| 基于k最近邻的激光雷达飞机尾涡识别 |
| |
| 引用本文： | 潘卫军,吴郑源,张晓磊. 基于k最近邻的激光雷达飞机尾涡识别[J]. 激光技术, 2020, 44(4): 471-477. DOI: 10.7510/jgjs.issn.1001-3806.2020.04.013 |
| |
| 作者姓名： | 潘卫军 吴郑源 张晓磊 |
| |
| 作者单位： | 中国民用航空飞行学院 空中交通管理学院，广汉 618307 |
| |
| 基金项目： | 民航局安全能力建设基金资助项目;国家自然科学基金;研究生科研创新项目 |
| |
| 摘 要： | 为了对脉冲多普勒激光雷达探测径向速度场进行识别，建立了基于k最近邻(KNN)方法的分类模型。结合飞机尾涡Hallock-Burnham模型和脉冲多普勒激光雷达特性，提取脉冲多普勒激光雷达探测径向速度场的特征参量。基于现有测试数据，在非均匀背景风场下利用KNN方法对飞机尾涡进行识别。结果表明，以准确率(ACC)和ROC曲线下的面积(AUC)作为评估标准，所提出的方法对尾流识别所获得的ACC和AUC分别为0.772和0.855。该方法可有效地识别飞机尾涡并具备较好的鲁棒性。
|
| 关 键 词： | 激光技术 多普勒激光雷达 k最近邻 特征提取 尾涡识别 |
| 收稿时间： | 2019-07-17 |
Identification of aircraft wake vortex based on k-nearest neighbor |
| |
| Abstract: | To identify aircraft wake vortex by pulsed doppler LiDAR's characteristics, a classification model based on k-nearest neighbor (KNN) was established in this paper. This approach by combining Hallock-Burnham model with pulsed doppler LiDAR's characteristics to extract the feature parameters of radial velocity of wind field was pursued. Based on the test dataset, the KNN was employed to identify aircraft wake vortex in the context of nonuniform wind field. The performance of the proposed method was evaluated in terms of the accuracy (ACC) and the area under ROC curve (AUC). The ACC and the AUC of our technique on test dataset are 0.772 and 0.855, respectively. Experimental results are presented to illustrate the validity and robustness of the proposed approach to aircraft wake vortex. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《激光技术》浏览原始摘要信息 |
|
点击此处可从《激光技术》下载免费的PDF全文 |
