A Complementary Filter-Based All-Parameters Estimation for Triaxis Gyroscopes and Optical Angular Encoders With Intrinsic Eccentricity
Outfield tests for inertial sensors have drawn growing attention in recent years since it will provide more practical application conditions. The traditional field tests carried out with simple ancillary equipment cannot maintain high accuracy testing results. In this paper, we implemented the outfi...
Gespeichert in:
Veröffentlicht in: | IEEE sensors journal 2021-02, Vol.21 (4), p.5060-5069 |
---|---|
Hauptverfasser: | , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Outfield tests for inertial sensors have drawn growing attention in recent years since it will provide more practical application conditions. The traditional field tests carried out with simple ancillary equipment cannot maintain high accuracy testing results. In this paper, we implemented the outfield tests on an outfield testing platform with a high-accuracy turntable. However, the platform's movement and deformation in mechanical structure will induce an unknown intrinsic eccentricity to the optical angular encoders (OAE) in the turntable, which will have a negative impact on the testing accuracy. To estimate the unknown eccentricity, the limitation brought by the outfield condition is the difficulty in acquiring and applying the additional standard sensors. Therefore, we propose an all-parameters estimation method is to identify the eccentricity with a triaxis gyroscope under tests synchronously based on a designed complementary filter with the Gauss-Newton. This method will reduce the dependence of additional references. Then, several rules are discussed to design rotation schemes. Finally, both simulations and experiments are performed to evaluate the proposed method. Experiments result show that the method can improve the accuracy of the optical angular encoder and gyroscopes. |
---|---|
ISSN: | 1530-437X 1558-1748 |
DOI: | 10.1109/JSEN.2020.3034325 |