的应用[J].仪器仪表学报,2005,26(8):211-214
[29]陶本藻.卡尔曼滤波模型误差的识别[J].地壳形变与地震,1999,19,15-20
[30]田晓宇,李明干,刘沛:基于Kalman滤波的神经网络学习算法及应用[J].计算机与数字工程,2005,33(2):40-42
[31]万德钧,房建成.惯性导航初始对准[M].南京:东南大学出版社,1998
[32]王国富,朱建军.含有粗差观测值的自适应滤波[J].测绘通报,2004,4:19-21
[33]王甫红,刘基余.星载GPS载波相位测量的周跳探测方法研究[J].武汉大学学报:信息科学版,2004,29(9):772-774
[34]王新洲.非线性模型线性近似的容许曲率[J].武汉测绘科技大学学报,1997(2):119-121
[35]王新洲.非线性模型参数估计的直接解法[J].武汉测绘科技大学学报,1999a(1):64-67
[36]王新洲.非线性模型能否线性化的实用判据[J].武汉测绘科技大学学报,1999b(2):145-148
[37]王新洲.非线性模型平差中单位权方差的估计[J].武汉测绘科技大学学报,2000(4):358-361
[38]王新洲,邓新升.大坝变形预报的模糊神经网络模型[J].武汉大学学报,信息科学 版,2005,30(7):588-591
[39]王耀南.基于Kalman滤波的神经网络快速学习算法及应用[J].系统工程与电子技术,1997,(3):44-47
[40]王志刚,芮国瑞,胡昊.UPF算法及其在目标跟踪中的应用[J].雷达科学与技术,2005,3(1):20-24
[41]魏海坤.神经网络结构设计的理论与方法.国防工业出版社,2005
[42]吴玲,卢发兴,刘忠.UKF算法及其在目标被动跟踪中的应用[J].系统工程与电子技术,2005,27(1):49-51
[43]夏克文,李昌彪,沈钧毅.前向神经网络隐含层节点数的一种优化算法[J].计算机科学,2005,32(10):143-145
[44]熊伟,何友,张晶炜.多传感器多目标粒子滤波算法[J].光电工程,2005,32(4):1-4
[45]徐培亮.非线性函数的协方差传播公式[J].武汉测绘科技大学学报,1986(2):92-99
[46]许丽佳,陈阳丹,崔平远.基于神经网络的组合导航状态估计[J].中国惯性技术学报,2004,12(2):40-46
[47]杨建刚.人工神经网络使用教程[M].浙江大学出版社,2001
[48]杨旭,程杨,曹喜滨,等.粒子滤波在卫星轨道确定中的应用[J].控制理论与应用,2005,22(4):573-577
[49]杨元喜.参数M估计及验后精度综述[J].测绘通报,1991(6):6-10
[50]杨元喜.相关观测抗差估计[A].抗差估计论文集[C],测绘出版社,1992c,14-21
[51]杨元喜.抗差估计理论及其应用[M].北京:八一出版社,1993
[52]杨元喜.抗差估计的概念及其任务[J].测绘通报,1994(4):36-39
[53]杨元喜.异常影响诊断与抗差估计[J].测绘通报,1994(5):34-36
[54]杨元喜.等价权原理-参数平差模型的抗差最小二乘解[J].测绘通报,1994(6):33-35
[55]杨元喜.动态系统的抗差Kalman滤波[J].郑州测绘学院学报,1997,14(2):79-84
[56]杨元喜.Kalman滤波异常检验[J].测绘科学与工程,2005,25(4):1-4
[57]杨元喜.自适应动态导航定位[M].测绘出版社,2006
[58]杨元喜,何海波,徐天河.论动态自适应滤波[J].测绘学报,2001,30(4):293-298
[59]杨元喜,高为广.基于方差分量估计的自适应融合导航[J].测绘学报,2004a,33(1): 22-26
[60]杨元喜、高为广.基于多传感器观测信息抗差估计的自适应融合导航.武大学报信息科学版,2004b,29(10),885-888
[61]杨元喜,文援兰.卫星精密轨道综合自适应抗差滤波技术[J].中国科学(D辑),2003,32(11):1112-1119
[62]杨元喜,徐天河.基于移动开窗法协方差估计和方差分量估计的自适应滤波[J].武汉大学学报,信息科学版,2003,28(6):714-718
[63]杨元喜,张双成,高为广.GPS导航解算中的几种非线性Kalman滤波的理论分析与比较[J].测绘工程,2005,14(3):4-7
[64]姚剑敏.粒子滤波跟踪方法研究[D].中科院博士论文,2004
[65]俞济祥.卡尔曼滤波及其在惯性导航中的应用[M].西安:航空专业教材编审组,1984
[66]袁信,俞济祥,陈哲.导航系统[M].北京:航空工业出版社,1992
[67]张鹏,王金城.自适应滤波算法的神经网络实现.电气传动自动化,2003,25(5):28-29
[68]张红梅,邓正隆,高玉凯.UKF在基于修正罗德里格参数的飞行器姿态确定中的应用[J].宇航学报,2005,26(2):164-167
[69]张勤.非线性最小二乘理论及其在GPS定位中应用研究[D].武汉大学博士论文,2002
[70]张友民,戴冠中,张洪才.一种前馈神经网络的卡尔曼滤波学习方法[J].信息与控制,1994,23(2):113-118
[71]赵启林,卓家寿.BP网络的最大误差学习算法[J].河海大学学报,2000,28(1):113-115
[72]郑高峰.带动量项的BP神经网络收敛性分析[D].大连理工大学硕士论文,2005
[73]周志华,曹存根.神经网络及其应用[M].北京:清华大学出版社,2004
[74]朱利锋,鲍其莲,张炎华.捷联惯导系统初始对准的H_∞滤波及卡尔曼滤波比较研究[J].中国惯性技术学报,2005,13(3):4-9
[75]Athanasios Dermanis and Fernando Sanso.Nonlinear Estimation Problems for Nonlinear Models[J].Manuscripa Geodatica,1995(20):110-122
