Geomagnetic Survey Interpolation with the Machine Learning Approach Igor Aleshin12 Kirill Kholodkov1 Ivan Malygin1 Roman

2025-04-29 0 0 6.31MB 8 页 10玖币

侵权投诉

Geomagnetic Survey Interpolation with the Machine

Learning Approach

Igor Aleshin1,2, Kirill Kholodkov1, Ivan Malygin1, Roman

Shevchuk1,2, Roman Sidorov2

1) Schmidt Institute of Physics of the Earth of the Russian Academy of Sciences,

123242 Moscow, Russia

keir@ifz.ru

2) Geophysical Center of the Russian Academy of Sciences, 119296 Moscow, Russia

1 Abstract

This paper portrays the method of UAV magnetometry survey data interpola-

tion. The method accommodates the fact that this kind of data has a spatial

distribution of the samples along a series of straight lines (similar to maritime

tacks), which is a prominent characteristic of many kinds of UAV surveys. The

interpolation relies on the very basic Nearest Neighbours algorithm, although

augmented with a Machine Learning approach. Such an approach enables the

error of less than 5 percent by intelligently adjusting the Nearest Neighbour

algorithm parameters. The method was pilot tested on geomagnetic data with

Borok Geomagnetic Observatory UAV aeromagnetic survey data.

2 Introduction

The advent of UAVs for aerial magnetometry allowed for better area coverage

and faster surveys over the traditional foot-borne magnetometry[3]. However,

the data set obtained by geomagnetic survey using unmanned aerial vehicles

(UAVs) is characterized by a high degree of spatial heterogeneity and anisotropy.

It is due to the nature of the way the measurements are performed. We used a

multirotor UAV with a quantum rubidium magnetometer suspended beneath.

The ground speed of the UAV was approximately 3-5 m/s (6-10 kt) and the

magnetic ﬁeld was sampled at 10 Hz. The suspended magnetometer’s posi-

tion was recorded with a GNSS receiver at the same rate. Therefore, the ﬁeld

sampling points were recorded at spans of about ten-to-twenty centimeters (4-8

inches) from each other. Usually, the magnetometer survey consists of a series

of straight lines. The distance between these lines is based on the characteristic

spatial size of the ﬁeld anomalies of interest and implemented in a ﬂight plan for

the UAV. In this work, this distance was approximately 50 meters (160 ft.) (see

Fig. 2) which is two times more than the distance between neighboring points

along the line. Such spatial distribution of data aggravates the direct application

of general data processing methods. The purpose of this research is to form a

methodology for organizing aerial geomagnetic surveys, taking into account the

widespread use of multirotor UAVs, and develop the procedures for processing

the data obtained through such surveys. Be it noted that the equivalent situa-

tion occurs during the cross-borehole electromagnetic imaging. In our paper [1]

arXiv:2210.03379v1 [physics.geo-ph] 7 Oct 2022

it was shown that the problem can be solved with a scale transformation of one

of the axis. Similarly, we used the method for processing geomagnetic survey

results in this paper. As a test range, we used UAV geomagnetic measurements

in the vicinity of the Borok Geomagnetic Observatory. The ﬁrst section provides

a brief description of the measurement site, the equipment used, and the survey

arrangement.

3 Survey

Figure 1: Magnetic anomaly map superposed over the Microsoft Bing satellite

imagery. The Geomagnetic Observatory “Borok” is marked with the label BOX

(which is the IAGA code of the site). The car symbol shows the position of

the motor vehicle with UAV control equipment. Cross depicts the place where

take-oﬀ, landing, and tethering procedures were performed.

The measurements were carried out above the land directly adjacent to the

Borok Geomagnetic Observatory [4] (see ﬁg. 1), which participates in the in-

ternational INTERMAGNET [5], [6] network. The location of the observatory

is characterized by the absence of signiﬁcant sources of electromagnetic distur-

bances. However, this does not limit the presence of magnetic anomalies caused

by nearby residential and commercial buildings. This circumstance makes the

place perfect for practicing the technique of aeromagnetic survey with UAVs.

In the future, the obtained results can be partially compared with ground-based

measurements, as well as used for a complete assessment of the magnetic situ-

ation around the observatory. Due to the small area of possible anomalies, it

is convenient to change geodetic latitude ϕand longitude λto local cartesian

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

GeomagneticSurveyInterpolationwiththeMachineLearningApproachIgorAleshin1;2,KirillKholodkov1,IvanMalygin1,RomanShevchuk1;2,RomanSidorov21)SchmidtInstituteofPhysicsoftheEarthoftheRussianAcademyofSciences,123242Moscow,Russiakeir@ifz.ru2)GeophysicalCenteroftheRussianAcademyofSciences,119296Moscow,Russia...

展开>> 收起<<

Geomagnetic Survey Interpolation with the Machine Learning Approach Igor Aleshin12 Kirill Kholodkov1 Ivan Malygin1 Roman.pdf

共8页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Geomagnetic Survey Interpolation with the Machine Learning Approach Igor Aleshin12 Kirill Kholodkov1 Ivan Malygin1 Roman

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: