基于UKF的机动目标跟踪算法  

The Algorithm of Maneuvering Target Tracking based on UKF

作　　者：张苗辉[1] 杨一平[1] 刘先省[1] 

机构地区：[1]河南大学先进控制与智能信息处理研究所

出　　处：《火力与指挥控制》2007年第8期37-39,共3页Fire Control & Command Control

基　　金：河南省高校杰出科研人才创新工程基金资助项目(2003KYCX003);河南省高校创新人才培养工程项目;河南省自然科学基金资金项目(0411010400)

摘　　要：基于EKF在处理非线性问题时采用系统预测状态的局部线性化方法来近似系统状态演化方程所产生的不良影响,提出了一种用UKF方法解决机动目标方位和速度的跟踪及估计问题的算法。按实际的非线性模型演化时,该算法能够很好地对非线性函数的后验均值与方差进行拟合,并充分利用了传感器每次量测带来的信息并进一步优化了测量方差。仿真试验表明,该算法能很好地对机动目标的速度和航迹进行估计和跟踪。Based on the shortcoming of using EKF to deal with nonlinearity problems, which adopts the part linearization to estimate approximately the system state equation. This paper presents a new algorithm, which utilizes UKF to track the azimuth of maneuvering target, or estimate the velocity. When some parameters undergo the nonlinearity model, this algorithm can exactly present the posterior distribution, which is able to make full use of sensor measurement message every time, and optimize measurement variance. The simulation results show that this algorithm can exactly estimate the track of the maneuvering target.

关 键 词：UKF 非线性系统 机动目标跟踪 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]