Hostname: page-component-7b7567c8c-tnkh4 Total loading time: 0 Render date: 2026-11-16T11:15:33.886Z Has data issue: false hasContentIssue false

Offline Calibration for MEMS Gyroscope G-sensitivity Error Coefficients Based on the Newton Iteration and Least Square Methods

Published online by Cambridge University Press:  11 October 2017

Li Xing
Affiliation:
(Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China)
Zhi Xiong*
Affiliation:
(Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China)
Jian-ye Liu
Affiliation:
(Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China)
Wei Luo
Affiliation:
(Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China)
Ya-zhou Yue
Affiliation:
(Flight Automatic Control Research Institute, Xi'an, China)
Rights & Permissions [Opens in a new window]

Abstract

With the improvement of the bias instability of Micro-Electromechanical Systems (MEMS) gyroscopes, the g-sensitivity error is gradually becoming one of the more important factors that affects the dynamic accuracy of a MEMS gyroscope. Hence there is a need for correcting the g-sensitivity error. However, the traditional calibration of g-sensitivity error uses a centrifuge. The calibration conditions are harsh, the process is complex and the cost is relatively high. In this paper, a fast and simple method of g-sensitivity error calibration for MEMS gyroscopes is proposed. With respect to the bias and random noise of a MEMS gyroscope, the g-sensitivity error magnitude is relatively small and it is simultaneously coupled with the Earth's rotation rate. Therefore, in order to correct the g-sensitivity error, this work models the calibration for g-sensitivity error coefficients, designs an (8+N)-position calibration scheme, and then proposes a fitting method for g-sensitivity error coefficients based on the Newton iteration and least squares methods. Multi-group calibration experiments designed on a MEMS Inertial Measurement Unit (MEMS IMU) product demonstrate that the proposed method can calibrate g-sensitivity error coefficients and correct the g-sensitivity error effectively and simply.

Information

Type
Research Article
Copyright
Copyright © The Royal Institute of Navigation 2017 
Figure 0

Figure 1.Silicon MEMS gyroscope internal simple structure.

Figure 1

Figure 2.Silicon MEMS gyroscope equivalently simplified kinetic model

Figure 2

Figure 3.(8 + N(N ≥ 1))-position calibration experiment scheme.

Figure 3

Figure 4.(8 + N(N ≥ 1)) calibration positions diagram.

Figure 4

Figure 5.The other seven positions of the STIM300 turntable experiment.

Figure 5

Figure 6.The MEMS IMU product STIM300 and its data acquisition equipment.

Figure 6

Figure 7.MEMS gyroscope raw output of STIM300 at 16 positions.

Figure 7

Figure 8.MEMS accelerometer raw output of STIM300 at 16 positions.

Figure 8

Table 1.Calibration values of g-sensitivity error coefficients in MEMS gyroscope of STIM300.

Figure 9

Figure 9.Comparison results between Δωie_before and Δωie_after at each turntable position.

Figure 10

Table 2.Mean and standard deviation comparison of Δωie_before and Δωie_after

Figure 11

Figure 10.MEMS accelerometer raw output at four arbitrary positions.

Figure 12

Table 3.Comparison Results of Δωie_before and Δωie_after at four arbitrary positions.

Figure 13

Table 4.The g-sensitivity error coefficients calibrated results comparison in five group experiments.

Figure 14

Table 5.The MEMS gyroscope bias calibrated results comparison in five group experiments.

Figure 15

Figure 11.Comparison between Δωie_before and Δωie_after in five group experiments for analysing the repeatability of the coefficients calibrated.