KR102172543B1 - Apparatus estimating sag for power transmission line - Google Patents

Abstract

An apparatus for estimating a transmission line path is disclosed. The apparatus for estimating the direction of a transmission line of the present invention includes a three-dimensional point cloud data conversion unit for converting Lada point group data measured in relative coordinates with respect to a position of an unmanned aerial vehicle into three-dimensional point group data in an absolute coordinate axis; A down-sampling unit generating a 3D point cloud map by down-sampling the 3D point cloud data converted by the 3D point cloud data conversion unit; And an ear degree estimating unit configured to detect a power transmission line using the three-dimensional point cloud data down-sampled by the down sampling unit and estimate an ear degree of the detected transmission line.

Description

Transmission line path estimation device {APPARATUS ESTIMATING SAG FOR POWER TRANSMISSION LINE}

The present invention relates to a transmission line direction estimation apparatus, and more particularly, by detecting 3D point cloud data using an unmanned aerial vehicle (UAV) equipped with a LiDAR sensor, and detecting the 3D point group The present invention relates to a transmission line direction estimation apparatus that automatically estimates the direction of a transmission line by mapping data in real time using a signal processing technique.

It is very important to perform continuous health evaluation and maintenance work of transmission lines in order to stably transmit power produced by power plants. Transmission lines are deteriorated because they are continuously exposed to various surrounding environments such as stochastic wind loads and temperature changes, and fatigue damage is likely to occur in weak areas such as line connections. Due to this problem, various studies have been continuously conducted for periodic inspection and soundness evaluation of transmission lines.

As deterioration progresses, the ear canal sag due to a decrease in the stiffness of the transmission line progresses, so the ear canal is an important measure of the integrity and deterioration of the transmission line.

Currently, as a method of measuring the ear canal of a transmission line, a direct method, which is measured by seeing a tangent to an electric wire, and an indirect method, which is calculated from the measured value using physical properties of the electric wire, have been developed. The direct method includes the isotonic method, the head length method, the horizontal ear canal method, and the angle method, and the indirect method includes the tensiometer method, the short vibration period measurement method, and the mechanical shock wave method, but the indirect method has relatively low accuracy compared to the direct method. Therefore, in the field, direct method is mainly used for reliable ear canal measurement. However, since the direct method mainly measures the ear canal by boarding a transmission line, a professional manpower capable of boarding a transmission line is required, and there is a disadvantage that such experts are exposed to danger. In order to compensate for the shortcomings of the direct method, a measuring instrument capable of measuring the ear canal from the ground was developed, but since it measures the ear canal of the transmission line from the ground, there is a limit to measuring the ear canal in mountainous areas where access to the transmission line is difficult.

The background technology of the present invention is disclosed in Korean Patent Laid-Open Publication No. 10-2018-0032804 (2018.04.02),'An apparatus and method for measuring ear canal of a transmission line'.

The present invention was invented to improve the above-described problem, and an object according to an aspect of the present invention is to detect 3D point cloud data using an unmanned aerial vehicle (UAV) equipped with a LiDAR sensor. In addition, an object of the present invention is to provide an apparatus for estimating an ear degree of a transmission line by mapping the detected 3D point cloud data in real time by a signal processing technique.

According to an aspect of the present invention, there is provided an apparatus for estimating the direction of a transmission line including: a 3D point cloud data conversion unit for converting Lada point group data measured in relative coordinates with respect to an unmanned aerial vehicle position into 3D point group data in an absolute coordinate axis; A down-sampling unit for down-sampling the 3D point cloud data converted by the 3D point cloud data conversion unit to generate a 3D point cloud map; And an ear degree estimating unit for detecting a power transmission line using the three-dimensional point cloud data down-sampled by the down sampling unit and estimating an ear degree of the detected transmission line.

The 3D point cloud data conversion unit of the present invention is characterized in that converting the LIDAR point cloud data into 3D point cloud data by applying absolute position and attitude information of the unmanned aerial vehicle to the LIDAR point cloud data.

The 3D point cloud data conversion unit of the present invention adds the product of the rotation matrix representing the attitude of the unmanned aerial vehicle and the unmanned aerial vehicle coordinate axis to the absolute position coordinates of the absolute coordinate axis of the unmanned aerial vehicle measured by the GPS sensor to convert the Lada point cloud data into 3D point cloud data. It is characterized by converting.

The down-sampling unit of the present invention generates a 3D point cloud map by down-sampling the 3D point cloud data in the absolute coordinate axis converted by the 3D point cloud data conversion unit using a probabilistic down-sampling technique.

The down-sampling unit of the present invention converts the three-dimensional space of the three-dimensional point cloud data converted by the three-dimensional point cloud data conversion unit into pixels at set intervals, and sensor data is occupied by a probabilistic expression in each of the generated pixels. A 3D point cloud map is generated by updating the probability and the probability of the sensor data coming in from 1 second to t seconds according to whether sensor data exists in the pixel at t seconds.

The down sampling unit of the present invention is characterized in that the 3D point cloud data converted by the 3D point cloud data conversion unit is filtered and down-sampled using a distance filter.

The down-sampling unit of the present invention is characterized in that filtering the 3D point cloud data by extracting data in which the relative distance between the lidar data and the unmanned aerial vehicle is less than a preset threshold value from the 3D point cloud data.

The threshold value of the present invention is characterized in that it is greater than the distance from the unmanned aerial vehicle to the most distant transmission line or smaller than the portion where the transmission tower and the ground contact each other.

The down-sampling unit of the present invention pixelates a three-dimensional space in data in which the relative distance between the lidar data and the unmanned aerial vehicle in the three-dimensional point cloud data is less than a preset threshold at set intervals, and uses a probabilistic expression for each of the generated pixels. The 3D point cloud map is generated by updating a probability that sensor data is occupied and a probability of sensor data coming in from 1 second to t seconds according to whether sensor data exists in a pixel at t second.

The ear canal estimating unit of the present invention comprises: an ambient environment removing unit for removing ambient environment data from the 3D point cloud data down-sampled by the down sampling unit; A transmission line recognition unit for recognizing a location of a transmission line by using 3D point cloud data on the transmission line detected by the surrounding environment removal unit; A transmission line point group data detector configured to detect transmission line point group data by removing data other than the transmission line recognized by the transmission line recognition unit; An individual transmission line detection unit for detecting a degree of deflection of an individual transmission line or catenary line from the transmission line point group data detected by the transmission line point group data detection unit; And an individual transmission line path detection unit detected by the individual transmission line detection unit.

The surrounding environment removal unit of the present invention classifies point group data within a set distance from a transmission line and a transmission tower from the down-sampled three-dimensional point cloud data by the down sampling unit into clusters, and collects the surrounding environment data from the transmission line and the transmission tower. It is characterized by separating.

The transmission line recognition unit of the present invention is characterized in that it recognizes the location of the transmission line by orthogonally projecting the three-dimensional point group data of the transmission line detected by the surrounding environment removal unit to the ground surface and the horizontal plane.

