基于线性均方估计的数据融合理论

The theory of multi-sensor system data fusion based on linear least square estimation

提出数据的统一线性融合模型 ,使测量数据、先验信息和预测信息均用数据的统一融合模型进行描述 ;提出数据的信息量的概念 ,指出信息量越大的数据 ,精度越高 ;提出和证明了基于统一线性融合模型的数据最优融合定理和信息量分解定理 ;揭示数据线性融合的本质规律。基于数据融合理论 ,研究多传感器系统联合Kalman滤波 ,指出信息量分享系数取值的不同只能够影响对应子系统状态的估计精度 ,不能够改变全局估计精度 ,得出与文献 [11～ 15

In this paper,an unified linear fusion model for data fusion is presented,and it can describe varied data including measuring data,apriority data and forecasting data.Based on the unified linear fusion model,both the optimum fusion estimation theorem and the information magnitude decomposing theorem are proposed and proved;and draw a conclusion that “the more measurements are used,the higher accuracy is obtained”.Based on researching multi sensor system federal Kalman filter,point out the fact that the allocation of information can not change global estimating accuracy,but local estimating accuracy,which draws a reverse conclusion to literature[11～15].