基于气象无人机飞行状态的“野值”判定及剔除方法研究

气象无人机探测过程中误差较大的数据统称为野值数据,研究如何判定并能够成功剔除野值就成为了提高气象无人机测风精度的重要方法。首先通过判定并剔除气象无人机飞行状态中的加速度和角速度的野值数据,进而根据有效的加速度和角速度信息确定无人机飞行状态,建立了匀速直线、变速、转动、其它非惯性运动等四种飞行状态下的野值评判标准,并设定了相应的阈值,给出了基于不同飞行状态的野值判定、剔除方法,通过仿真分析结果表明,该方法明显剔除了误差较大的数据,成功降低由于测量总压波动所引起的风速误差,对提高气象无人机测风精度具有重要参考价值。

Research on the method for determining and removing outliers based on flight status of meteorological UAVs

The data with larger errors are called outliers in detection processes of meteorological UAVs. How to determine and remove outliers is one of the key technologies to increase the measuring accuracy of meteorological UAVs. Firstly, the outliers in acceleration and angular velocity information are determined andremoved, so the flight status of a meteorological UAV can be built based on the clean data of acceleration and angular velocity. Then four criteria of outliers are established and the corresponding thresholds are given based on four flight status of uniform straight line motion, ,non uniform straight line motion, rotational motion and others. Finally four methods for determining and removing outliers in different flight status are presented. A simulation analysis is conducted and the results show that these methods can remove the data with large errors, reduce the wind speed measuring errors from total pressure, and increase the wind measuring accuracy of meteorological UAVs effectively.