一种带有色量测噪声的非线性系统辨识方法

黄玉龙; 张勇刚; 李宁; 赵琳

doi:10.16383/j.aas.2015.c150278

一种带有色量测噪声的非线性系统辨识方法

doi: 10.16383/j.aas.2015.c150278

1.
哈尔滨工程大学自动化学院哈尔滨 150001

基金项目:

国家自然科学基金(61001154,61201409,61371173),中国博士后科学基金(2013M530147,2014T70309),黑龙江省博士后基金(LBH-Z13052,LBH-TZ0505),哈尔滨工程大学中央高校基本科研业务费专项基金(HEUCFQ20150407)资助

详细信息

作者简介:
黄玉龙哈尔滨工程大学自动化学院博士研究生.主要研究方向为惯性导航,滤波算法,组合导航.E-mail:heuedu@163.com

李宁哈尔滨工程大学自动化学院副教授.主要研究方向为自适应滤波,组合导航.E-mail:ningli@hrbeu.edu.cn

赵琳哈尔滨工程大学自动化学院教授.主要研究方向为惯性导航,卫星导航,组合导航.E-mail:zhaolin@hrbeu.edu.cn

通讯作者:
张勇刚哈尔滨工程大学自动化学院研究员.2007年获得英国Cardiff大学博士学位.主要研究方向为光纤陀螺,惯性导航,滤波算法,组合导航.本文通信作者.E-mail:zhangyg@hrbeu.edu.cn

计量
- 文章访问数: 2059
- HTML全文浏览量: 76
- PDF下载量: 954
- 被引次数: 11
出版历程
- 收稿日期: 2015-05-11
- 修回日期: 2015-08-18
- 刊出日期: 2015-11-20

An Identification Method for Nonlinear Systems with Colored Measurement Noise

1.
College of Automation, Harbin Engineering University, Harbin 150001

Funds:

Supported by National Natural Science Foundation of China (61001154, 61201409, 61371173), China Postdoctoral Science Foundation (2013M530147, 2014T70309), Heilongjiang Postdoctoral Fund (LBH-Z13052, LBH-TZ0505), and Fundamental Research Funds for the Central Universities of Harbin Engineering University (HEUCFQ20150407)

摘要

摘要: 利用最大似然判据, 本文提出了一种带有色量测噪声的非线性系统辨识方法. 首先, 利用量测差分方法将有色量测噪声白色化, 获得新的量测方程, 从而将带有色量测噪声的非线性系统辨识问题转化成带白色量测噪声和一步延迟状态的非线性系统辨识问题. 其次, 利用期望最大化(Expectation maximization, EM)算法提出了一种新的基于最大似然估计的非线性系统辨识方法, 该算法由期望步骤(Expectation step, E-step)和最大化步骤(Maximization step, M-step)两部分组成. 在期望步骤中, 基于当前估计的参数并利用带有色量测噪声的高斯近似滤波器和平滑器, 近似计算完整的对数似然函数的期望. 在最大化步骤中, 近似计算的似然函数期望值被最大化, 并且通过解析更新获得噪声参数估计, 通过Newton更新方法获得模型参数的估计. 最后, 数值仿真验证了本文提出算法的有效性.
- 非线性系统辨识 /
- 最大似然判据 /
- 有色量测噪声 /
- 期望最大化算法 /
- 量测差分方法 /
- 非线性状态估计器
Abstract: In this paper, an identification method for nonlinear systems with colored measurement noise is proposed by using the maximum likelihood criterion. Firstly, the colored measurement noise is decorrelated based on the measurement differencing approach, and a new measurement equation is derived. Thus, the nonlinear system identification problem with colored measurement noise is transformed into the nonlinear system identification problem with white measurement noise and one-step delayed state. Secondly, a new nonlinear system identification method with maximum likelihood estimation is proposed based on the expectation maximization (EM) algorithm, which consists of expectation step (E-step) and the maximization step (M-step). In the E-step, the expectation of the complete data log-likelihood function is approximately calculated based on currently estimated parameters and the Gaussian approximated filter and smoother for nonlinear system with colored measurement noise. In the M-step, the approximately calculated expectation value is maximized, and noise parameter estimations are updated analytically and model parameter estimations are updated approximately by using Newton method. Finally, the efficiency of the proposed algorithm is illustrated in numerical simulations.
- Nonlinear system identification /
- maximum likelihood criterion /
- colored measurement noise /
- expectation maximization (EM) algorithm /
- measurement differencing approach /
- nonlinear state estimator

HTML全文

参考文献(44)

