两轴平台稳定系统中MEMS陀螺误差建模与分析

发布时间：2019-03-16 07:55

【摘要】：平台稳定系统能隔离运动载体对探测系统的扰动，实时监测平台姿态和位置的变化，同时施加逆向力矩对外界干扰进行补偿，精确保持探测器动态姿态基准不变，并通过平台上的光电探测器实现对定目标的监测或运动目标的自动跟踪。陀螺仪是平台稳定系统中的核心敏感元件，其测量精度直接影响平台的稳定精度。MEMS陀螺具有重量轻、体积小、低价格、低功耗及高可靠性等优点，广泛应用于军事装备和民用设施中。但由于MEMS陀螺加工材料在微米量级以及加工工艺水平等方面的制约，目前国内外MEMS陀螺测量精度水平仍然较低，提高MEMS陀螺的测量精度是促进平台稳定系统小型化的关键。本文设计了应用于微小型光电吊舱两轴平台稳定系统中MEMS陀螺信号采集系统，并在采集系统中加入了以数字电位器为核心的陀螺零位调整电路，，通过DSP实时控制数字电位器输出，实现对陀螺零位中心的调整。环境温度变化是影响MEMS陀螺测量精度的最主要因素之一。由于MEMS陀螺微型化尺寸和内部构造，其性能对环境温度的变化十分灵敏。其中，温度变化对陀螺零位输出和标度因数影响最大。因此，本文设计了获取陀螺零位特性的全温测试实验与电源模拟陀螺零位变化时调整数字电位器的实验，研究获得陀螺零位补偿模型，并通过对数字电位器的控制实现零位中心的调整。温度补偿实验表明，采用全温测试以及数字电位器补偿的方法，陀螺在全温范围内的零位稳定精度得到较大的提高，具有一定的实用价值。设计了陀螺在不同温度下的转速标定实验，采用了三种标度因数拟合方法，通过拟合曲线残差的对比，得出按转速分段拟合方法残差最小，效果最佳。陀螺随机漂移误差也是陀螺测量误差的主要来源之一。利用时间序列分析法对MEMS陀螺随机漂移采样数据进行ARMA建模，通过最小AIC准则确定AR(1)模型为最佳误差模型；在该模型的基础上，采用Kalman滤波方法对随机漂移数据进行滤波，对滤波前后陀螺随机误差进行Allan方差分析。结果表明，Kalman滤波方法在陀螺随机误差滤波中有效、可行。
[Abstract]:The platform stability system can isolate the disturbance of the moving carrier to the detection system, monitor the change of attitude and position of the platform in real time, and at the same time apply the reverse torque to compensate the external disturbance, so as to keep the dynamic attitude reference of the detector exactly unchanged. And through the photoelectric detector on the platform to achieve target monitoring or automatic tracking of moving targets. Gyroscope is the core sensitive element in the platform stabilization system, and its measurement precision directly affects the stability precision of the platform. MEMS gyroscope has the advantages of light weight, small volume, low price, low power consumption and high reliability, etc. It is widely used in military equipment and civilian facilities. However, due to the limitation of MEMS gyroscope machining materials in micrometer order and processing level, the measurement accuracy of MEMS gyroscope is still low at home and abroad. Improving the measuring accuracy of MEMS gyroscope is the key to promote the miniaturization of platform stabilization system. In this paper, a MEMS gyroscope signal acquisition system is designed for the two-axis platform stabilization system of micro-miniature photoelectric pods, and a gyro zero adjustment circuit with digital potentiometer as the core is added to the acquisition system. The zero center of gyro is adjusted by real-time control of digital potentiometer output by DSP. The change of ambient temperature is one of the most important factors affecting the measurement accuracy of MEMS gyroscope. Due to the miniaturized size and internal structure of MEMS gyroscope, its performance is sensitive to the variation of ambient temperature. Among them, the temperature change has the greatest influence on the zero-position output and scale factor of gyro. Therefore, this paper designs the full-temperature test experiment to obtain the zero-position characteristic of gyro and the experiment of adjusting digital potentiometer when the zero-position change of gyro is simulated by the power supply, and obtains the zero-position compensation model of gyro. By controlling the digital potentiometer, the zero-bit center can be adjusted. The temperature compensation experiment shows that the zero-position stability precision of gyro in the whole temperature range is greatly improved by the method of full-temperature measurement and digital potentiometer compensation, which has a certain practical value. The rotating speed calibration experiment of gyro at different temperature is designed. Three kinds of scale factor fitting method are adopted. By comparing the residual error of fitting curve, it is concluded that the residual error of rotating speed segment fitting method is the least and the effect is the best. Gyro random drift error is also one of the main sources of gyro measurement error. The time series analysis method is used to model the random drift sampling data of MEMS gyroscope by ARMA. The AR (1) model is determined to be the best error model by the minimum AIC criterion. On the basis of this model, the random drift data is filtered by Kalman filtering method, and the random error of gyro is analyzed by Allan variance before and after filtering. The results show that the Kalman filtering method is effective and feasible in gyro random error filtering.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TN965;TH-39

【参考文献】