摘要
针对多传感器数据配准误差难以估计的问题,基于扩展卡尔曼滤波的基本思想,将目标运动模型和传感器配准误差模型组合在一个状态方程中,研究了扩展维度后的观测向量与状态向量间的关系,构造了简化的观测矩阵,在此基础上,利用扩展卡尔曼滤波方程对三维坐标系下配准误差进行估计,解决了配准误差难以估计的问题。数值试验表明,该方法能在有效估计目标运动状态的同时,估计出目标的传感器的配准误差,且算法复杂度较低、可实现性强。
The multi sensor data registration error is difficult to estimate. Based on the basic idea of extended Kalman filter, the sensor registration errors are incorporated into an augmented dynamic model. The relationship between observed vector and system vector is studied and observed matrix is constructed, on foundation of which the Kalman filter is used to estimate the registration errors, which resolves the difficult problem of registration error. Computer simulations show that system states and registration errors are effectively estimated with less complex algorithm and better implementability.
出处
《电子科技》
2017年第6期34-38,42,共6页
Electronic Science and Technology
基金
国防预先研究项目(12100201)
关键词
数据配准
随机误差
配准误差
观测矩阵
扩展卡尔曼滤波
data registration
random error
registration error
observed matrix
extended Kalman filter
