摘要
针对多个传感器对某一特性指标进行测量实验的数据融合问题,从稳健性角度,利用统计理论中的最小一乘估计,提出一种多传感器数据的融合方法。该方法基于自适应加权,以最小化传感器测量数据的绝对偏差为目标函数,通过求解条件极值问题,得到各传感器数据的权数,从而给出融合结果。仿真实例表明方法的有效性和较好的稳健性。
Due to data fusion of multi-sensor experiment on some characteristic index,applied the least absolute deviation estimation in the theory of statistic,a fusion method for multi-sensor data is brought forward from the angle of stability. Based on the adaptive weight,the object function of the method is to minimize the absolute deviation of sensor＇s measurement,the fusion weights are obtained by solving the conditioned extremum problem,so the result of fusion is given. Simulation examples show that the method is not only effective,but also very robust.
出处
《计算机工程》
CAS
CSCD
北大核心
2010年第2期257-259,共3页
Computer Engineering
基金
国家自然科学基金资助项目(10626029)
江西省自然科学基金资助项目(0611082
2007GQS0074)
江西省教育厅科技基金资助项目(GJJ08350)
关键词
多传感器
数据融合
最小一乘估计
最小二乘估计
稳健性
multi-sensor
data fusion
least absolute deviation estimation
least square estimation
robustness