The transmission line recognition unit of the present invention makes a straight line by selecting any two points from point group data on the ground surface and the horizontal horizontal plane, and repeats the process of calculating the number of point group data included in a distance perpendicular to the created straight line. It is characterized in that a straight line having the largest value of the number of point cloud data included in a distance perpendicular to the straight line is estimated as a transmission line.

domain으로 변환하고, 변환된

domain에서 생기는 선들이 가장 많이 겹치는 점을 선택하며, 선택된 점을 cartesian domain으로 변환하여 직선을 검출한 후, 검출된 직선과 수직한 거리에 포함된 점군 데이터의 개수의 값이 가장 큰 직선을 송전선로로 추정하는 것을 특징으로 한다.The transmission line recognition unit of the present invention receives the mode data of the horizontal plane

converted to domain, converted

The point where the lines generated in the domain overlap the most is selected, the selected point is converted into a cartesian domain to detect a straight line, and then the straight line with the largest value of the number of point group data included in the distance perpendicular to the detected straight line is used as a transmission line. It is characterized in that it is estimated to be.

축과 평행한 철탑을 검출하고 두 개의 송전철탑(경간) 양측 측정 데이터를 제거하여 송전선로 점군 데이터를 추출하는 것을 특징으로 한다.The transmission line point group data detection unit of the present invention uses the three-dimensional point group data detected by the transmission line recognition unit in the transmission line-elevation plane.

It is characterized in that it detects a pylon parallel to an axis and removes measurement data on both sides of two transmission pylons (spans) to extract point cloud data of a transmission line.

The individual transmission line detection unit of the present invention is characterized in that using the transmission line point group data detected by the transmission line point group data detection unit to detect individual transmission lines in the transmission line-elevation plane, and to estimate the degree of deflection of the catenary line.

축으로의 거리에 포함된 점들의 개수를 카운트하는 과정을 반복하여

축으로의 거리에 포함된 점들의 값이 가장 큰 곡선을 송전선이 포함된 곡선으로 추정하는 것을 특징으로 한다.The individual transmission line detection unit of the present invention selects an arbitrary three points from the transmission line point group data detected by the transmission line point group data detection unit to create a quadratic polynomial curve, and from the generated curve

Repeat the process of counting the number of points included in the distance to the axis

It is characterized by estimating a curve with the largest value of points included in the distance to the axis as a curve including a transmission line.

축과 평행한 거리를 측정하여 이도를 검출하는 것을 특징으로 한다.The island detection unit of the present invention calculates the number of subsets of the set in which the secondary polynomial is stored from the individual transmission line data detected by the transmission line detection unit, and draws a plurality of straight lines using both end points of the transmission line indicated by the curve. After making, from the center of the straight line

It is characterized in that the ear canal is detected by measuring a distance parallel to the axis.

축과 평행한 거리를 측정하여 이도를 검출하는 것을 특징으로 한다. The island detection unit of the present invention calculates the number of subsets of the catenary equation set from the individual transmission line data detected by the transmission line detection unit, and creates a plurality of straight lines using both end points of the transmission line indicated by the curve, Of the transmission line from the center of the straight line

It is characterized in that the ear canal is detected by measuring a distance parallel to the axis.

In an apparatus for estimating the direction of a transmission line according to an aspect of the present invention, 3D point cloud data is detected using an unmanned aerial vehicle equipped with a lidar sensor, and the detected 3D point cloud data is mapped in real time with a signal processing technique to automatically Estimate the ear canal.

The transmission line path estimation apparatus according to another aspect of the present invention prevents a phenomenon that the transmission line is deteriorated so that the transmission line cannot withstand excessive tension at the transmission line support in an extreme environment and is disconnected.

The transmission line roadway estimation apparatus according to another aspect of the present invention prevents an administrator who measures the roadway of a transmission line from being exposed to danger because it measures the roadway of a transmission line through an unmanned aerial vehicle at a distance.

The transmission line road estimating apparatus according to another aspect of the present invention can calculate the ear degree of a transmission line even in an extreme environment, and diagnose the safety of the transmission line or measure the growth rate of the surrounding trees and the change of the ear canal of the transmission line through periodic measurement. Because it is estimated, it can be used as a method of preventive diagnosis of transmission lines.

1 is a block diagram of an apparatus for estimating a transmission line path according to an embodiment of the present invention.
2 is a view showing an unmanned aerial vehicle-LIDAR platform and coordinate axes according to an embodiment of the present invention.
3 is a diagram showing three-dimensional Lada point group data converted to an absolute coordinate axis according to an embodiment of the present invention.
4 is a diagram showing a down-sampling and 3D map generation algorithm according to an embodiment of the present invention.
5 is a diagram showing an embodiment of an algorithm for down-sampling and creating a 3D point cloud map according to an embodiment of the present invention.
6 is a diagram showing a distance filter and a down-sampling algorithm according to an embodiment of the present invention.
7 is a diagram showing an embodiment of a distance filter and a down-sampling algorithm according to an embodiment of the present invention.
8 is a diagram showing an algorithm for removing an environment around a transmission line according to an embodiment of the present invention.
9 is a diagram showing an embodiment of an algorithm for removing an environment around a transmission line according to an embodiment of the present invention.
10 is a diagram showing point group data of a transmission line on a ground surface and a horizontal plane according to an embodiment of the present invention.
11 is a diagram showing an example of an algorithm for recognizing a transmission line in a ground surface and a horizontal plane according to an embodiment of the present invention.
FIG. 12 is a diagram illustrating a result of application of a transmission line recognition algorithm on the ground surface and horizontal plane of FIG. 11 according to an embodiment of the present invention.
13 is a diagram showing another example of an algorithm for recognizing a transmission line in a ground surface and a horizontal plane according to an embodiment of the present invention.
14 is a diagram showing a result of application of a transmission line algorithm in the ground surface and horizontal plane of FIG. 12 according to an embodiment of the present invention.
15 is a diagram showing point group data of a transmission line and a pylon in a transmission line-elevation plane according to an embodiment of the present invention.
16 is a diagram illustrating an algorithm for detecting a pylon and removing data on both sides of a span according to an embodiment of the present invention.
FIG. 17 is a diagram showing a result of removing other cluster information except for a span selected by using the algorithm shown in FIG. 16 in a transmission line-altitude plane according to an embodiment of the present invention.
18 is a diagram illustrating an algorithm for detecting individual transmission lines in a transmission line-elevation plane according to an embodiment of the present invention.
19 is a diagram showing an example of application of individual transmission line extraction by applying an individual transmission line detection algorithm in a transmission line-elevation plane according to an embodiment of the present invention.
20 is a view showing an island view of a transmission line according to an embodiment of the present invention.
21 is a diagram showing an example of an algorithm for estimating a transmission line path according to an embodiment of the present invention.
FIG. 22 is a diagram showing a result of application of the algorithm for estimating the path of the transmission line of FIG. 21 according to an embodiment of the present invention.
23 is a diagram showing another example of an algorithm for estimating a transmission line path according to an embodiment of the present invention.
FIG. 24 is a diagram showing a result of application of the algorithm for estimating the path of the transmission line of FIG. 23 according to an embodiment of the present invention.
25 to 27 are diagrams showing transmission line detection and estimating results for two transmission lines according to an embodiment of the present invention.

Hereinafter, an apparatus for estimating a transmission line island according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In this process, the thickness of the lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of users or operators. Therefore, definitions of these terms should be made based on the contents throughout the present specification.

1 is a block diagram of an apparatus for estimating a transmission line path according to an embodiment of the present invention.

Referring to FIG. 1, the apparatus for estimating the ear canal of a transmission line according to an embodiment of the present invention includes a sensing unit 10, a 3D point cloud data conversion unit 20, a down-sampling unit 30, and an ear canal estimating unit 40. Include.

The sensing unit 10 includes a lidar sensor 11, a ground positioning system (GPS) sensor 12, and an inertia measurement unit (IMU) sensor 13.

