摘要
针对目前小样本容量的捷联惯组误差系数预测精度不高的问题,采用最小二乘支持向量机(LS-SVM)对捷联惯组误差系数进行了预测研究,并以某型捷联惯组的某项陀螺漂移误差系数的历史数据为例进行了预测。结果表明,最小二乘支持向量机具有优秀的小样本数据学习能力和预测能力。
Considering to the low prediction precision in small sample of the error coefficients of strapdown Inertial Measurement Unit（SIMU）,Least Squares Support Vector Machines（LS-SVM） is adopted to predict the error coefficients of the SIMU,and prediction of some gyroscope drift error coefficient of some SIMU is proposed.The simulation indicates that the method presented has excellent learning ability and can provide more accurate data prediction.
出处
《科学技术与工程》
2009年第7期1989-1991,2004,共4页
Science Technology and Engineering
关键词
捷联惯性测量组合
支持向量机
最小二乘支持向量机
漂移误差系数
预测
strapdown inertial measure unit support vector machines least squares support vector machines drift error coefficients prediction