[1]	Zhang Yong-Gang, Huang Yu-Long, Zhao Lin. A general framework solution of Gaussian filter with multiple step randomly delayed measurements. Acta Automatica Sinica, 2015, 41(1):122-135(张勇刚, 黄玉龙, 赵琳. 一种带多步随机延迟量测高斯滤波器的一般框架解. 自动化学报, 2015, 41(1):122-135)
[2]	[2] Lei M, van Wyk B J, Qi Y. Online estimation of the approximate posterior Cramer-Rao lower bound for discrete-time nonlinear filtering. IEEE Transactions on Aerospace and Electronic Systems, 2011, 47(1):37-57
[3]	Zhang Yong-Gang, Huang Yu-Long, Li Ning, Zhao Lin. Conditional posterior Cramr-Rao lower bound for nonlinear sequential Bayesian estimation with one-step randomly delayed measurements. Acta Automatica Sinica, 2015, 41(3):559-574(张勇刚, 黄玉龙, 李宁, 赵琳. 带一步随机延迟量测非线性序列贝叶斯估计的条件后验克拉美罗下界. 自动化学报, 2015, 41(3):559-574)
[4]	[4] Arasaratnam I, Haykin S. Cubature Kalman filters. IEEE Transactions on Automatic Control, 2009, 54(6):1254-1269
[5]	[5] Arasaratnam I, Haykin S. Cubature Kalman smoothers. Automatica, 2011, 47(10):2245-2250
[6]	[6] Sarmavuori J, Srkk S. Fourier-Hermite Kalman filter. IEEE Transactions on Automatic Control, 2012, 57(6):1511-1515
[7]	[7] Sarmavuori J, Srkk S. Fourier-Hermite Rauch-Tung-Striebel smoother. In:Proceedings of the 20th European Signal Processing Conference. Bucharest, Romania:IEEE, 2012. 2109-2113
[8]	[8] Jia B, Xin M, Cheng Y. High-degree cubature Kalman filter. Automatica, 2013, 49(2):510-518
[9]	[9] Jia B, Xin M. A new class of nonlinear Rauch-Tung-Striebel cubature Kalman smoothers. ISA Transactions, 2015, 55(3):72-80
[10]	Zhang Yong-Gang, Huang Yu-Long, Wu Zhe-Min, Li Ning. A high order unscented Kalman filtering method. Acta Automatica Sinica, 2014, 40(5):838-848(张勇刚, 黄玉龙, 武哲民, 李宁. 一种高阶无迹卡尔曼滤波方法. 自动化学报, 2014, 40(5):838-848)
[11]	Jia B, Xin M, Cheng Y. Sparse-grid quadrature nonlinear filtering. Automatica, 2012, 48(2):327-341
[12]	Dunk J, Straka O, imandl M. Stochastic integration filter. IEEE Transactions on Automatic Control, 2013, 58(6):1561-1566
[13]	Zhang X C. A novel cubature Kalman filter for nonlinear state estimation. In:Proceedings of the 2013 IEEE 52nd Annual Conference on Decision and Control. Firenze:IEEE, 2013. 7797-7802
[14]	Zhang Y G, Huang Y L, Li N, Zhao L. Embedded cubature Kalman filter with adaptive setting of free parameter. Signal Processing, 2015, 114:112-116
[15]	Wang S Y, Feng J C, Tse C K. Spherical simplex-radial cubature Kalman filter. IEEE Signal Processing Letters, 2014, 21(1):43-46
[16]	Chang L B, Hu B Q, Li A, Qin F J. Transformed unscented Kalman filter. IEEE Transactions on Automatic Control, 2013, 58(1):252-257
[17]	Ljung L. Perspectives on system identification. In:Proceedings of the 17th IFAC World Congress. Seoul, Korea:IFAC, 2008. 7172-7184
[18]	Gaperin M, Juričić D. Application of unscented transformation in nonlinear system identification. In:Proceedings of the 18th IFAC World Congress. Milano, Italy:IFAC, 2011. 4428-4433
[19]	Zhao Lin, Wang Xiao-Xu, Sun Ming, Ding Ji-Cheng, Yan Chao. Adaptive UKF filtering algorithm based on maximum a posterior estimation and exponential weighting. Acta Automatica Sinica, 2010, 36(7):1007-1019(赵琳, 王小旭, 孙明, 丁继成, 闫超. 基于极大后验估计和指数加权的自适应UKF滤波算法. 自动化学报, 2010, 36(7):1007-1019)
[20]	Zhao L Q, Wang J L, Yu T, Jian H, Liu T J. Design of adaptive robust square-root cubature Kalman filter with noise statistic estimator. Applied Mathematics and Computation, 2015, 256:352-367
[21]	Wang Lu, Li Guang-Chun, Qiao Xiang-Wei, Wang Zhao-Long, Ma Tao. An adaptive UKF algorithm based on maximum likelihood principle and expectation maximization algorithm. Acta Automatica Sinica, 2012, 38(7):1200-1210(王璐, 李光春, 乔相伟, 王兆龙, 马涛. 基于极大似然准则和最大期望算法的自适应UKF算法. 自动化学报, 2012, 38(7):1200-1210)
[22]	Bavdekar V A, Deshpande A P, Patwardhan S C. Identification of process and measurement noise covariance for state and parameter estimation using extended Kalman filter. Journal of Process Control, 2011, 21(4):585-601
