IUKF neural network modeling for FOG temperature drift
IUKF neural network modeling for FOG temperature drift
复制标题
FOG 温度漂移的 IUKF 神经网络建模
DOI：
10.1109/jsee.2013.00097
复制
发表时间：
2013-10-01
影响因子：
2.1
通讯作者：
He, Hongyang
中科院分区：
文献类型：
--
作者：
Zha, Feng;Xu, Jiangning;He, Hongyang
建立了一种基于迭代无迹卡尔曼滤波器（IUKF）算法的新型神经网络来建模和补偿由温度引起的光纤陀螺（FOG）偏置漂移。在网络中，FOG 温度及其梯度被设置为输入，FOG 偏置漂移被设置为预期输出。建立了用IUKF算法训练的2-5-1网络。 IUKF算法是在无迹卡尔曼滤波器（UKF）的基础上发展起来的。将隐藏层的权重和偏置向量设置为UKF的状态，并根据网络架构推导了其过程和测量方程。为了解决UKF算法中不可避免的状态均值和协方差的估计偏差，在测量更新后在UKF中引入迭代计算。而测量噪声R在迭代前被扩展到状态向量中，以满足估计和测量噪声的统计正交性。 IUKF算法由于其状态扩展和迭代可以为神经网络提供优化估计。 FOG 的温升 (-20-20°C) 和下降 (70-20°C) 测试在恒温器中进行。采用神经网络建立温度漂移模型，并分别采用BP、UKF和IUKF算法进行训练。结果证明，与反向传播（BP）和UKF网络模型相比，所提出的模型具有更高的精度。
A novel neural network based on iterated unscented Kalman filter (IUKF) algorithm is established to model and compensate for the fiber optic gyro (FOG) bias drift caused by temperature. In the network, FOG temperature and its gradient are set as input and the FOG bias drift is set as the expected output. A 2-5-1 network trained with IUKF algorithm is established. The IUKF algorithm is developed on the basis of the unscented Kalman filter (UKF). The weight and bias vectors of the hidden layer are set as the state of the UKF and its process and measurement equations are deduced according to the network architecture. To solve the unavoidable estimation deviation of the mean and covariance of the states in the UKF algorithm, iterative computation is introduced into the UKF after the measurement update. While the measurement noise R is extended into the state vectors before iteration in order to meet the statistic orthogonality of estimate and measurement noise. The IUKF algorithm can provide the optimized estimation for the neural network because of its state expansion and iteration. Temperature rise (-20-20°C) and drop (70-20°C) tests for FOG are carried out in an attemperator. The temperature drift model is built with neural network, and it is trained respectively with BP, UKF and IUKF algorithms. The results prove that the proposed model has higher precision compared with the back-propagation (BP) and UKF network models.