A Vondrak Low Pass Filter for IMU Sensor Initial Alignment on a Disturbed Base

Sensors

Zengke Li¹  Jian Wang¹  Jingxiang Gao¹  Binghao Li² 

[1] School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 22116, China; E-Mails:;School of Surveying and Spatial Information Systems, The University of New South Wales, Sydney 2052, Australia; E-Mail:

关键词: initial alignment;    Vondrak filter;    genetic algorithms;    disturbed base;   

DOI  :  10.3390/s141223803

来源: mdpi

【 摘 要 】

The initial alignment of the Inertial Measurement Unit (IMU) is an important process of INS to determine the coordinate transformation matrix which is used in the integration of Global Positioning Systems (GPS) with Inertial Navigation Systems (INS). In this paper a novel alignment method for a disturbed base, such as a vehicle disturbed by wind outdoors, implemented with the aid of a Vondrak low pass filter, is proposed. The basic principle of initial alignment including coarse alignment and fine alignment is introduced first. The spectral analysis is processed to compare the differences between the characteristic error of INS force observation on a stationary base and on disturbed bases. In order to reduce the high frequency noise in the force observation more accurately and more easily, a Vondrak low pass filter is constructed based on the spectral analysis result. The genetic algorithms method is introduced to choose the smoothing factor in the Vondrak filter and the corresponding objective condition is built. The architecture of the proposed alignment method with the Vondrak low pass filter is shown. Furthermore, simulated experiments and actual experiments were performed to validate the new algorithm. The results indicate that, compared with the conventional alignment method, the Vondrak filter could eliminate the high frequency noise in the force observation and the proposed alignment method could improve the attitude accuracy. At the same time, only one parameter needs to be set, which makes the proposed method easier to implement than other low-pass filter methods.

【 授权许可】

CC BY   
© 2014 by the authors; licensee MDPI, Basel, Switzerland.