三状态样条Kalman滤波与目标机动检测  

Three-state Spline Kalman Filter and Target Maneuvering Detection

作　　者：汪雄良[1] 朱炬波[1] 王春玲[1] 

机构地区：[1]国防科技大学理学院,长沙410073

出　　处：《现代雷达》2004年第4期21-23,28,共4页Modern Radar

基　　金：国家自然科学基金 (69872 0 3 9);教育部振兴计划高校骨干教师资助计划项目

摘　　要：基于样条方法和Kalman滤波理论 ,给出了一种新的三状态样条Kalman滤波方法及对目标进行机动检测的方法。该滤波方法用样条函数建立目标运动模型 ,应用Kalman滤波进行状态估计 ;该机动检测方法通过检测滤波新息来对目标机动发生或消除进行判断。由于考虑了加速度的实时估计 ,该滤波方法尤其适用于对弹道式目标的机动跟踪。仿真计算与实测数据计算表明 :该滤波具有较高的精度 ;Basing on two state spline filter and Kalman filter theory, this paper presents a novel three state spline Kalman filter and maneuvering detection method. The three state spline Kalman filter uses spline function to establish target motion model, then uses Kalman filter to make state estimation. By detecting the new information from the filtering process, this maneuvering detection method can estimate whether the target is maneuvering or not. Because of the taking into consideration of the real time acceleration estimation, this filter method is applicable to maneuverable tracking of the target using its trajectory data. The simulation and actual data calculation result show that this approach of maneuvering detection has high precision and good performance.

关 键 词：三状态样条 KALMAN滤波 机动检测 机动目标 雷达 

分 类 号：TN959[电子电信—信号与信息处理]