基于最小二乘原理的多传感器加权融合算法

A Weighted Fusion Algorithm of Multi-sensor Based on the Principle of Least Squares

以存在随机扰动环境中的不同参数多传感器为研究对象 ,基于最小二乘原理 ,提出了一种加权融合算法 ,推导出各传感器的权系数与测量方差的关系。并且根据测量信息 ,提出了一种方差估计学习算法 ,实现对各传感器测量方差的估计 ,从而对各传感器的权值进行合理的分配。该算法简单 ,能快速。

Taking multi sensor of variant parameters in the environment with stochastic noise as object, the paper proposes a weighted fusion algorithm based on the principle of least squares and deduces the relationship among the weights of respective sensor and the variance of the measured error. The paper also proposes a new learning algorithm for variance estimation. Using the measurements, the learning algorithm can estimate the variance of respective sensor, and then is able to distribute the sensor's weights properly. The algorithm proposed is simple and able to estimate the state of the variable to be measured quickly and exactly.