| 摘要: |
| 随着无人机技术在航空磁测领域的迅速发展，需要使用航磁补偿技术对无人机载体磁场干扰进行补偿，以获取准确的磁场信息。通过分析四旋翼无人机电机因姿态角变化率变化而产生的非线性磁场，提出了一种考虑非线性磁场因素的磁补偿方法。该方法在Tolles-Lawson(T-L)模型补偿的基础上，使用一维卷积神经网络（1DCNN）建立非线性磁场模型，并对T-L模型补偿的结果进行二次补偿。通过实际飞行数据验证表明，该方法能够准确估计无人机载体的干扰磁场，具有较高的补偿精度和较强的鲁棒性，使用该方法的磁补偿改善比达到了20以上。 |
| 关键词: 航磁补偿 神经网络 航空磁测 磁干扰模型 电机磁场 |
| DOI： |
|
| 基金项目: |
|
| A one-dimentional convolutional neural network aeromagnetic compensation method based on Tolles-Lawson model |
| LIU Qiang,WANG Tao,ZHANG Xiaoming,LIU Jun,ZHANG Ge,WANG Yaguo |
| (School of Instrumentation and Electronics, North University of China, Taiyuan 030051, China;Military Representative Office for Land Equipment in Changzhi Region, Changzhi， Shanxi 046000, China) |
| Abstract: |
| With the rapid development of drone technology in the field of airborne magnetic surveys, it is necessary to use aeromagnetic compensation to compensate for the magnetic interference of unmanned aerial vehicle (UAV) carriers to obtain accurate magnetic field information. By analyzing the nonlinear magnetic fields generated by the motors of quadrotor UAVs due to changes in the rate of attitude angles, a magnetic compensation method that considers nonlinear magnetic field factors is proposed. Based on the Tolles-Lawson (T-L) model compensation, this method establishes a nonlinear magnetic field model using a one-dimensional convolutional neural network (1DCNN) and performs secondary compensation on the results of the T-L model compensation. Validation by actual flight data shows that this method can accurately estimate the interference magnetic field of the UAV carrier, achieving high compensation accuracy and strong robustness. And the magnetic compensation improvement ratio using this method exceeds 20. |
| Key words: Aeromagnetic compensation Neural network Airborne magnetic survey Magnetic interference model Motor magnetic field |
