基于扩维贴近度的多传感器一致可靠性融合方法

摘要

对于先验知识未知的多传感器对某一目标特性进行多次测量的情形,提出一种基于模糊贴近度函数的融合算法.首先视各传感器的测量值为模糊集合,利用最大最小贴近度定义贴近度矩阵,并将前一次的融合值假设为观测值,参与当前次的融合运算,再通过扩维来度量不同时刻传感器测量值间的综合相近程度,然后参照已有的信噪比法定义了一致可靠性测度,进而得到多传感器数据的融合公式.应用实例及仿真验证了该方法的有效性.结果表明该算法计算简单,融合结果具有很好的精确性和稳健性.%In the case of multi-sensors measurement of many times on some characteristic index with unknown prior knowledge,a simple fusion method based on the fuzzy degree of nearness is proposed. This method views all the sensors' observation values as a fuzzy set and defines the nearness degree matrix by the maximum and minimum principle. This method views the previous fusion value as an observation values and uses it in the current time fusion arithmetic. Then the nearness degree matrix's dimension is augmented to measure the mutual integrated nearness degree of sensors at different times. The final fusion expression is obtained based on the defined consistent reliability measurement. Applied example and simulations prove that the proposed method is simple and effective. The compared analyses show that this method not only has higher fusion precision, but also has excellent ability of stableness.