基于抗差EKF的机载单站无源定位算法  

Airborne Single-Observer Passive Location Algorithm Based on Robust EKF

作　　者：张艺航 陈树新[1] 吴昊[1] 张衡阳[1] 

机构地区：[1]空军工程大学信息与导航学院,陕西西安710077

出　　处：《空军工程大学学报（自然科学版）》2014年第4期75-78,87,共5页Journal of Air Force Engineering University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61202490);陕西省自然科学基金资助项目(2010JQ8010)

摘　　要：机载单站无源测向定位中,若观测值中存在异常误差,采用EKF算法会导致定位结果有偏。为增强算法抗差性,将抗差EKF算法引入到机载单站无源测向定位中。首先,建立了机载单站无源测向定位模型,依据标准化观测残差构建了抗差等价权矩阵,对异常观测进行降权处理。其次,针对抗差EKF算法效率低的特点,构建了卡方统计量,提出了基于卡方检验的抗差EKF定位算法,即先利用卡方检验判别出含异常误差的观测值,再调用抗差EKF算法进行定位解算。最后,通过在观测值中加入单个异常误差和连续异常误差,测试了算法对不同异常误差的抵抗能力。仿真表明,抗差EKF算法能较好地削弱异常误差对位置估值的影响,并且基于卡方检验的抗差EKF定位算法能提高定位的效率。In the application of airbone single-observer passive wireless radio frequency inj ection location,if there has outliers in the observed value,the application of the EKF algorithm will make the location estimation biased.Therefore,a robust EKF algorithm is applied to airborne single-observer passive wireless radio frequency inj ection location system.Firstly,airbone single-observer passive bearing location model is build,and the robust equivalent weight matrix built by standardized observation residuals is used to resist outliers.Then,in order to improve the efficiency of passive location system,the robust EKF based on chisquare test is proposed by creating a chi-square statistical model.It can distinguish outliers through chi-square test,and then call the robust EKF to complete location solution.Through rej ecting single outlier and continuing outliers to test the ability of the algorithm.Simulation result shows that the suggested algorithm has great outlier suppression ability and the robust EKF based on chi-square test is fast.

关 键 词：无源定位 单站定位 抗差EKF滤波 定位精度 

分 类 号：TN971[电子电信—信号与信息处理]