论文部分内容阅读
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。
The temperature drift error of FOG is one of the main error sources that affect its output accuracy. In practice, the temperature drift error of FOG must be properly compensated. The traditional least square method and other linear compensation methods are difficult to meet the requirements of the accuracy of the compensation and the applicability is poor. By using BP neural network and RBF neural network respectively, a nonlinear fiber optic gyroscope temperature drift error model is established, which can effectively improve the compensation accuracy. The data is compared with the least squares model and the neural network compensation model to verify the effectiveness of the nonlinear compensation algorithm based on neural network in FOG temperature drift compensation.