The lidar sensor 11 is installed on the unmanned aerial vehicle and measures lidar point group data measured in relative coordinates with respect to the position of the unmanned aerial vehicle.

The GPS sensor 12 measures the absolute position of the unmanned aerial vehicle.

The IMU sensor 13 measures the posture of the unmanned aerial vehicle.

The 3D point cloud data conversion unit 20 converts the LIDAR point cloud data measured in relative coordinates with respect to the position of the unmanned aerial vehicle, and the absolute position of the unmanned aerial vehicle measured by the GPS sensor 12 and the IMU sensor 13 respectively mounted on the unmanned aerial vehicle. The posture information is mixed and converted into 3D point cloud data in the absolute coordinate axis.

FIG. 2 is a diagram showing an unmanned aerial vehicle-LIDAR platform and coordinate axes according to an embodiment of the present invention, and FIG. 3 is a diagram illustrating three-dimensional LaDAR point cloud data converted to an absolute coordinate axis according to an embodiment of the present invention.

)에 대한 상대변위 좌표

, GPS 센서(12)에서 측정되는 무인항공기 절대좌표축 (

)에 대한 절대위치 좌표를

로 각각 정의하면 라이다 센서(11)에서 측정되는 상대변위 점군데이터는 아래의 수학식 1을 이용하여 절대좌표축 (

)에 대한 절대위치 좌표로 변환 가능하다.2, the relative coordinate axis of the lidar measured by the lidar sensor 11 (

) Relative displacement coordinates

, The absolute coordinate axis of the unmanned aerial vehicle measured by the GPS sensor 12 (

) Of the absolute position

If each defined as, the relative displacement point group data measured by the lidar sensor 11 is the absolute coordinate axis (

) Can be converted to absolute position coordinates.

는 라이다 센서(11)에서 측정되는 데이터의 절대좌표축 (

)에 대한 좌표이다.

은 무인항공기 자세를 나타내는 회전행렬로 IMU 센서(13)의 사원수 데이터

,

,

,

를 이용하여 하기와 같이 계산한다. In Equation 1

Is the absolute coordinate axis of the data measured by the lidar sensor 11 (

).

Is a rotation matrix indicating the attitude of the unmanned aerial vehicle, and the quaternion data of the IMU sensor 13

,

,

,

It is calculated as follows using.

은 무인항공기 좌표축(

)에 대한 라이다의 위치정보이다. 라이다 센서(11)는 무인항공기 y축을 기준으로

°회전되어 있으므로, 아래의 수학식 3을 이용하여 라이다 좌표축에 대하여 계산되는 상대좌표

을 무인항공기 좌표축에 대한 상대좌표로 변환 가능하다.

Is the UAV coordinate axis (

) Is the location information of Lida. The lidar sensor 11 is based on the y-axis of the unmanned aerial vehicle.

Since ° is rotated, the relative coordinates calculated with respect to the LiDAR coordinate axis using Equation 3 below

Can be converted to relative coordinates with respect to the UAV coordinate axis.

은 3차원 회전 행렬로 아래의 수학식 4를 이용하여 계산한다.here

Is a three-dimensional rotation matrix and is calculated using Equation 4 below.

축을 기준으로 90°회전되어 있으므로 3차원 회전 행렬

은

이다.In one embodiment of the present invention, the lidar sensor 11 is an unmanned aerial vehicle

It is rotated by 90° about the axis, so a three-dimensional rotation matrix

silver

to be.

)에 대한 위치좌표로 변환 가능하다.Therefore, considering the absolute position of the unmanned aerial vehicle, the posture, and the installation position of the lidar for the unmanned aerial vehicle, the LIDA point group data is calculated using the following equation (5).

) Can be converted into position coordinates.

)에 대해 측정된 3차원 점군 데이터를 절대 좌표축 (

)으로 변환한 데이터는 도 3 과 같다.The relative coordinate axis measured in the lidar using Equation 5 (

), the measured 3D point cloud data is converted to the absolute coordinate axis (

The data converted to) are shown in FIG. 3.

)에서의 3차원 라이다 점군 데이터를 확률론적 다운-샘플링(down-sampling) 기법을 적용하여 3차원 점군 맵을 제작하고, 송전선로가 포함된 설정 거리 이하 데이터를 실시간으로 추출하여 다운-샘플링한다.The down-sampling unit 30 is an absolute coordinate axis calculated by the 3D point cloud data conversion unit 20 (

), a 3D point cloud map is produced by applying a probabilistic down-sampling technique to the 3D LIDAR point cloud data in ), and down-sampled by extracting data below the set distance including the transmission line in real time. .

FIG. 4 is a diagram showing a down-sampling and 3D map generation algorithm according to an embodiment of the present invention, and FIG. 5 is a diagram showing an embodiment of a down-sampling and 3D point cloud map generation algorithm according to an embodiment of the present invention. 6 is a diagram showing a distance filter and a down-sampling algorithm according to an embodiment of the present invention, and FIG. 7 is a diagram showing an embodiment of a distance filter and a down-sampling algorithm according to an embodiment of the present invention. .

)에서의 3차원 라이다 점군 데이터를 확률론적 다운-샘플링 기법을 적용하여 3차원 점군 맵을 제작한다. 라이다 센서(11)에서 측정되는 데이터는 초당 수 MB로 방대하기 때문에 연산장치 및 저장장치에 상당한 부하를 발생시킨다. 따라서 확률론적 다운-샘플링 기법을 이용하여 정확도는 유지하며 연산 데이터를 감소시키면 본 실시예의 송전선로 이도 추정 장치는 무인항공기에 장착 가능한 경량 연산장치가 될 수 있다. First, the down-sampling unit 30 is the absolute coordinate axis calculated by the 3D point cloud data conversion unit 20 (

), a 3D point cloud map is produced by applying a probabilistic down-sampling technique to the 3D LIDAR point cloud data. Since the data measured by the lidar sensor 11 is enormous at several MB per second, a significant load is generated on the computing device and the storage device. Therefore, by using a stochastic down-sampling technique to maintain accuracy and to reduce computational data, the transmission line roadway estimation apparatus of this embodiment can be a lightweight computational device that can be mounted on an unmanned aerial vehicle.

)으로 화소화(Voxelization)하고 각 화소를 아래의 수학식 6과 같은 확률론적 표현으로 차지(Occupy)되어 있는지 혹은 비어(Free)있는지를 나타낸다. The process of creating a 3D point cloud map involves setting a 3D space at a set interval (

), and whether each pixel is occupied or free by a probabilistic expression as shown in Equation 6 below.

은 하나의 화소가 차지되어 있을 확률이고

은 하나의 화소에 1초부터 t초까지 센서 데이터가 측정될 확률이다. 수학식 6에 log를 이용하여 정리하면 수학식 7과 같다.here,

Is the probability that one pixel is occupied

Is the probability that sensor data is measured from 1 second to t seconds in one pixel. If it is summarized using log in Equation 6, it is shown in Equation 7.

은 하나의 화소에서 시간 t초 때 라이다 데이터가 있으면 +

, 무인항공기와 라이다 데이터 사이에 있는 라이다 센서(11)의 레이저가 지나간 화소들은 -

으로

를 업데이트하고 3차원 점군 데이터 맵을 생성할 수 있다. Equation 7 can be simplified using log on the probability that one pixel is occupied according to sensor data from 1 second to t seconds,

Is + if there is lidar data at time t seconds in one pixel

, Pixels passed by the laser of the lidar sensor 11 between the UAV and the lidar data are-

to

And create a 3D point cloud data map.

