基于鲁棒最小二乘支持向量机的气动参数拟合

最小二乘支持向量机（LS－SVM）比标准支持向量机具有更高的计算效率,但是却散失了标准支持向量机的稀疏特性,而且当考虑异常值或者误差变量的高斯假设不成立时,会导致不稳健的估计结果。为了克服这两个缺点,在飞行器的气动参数拟合计算中引入了一种鲁棒最小二乘支持向量机（RLS－SVM）,该方法通过加权的支持向量机来获得鲁棒估计,并通过对支持值谱进行剪枝最终得到稀疏解。仿真结果表明：RLS－SVM方法简单,学习速度快,拟合精度高,鲁棒性强,是一种在飞行器轨迹计算中值得推广和采用的方法。

Aerodynamic parameter fitting based on robust least squares Support Vector Machines

LS－SVM is computationally more efficient than the standard SVM,but unfortunately the sparseness of standard SVM is lost,another problem is that LS－SVM might lead to estimates which are less robust with respect to outliers on the data or when the assumption of a Gaussian distribution for the error variables is not realistic.In this paper,a new modified version,robust LS－SVM,which can obtain robust estimates by applying a weighted LS－SVM and achieve the sparseness by doing pruning based upon the support value spectrum,is adopted and introduced into the aerodynamic parameter fitting.The simulation results indicate that robust LS－SVM has a better fitting of aerodynamic parameter with features of simplicity,precision,robustness and rapidness to learn.It is a good method which is worthy of popularizing and promoting in calculating of aircraft’s flight trajectory．