当前位置>主页 > 期刊在线 > 通信工程 >

通信工程

新型自适应容积卡尔曼滤波算法及其在目标跟踪中的应用
黄硕,李冠男,荆涛,曹洁
(陆军装甲兵学院信息通信系,北京 100072)
摘要点击次数:196    

摘  要:针对标准容积卡尔曼滤波(CKF)在目标跟踪中出现的问题,根据系统噪声统计特性不准确或未知的特点,提出一种基于协方差匹配原则的自适应容积卡尔曼滤波算法。该算法通过利用新息序列与残差序列来实现对观测噪声协方差和过程噪声协方差矩阵的实时跟踪,进而进行在线调整噪声统计特性,能够有效的改善由于噪声特性未知所引起的滤波发散相关问题。将该算法应用在目标跟踪仿真实验中,结果表明,与标准CKF算法相比,在系统噪声统计特性未知的情况下,该算法具有更好的实时性,不仅抑制了滤波器的发散问题,而且提高了对目标的跟踪精度。


关键词:容积卡尔曼滤波算法;协方差匹配;自适应滤波;目标跟踪


作者介绍:

黄硕,男,辽宁辽中人,陆军装甲兵学院,硕士研究生,本科。研究方向:信号与信息处理。


中图分类号:TN713;TN953;V557     文献标识码:A 文章编号:2096-4706(2018)02-0000-05

NewAdaptive Cubature Kalman Filter Algorithm and Its Application in Target Tracking

HUANG Shuo,LI Guannan,JING Tao,CAO Jie

(Department of InformationEngineering,Army Armored Military Academy,Beijing  100072,China)

AbstractIn view of the problem of standard volume Calman filter (CKF) in target tracking,an adaptive volumeCalman filtering algorithm based on covariance matching principle is proposedin the light of the inaccurate or unknown characteristics of the statisticalcharacteristics of the system noise.The algorithm realizes the real-timetracking of the covariance of the observed noise and the covariance matrix ofthe process noise by using the new interest sequence and the residual sequence,and then adjusts the statistical characteristics of the noise online,and can effectively improve the filtering anddivergence problem caused by the unknown noise characteristics.The algorithm isapplied to the target tracking simulation experiment. The results show that,compared with the standard CKF algorithm,thealgorithm has better real-time performance in the case of unknown noisestatistical characteristics of the system,which notonly inhibits the divergence of the filter, but alsoimproves the tracking precision of the target.

Keywordscubature Kalman filter;covariancematching;adaptive filter;targettracking


参考文献:

[1] Einicke GA,White L B. Robust extendedKalman filtering [J].Signal Processing IEEE Transactions on,1999,47(9):2596-2599.

[2] Julier S,Uhlmann J,Durrant-whyte F H. A new method for the nonlinear transformation ofmeans and covariances in filters and estimators [J].IEEE Trans.on Automat.contr,2000,45(3):477-482.

[3]Julier S J,Uhlmann J K. Unscentedfiltering and nonlinear estimation [J]. Proceedings of the IEEE,2004,92(3):401-422.

[4] Arasaratnam I,Haykin S.Cubature Kalman Filters [J]. IEEE Transactions on Automatic Control,2009,54(6):1254-1269.

[5] Nielsen R O. Accuracy of angle estimation withmonopulse processing using two beams [J].Aerospace & Electronic SystemsIEEE Transactions on,2001,37(4):1419-1423.

[6] 于浛,魏喜庆,宋申民,等.基于自适应容积卡尔曼滤波的非合作航天器相对运动估计[J].航空学报,2014,35(8):2251-2260.

[7] 丁家琳,肖建.基于极大后验估计的自适应容积卡尔曼滤波器[J].控制与决策,2014,29(2):327-334.

[8] Salahshoor K,MosallaeiM,Bayat M. Centralized and decentralized process andsensor fault monitoring using data fusion based on adaptive extended Kalmanfilter algorithm [J].Measurement,2008,41(10):1059-1076.

[9] 石勇,韩崇昭.自适应UKF算法在目标跟踪中的应用 [J].自动化学报,2011,37(6):755-759.

[10] 易令.高速高机动目标跟踪算法研究 [D].成都:电子科技大学,2006.