基于补偿卡尔曼滤波的姿态估计算法实现

摘要

九轴的姿态测量单元包括加速度计、磁力计和陀螺仪,它采用微机电加工工艺,价格低廉但精度较差.为了获得更好的滤波稳定性和估计精度,基于补偿卡尔曼滤波原理,讨论了姿态估计算法的实现.具体过程如下:基于补偿卡尔曼滤波原理,讨论九轴测量信息的融合方法;利用加速度计和磁力计估算姿态初始值;使用手动转台实现静态标定,纠正零偏;使用高精度陀螺仪STIM210实现动态标定,设置滤波系数.结果表明,估计算法具有良好性能.%9-axis orientation measurement unit consists of a 3-axis accelerometer,a 3-axis magnetometer and a 3-axis gyroscope.Its price is low for using MEMS technology,but the accuracy is poor.In order to obtain better filtering stability and estimation accuracy,the implementation of the orientation estimation algorithm based on complementary Kalman filter is discussed.The specific implementation process is as follows:the fusion of the 9-axis signal is discussed based on the principle of complementary Kalman filter,the initial orientation is estimated by using the measurements of the accelerometer and magnetometer,the bias is corrected by static calibration with manual turntable,the filter coefficients are set by dynamic calibration with a high-precision gyroscope STIM210.Simulations demonstrate that the new method achieves desired performance under the condition of large initial errors.