MEMS陀螺随机误差补偿在提高姿态参照系统精度中的应用

为提高航姿系统姿态角测量精度,论文对MEMS陀螺随机误差的测试和补偿进行了研究。运用Allan方差方法分析了MEMS陀螺各项随机误差的特性,建立了针对MEMS陀螺的随机误差补偿模型。采用Allan方差辨识算法完成对各项随机误差的分离,获取速率随机游走(RRW)和角度随机游走(ARW)噪声方差值。通过设计的卡尔曼滤波器对MEMS陀螺随机误差进行实时估计补偿,利用姿态解算算法得到随机误差补偿前和补偿后各3个方向的姿态角。实验结果表明经误差补偿后系统的静态偏航角测量精度提高了3倍、横滚角提高了4倍、俯仰角提高了12倍。

Compensating for Random Noises in a Low-Precision MEMS Gyroscope for Improving Accuracy of Attitude Reference System

Aim.To our knowledge,there does not exist,in the open literature,any compensation method suitable for improving a low-precision MEMS gyroscope.We now apply the Allan variance method,we believe,successfully to such compensation.In section 1 of the full paper,we extract the main random noises of the gyroscope such as rate random walk(RRW) and angle random walk(ARW) as shown in Fig.2.Section 3 has three subsections.In subsection 3.2,we give the calculation results of random noises of the gyroscope in Table 2.In subsection 3.3,we perform the real-time estimation of and compensation for the random noises and simulate the yaw angle,roll angle and pitch angle of an attitude reference system.The simulation and experimental results,given in Figs.7 through 9 and Table 3,show preliminarily that our compensation method reduces the yaw angle error to one third of that uncompensated for,the roll angle error to one fourth and the pitch angle error to one twelfth.