The 3D LiDAR point cloud data of FIG. 3 presented as an application example has a capacity of about 9 GB, but when down-sampling is performed, the capacity is reduced to 341 KB (99.99% reduction), which significantly reduces the load on the storage and processing unit. A high-performance computing device is required to perform many calculations, and power consumption is also increased. Therefore, if the input data is reduced by applying such a down-sampling technique, it is possible to use an ultra-lightweight and low-performance computing device. In the case of unmanned aerial vehicles, considering that the weight and power consumption of the installed equipment have a great influence on the flight time, data reduction through the application of this probabilistic down-sampling technique is essential.

In addition, since the probabilistic down-sampling technique simultaneously removes noise information measured through probabilistic updating of one pixel, it is possible to improve accuracy when detecting a transmission line and estimating an ear canal.

는 1.0 m,

는 0.85,

는 -0.4를 사용하였다.4 shows the down-sampling algorithm proposed in an embodiment of the present invention, and the application result is shown in FIG. As shown in FIG. 5, when the down-sampling algorithm is used, noise is removed from the 3D point cloud data of FIG. 3, and pixelated surrounding environment data remains. As described above, although the amount of data used is very small, it can be observed that there is no significant difference from FIG. 3 using all data. In this example

Is 1.0 m,

Is 0.85,

Used -0.4.

)에서의 3차원 라이다 점군 데이터를 거리 필터 사용하여 필터링하고 다운 샘플링한다.In addition, the down-sampling unit 30 is the absolute coordinate axis calculated by the 3D point cloud data conversion unit 20 (

), the 3D LiDAR point cloud data is filtered and downsampled using a distance filter.

)으로 화소화(Voxelization)하고 송전선과 주변 환경을 분할하기 위해, 도 6 에 도시된 바와 같이 주변 환경에서 측정된 라이다 데이터와 무인 비행체의 상대거리

(

)에 대해

를 충족시키는 데이터를 우선 추출한다. In this case, the down-sampling unit 30 divides the 3D space at a constant interval (

), and in order to divide the transmission line and the surrounding environment, the relative distance between the LIDAR data measured in the surrounding environment and the unmanned aerial vehicle as shown in FIG.

(

)About

First, data that satisfies is extracted.

,

, 및

는 무인항공기 상대좌표축(

) 또는 라이다 상대좌표축(

)을 기준으로 측정한 무인항공기에 대한 주변 환경의 상대거리(Relative Displacement, RD)이다. 후술한 이도 추정부(40)에서 수행되는 송전선로 주변 환경정보 제거 알고리즘이 점과 점사이의 거리에 따라 3차원 점군 데이터를 분류하여 군집을 생성하기 때문에 송전선로 주변 환경을 완벽하게 제거하기 위해, 다운 샘플링부(30)가 데이터를 추출할 때 임계값

를 적절하게 선정하는 점은 매우 중요하다. here

,

, And

Is the relative coordinate axis of the unmanned aerial vehicle (

) Or LiDAR relative coordinate axis (

) Is the relative distance (Relative Displacement, RD) of the surrounding environment to the UAV. In order to completely remove the environment surrounding the transmission line, since the algorithm for removing environment information around the transmission line, which will be described later, classifies the 3D point cloud data according to the distance between the point and the point to create a cluster, Threshold value when the down-sampling unit 30 extracts data

It is very important to select appropriately.

는 무인항공기로부터 가장 원거리 송전선까지 거리보다는 큰 값이 선정되어야 송전선이 모두 추출된다. 반면, 송전철탑과 지면이 맞닿는 부분보다 작은 값이 선정되어야 송전선로와 연결된 지면이 제거될 수 있으므로, 이도 추정부(40)의 송전선로와 주변 환경 분류를 통한 군집화 과정에서 송전철탑과 연결된 지면이 완벽하게 분리될 수 있다. Threshold

All transmission lines are extracted only when a value larger than the distance from the unmanned aerial vehicle to the most distant transmission line is selected. On the other hand, since the ground connected to the transmission line can be removed only when a value smaller than the part where the transmission tower and the ground contact is selected, the ground connected to the transmission tower in the clustering process through classification of the transmission line and surrounding environment of the island estimation unit 40 It can be completely separated.

는 점검하는 송전선로의 종류에 의존적이다. 본 실시예에서는 거리 필터 알고리즘을 적용하여 추출된 데이터에 대하여 확률론적 다운-샘플링 기법으로 각 화소가 차지(Occupy)되어 있는지 혹은 비어(free)있는지를 수학식 7과 같은 확률론적 표현을 통해 업데이트하여 연산처리 및 저장장치들의 부하를 감소시켰다.Threshold value when considering that the height of the transmission line is different depending on the transmission voltage

Depends on the type of transmission line being checked. In this embodiment, whether each pixel is occupied or free by a probabilistic down-sampling technique for data extracted by applying a distance filter algorithm is updated through a probabilistic expression such as Equation 7 Reduced load on computational processing and storage devices.

값은 0.5,

값은 50,

값은 0.85,

값은 -0.4를 사용하였다.6 shows a distance filter and a down-sampling algorithm proposed in this embodiment, and the application result is shown in FIG. 7. When a distance filter and a down-sampling algorithm as shown in FIG. 7 are used, most of the surrounding environment is removed from the 3D point cloud data of FIG. 3, and only a transmission line and some surrounding environment data remain. In this example

The value is 0.5,

The value is 50,

The value is 0.85,

The value was -0.4.

The ear canal estimating unit 40 estimates an ear canal of an individual transmission line by using the 3D point cloud data obtained by the down-sampling unit 30. The ear canal estimating unit 40 includes an ambient environment removal unit 41, a transmission line recognition unit 42, a point group data removal unit 43, a transmission line detection unit 44, and an ear canal detection unit 45.

8 is a diagram showing an algorithm for removing an environment around a transmission line according to an embodiment of the present invention, and FIG. 9 is a diagram showing an embodiment of an algorithm for removing an environment around a transmission line according to an embodiment of the present invention.

First, the surrounding environment removal unit 41 removes the surrounding environment except for the transmission line and the transmission tower. As described above, in the data to which the distance filter and down-sampling are applied by the down-sampling unit 30, the transmission line and the transmission tower are separated as shown in FIG. 7, but the surrounding environment data existing within a certain distance exists. Therefore, by classifying 3D point cloud data within a certain distance into clusters, it is possible to separate and cluster the transmission line and the surrounding environment.

, 분류 검사를 위한 집합

, 분류된 데이터를 저장하는 집합

, 그리고

로 선정한다,

에서 임의의 한 점을 골라

으로 이도한다.

에 있는 임의의 한 점을 중심으로 반지름

인 구를 선정 후,

에 속한 데이터들 중 구 안에 포함된 데이터들을

으로 이도한다. 다시

에 속한 임의의 한 점을 골라 상기의 방법을 반복하고, 더 이상 만족하는 점군 데이터가 없을 시

의 데이터를

의 부분집합

으로 이도하고

에 1을 더해준다. 이후 다시

에서 임의의 한 점을 골라

으로 이도하는 작업을 반복한다.

에 점군 데이터가 남아있지 않을 경우, 밀도(

)가 가장 낮은

(

)를 송전선로로 선택한다. The surrounding environment removal unit 41 is the down-sampled 3D point cloud data

, Assembly for classification inspection

, A set to store classified data

, And

Select as,

Pick a random point from

I do it.

