中文摘要：

SVM是一种理论依据充分的机器学习新算法，主要用于对有限个样本的分类识别和回归建模。将一条弹道视为一个训练样本，利用SVM回归方法在不同弹道之间归纳弹道落点的规律，从而可以对未知弹道的落点进行预测。将SVM引入弹道外推是重要创新，构建具有代表性的训练样本，以及统一样本空间维数等是技术创新。仿真实验表明，SVM可以提高弹道外推精度，同时缩短外推时间。

英文摘要：

SVM is a new machine learning method with sufficient theoretical motivation, and it is mainly used for classification recognition and regression modeling in terms of finite samples. In this paper, each trajectory is viewed as a training sample, and SVM regression method is applied to induce the law of the trajectory falling points. Therefore, the falling points can be predicted. One of the most important contribution in this paper is that we introduce SVM into the field of trajectory extrapolation of radar,and the another contribution lies in we apply SVM using many techniques, which includes constructing representative samples and unifying the dimension of input space. The simulation experiments demonstrate that SVM can improve the accuracy of trajectory prediction while reduce the prediction time.