Research On Close-Loop Control And Temperature Compensation Technologies For Fiber Optic Gyroscope

Fiber Optic Gyroscope(FOG) is a new type of angular rate sensor, which has the characteristics of high precision, small size, low power, wide dynamic range and good reliability etc. It has broad application prospects in the field of navigation and guidance, geodetic engineering, satellite positioning and robot control etc.High precision and low cost are the main future directions of FOG. However, temperature characteristics has become a primary factor that constraint the engineering of high precision FOG. The thesis depends on an advanced research project for a certain weapon. The main purpose is to solve the difficult problems of poor temperature performance in the entire temperature range of FOG which would be applied in the Strapdown Inertial Navigation System. The main work and conclusions of the thesis can be reflected in the following aspects:1. By researching the fundamental principle of FOG, some main factors that causing the temperature drift are analyzed, and some technologies for restrain the FOG's output drift are compared.2. The dynamic model of the system is deduced according to the in-depth analysis of the basic structure and operational principle of digital closed-loop FOG. In order to eliminate the steady-state error and improve the dynamic performance of FOG, a method of designing the digital controller in the detection circuit based on PID algorithm with partial differential is proposed. An algorithm which can adaptively sdjust the controller's parameters is designed and implement, and the control effect is validated by simulation.3. Based on the in-depth analysis of mechanics that causing the temperature drift and some numerous experimental data during high/low temperature environments, a new compensation model for FOG's temperature drift is established according to the Multi Linear Regression theory. The compensation effects under different temperature conditions are validated by simulation. The results indicate that, after compensation the bias stability of FOG can be rifely enhanced by 60%~70%, and the model shows good commonality. At the same time, the Radial Basis Function neural network is used to implement the identification of FOG's temperature drift. After compensation with the identification model, the bias stability of FOG can be improved by about 80%.4. According to the Allan variance theory, the components of FOG's output drift before/after temperature compensation is analyzed. In order to avoid the probable disadvantage of negative noise coefficients fitted, a sub fitting arithmetic is used to get the more accurate noise coefficients. Finally, based on the ARMA modeling of FOG's output drift, the output signals of FOG are processed by Kalman filter. The results indicate that, after filtering both the quantization noise and the random walk coefficients of FOG can be weakened by about 60%, and the bias stability can be further improved.