Robust Gaussian Approximate Filter and Smoother with Colored Heavy Tailed Measurement Noise
-
摘要: 为了解决带有色厚尾量测噪声的非线性状态估计问题，本文提出了新的鲁棒高斯近似（Gaussian approximate，GA）滤波器和平滑器.首先，基于状态扩展方法将量测差分后带一步延迟状态和白色厚尾量测噪声的非线性状态估计问题，转化成带厚尾量测噪声的标准非线性状态估计问题.其次，针对量测差分后模型中的噪声尺度矩阵和自由度（Degrees of freedom，DOF）参数未知问题，设计了新的高斯近似滤波器和平滑器，通过建立未知参数和待估计状态的共轭先验分布，并利用变分贝叶斯方法同时估计未知的状态、尺度矩阵、自由度参数.最后，利用目标跟踪仿真验证了本文提出的带有色厚尾量测噪声的鲁棒高斯近似滤波器和平滑器的有效性以及与现有方法相比的优越性.
-
关键词:
- 非线性状态估计 /
- 有色厚尾量测噪声 /
- 变分贝叶斯 /
- Student's t 分布 /
- 状态扩展方法
Abstract: In this paper, new robust Gaussian approximate (GA) filter and smoother are proposed to solve the problem of nonlinear state estimation with colored heavy tailed measurement noise. Firstly, the nonlinear state estimation problem with one-step delayed state and white heavy tailed measurement noise after measurement differencing is transformed into a standard nonlinear state estimation problem with heavy tailed measurement noise based on the state augmentation approach. Secondly, new GA filter and smoother are designed for the problem of unknown scale matrix and degrees of freedom (DOF) parameter of noise of the model after measurement differencing. The state, scale matrix and DOF parameter are estimated simultaneously by building the conjugate prior distributions for unknown parameters and estimated state and using variational Bayesian approach. Finally, the efficiency and superiority of the proposed robust GA filter and smoother with colored heavy tailed measurement noise, as compared with existing method, are shown in the simulation of target tracking. -
图 1 高斯分布、Student's t 分布、野值干扰导致的厚尾非高斯分布示意图
Fig. 1 Diagrams of Gaussian distribution,Student's t distribution and heavy tailed non-Gaussian distribution induced by outlier interference
图 3 尺度矩阵 R 和自由度参数 $\nu$ 的滤波估计
Fig. 3 The filtering estimations of scale matrix R and degree of freedom parameter $\nu$
图 7 当$\phi=0.1,0.2,\cdots,0.9$时,不同方法的位置ARMSE
Fig. 7 ARMSEs of the position of different methods when $\phi=0.1,0.2,\cdots,0.9$
图 8 当$\phi=0.1,0.2,\cdots,0.9$时,不同方法的速度ARMSE
Fig. 8 ARMSEs of the velocity of different methods when $\phi=0.1,0.2,\cdots,0.9$
图 9 当$\phi=0.1,0.2,\cdots,0.9$时,不同方法的转弯速率ARMSE
Fig. 9 ARMSEs of the turn rate of different methods when $\phi=0.1,0.2,\cdots,0.9$
图 10 当p = 0.05, 0.1,…, 0.3 时, 不同方法的位置ARMSE
Fig. 10 ARMSEs of the position of different methods when p = 0.05, 0.1,…, 0.3
图 11 当$p = 0.05, 0.1,…, 0.3 时,不同方法的速度ARMSE
Fig. 11 ARMSEs of the velocity of different methods when p = 0.05, 0.1,…, 0.3
图 12 当p = 0.05, 0.1,…, 0.3 时,不同方法的\\转弯速率ARMSE
Fig. 12 ARMSEs of the turn rate of different methods when p = 0.05, 0.1,…, 0.3
图 13 当N = 1, 2,…,10时,不同方法的位置ARMSE
Fig. 13 ARMSEs of the position of different methods when N = 1, 2,…,10
图 14 当N = 1, 2,…,10时,不同方法的速度ARMSE
Fig. 14 ARMSEs of the velocity of different methods when N = 1, 2,…,10
图 15 当N = 1, 2,…,10时,不同方法的转弯速率ARMSE
Fig. 15 ARMSEs of the turn rate of different methods when N = 1, 2,…,10
