小型无人直升机模型辨识数据处理方法研究

针对小型无人直升机模型频域辨识过程中的姿态角速率测量误差，提出了一种飞行数据处理方法。该方法采用飞行试验扫频测试技术，确保激励信号能够满足不同频段下模型辨识对飞行数据的需求；设计基于有色噪声的卡尔曼滤波器以降低紊流风场对飞行测量数据的影响，同时，对飞行测量数据使用数据预处理的方法以剔除测量噪声、野值、直流成分和低频分量。在小型无人直升机系统各通道中进行验证，验证结果表明，所提出的飞行数据处理方法能够满足小型无人直升机模型辨识对姿态角速率数据精度的要求，为精确建模提供了较高质量的飞行数据。

Research on data processing method for model identification of small unmanned helicopter

A flight data processing method is proposed for the measurement error of attitude angular rate in frequency domain identification of small unmanned helicopter model. The method adopts the flight test frequency sweep test technology to ensure that the excitation signal can meet the requirements of flight data for model identification in different frequency bands. The Kalman filter based on colored noise was designed to reduce the impact of turbulent wind field on flight measurement data, and a method of data preprocessing was used for the flight measurement data to eliminate measurement noise, outliers, DC components and low frequency components. The verification was performed in each channel of the small unmanned helicopter system, the verification results show that the proposed flight data processing method can meet the requirements of attitude angular rate data accuracy for the identification of small unmanned helicopter model and provide high quality flight data for accurate modeling.