基于序贯最小二乘的多传感器误差配准方法  

RLS-based registration algorithm for multi-sensors

作　　者：赵杰[1] 江晶[2] 周样晶[1] 韩正国[1] 

机构地区：[1]空军雷达学院研究生管理大队 [2]空军雷达学院信息与指挥自动化系,湖北武汉430019

出　　处：《传感器与微系统》2007年第8期99-101,共3页Transducer and Microsystem Technologies

摘　　要：为实时估计多传感器系统偏差,针对广义最小二乘(GLS)配准方法不能实时估计传感器偏差的问题,提出了基于序贯最小二乘的多传感器误差估计方法,该方法在GLS配准模型基础上,采用最小二乘的序贯方法来估计系统偏差,不必存储过去的测量数据,能够实时估计系统偏差。仿真结果表明了该方法的有效性。To estimate systemic errors of multi-sensors in real-time, the registration algorithm based on recursive least squares （RLS） for multi-sensors is presented, concerning for the generalized least-squares estimation （GLS） algorithm being not used for estimating sensors error in real-time. The RLS is employed for estimating systematic error in this algorithm, based on the GLS registration model, and the errors are estimated in real-time while the previously measured data didn＇ t need to be stored. By simulations it shows that the algorithm is effective.

关 键 词：多传感器 误差 配准 序贯最小二乘 

分 类 号：TN957[电子电信—信号与信息处理] TP212[电子电信—信息与通信工程]