Radius around any point in

After selecting the population,

The data contained in the sphere among the data belonging to

I do it. again

Repeat the above method by selecting an arbitrary point belonging to and when there is no more satisfactory point cloud data.

The data of

Subset of

I do it with

Add 1 to Afterwards again

Pick a random point from

Repeat the task to lead.

If there are no point cloud data remaining in the density (

) Is the lowest

(

) As the transmission line.

This algorithm considers the morphological characteristics of the transmission line. Since the transmission line is thin and long when viewed from the entire space, it has a relatively small density compared to other 3D point cloud data that display environmental information. On the other hand, the point group data of the surrounding environment other than the transmission line are clustered at a high density.

값을 5 m로 선정하였다.Fig. 8 shows the above-described topography removal algorithm around the transmission line, and the application result is shown in Fig. 9. Through FIG. 9, it is possible to observe that the topographic information around the transmission line and the pylon remaining in FIG. 7 is removed, and only information on the transmission line and the pylon connected to the line remains. In the practical application example of FIG. 8

The value was chosen as 5 m.

평면)으로 정사영하여 송전선로가 존재하는 위치를 인지한다. The transmission line recognition unit 42 stores the three-dimensional point group data of the transmission line from which the surrounding environment has been removed by the surrounding environment removal unit 41, as shown in FIG.

Plane) to recognize the location of the transmission line.

FIG. 10 is a view showing point group data of transmission lines on the ground and horizontal planes according to an embodiment of the present invention, and FIG. 11 is a diagram showing an example of an algorithm for recognizing transmission lines on the ground and horizontal planes according to an embodiment of the present invention. 12 is a diagram showing the result of application of the transmission line recognition algorithm on the ground surface and horizontal plane of FIG. 11 according to an embodiment of the present invention, and FIG. 13 is a diagram showing the application result of the transmission line recognition algorithm on the ground surface and the horizontal plane according to an embodiment of the present invention. A diagram showing another example of a transmission line recognition algorithm, and FIG. 14 is a diagram showing a result of application of the transmission line algorithm on the ground surface and horizontal plane of FIG. 12 according to an embodiment of the present invention.

평면)으로 정사영하게 되면 도 10 과 같이 직선으로 도시된다. 지표면과 수평 평면(도 2 의

평면)에서 송전선로가 꺾이는 위치는 단지 송전철탑에 의해서만 존재한다. 따라서 3차원 평면 데이터를 지표면과 수평 평면(도 2 의

평면)으로 투영하면 송전선로 인지가 용이하다. 본 실시예에서는 지표면과 수평 평면(도 2 의

평면)에서 송전선로 위치를 인지하는 방법을 도 11 및 도 13을 참조하여 두 가지로 제시한다.Since the transmission line sags to the surface by gravity, the surface and the horizontal plane (Fig. 2)

When projecting orthogonal to the plane), it is shown as a straight line as shown in FIG. 10. Ground and horizontal planes (Fig. 2

The position where the transmission line is bent in the plane) exists only by the transmission tower. Therefore, the 3D plane data is converted to the ground surface and the horizontal plane (Fig.

It is easy to recognize the transmission line by projecting it on a plane). In this embodiment, the ground surface and the horizontal plane (Fig. 2

In the plane), two methods of recognizing the location of the transmission line are presented with reference to FIGS. 11 and 13.

로 선정한다. 이후, 지표면과 수평한 면(

평면)에서의 점군 데이터에서 임의의 두 점을 선택하여 직선을 만들고, 만들어진 직선과 수직한 거리

에 포함된 점군 데이터의 개수

을 계산한다. 이 작업을

번 반복하여 거리 내에 존재하는 점군 데이터의 개수

의 값이 가장 큰 직선을 송전선로로 추정한다. 측정된 3차원 점군 데이터는 여러 개의 송전선로를 포함할 수 있기 때문에

의 개수가 임계값

이상일 경우 검출된 직선과 수직한 거리

에 포함된 점들을

의 부분집합

으로 옮긴 후

에 1을 더해주고 알고리즘을 반복하며,

의 개수가 임계값

이하인 경우

을 저장한 후 알고리즘을 종료한다. Referring to FIG. 11, the first method of recognizing the location of a transmission line is first

Select as. After that, the surface and the horizontal plane (

Plane), a straight line is created by selecting any two points from the point group data, and the distance perpendicular to the created straight line

Number of point cloud data included in

Calculate Do this

Number of point cloud data that exist within the distance by repeating once

The straight line with the largest value of is estimated as the transmission line. Because the measured 3D point cloud data can include multiple transmission lines,

The number of thresholds

In case of abnormality, the distance perpendicular to the detected straight line

The points included in

Subset of

After moving to

Add 1 to and repeat the algorithm,

The number of thresholds

Below

After saving, the algorithm ends.

값은 3,

값은 1000, 임계값

는 50으로 설정하여 송전선로를 추출하였다. 4개의 송전선로가 순차적으로 완벽하게 추출되는 것이 도 12 를 통하여 확인 가능하다.The result values of the algorithm for extracting the transmission line by using the algorithm presented in FIG. 11 are shown in FIG. 12. The application result of FIG. 12 is a variable in the algorithm of FIG.

The value is 3,

Value is 1000, threshold

Was set to 50 to extract the transmission line. It can be confirmed through FIG. 12 that the four transmission lines are sequentially and completely extracted.

평면)에서 송전선로 위치를 인지하는 방법은 도 11 에 도시된 알고리즘을 이용해서도 추출이 가능하다. 먼저

로 선정하고, 지표면과 수평면(그림 1의

평면)의 모든 데이터를 아래의 수학식 8을 이용하여

domain으로 변환한다.Surface and horizontal plane (Fig. 1)

The method of recognizing the location of the transmission line in the plane) can also be extracted using the algorithm shown in FIG. 11. first

And the ground surface and the horizontal plane (Fig. 1)

Plane) using Equation 8 below

convert to domain.

domain에서 생기는 선들이 가장 많이 겹치는 점을 선택하고, 그 점을 다시 cartesian domain으로 변환하여 직선을 검출한다. 이후, 도 11 과 마찬가지로 직선과 수직한 거리

에 포함된 데이터의 개수(

)를 카운트한 후

의 개수가 임계값

이상일 경우 검출된 직선과 수직한 거리

에 포함된 점들을

의 부분집합

으로 옮긴 후

에 1을 더해주고 알고리즘을 반복하며,

의 개수가 임계값

이하인 경우

을 저장한 후 알고리즘을 종료한다. 도 13의 적용결과는 도 14 에 도시되었으며, 도 12 의 결과와 육안으로 판단하기 어려울 만큼 유사한 것이 관찰 가능하다. 도 14 의 적용예에서는

,

값은 1000, 1000,

값은 3, 임계값

는 50으로 선정하였다. 따라서 본 발명의 일 실시예에서 제안하는 두 가지 방법들은 모두 지표면과 수평한 면에서 송전선로를 검출하는 것이 가능하다.Following

The point where the lines generated in the domain overlap the most is selected, and the point is converted back to the cartesian domain to detect the straight line. Thereafter, as in FIG. 11, the distance perpendicular to the straight line

The number of data contained in (

) After counting

The number of thresholds

In case of abnormality, the distance perpendicular to the detected straight line

The points included in

Subset of

After moving to

Add 1 to and repeat the algorithm,

The number of thresholds

Below

After saving, the algorithm ends. The application result of FIG. 13 is shown in FIG. 14, and it is possible to observe that the result of FIG. 12 is similar enough to be difficult to judge with the naked eye. In the application example of Fig. 14

,

Values are 1000, 1000,

Value is 3, threshold

Was selected as 50. Therefore, both methods proposed in an embodiment of the present invention are capable of detecting a transmission line in a horizontal plane with a ground surface.