[23]	Schn T B, Wills A, Ninness B. System identification of nonlinear state-space models. Automatica, 2011, 47(1):39-49
[24]	Vnnen V. Gaussian Filtering and Smoothing based Parameter Estimation in Nonlinear Models for Sequential Data[Master dissertation], Aalto University, Finland, 2012
[25]	Kokkala J, Solin A, Srkk S. Expectation maximization based parameter estimation by sigma-point and particle smoothing. In:Proceedings of the 17th International Conference on Information Fusion. Salamanca, Spain:IEEE, 2014. 1-8
[26]	Wu W R, Chang D C. Maneuvering target tracking with colored noise. IEEE Transactions on Aerospace and Electronic Systems, 1996, 32(4):1311-1320
[27]	Srinivasan S, Aichner R, Kleijn W B, Kellermann W. Multi-channel parametric speech enhancement. IEEE Signal Processing Letters, 2006, 13(5):304-307
[28]	Jamoos A, Grivel E, Bobillet W, Guidorzi R. Errors-in-variables-based approach for the identification of AR time-varying fading channels. IEEE Signal Processing Letters, 2007, 14(11):793-796
[29]	Mahmoudi A, Karimi M, Amindavar H. Parameter estimation of autoregressive signals in presence of colored AR(1) noise as a quadratic eigenvalue problem. Signal Processing, 2012, 92(4):1151-1156
[30]	Wang Xiao-Xu, Liang Yan, Pan Quan, Zhao Chun-Hui, Li Han-Zhou. Unscented Kalman filter for nonlinear systems with colored measurement noise. Acta Automatica Sinica, 2012, 38(6):986-998(王小旭, 梁彦, 潘泉, 赵春晖, 李汉舟. 带有色量测噪声的非线性系统Unscented卡尔曼滤波器. 自动化学报, 2012, 38(6):986-998)
[31]	Wang X X, Pan Q. Nonlinear Gaussian filter with the colored measurement noise. In:Proceedings of the 17th International Conference on Information Fusion. Salamanca, Spain:IEEE, 2014. 1-7
[32]	Wang X X, Liang Y, Pan Q, Zhao C H, Yang F. Nonlinear Gaussian smoothers with colored measurement noise. IEEE Transactions on Automatic Control, 2015, 60(3):870-876
[33]	Yuan G N, Xie Y J, Song Y, Liang H B. Multipath parameters estimation of weak GPS signal based on new colored noise unscented Kalman filter. In:Proceedings of the 2010 IEEE International Conference on Information and Automation. Harbin, China:IEEE, 2010. 1852-1856
[34]	Titterton D H, Weston J L. Strapdown Inertial Navigation Technology. London:Peter Peregrinus, Ltd., 1997.
[35]	Goto S, Nakamura M, Uosaki K. On-line spectral estimation of nonstationary time series based on AR model parameter estimation and order selection with a forgetting factor. IEEE Transactions on Signal Processing, 1995, 43(6):1519-1522
[36]	Giurcăneanu C D, Razavi S A. AR order selection in the case when the model parameters are estimated by forgetting factor least-squares algorithms. Signal Processing, 2010, 90(2):451-466
[37]	Kok M, Dahlin J, Schn T B, Wills A. Newton-based maximum likelihood estimation in nonlinear state space models. In:Proceedings of the 17th IFAC Symposium on System Identification. IFAC, 2015
[38]	Poyiadjis G, Doucet A, Singh S S. Particle approximations of the score and observed information matrix in state space models with application to parameter estimation. Biometrika, 2011, 98(1):65-80
[39]	Li X R. Measure of nonlinearity for stochastic systems. In:Proceedings of the 15th International Conference on Information Fusion. Singapore:IEEE, 2012. 1073-1080
[40]	Bucy R S, Senne K D. Digital synthesis of non-linear filters. Automatica, 1971, 7(3):287-298
[41]	Dunik J, Simandl M, Straka O. Unscented Kalman filter:aspects and adaptive setting of scaling parameter. IEEE Transactions on Automatic Control, 2012, 57(9):2411-2416
[42]	Einicke G A, White L B. Robust extended Kalman filtering. IEEE Transactions on Signal Processing, 1999, 47(9):2596-2599
[43]	Li W L, Jia Y M. H-infinity filtering for a class of nonlinear discrete-time systems based on unscented transform. Signal Processing, 2010, 90(12):3301-3307
[44]	Jia B, Xin M. Sparse-grid quadrature H_{} filter for discrete-time systems with uncertain noise statistics. IEEE Transactions on Aerospace and Electronic Systems, 2013, 49(3):1626-1636