基于RBF神经网络的光纤陀螺启动补偿及应用

光纤陀螺对温度比较敏感，由于温度引起的零偏漂移是光纤陀螺工作尤其是启动过程中的一种较大误差。文中为了减小光纤陀螺启动过程的零偏漂移、缩短启动时间，提出了对光纤陀螺启动过程进行补偿的方案。该方案以光纤陀螺温度和温度变化率为输入、光纤陀螺漂移为输出建立二输入单输出的RBF神经网络，用于陀螺启动过程补偿。在室温下对某型号光纤陀螺启动漂移进行了补偿，试验结果表明该方法能有效减小陀螺的启动温度漂移，缩短陀螺启动时间。将该方案运用到某型号的光纤陀螺寻北仪上，常温试验表明，该方案大大缩短了寻北仪的准备时间，提高了寻北精度。

Application and compensation for startup phase of FOG based on RBF neural network

Fiber optic gyroscope(FOG) is sensitive to temperature, and there is a certain temperature drift error in the working process of FOG especially in the startup phase. In this paper, to reduce the bias drift in the startup phase of FOG and shorten the startup time, a scheme based on radial basis function (RBF) neural networks was designed to compensate the drift in the startup phase of FOG. The model took the temperature of FOG and the temperature change rate as the inputs and used the bias drift of FOG as the output. In the room temperature, the RBF neural network was used to compensate the startup drift of FOG, and the experiment shows that the method can effectively reduce the temperature drift and shorten the startup time of FOG. This method is used in a certain type of FOG north finder and can greatly reduce the preparation time, and so improves the north-seeking accuracy.