FIG. 15 is a view showing point group data of a power transmission line and a pylon in an elevation plane according to an embodiment of the present invention, and FIG. 16 is a view showing an algorithm for detecting and removing data on both sides of a span according to an embodiment of the present invention. 17 is a view showing a result of removing other cluster information except for a span selected using the algorithm presented in FIG. 16 in the transmission line-elevation plane according to an embodiment of the present invention.

평면)에서 관찰하면, 송전선로를 지지하는 송전철탑 데이터가 포함되는 것을 관찰 가능하다. 이도는 철탑 사이의 송전선로의 처짐으로 정의되기 때문에 송전철탑의 위치 정보가 필요하여 이도 추정시 정확도를 향상시키기 위해서는 송전철탑 데이터를 제거해야 한다. Transmission line data extracted by the transmission line recognition unit 42 is converted into a transmission line-altitude plane (Fig.

When observed from the plane), it is possible to observe that the transmission tower data that supports the transmission line is included. Since the island is defined as the deflection of the transmission line between the towers, the location information of the transmission towers is required. In order to improve the accuracy when estimating the islands, the transmission tower data must be removed.

In addition, as shown in FIG. 15, the extracted data other than the transmission line for estimating the island must be removed from both axes of the spans of the two transmission towers to be detected to reduce the error when extracting the transmission line.

평면)에서 도 16 에 도시된 알고리즘을 적용하여 선정된 송전철탑 이외의 모든 점군 데이터를 제거한다. 제안 알고리즘의 기본적인 전제는 송전철탑은 직립하기 때문에, 송전선로 점군 데이터는 고도에 대하여 수평으로 존재한다는 것이다. 도 15 에는 송전선로 인지부(42)에서 검출한 직선 중 첫 번째 루프를 통해 추출된 송전선로 데이터(도 12 (a))를 이용하여 송전선로-고도 평면(도 12 의

평면)에서

축과 평행한 철탑을 검출하고 두 개의 송전철탑(경간) 양측 측정 데이터를 제거하는 단계를 도시하였다.Therefore, the point cloud data removal unit 43 converts the transmission line and transmission tower data detected by the transmission line recognition unit 42 into a transmission line-altitude plane (Fig. 12).

In the plane), all point cloud data other than the selected transmission tower is removed by applying the algorithm shown in FIG. 16. The basic premise of the proposed algorithm is that since the transmission tower is upright, the transmission line point cloud data exists horizontally with respect to the altitude. In FIG. 15, the transmission line-elevation plane (Fig. 12) using the transmission line data extracted through the first loop among the straight lines detected by the transmission line recognition unit 42

Flat)

A step of detecting a pylon parallel to the axis and removing measurement data on both sides of the two transmission pylons (spans) is shown.

평면)에서 철탑은 직립하여 있으므로 도 16에서는

모델로 사용한다. 도 16에 도시된 본 발명의 실시예에서 제안하는 알고리즘에서는 임의로 한 점을 선택하여

축과 평행한 직선을 만들고, 생성된 직선과 수직한 거리

에 포함된 점들의 개수(

)를 계산한다. 앞선 작업을

번 반복하여

의 값이 가장 큰 직선이 철탑이 포함된 직선으로 추정하고, 이 직선과 수직한 거리

에 포함된 점들을

로 옮긴다. 일회 운행 시 한 개의 경간을 측정하는 반면, 송전선로 점군 데이터에는 2개의 철탑이 존재한다. 따라서 두 철탑 사이 정보를 제외한 모든 정보를 제거하고

을 저장 후 및 알고리즘을 종료한다. 도 17 의 적용예에서

값은 3,

값은 점군 데이터의 개수로 선정하였다.As described above, the transmission line-elevation plane (Fig.

Flat), the tower is upright, so in Fig.

Use as a model. In the algorithm proposed in the embodiment of the present invention shown in FIG. 16, a point is randomly selected and

Create a straight line parallel to the axis, and the distance perpendicular to the generated straight line

The number of points in (

) Is calculated. Work ahead

Repeatedly

The straight line with the largest value of is estimated to be the straight line containing the pylon, and the distance perpendicular to this line

The points included in

Move to. While one span is measured for one operation, there are two pylons in the point cloud data of a transmission line. Therefore, remove all information except the information between the two pylons

And then the algorithm ends. In the application example of Fig. 17

The value is 3,

The value was selected by the number of point group data.

A transmission line is composed of several transmission lines. In the case of the 154 kV transmission line measured for this example, there are 6 transmission lines, 3 on the left and right sides, and 2 overhead branch lines, 1 on the left and right, respectively, and at least 6 on the 345 kV and 765 kV transmission lines. There are up to 12 transmission lines and 2 overhead branch lines. Therefore, accurate soundness evaluation is possible only by extracting all individual transmission lines and estimating the angle of each transmission line.

The transmission line detection unit 44 detects individual transmission lines by using the point group data of the transmission line extracted from the point group data removal unit 43.

18 is a diagram showing an algorithm for detecting individual transmission lines in a transmission line-elevation plane according to an embodiment of the present invention, and FIG. 19 is a diagram showing an algorithm for detecting individual transmission lines in a transmission line-elevation plane according to an embodiment of the present invention. This is a diagram showing an example of application of individual transmission line extraction.

The deflection of the transmission line follows the general catenary equation, and the catenary equation is shown in Equation 9.

는 송전선의 처진 정도를 나타내는 계수이며

와

는 송전선 중심의 위치를 나타낸다. 수학식 9를 테일러급수를 이용하여 2차 항까지 근사하면 수학식 10과 같다. Variable in Equation 9

Is a coefficient indicating the degree of deflection of the transmission line

Wow

Indicates the location of the center of the transmission line. If Equation 9 is approximated to the quadratic term using the Taylor series, it is equal to Equation 10.

와 같은 2차 다항식으로 커브 피팅(curve fitting)을 하여 변수

,

,

를 추출하면, 수학식 10과 같이 현수선 방정식과 테일러급수로 표현된 근사식 사이의 관계는

,

,

와 같다. 따라서 현수선의 처짐 정도는 현수선 방정식 및 2차 방정식으로 표현이 가능하다는 것을 유추할 수 있다. So the transmission line

Variable by performing curve fitting with a quadratic polynomial such as

,

,

When is extracted, the relationship between the catenary equation and the approximate equation expressed in Taylor series as in Equation 10 is

,

,

Same as Therefore, it can be inferred that the degree of deflection of the catenary can be expressed by the catenary equation and the quadratic equation.

18 is an algorithm proposed in an embodiment of the present invention for detecting individual transmission lines in the transmission line shown in FIG. 17(c) extracted by using the algorithm shown in FIG. 16.