表 1 5种初值选取
Table 1 Five choices of initial values
u0 U0 a0 b0 初值1 4 ∑0 5 1 初值2 4 ∑0 10 1 初值3 4 ∑0 1 1 初值4 4 5∑0 5 1 初值5 4 0:5∑0 5 1 
下载: 导出CSV
表 2 提出的CKF和CKS在不同初值下的位置ARMSE、速度ARMSE、转弯速率ARMSE
Table 2 ARMSEs of position,velocity,and turn rate from the proposed CKF and CKS under different initial values
位置ARMSE(m) 速度ARMSE(m/s) 转弯速率ARMSE(Deg/s) 提出的CKF(初值1) 30.5676 13.4203 1.3568 提出的CKF(初值2) 30.7507 13.4447 1.3639 提出的CKF(初值3) 32.0552 13.7149 1.3736 提出的CKF(初值4) 30.8171 13.4256 1.3649 提出的CKF(初值5) 31.1468 13.5550 1.3702 提出的CKS(初值1) 18.0328 4.8804 0.6006 提出的CKS(初值2) 17.8937 4.8634 0.5996 提出的CKS(初值3) 18.4848 4.9462 0.6041 提出的CKS(初值4) 17.9099 4.8710 0.5999 提出的CKS(初值5) 18.2740 4.9138 0.6024
下载: 导出CSV
-
[1] 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 http://www.aas.net.cn/CN/abstract/abstract18590.shtmlZhang Yong-Gang, Huang Yu-Long, Zhao Lin. A general framework solution to Gaussian filter with multiple-step randomly-delayed measurements. Acta Automatica Sinica, 2015, 41(1):122-135 http://www.aas.net.cn/CN/abstract/abstract18590.shtml [2] 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 http://www.aas.net.cn/CN/abstract/abstract18634.shtmlZhang Yong-Gang, Huang Yu-Long, Li Ning, Zhao Lin. Conditional posterior Cramér-Rao lower bound for nonlinear sequential Bayesian estimation with one-step randomly delayed measurements. Acta Automatica Sinica, 2015, 41(3):559-574 http://www.aas.net.cn/CN/abstract/abstract18634.shtml [3] 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 doi: 10.1109/TAES.2011.5705658 [4] Arasaratnam I, Haykin S. Cubature Kalman filters. IEEE Transactions on Automatic Control, 2009, 54(6):1254-1269 doi: 10.1109/TAC.2009.2019800 [5] Arasaratnam I, Haykin S. Cubature Kalman smoothers. Automatica, 2011, 47(10):2245-2250 doi: 10.1016/j.automatica.2011.08.005 [6] Sarmavuori J, Särkkä S. Fourier-Hermite Kalman filter. IEEE Transactions on Automatic Control, 2012, 57(6):1511-1515 doi: 10.1109/TAC.2011.2174667 [7] Sarmavuori J, Särkkä S. Fourier-Hermite Rauch-Tung-Striebel smoother. In:Proceedings of the 20th European Signal Processing Conference. Bucharest, Romania:IEEE, 2012. 2109-2113 http://cn.bing.com/academic/profile?id=412252df135506ae304efcae333e98ee&encoded=0&v=paper_preview&mkt=zh-cn [8] Jia B, Xin M, Cheng Y. High-degree cubature Kalman filter. Automatica, 2013, 49(2):510-518 doi: 10.1016/j.automatica.2012.11.014 [9] Jia B, Xin M. A new class of nonlinear Rauch-Tung-Striebel cubature Kalman smoothers. ISA Transactions, 2015, 55(3):72-80 http://cn.bing.com/academic/profile?id=70d66a19e22d2c31f91fe80f4f81244e&encoded=0&v=paper_preview&mkt=zh-cn [10] 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 http://www.aas.net.cn/CN/abstract/abstract18352.shtmlZhang 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 http://www.aas.net.cn/CN/abstract/abstract18352.shtml [11] Jia B, Xin M, Cheng Y. Sparse-grid quadrature nonlinear filtering. Automatica, 2012, 48(2):327-341 doi: 10.1016/j.automatica.2011.08.057 [12] Duník J, Straka O, Šimandl M. Stochastic integration filter. IEEE Transactions on Automatic Control, 2013, 58(6):1561-1566 doi: 10.1109/TAC.2013.2258494 [13] Zhang X C. A novel cubature Kalman filter for nonlinear state estimation. In:Proceedings of the 52nd IEEE Conference on Decision and Control. Florence, Italy:IEEE, 2013. 