基于神经网络的光纤陀螺温度漂移误差建模与补偿

光纤陀螺（FOG）温度漂移误差是影响其输出精度的主要误差源之一，在实际应用中必须对光纤陀螺温度漂移误差进行适当补偿。传统的最小二乘法等线性补偿方法很难满足补偿精度的要求且适用性较差，利用BP及RBF神经网络分别建立非线性光纤陀螺温度漂移误差模型，可以有效提高补偿精度，使用FOG温箱实测数据对最小二乘模型及神经网络补偿模型进行了测试对比，验证了基于神经网络的非线性补偿算法在FOG温度漂移补偿中的有效性。

Modeling and Compensation Algorithm of FOG Temperature Drift with Neural Network

FOG temperature drift is one of the major error sources that affect the FOG output precision,which is must be compensated in pratical application.Traditional least square method show low accuracy and poor aplicablity in the application of FOG temperature drift compensation.Modeling and compensation with BP or RBF neural network can improve the compensation accuracy effectively. Verifying and comparison the least square method with neural network compensation model, results show that this non-linear model based on neural network can improve the FOG temperature drift error compensation accuracy effectively.