[76]Blank T B,Brown S D.Adaptive,global,extended Kalman filters for training feed-forward neural networks[J].Journal of Chemometrics,1994,(8):391-407
[77]Bucy R S,Renne K D.Digital synthesis of nonlinear filter[J].Automatic,1971,7(3): 287-289
[78]Butes. D. M. and Watte. D.G . Nonlinear Regression Analysis and Its Applications[J]. John Wiley & Sons, Inc, 1988
[79]Cichocki A and Unbehauen R. Switched-capacitor artificial neural networks for nonlinear optimization with constraints [J], Proc. IS-CAS-90 Symp., May, 1990, pp. 2809-2812
[80]Cichocki A and Unbehauen R. Neural networks for solving systems of linear equations and related problems[J]. IEEE Trans. Circuits and Systems-1: Fundamental Theory and Applications, 1992, 39(2): 124-138
[81]Dah Jing Jwo. GPS navigation solutions by analogue neural network least-squares processors. The Journal of navigation, 2005, 58:105-118
[82]E A. Wan, R van der Merwe.. The unscented Kalman filter for nonlinear estimation[J]. in Proc. of IEEE Symposium 2000, Lake Louise Alberta, Canada, Oct. 2000
[83]Hopfield J J. Neural networks and physical systems with emergent collective computational abilities[J], Proceedings of the National Academy of Sciences, 1982, 79: 2554-2558
[84]Hopfield J J. Neurons with graded response have collective computational properties like those of two-state neurons[J], Proc. Nat. Acad. Sci U.S.A., 1984 81: 3088-3092
[85]Kazufumi Ito Kaiqi Xiong. Gaussian filters for nonlinear filtering problems[J]. IEEE Transactions on Automatic Control, 2000, 45(5): 910-927
[86]Koch K. R. Einfuehrung in Die Bayes-Statistik [M]. Springer, Berlin Heidelberg, 2000, New York
[87]Koch K R, Yang Y . Robust Kalman Filter for Rank Deficient Observation Model[J]. Journal of Geodesy, 1998, 72(8): 436-441
[88]Lary D J., Mussa H Y.. Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations [J]. Atmos. Chem. Phys. Discuss., 2004,4: 3653-3667
[89]Lu G, Lachapelle G. Reliability analysis for kinematic GPS position and velocity estimation[J]. Proceedings of IAG Symposium 107 on Kinematic Systems in Geodesy, Surveying and Remote sensing. New York :Springer Verlag, 1990
[90]Miima J. B., Niemeier W. Kraus B. Aneural network approach to modeling geodetic deformation[C]. First International Symposium on Robust Statistics and Fuzzy Techniques in Geodesy and GIS, 2001
[91]R van der Merwe, Wan E A. Efficient derivative-free Kalman filters for online learning. In European Symposium on Artificial Neural Networks, Bruges, Belgium, 2001
