基于小波支持向量机的陀螺随机漂移预测模型研究

通过小波变换抑制各种干扰噪声，预处理后的陀螺漂移数据采用支持向量机的方法建立陀螺漂移预测模型。试验得到的陀螺漂移数据对提出的模型进行验证。结果表明，相对于独立的支持向量机模型（SVM）和径向基神经网络模型（RBF），提出模型得到的陀螺随机漂移预测精度更高。

Gyroscope Random Prediction Drift Model Based on Wavelet-support Vector Machine

Based on the analysis of the existing models, an improved gyroscope random drift model wavelet transform- support vector machine （WT- SVM） is provided. Firstly, disturbing impassive noises is eliminated by wavelet transform. The pretreatment data is used to construct gyroscope random drift model with method of support vector machine. The numerical results indicate the proposed model can reduce the drift model error and improve the model accuracy. And the performance of proposed model is more effective than RBF and support vector machine model in prediction accuracy.