现代电子技术

单一传感器的数据容易受噪声、误差或环境因素等影响，导致采集信息不完整，难以实现飞机系统的稳定控制。为此，提出一种新的机载多传感器数据自适应融合技术。在分析不同类型机载传感器的基础上，去除各个传感器的测量数据最高值和最低值，再利用格拉布斯准则删除中值和平均值差距过大的异常值，获取正常机载传感器测量数据。然后求取传感器自相关函数与任意传感器互相关函数平均值的差，通过这种自适应方式获取正常机载传感器测量数据值的均方误差，并依据均方误差确定传感器加权因子，利用加权因子估算传感器融合结果。实验结果表明：所提技术能够有效融合不同类型机载传感器的数据信号；且基于融合后的高精度数据，飞机飞行状态的经度、纬度以及高度误差三种飞机飞行位置误差可保持在0 m附近，能够为飞行器实时提供更精确的机载设备系统工作参数，确保飞机飞行过程的稳定性与可靠性。

 The data from a single sensor is easily affected by noise, error and environmental factors, resulting in incomplete information collection and making it difficult to achieve stable control of the aircraft system. On this basis, a new adaptive fusion technology for airborne multi-sensor data is proposed. On the basis of analyzing different types of airborne sensors, the highest and lowest values of the measurement data of each sensor are removed, and then the Grubbs criterion is used to remove outliers with excessive deviation between the median and mean values, so as to obtain normal airborne sensor measurement data. The difference between the autocorrelation function of sensors and the average value of the cross-correlation functions of arbitrary sensors are calculated to obtain the mean square error of normal airborne sensor measurement data values. Based on the mean square error, the sensor weighting factor is obtained, and the sensor fusion result is estimated by means of the weighting factor. The experimental results show that this technology can effectively fuse data signals from different types of onboard sensors, and the three aircraft flight position errors, namely the longitude, latitude and altitude errors of the aircraft flight state can be kept near 0 m based on fused high-precision data, providing real-time and more accurate airborne equipment system operating parameters for aircraft piloting, and ensuring the stability and reliability of the aircraft during flight.