基于强跟踪滤波器的MACA-MIE模型及跟踪算法  

MACA-MIE model and tracking algorithm based on strong tracking filter

作　　者：周政[1] 刘进忙[1] 

机构地区：[1]空军工程大学防空反导学院,西安710051

出　　处：《控制与决策》2013年第1期100-104,共5页Control and Decision

基　　金：国家自然科学青年基金项目(61102109);陕西省自然科学基金项目(2010JM8013)

摘　　要：结合自适应常加速模型(ACA)、改进输入估计(MIE)和强跟踪滤波器,提出一种新的自适应目标跟踪模型和算法.该算法通过扩展ACA模型状态矢量和改进状态噪声协方差调整方法,利用MIE和强跟踪滤波器,实现了机动加速度方差和状态预测协方差依据残差信息的实时完全自适应调整,在缺乏目标加速度先验知识的情况下,能够实时高精度跟踪目标突变状态、弱机动和非机动状态.仿真实验表明,相比ACA模型和MIE,该算法具有更好的机动状态和非机动状态跟踪性能.A new adaptive target tracking algorithm is proposed based on adaptive constant accelaraion（ACA） model, modified input estimation（MIE） and strong tracking filter. By extending ACA model state vector and improving adjusting method of state noise covariance, the proposed algorithm can adjust the accelaration variance and the state-estimation error covariance to the change of filtering residual thorough adaptively and timely under the support of MIE and strong tracking filter. The proposed algorithm is able to track the state of mutation, low maneuver and non-maneuver accurately in real time when lacking the information on the acceleration. The simulation shows that the proposed algorithm has better performance than ACA model and MIE in scenarios of maneuver and non-maneuver.

关 键 词：机动目标跟踪 自适应常加速模型 改进输入估计 卡尔曼滤波 强跟踪滤波器 

分 类 号：TN953[电子电信—信号与信息处理]