大工至善|大学至真分享 http://blog.sciencenet.cn/u/lcj2212916

博文

[转载]【无人机】【2013.06】无人机在帮助保护地形崎岖且恐怖分子活跃边界方面的效力

已有 286 次阅读 2021-1-14 21:33 |系统分类:科研笔记|文章来源:转载

图片


本文为美国海军研究生院(作者:Begum Y. Ozcan)的硕士论文,共122页。

 

边境安全对大多数国家都很重要。土耳其自20世纪80年代以来一直与恐怖组织发生冲突,至今已有4万多人丧生,其中包括土耳其士兵和平民。土耳其与伊拉克边境的空旷和开放,加上该地区崎岖的地形,为恐怖组织运送物资和人员创造了通道。无人机的技术能力可以用来提高边境保护的覆盖范围。然而,它们的效力在很大程度上取决于该区域的特点。

 

在这项研究中,87公里的土-伊边界由MapAware Non Uniform Automata(MANA)建模,以检验无人机对从伊拉克北部进入土耳其的恐怖分子侦查和分类的潜在影响。利用描述性统计、逐步线性回归、套索回归、回归树和随机森林对103200次模拟恐怖袭击的结果进行了分析。无人机的使用被认为在该地区有效地侦查和分类恐怖分子。分析技术表明,最重要的因素是无人机的探测和分类性能,以及恐怖分子的反侦查能力。因此,土耳其(以及试图保护类似地形的国家)应该购买(或建造)并使用带有复杂传感器的难以被探测的无人机。

 

Border security is of great importance tomost countries. Turkey has been in conflict with terrorist groups since the1980s. Up to now, more than 40,000 people have been killed, including Turkishsoldiers and civilians. The porosity and openness of Turkey’s Iraq border, combinedwith the rugged topography of the region, creates a passage for terroristgroups to move materiel and personnel. Technical capabilities of UnmannedAerial Vehicles UAVs can be used to improve coverage along borders. However,their effectiveness is highly dependent on the characteristics of the region.In this study, 87 km of the Turkey-Iraq border is modeled in Map Aware NonUniform Automata (MANA) to examine the potential impact of UAVs on detectingand classifying terrorists seeking passage from Northern Iraq into Turkey. Theresults from the 103,200 simulated terrorist incursions are analyzed usingdescriptive statistics, stepwise linear regression, lasso regression,regression trees, and random forests. The use of UAVs is found to be efficientin the detection and classification of terrorists in this region. The analysistechniques reveal that the most significant factors are the UAV’s detection andclassification performance, as well as the terrorists’ counter detectioncapabilities. Thus, Turkey (and countries trying to secure similar terrain)should purchase (or build) and employ hard-to-detect UAVs with sophisticatedsensors.

 

1.  引言

2. MANA

3. 场景和模型开发

4. 实验设计与模型运行

5. 数据分析

6. 结论

附录构成设计点的因变量

附录所有因素的散点图矩阵

附录C T检验的详细比较报告

附录相互作用模型的统计结果

附录树结构


更多精彩文章请关注公众号:205328s611i1aqxbbgxv19.jpg




http://blog.sciencenet.cn/blog-69686-1267231.html

上一篇:[转载]【电信学】【2018.01】在芬兰私人和公共停车场实施5G V2X服务
下一篇:[转载]【计算机科学】【2016.07】基于点云数据和CAD模型的夹具自动设置

0

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2021-2-26 10:06

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部