평면)에서 송전선의 점군 데이터는 앞서 기술한 바와 같이

2차 다항식 모델로 도 18 에 제시되는 알고리즘을 사용하여 추출할 수 있다. 먼저

로 선정한다. 이후, 임의로 세 점을 선택하여 2차 다항식 곡선을 만들며 생성된 곡선으로부터

축으로의 거리

에 포함된 점들의 개수(

)를 센다. 앞선 작업을

번 반복하여

의 값이 가장 큰 곡선을 송전선이 포함된 곡선으로 추정한다. 하나의 송전선 점군 데이터에는 여러 개의 송전선을 포함할 수 있으므로

가 임계값

보다 크면 곡선으로부터

축으로의 거리

에 포함된 점들을 제거하고, 현수선 방정식의 변수

,

,

는

의 부분집합

, 2차 다항식의 변수

,

,

는

의 부분집합

에 저장하며

에 1을 더한다.

가 임계값

보다 작으면 더 이상 송전선이 없다고 판명하여 각각의 현수선 방정식의 변수가 저장된

및 2차 다항식의 변수가 저장된

를 저장하고 알고리즘을 종료한다. 도 19 의 적용예에서

값은 2,

값은 1000,

값은 300으로 선정하였다.Transmission line-elevation plane (Fig. 12

Plane), the point cloud data of the transmission line is

As a quadratic polynomial model, it can be extracted using the algorithm shown in FIG. 18. first

Select as. Then, randomly select three points to create a quadratic polynomial curve, and from the generated curve

Distance to axis

The number of points in (

). Work ahead

Repeatedly

The curve with the largest value of is estimated as the curve containing the transmission line. Since one transmission line point cloud data can contain multiple transmission lines,

Fall threshold

Greater than from the curve

Distance to axis

Remove the points included in the catenary equation

,

,

Is

Subset of

, Variables of quadratic polynomial

,

,

Is

Subset of

To

Add 1 to

Fall threshold

If it is less than, it is determined that there are no more transmission lines, and the variables of each catenary equation are stored.

And the variables of the quadratic polynomial are stored

Save and exit the algorithm. In the application example of Fig. 19

Value is 2,

Value is 1000,

The value was selected as 300.

The ear canal detection unit 45 detects the ear canal of individual transmission lines extracted by the power transmission line detection unit 44.

FIG. 20 is a diagram showing a diagram of a transmission line according to an embodiment of the present invention, FIG. 21 is a diagram showing an example of an algorithm for estimating a transmission line diagram according to an embodiment of the present invention, and FIG. Fig. 21 is a view showing the application result of the transmission line direction estimation algorithm according to an embodiment, Fig. 23 is a view showing another example of the transmission line direction estimation algorithm according to an embodiment of the present invention, and Fig. 24 is the present invention A diagram showing a result of application of the algorithm for estimating the path of a transmission line according to an embodiment of FIG.

을 이용하여 이도를 산출하는 알고리즘이고, 도 21 은 현수선 방정식

을 이용하여 이도를 산출하는 알고리즘이다. 상술한 바와 같이 현수선 방정식은 테일러급수를 이용하여 2차 다항식 곡선으로 근사가 가능하므로 두 개의 방법 모두 적용이 가능하다. The angle of the transmission line is defined as shown in FIG. 20. In an embodiment of the present invention, an algorithm for estimating the ear degree of the power transmission line shown in FIGS. 21 and 23 is proposed for calculating the degree of the individual power transmission line. 21 is a quadratic polynomial curve calculated by the algorithm of FIG. 18

It is an algorithm for calculating the ear canal using, and FIG. 21 is a catenary equation

It is an algorithm that calculates the ear canal using. As described above, since the catenary equation can be approximated by a quadratic polynomial curve using the Taylor series, both methods can be applied.

의 부분집합의 개수 N을 계산하고, 곡선으로 나타낸 송전선의 양 끝점을 이용하여 N개의 직선을 만든다. 이후, 이도 검출부(45)는 직선의 중심으로부터 송전선의

축과 평행한 거리를 측정하여 N개의 이도를 저장한다. 도 22 는 이도 검출부(45)가 첫 번째 루프에서 추출한 송전선(도 19 의 (a))을 이용하여 송전선의 이도를 도 21 에 제시된 알고리즘을 이용하여 산출한 결과이다.The island detection unit 45 is a set in which the second order polynomials detected by the transmission line detection unit 44 are stored.

Calculate the number N of subsets of, and make N straight lines using both end points of the transmission line indicated by the curve. Thereafter, the ear canal detection unit 45 of the transmission line from the center of the straight line

Measure the distance parallel to the axis and store N ear canals. FIG. 22 is a result of calculating the ear degree of the transmission line by using the transmission line (FIG. 19(a)) extracted by the ear canal detection unit 45 from the first loop using the algorithm shown in FIG. 21.

를 이용하여 이도를 측정한다. 도 24 는 이도 검출부(45)가 첫 번째 루프에서 추출한 송전선(도 19 의 (a))을 이용하여 송전선의 이도를 도 23 의 알고리즘을 이용하여 산출한 결과이다.The algorithm of FIG. 23 proceeds like the algorithm of FIG. 21 described above, but the set of catenary equations obtained in FIG. 18

Measure the ear canal using. FIG. 24 is a result of calculating the ear degree of the power transmission line by using the power transmission line (FIG. 19 (a)) extracted by the ear canal detection unit 45 from the first loop using the algorithm of FIG. 23.

25 to 27 are diagrams showing transmission line detection and estimating results for two transmission lines according to an embodiment of the present invention.

)에서의 라이다 점군 데이터를 입력 데이터로 본 발명의 일 실시예를 적용하였을 때 두 개의 송전선로에 대한 송전선 검출 및 이도 추정 결과를 보여준다. 두 가지 송전선로는 도 12 의 (a), (b) 두 가지 송전선로이며 성공적으로 송전선을 검출하고 송전선의 이도를 추정함을 관찰 가능하다.25 and 26 and Tables 1 and 2 below are the absolute coordinate axes of FIG.

) Shows the results of transmission line detection and island road estimation for two transmission lines when an embodiment of the present invention is applied to the LIDAR point cloud data as input data. The two transmission lines are two transmission lines (a) and (b) of Fig. 12, and it can be observed that the transmission line is successfully detected and the angle of the transmission line is estimated.

Estimation result of power transmission line (from the top transmission line to transmission line number 1) Transmission line number Ear degree (m) One 6.2270 2 8.2327 3 7.9032 4 8.2228

Estimation result of power transmission line (from the top transmission line to transmission line number 1) Transmission line number Ear degree (m) One 4.3390 2 6.8121 3 6.1605 4 6.6797

27 and Table 3 are examples of inputting 3D point cloud data of a transmission line different from the previous embodiment.As a result of applying the algorithm according to an embodiment of the present invention, as in the previous embodiment, a transmission line was successfully detected and the degree of the transmission line It can be seen that it is estimated.

Estimation result of power transmission line (from the top transmission line to transmission line number 1) Transmission line number Ear degree (m) One 17.9485 2 20.0757 3 21.2538 4 19.7515

As described above, the apparatus for estimating the direction of a transmission line according to an embodiment of the present invention detects 3D point cloud data using an unmanned aerial vehicle equipped with a lidar sensor 11, and maps the detected data in real time with a signal processing technique. Automatically estimates the angle of the transmission line.

In addition, the transmission line direction estimation apparatus according to an embodiment of the present invention prevents a phenomenon in which the transmission line is disconnected without sustaining excessive tension at the transmission line support in an extreme environment due to deterioration of the transmission line.