7797-7802 http://www.baidu.com/link?url=afvSkw6vNP-At3OO_bxK6vVGSOkyDA47uE6E0Qjzv0zQP9bAEW4rFBquM7F5BQzVVVigBdiShi5nlS0PAiZFxBLWVNAa1SjlTf_nzz7USjFUTIit8in3LkKeiBfX3cCgoVmtI2scQ4V5wafbs9XPPGxAQN8WFL9hqpBcixHHK3GLsVE7gKnZ59dm498VREDjfLGRgUlq4hcX0aKDf0xU_neo_g-wUSBG1m6fqvcinPLvtdpEPQgICqlBHxum_FUukWPRsZTZs6FGmgnnw-YMW0O-Au0ECuyZerXONp0PYLUu8_zqBlMVeU5hukczuhMArG0RO_BGCWgWwxRrN_bD7K&wd=&eqid=b91cc5c40003cce3000000055882b48d [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 doi: 10.1016/j.sigpro.2015.02.022 [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 doi: 10.1109/LSP.2013.2290381 [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 doi: 10.1109/TAC.2012.2204830 [17] Zhang Y G, Huang Y L, Li N, Zhao L. Interpolatory cubature Kalman filters. IET Control Theory and Applications, 2015, 9(11):1731-1739 doi: 10.1049/iet-cta.2014.0873 [18] 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 doi: 10.3724/SP.J.1004.2012.00986Wang 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 doi: 10.3724/SP.J.1004.2012.00986 [19] 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 http://cn.bing.com/academic/profile?id=9194221a81f1a46e620b8b92f7a2fb5c&encoded=0&v=paper_preview&mkt=zh-cn [20] 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 doi: 10.1109/TAC.2014.2337991 [21] Piché R, Särkkä S, Hartikainen J. Recursive outlier-robust filtering and smoothing for nonlinear systems using the multivariate student-t distribution. In:Proceedings of the 2012 IEEE International Workshop on Machine Learning for Signal Processing. Santander, Spain:IEEE, 2012. 1-6 http://www.baidu.com/link?url=ApK7VJgVmWCzC3nf5uc3a4Xdpw67Lpk_Ti0gvTT1Src6om3tzouoIoAI5SHKugtfrPytWtkY9ny5pDLXtK-MhDQEBQbsQcEeyCFkyDPp3eNcHjpKhqeeatuxir9rRkbjc5j_KRyuxhn0c08ED-VlVqM9lusTYl8p8jc04nx07FUeCbCpi_rSoQof07_cyMddrgrNEXXFWEoVQl89jw6A2vXwT6bmEx_9e7L8HWvNpEdLHc-mLpehwy2PoXhfr0R7U5V69rzYtaRTxgZV0R441XtDVs0_VBpIImVZsYdU4Ce3YBoyuA74o-BIJ8rsSMhmmjfo463zq4YSsgy31Gtcnymgl6PZFHXuYYIWEmrOBcrQJbLaiBStAVTCDNY4eQEaUBzSebexZaGoTcNalEO3ZK&wd=&eqid=bf8c088c000404f8000000055882b4ac [22] Agamennoni G, Nebot E M. Robust estimation in non-linear state-space models with state-dependent noise. IEEE Transactions on Signal Processing, 2014, 62(8):2165-2175 doi: 10.1109/TSP.2014.2305636 [23] Simon D. Optimal State Estimation. New Jersey:John Wiley and Sons, Inc., 2006. [24] 黄玉龙, 张勇刚, 李宁, 赵琳. 一种带有色量测噪声的非线性系统辨识方法. 