Modeling and Compensation Method for Temperature Error of FOG based on the SVM
Guochen Wang, Wei Gao, Bo Zhao, Runfeng Zhang
Available Online July 2015.
- https://doi.org/10.2991/lemcs-15.2015.192How to use a DOI?
- Component; Fiber Optical Gyroscope (FOG); Temperature Error; Support Vector Machine (SVM); Bias Stability; Error Compensation
- The performance of the Fiber Optical Gyroscope (FOG) is susceptible to the external environment temperature, and the temperature effect of the FOG has nonlinear characteristics. Therefore, it is necessary to develop an outstanding temperature compensation method with high precision and good versatility. In this paper, the severe influence of environmental temperature on the FOG’s accuracy was analyzed in detailed. Then the temperature error model of the FOG was established. Based on the function fitting capability of the Support Vector Machine (SVM), a new modeling and compensation method of the gyroscope’s temperature error was proposed. To verify the advantages of this novel method, we compared it with other two compensation methods, which are based on the classical statistical estimation parameters (polynomial regression) and the neural network, respectively. Experiment results showed that the proposed method can decrease the temperature error of the FOG by 75% and improve the adaptability of the environmental temperature. With this novel method, the gyro bias stability can meet the practical need of the temperature compensation which requires high-precision FOG.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Guochen Wang AU - Wei Gao AU - Bo Zhao AU - Runfeng Zhang PY - 2015/07 DA - 2015/07 TI - Modeling and Compensation Method for Temperature Error of FOG based on the SVM BT - International Conference on Logistics Engineering, Management and Computer Science (LEMCS 2015) PB - Atlantis Press SP - 972 EP - 977 SN - 1951-6851 UR - https://doi.org/10.2991/lemcs-15.2015.192 DO - https://doi.org/10.2991/lemcs-15.2015.192 ID - Wang2015/07 ER -