海参捕捞装置导航中低成本陀螺仪的降噪研究与试验

Research on noise reduction and its experiments of low-cost gyroscopes for sea cucumber fishing device navigation

关键词:  MEMS陀螺仪  惯性导航  AR模型  Kalman滤波  海参捕捞

Key words:  MEMS gyroscope  inertial navigation  AR model  Kalman filtering  sea cucumber fishing

摘要:

针对MEMS陀螺仪输出信号中含有的随机漂移噪声造成海参捕捞装置惯性导航精度明显下降的问题，采用时间序列分析法和Kalman滤波算法，对MEMS陀螺仪随机漂移噪声的削减问题进行研究。以MEMS惯性测量单元的实测数据和三维电子罗盘测量的姿态角为样本，对本研究的低成本陀螺仪的降噪效果进行测试，试验结果表明：1）经过降噪处理后的陀螺仪随机漂移信号的方差比陀螺仪原始采样信号的方差降低1个数量级，显著改善了陀螺仪随机漂移数据的精密度；2）以高精度三维电子罗盘实测的姿态角作为参考基准，将降噪后的陀螺仪随机漂移数据导入捷联惯导姿态更新算法程序，在300 s时间内，解算出的俯仰角、横滚角和航向角的均方根误差RMSE全部小于1°；与降噪前相比，相应的俯仰角、横滚角、航向角的RMSE分别降低了170.5、97.6和42.5倍，明显提高了惯性导航精度。

Abstract:

In order to solve the problem that the significant reduction of inertial navigation accuracy of sea cucumber fishing device caused by random drift noise in the MEMS gyroscope output signal, the reduction issue of random drift noise of the MEMS gyroscope is studied by using time series analysis method and Kalman filtering algorithm. Taking the measured data of the MEMS inertial measurement unit and attitude angles measured by the three-dimensional electronic compass as samples, the noise reduction effect of the low-cost gyroscope is tested. The experimental results show that:1) After the noise reduction, the variance of the random drift signal of the gyroscope is reduced by one order of magnitude compared with the variance of the original sampling signal of the gyroscope, and the precision of the random drift data of the gyroscope is significantly improved. 2) Using the attitude angle measured by the high-precision three-dimensional electronic compass as a reference, the de-noised gyroscope random drift data are imported into the SINS attitude calculation program, and the RMSEs of the calculated pitch angle, roll angle, and heading angle are all less than 1° in 300 s; Compared with before noise reduction, the RMSEs of the corresponding pitch, roll, and heading angles are respectively reduced by 170.5, 97.6, and 42.5 times, which significantly improves the inertial navigation accuracy.