自动化学报, 2015, 41(11):1877-1892 http://www.aas.net.cn/CN/abstract/abstract18763.shtmlHuang Yu-Long, Zhang Yong-Gang, Li Ning, Zhao Lin. An identification method for nonlinear systems with colored measurement noise. Acta Automatica Sinica, 2015, 41(11):1877-1892 http://www.aas.net.cn/CN/abstract/abstract18763.shtml [25] Agamennoni G, Nieto J I, Nebot E M. An outlier-robust Kalman filter. In:Proceedings of the 2011 IEEE International Conference on Robotics and Automation (ICRA). Shanghai, China:IEEE, 2011. 1551-1558 http://www.baidu.com/link?url=RD-GOwDWbEBEUsRkmUvoEJ8s05LWHFkzVMwk-fHxlsbCThoE_m1ww2a-Mc2isuWNMZ445m62svwVpg8vn9-xIadwEGTVZ_fB5nSxLoerWnkavjd_jXgu0L-1KyKIIA5BNjpje2lhUE8txkgXy7v9OpkS58ykAHEGXiaRyDLMNnimneiawRG_ouwvGawH2XTB0Y57dc980O6X9LD0OhMm6PDfotRYfG05earsZTUfVjnD1QBpUD__6shQHna6xoBR04Mwp3LJrkyUphry7IcPob2nBorTpdr4v8uHkwUqK7u&wd=&eqid=bf8c088c0004081e000000055882b4d2 [26] Agamennoni G, Nieto J I, Nebot E M. Approximate inference in state-space models with heavy-tailed noise. IEEE Transactions on Signal Processing, 2012, 60(10):5024-5037 doi: 10.1109/TSP.2012.2208106 [27] Zhu H, Leung H, He Z S. A variational Bayesian approach to robust sensor fusion based on student-t distribution. Information Sciences, 2013, 221:201-214 doi: 10.1016/j.ins.2012.09.017 [28] Nurminen H, Ardeshiri T, Piché R, Gustafsson F. Robust inference for state-space models with skewed measurement noise. IEEE Signal Processing Letters, 2015, 22(11):1898-1902 doi: 10.1109/LSP.2015.2437456 [29] Roth M, Özkan E, Gustafsson F. A student's t filter for heavy tailed process and measurement noise. In:Proceedings of the 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). Vancouver, BC, Canada:IEEE, 2013. 5770-5774 http://www.baidu.com/link?url=ZdeDibgasMFldsIonX6rOu60u80Efc5qyHVPpOYpFgkGSLfIRqfaVw2jZ_QGPYXresS2iPyvJD18SeYfw2gm-W_jM9a_bF61TIivE7Zb7_Rt34_LW1GEjWIjL2qG7rb6xtiq7O62b_QI-OYCKuuuouQvcxszq16G90PS2zRIrxAE6B-cFAzAusP4FJ5BzmFGRcDXNw6AMFzWztDA1z64Bi6wVF_bj2FlPQVt-YYNgWYCJDRua9XHymqND4V2lYWX4cAseoJxlXGXByEDpD2EIZOM6g4RJaRHyolkPzrA6ZjU_KuDWClE9-A9iFjngi85jukkmZACtQGTbcXCQf-V4_&wd=&eqid=d6b91c7b00042324000000055882b529 [30] Xu D J, Shen C, Shen F. A robust particle filtering algorithm with non-Gaussian measurement noise using student-t distribution. IEEE Signal Processing Letters, 2014, 21(1):30-34 doi: 10.1109/LSP.2013.2289975 [31] 黄玉龙, 张勇刚, 李宁, 赵琳. 一种改进的高斯近似滤波方法. 自动化学报, 2016, 42(3):385-401 http://www.aas.net.cn/CN/abstract/abstract18828.shtmlHuang Yu-Long, Zhang Yong-Gang, Li Ning, Zhao Lin. An improved Gaussian approximate filtering method. Acta Automatica Sinica, 2016, 42(3):385-401 http://www.aas.net.cn/CN/abstract/abstract18828.shtml [32] Tzikas D G, Likas A C, Galatsanos N P. The variational approximation for Bayesian inference. IEEE Signal Processing Magazine, 2008, 25(6):131-146 doi: 10.1109/MSP.2008.929620 [33] Golub G H, Van Loan C F. Matrix Computations (4th edition). Baltimore, Maryland:The Johns Hopkins University Press, 2013. -
计量
- 文章访问数: 3691
- HTML全文浏览量: 279
- PDF下载量: 675
- 被引次数: 0
