RBF神经网络用于辨识光纤陀螺温度漂移

温度漂移是存在于光纤陀螺系统中使得输出信号产生较大偏置误差的一种不可忽略因素，准确地辨识漂移并有效地对其进行补偿直接关系到陀螺的测量精度。文中比较了前馈网络中的ＢＰ网络和径向基函数（ＲＢＦ）网络，采用ＲＢＦ网络进行温漂辨识，温漂辨识可以通过离线事先学习，因而在多种学习方法中选择了简单易行、精度高且运算速度快的正产郇小二乘（ＯＬＳ）法。通过仿真验证，采用ＲＢＦ网络及其ＯＬＳ学习算法可以快速、有效、高

Identification of Temperature Drift for FOG Using RBF Neural Networks

Temperature drift is a nonnegligible factor causing big biasing error in FOG. How to identify and compensate this error relates directly to the measurement accuracy. After comparing the features of BP and RBF networks, the latter was applied to identify the temperature drift as it can achieve the global optimum evaluation and has the linear weight combiner and fast learning. Of the different approaches of parameter learning, the OLS algorithm is prior by its simplicity, high accuracy and fast speed. The simulation results show that the RBF network based method with the OLS learning offers a powerful and successful procedure for fitting and compensating the temperature drift.