[92]Rudolph van der Merwe, Nando de Freitas, Arnaud Doucet, etc. The unscented particle filter[J]. Adv. Neural Inform, Process. Syst.,Dec. 2000
[93]Salzmann M. Real-time adaptation for model errors in dynamic systems[J]. Bulletin Geodesigue, 1995,69:81-89
[94]S J. Julier, J K. Uhlmann. A general method for approximating nonlinear transformation of probability distributions[J] . ( 1994 Aug ) [online] . available: http://www.robots.ox.ac.uk/～siju
[95]S J. Julier, J K. Uhlmann. A new approach for filtering nonlinear system[C]. in Proc. Am. Conf., Seattle, WA, 1995
[96]S J. Julier, J K. Uhlmann. A new approach for nonlinear transformation of means and covariance in filters and estimators[J]. IEEE Trans. Automatic Control, 2000, 45: 477-482
[97]S J. Julier, Simon J. H. F. Durrant-Whyte. Navigation and parameter estimation of high speed road vehicles[J]. In Robotics and Automation Conference, 1995, 101-105
[98]Sunahara Y. An approximate method of state estimation for nonlinear dynamical systems[J]. Joint Automatic Control Conf, Univ. of Colorado, 1969
[99]Tank D W and Hopfield J J. Simple 'neural' optimization networks: an A/D converter, signal decision circuits, and a linear programming circuit[J]. IEEE Trans. Circuits Syst., 1988, CAS-33: 533-562
[100]Teunissen P J G. Quality control in navigation systems[J]. IEEE Aerospace and Electronic Systems Magazine, 1990, 5(7): 35-41
[101]Teunissen P J G, Salzmann M A. A recursive slippage test for use in state-space filtering[J]. Manuscripta Geodaetica, 1989,14(6): 383-390
[102]T. Saito. The Nonlinear Least Squares of Condition Equations[J]. Bull, Geod, 1973
[103]Yang Y. Robust estimation for dependent observations[J]. Mans. Geod. 1994, 19: 10-17
[104] Yang Y, Gao W. Comparison of adaptive factors on navigation results[J]. The Journal of Navigation, 2005, 58(3): 471-478
[105]Yang Y,Gao W.An optimal adaptive Kalman filter[J].J Geod,2006,80:177-183.
[106]Yang Y,He H,Xu G.Adaptively robust filtering for kinematic geodetic positioning[J].Journal of Geodesy,2001,75(2/3):109-116
[107]Yang Y,Xu T,He H.On adaptively kinematic filtering[J].Selected Paper for English of Acta Geodetica Cartographica Sinica,2001b,25-32
[108]Yang Y.,Xu T.and Song L.Robust estimation of variance components with application in GPS network adjustment[J].Journal of Surveying Engineering,2005,131(4):107-112
[109]Zhang S and Constantinides A G.Langrange Programming Neural Networks[J],IEEE TRANSACTIONS ONCIRCUITS AND SYSTEMS- Ⅱ:ANALOG AND SIGNAL PROCESSING 1992,39(7):441-452