基于Kalman滤波的数据处理多尺度融合算法  

Algorithm of Multi-scale Fusion for Data Processing Based on Kalman Filter

作　　者：司长哲[1] 任松[1] 

机构地区：[1]西昌卫星发射中心技术部,四川西昌615000

出　　处：《计算机测量与控制》2008年第2期165-167,共3页Computer Measurement &Control

摘　　要：针对我国航天测量事后数据处理的现状,提出了一种基于Kalman滤波的数据处理多尺度融合算法;该算法以建立的系统动态模型方程为基础,将多台雷达的测量数据分别在不同尺度上逐次分解,然后在同一尺度上对所有雷达数据进行系统融合处理,从而提高测量数据的处理精度;用该算法处理某次卫星发射任务的理论弹道,计算结果表明:该算法处理的效果明显优于单台雷达直接进行Kal-man滤波处理的效果,并且融合尺度越小,处理精度改善的效果越明显;该算法与α-β-γ滤波算法相比,数据处理精度有较大提高。According to the actuality of data processing after aerospace measurement mission in China, an algorithm of multi-scale data fusion for data processing based on kalman filter is put forward. On the basis of system dynamic model equation, the algorithm decomposes step by step the measuring data of many radars in different scale, then the algorithm fuses all the data in the same scale, by the method the precision of data processing is improved. The algorithm has processed the theory trajectory of a satellite launch mission, the computing result demonstrates that the algorithm is more effective than processing data of a radar directly by Kalman filter, furthermore the scale is more smaller, the effect of precision melioration is more evident; contrasting the fusion algorithm with a-β-τfilter algorithm, the precision of data processing is improved markedly.

关 键 词：航天测量 数据处理 多尺度融合算法 KALMAN滤波 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]