In addition, since the transmission line direction estimation apparatus according to an embodiment of the present invention measures the direction of the transmission line through an unmanned aerial vehicle at a long distance, the manager who measures the direction of the transmission line is prevented from being exposed to danger.

In addition, the transmission line road estimating device according to an embodiment of the present invention can calculate the ear level of the transmission line even in an extreme environment, and diagnose the safety of the transmission line or periodically measure the growth rate of the surrounding trees and the path of the transmission line. Since the variation is estimated, it can be used as a method of preventive diagnosis of transmission lines.

The present invention has been described with reference to the embodiments shown in the drawings, but these are only exemplary, and those of ordinary skill in the art to which the present technology pertains, various modifications and other equivalent embodiments are possible. I will understand. Therefore, the true technical protection scope of the present invention should be determined by the following claims.

10: sensing unit 4 11: lidar sensor
12: GPS sensor 13: IMU sensor
20: 3D point cloud data conversion unit
30: down sampling unit 40: ear canal estimation unit
41: surrounding environment removal unit 42: transmission line recognition unit
43: point cloud data removal unit 44: transmission line detection unit
45: ear canal detection unit

Claims (19)

A 3D point cloud data conversion unit for converting the LIDAR point cloud data measured in relative coordinates with respect to the position of the unmanned aerial vehicle into 3D point cloud data in an absolute coordinate axis;
A down-sampling unit for down-sampling the 3D point cloud data converted by the 3D point cloud data conversion unit to generate a 3D point cloud map; And
An ear degree estimating unit for detecting a transmission line using the three-dimensional point cloud data down-sampled by the down-sampling unit, and estimating an ear degree of the detected transmission line,
The ear canal estimating unit includes an ambient environment removing unit for removing ambient environment data from the 3D point cloud data down-sampled by the down sampling unit; A transmission line recognition unit for recognizing a location of a transmission line by using 3D point cloud data on the transmission line detected by the surrounding environment removal unit; A transmission line point group data detector configured to detect transmission line point group data by removing data other than the transmission line recognized by the transmission line recognition unit; An individual transmission line detection unit for detecting a degree of deflection of an individual transmission line or catenary line from the transmission line point group data detected by the transmission line point group data detection unit; And an individual transmission line edge detection unit detected by the individual transmission line detection unit,
The transmission line recognition unit orthogonal projection of the three-dimensional point group data of the transmission line detected by the surrounding environment removal unit to the ground surface and a horizontal plane to recognize the location of the transmission line path estimation device. The apparatus of claim 1, wherein the 3D point cloud data conversion unit converts the LIDAR point cloud data into 3D point cloud data by applying absolute position and attitude information of the unmanned aerial vehicle to the LIDAR point cloud data. . The 3D point cloud data of claim 1, wherein the 3D point cloud data conversion unit adds the product of the rotation matrix indicating the attitude of the unmanned aerial vehicle and the unmanned aerial vehicle coordinate axis to the absolute position coordinates of the unmanned aerial vehicle absolute coordinate axis measured by the GPS sensor. An apparatus for estimating an island path of a transmission line, characterized in that converting it into point cloud data. The method of claim 1, wherein the down-sampling unit generates a 3D point cloud map by down-sampling the 3D point cloud data on the absolute coordinate axis converted by the 3D point cloud data conversion unit using a probabilistic down-sampling technique. A transmission line path estimation device. The method of claim 4, wherein the down-sampling unit converts the three-dimensional space of the three-dimensional point cloud data converted by the three-dimensional point cloud data conversion unit into pixels at set intervals, and generates sensor data as a probabilistic expression in each of the generated pixels. An apparatus for estimating a three-dimensional point cloud by updating the occupied probability and the probability of the sensor data coming in from 1 second to t seconds according to the presence of sensor data in the pixel at t seconds. The apparatus of claim 1, wherein the down-sampling unit filters and down-samples the 3D point group data converted by the 3D point group data conversion unit using a distance filter. The apparatus of claim 6, wherein the down-sampling unit extracts data in which the relative distance between the lidar data and the unmanned aerial vehicle is less than a preset threshold value from the three-dimensional point cloud data and filters the three-dimensional point cloud data. . 8. The apparatus of claim 7, wherein the threshold value is greater than a distance from the unmanned aerial vehicle to the most distant transmission line or smaller than a portion where the transmission tower and the ground contact each other. The method of claim 7, wherein the down-sampling unit pixelates the 3D space in the data in which the relative distance between the lidar data and the unmanned aerial vehicle in the 3D point cloud data is less than a preset threshold value at a set interval, and generates a probability for each of the generated pixels. A transmission line, characterized in that the 3D point cloud map is generated by updating the probability that sensor data is occupied and the probability of sensor data coming in from 1 second to t seconds in a theoretical expression according to whether sensor data exists in the pixel at t seconds. Rhoidometric device. delete The method of claim 1, wherein the surrounding environment removal unit classifies point group data within a set distance from a transmission line and a transmission tower from the down-sampled 3D point group data by the down sampling unit into clusters, and An apparatus for estimating an island road of a transmission line, characterized in that separating environmental data. delete The method of claim 1, wherein the transmission line recognition unit selects two arbitrary points from point group data on a horizontal surface and a horizontal plane to create a straight line, and calculates the number of point group data included in a distance perpendicular to the created straight line. A transmission line path estimation apparatus, characterized in that the straight line having the largest value of the number of point group data included in the distance perpendicular to the straight line is repeatedly estimated as the transmission line. The method of claim 1, wherein the transmission line recognition unit stores mode data on a horizontal plane horizontal to the ground surface.

converted to domain, converted

The point where the lines generated in the domain overlap the most is selected, the selected point is converted into a cartesian domain to detect a straight line, and then the straight line with the largest value of the number of point group data included in the distance perpendicular to the detected straight line is used as a transmission line. A transmission line road estimating device, characterized in that it is estimated to be. The transmission line-elevation plane of claim 1, wherein the transmission line point group data detection unit uses the three-dimensional point group data detected by the transmission line recognition unit.

A transmission line path estimation device, characterized in that for detecting a pylon parallel to an axis and removing measurement data on both sides of two transmission pylons (spans) to extract point group data of a transmission line. The method of claim 1, wherein the individual transmission line detection unit detects an individual transmission line in a transmission line-elevation plane by using the transmission line point group data detected by the transmission line point group data detection unit and estimates the degree of deflection of the catenary line. A device for estimating the road of transmission lines. The method of claim 16, wherein the individual transmission line detection unit selects a random three points from the transmission line point group data detected by the transmission line point group data detection unit to create a quadratic polynomial curve, and from the generated curve

Repeat the process of counting the number of points included in the distance to the axis

A transmission line road estimating device, characterized in that the curve in which the values of points included in the distance to the axis are the largest is estimated as a curve including the transmission line. The method of claim 1, wherein the island detection unit calculates the number of subsets of the set in which the secondary polynomial is stored in the individual transmission line data detected by the transmission line detection unit, and uses a plurality of ends of the transmission line indicated by a curve. After making four straight lines, from the center of the straight line,

A transmission line ear canal estimation device, characterized in that the ear canal is detected by measuring a distance parallel to the axis. The method of claim 1, wherein the island detection unit calculates the number of subsets of the catenary equation set from the individual transmission line data detected by the transmission line detection unit, and draws a plurality of straight lines using both end points of the transmission line indicated by the curve. After making, from the center of the straight line

A transmission line ear canal estimation device, characterized in that the ear canal is detected by measuring a distance parallel to the axis.

Images