摘要
硅微陀螺在中低精度的惯性稳定平台中具有广泛应用。通过多个硅微陀螺组成阵列,采用数据融合技术可提高硅微陀螺的输出精度。在卡尔曼滤波算法基础上,研究了一种考虑前若干个时刻量测输出值的滚动时域估计(MHE)算法,获得陀螺阵列输出的最优估计。采用2×2的ADIS16080组成陀螺阵列,基于TMS320F28335嵌入式计算平台开展试验研究。试验结果表明,单陀螺输出经卡尔曼滤波后,输出噪声方差的均值为4.179 4 s~(-2)。采用陀螺阵列信息融合后,方差降至4.071 7 s~(-2)。进一步采用预测长度为3的MHE算法可将输出噪声方差减少至2.127 3 s~(-2)。
Silicon micro gyroscopes have been widely used in medium-low precision inertial stability platform. The output precision of silicon micro gyroscopes array can be improved by data fusion. Based on Kalman filtering for gyroscopes output,a Moving Horizon Estimation( MHE) algorithm was studied which made use of several former measurements. The optimal output estimation can be acquired by MHE. Based on the TMS320 F28335 embedded platform,a gyroscope array with 2×2 ADIS16080 was built up. Experiments show that after Kalman filtering,the average variation for single gyroscope output noise is 4.1794 s~(-2). The variation can be reduced to 4.071 7 s~(-2) by gyroscope array data fusion. Furthermore,the noise variation after MHE with prediction horizon is 3 can be reduced to 2.127 3 s~(-2).
引文
[1]何昆鹏,程万娟,高延滨,等.虚拟陀螺技术在MEMS惯性导航系统中的应用[J].哈尔滨工程大学学报,2009,30(10):1123-1128.
[2]XU G,XU H,JIN W.Study on stabilization and tracking for SOTM[C].IEEE the Ninth International Conference on Electronic Measurement and Instruments.Piscataway,N.J:IEEE Press,2009.
[3]DEBRUIN J.Control systems for mobile satcom antennas[J].IEEE Control Systems,2008,28(1):86-101.
[4]DAVID S B,SCOTT R P.Combining multiple gyroscope outputs for increased accuracy[R].Pasadena,California:NASA,2003.
[5]KIM S,LEE B,LEE J,et al.A gyroscope array with linked beam structure[C].the 14th IEEE International Conference on Micro Electro Mechanical Systems.Piscataway,N.J:IEEE Press,2001.
[6]张印强,杨波,李婧,等.硅微阵列陀螺仪的仿真、设计与测试[J].传感技术学报,2012,25(1):29-32.