带相关噪声的观测融合稳态Kalman滤波算法及其全局最优性

邓自立; 顾磊; 冉陈键

doi:10.3724/SP.J.1146.2007.01530

带相关噪声的观测融合稳态Kalman滤波算法及其全局最优性

doi: 10.3724/SP.J.1146.2007.01530

基金项目:

国家自然科学基金(60374026)和黑龙江大学自动控制重点实验室(F04-01)资助课题

计量
- 文章访问数: 3554
- HTML全文浏览量: 89
- PDF下载量: 1028
- 被引次数: 0
出版历程
- 收稿日期: 2007-12-19
- 修回日期: 2008-04-14
- 刊出日期: 2009-03-19

Measurement Fusion Steady-State Kalman Filtering Algorithm with Correlated Noises and Global Optimdity

摘要

摘要: 对于带相关的输入白噪声和观测白噪声及相关观测白噪声的多传感器线性离散定常随机系统，用加权最小二乘(WLS)法提出了一种加权观测融合稳态Kalman滤波算法，可处理状态、白噪声和信号融合滤波、平滑、预报问题。基于稳态信息滤波器证明了它完全功能等价于集中式观测融合稳态Kalman滤波算法，因而它具有渐近全局最优性，且可减少计算负担。一个跟踪系统仿真例子验证了它的功能等价性。
- 多传感器信息融合;加权观测融合;相关噪声;稳态Kalman滤波;渐近全局最优性
Abstract: For the multisensor linear discrete time-invariant stochastic control systems with correlated input and measurement white noises, and with correlated measurement white muses, a weighted measurement fusion steady-state Kalman filtering algorithm is presented by using the Weighted Least Squares(WLS)method. It can handle the fused filtering , smoothing and prediction problems for the state, white noise and signal. Based on the steady-state information filter, it is proved that it is completely functionally equivalent to the centralized measurement fusion steady-state Kalman filtering algorithm, so that it has asymptotic global optimality, and can reduced the computational burden. A simulation examples for tracking systems verifies its functional equivalence.

HTML全文

参考文献(1)

Gan Q and Harris C J. Comparison of two measurementfusion methods for Kalman filter-based mutisensor datafusion[J].IEEE Trans. on Aerospace and Electronic Systems.2001, 37(1):273-279[2]Roecker J A and McGillen C D. Comparison of two-sensortracking methods based on state vector fusion andmeasurement fusion. IEEE Trans. on Aerospace andElectronic Systems, 1988, 21(4): 447-449.[3]Li X R, Zhu Y M, and Wang J, et al.. Optimal linearestimation fusion-part I: Unified fusion rules[J].IEEE Trans. onInformation Theory.2003, 49(9):2192-2208[4]Hall D L and Llinas J. An introduction to multisensor datafusion[J].Proc. IEEE.1997, 85(1):6-23[5]Deng Z L, Gao Y, and Mao L, et al.. New approach toinformation fusion steady-state Kalman filtering[J].Automatica.2005, 41(10):1695-1707[6]邓自立. 两种最优观测融合方法的功能等价性. 控制理论与应用, 2006, 23(2): 319-323.Deng Zi-li. On functional equivalence of two measurementfusion methods. Control Theory Applications, 2006, 23(2):319-323.[7]Roy S and Iltis R A. Decentralized linear estimation incorrelated measurement noise[J].IEEE Trans. on Aerospaceand Electronic System.1991, 27(6):939-941[8]欧连军, 邱红专, 张洪钺. 多个相关测量的融合算法及其最优性. 信息与控制, 2005, 34(6): 690-695.Ou Lian-jun, Qiu Hong-zhuan, and Zhang Hong-yue.Multiple correlated measurement fusion algorithm and itsoptimality. Information and Control, 2005, 34(6): 690-695.[9]惠玉松, 顾磊, 冉陈键, 邓自立. 基于稳态Kalman滤波的相关观测融合算法及其功能等价性.科学技术与工程, 2007,7(19): 4809-4814.Hui Yu-song, Gu Lei, Ran Chen-jian, and Deng Zi-li.Correlated measurements fusion methods based steady-stateKalman filtering and their functional equivalence. ScienceTechnology and Engineering, 2007, 7(19): 4809-4814.[10]Darouach M, Zasdzinshi M, and Onana A B, et al.. Kalmanfilteing with unknown inputs via optimal state estimation.Int. J. Systerms, 1995, 20(10): 2015-2028.[11]Kailath T A, Sayed H, and Hassibi B. Linear Estimation,Upper Sddle River, New Jersey : Prentice-Hall, 2000: 41-333.[12]邓自立. 最优估计理论及其应用建模、滤波、信息融合估计. 哈尔滨: 哈尔滨工业大学出版社, 2005: 150-154.Deng Zi-li. Optional Estimation Theory with Applications,Modeling, Filtering, and Information Fusion. Harbin: HarbinInstitute of Technology Press, 2005: 150-154.[13]Chui C K and chen G. Kalman Filterig with Real TimeApplications. New York: Spring-Verlag Berlin Heidelberg,1987: 91-95.[14]邓自立. 信息融合滤波理论及其应用. 哈尔滨: 哈尔滨工业大学出版社, 2007: 119-123.Deng Zi-li. Information Fusion Filterig Theory withApplications. Harbin: Harbin Institute of Technology Press,2007: 119-123.

施引文献

资源附件(0)

访问统计