Method for predicting earthquakes using drones

Abstract

The present invention provides an earthquake prediction method using a drone that includes a flight controller equipped with a main processor for controlling flight, a GPS module, and a battery, controls the drone to maintain a constant altitude by applying a PID control algorithm and measuring the distance from the ground using a LiDAR sensor, and performs a hovering flight upon reaching a measurement point using GPS data and IMU data. The method comprises the steps of: detecting a vibration signal at a measurement point using a seismometer mounted on the drone; measuring the distance from the ground at the measurement point using a distance measuring sensor mounted on the drone; generating an earthquake precursor signal by detecting a low-frequency sound wave at the measurement point using a sound wave measuring sensor mounted on the drone; measuring a geomagnetic value using a geomagnetic sensor mounted on the drone; receiving data of the vibration signal, distance measuring signal, earthquake precursor signal, and geomagnetic value, and then performing a preprocessing process for filtering and noise removal by applying an AI-based signal processing algorithm to the data; and applying the data that has undergone the preprocessing process to an artificial intelligence-based earthquake prediction model to generate earthquake prediction information for predicting the possibility of an earthquake.

Description

Technology Field

[0001] The present invention relates to a method for predicting earthquakes using data collected through a drone. Background Technology

[0002] The Korean Peninsula is known to be a region with a relatively low frequency of earthquakes because it is not part of the 'Ring of Fire,' where massive tectonic plates form boundaries on the Earth's surface, unlike Japan or Taiwan; however, there is a possibility that active faults may develop even within the tectonic plates if stress (stress on the crust) accumulates.

[0003] Several earthquakes that have occurred in recent years have once again highlighted the need for earthquake preparedness. A representative example is the magnitude 5.8 earthquake that occurred on September 12, 2016, 11 kilometers south of Gyeongju, Gyeongsang Province, at a depth of approximately 10 to 15 kilometers—the largest since earthquakes were first detected in the southern Korean Peninsula using modern equipment. Since then, a total of over 600 aftershocks of varying magnitudes have been observed as of March 27, 2017. Within a 100-kilometer radius of the epicenter of the Gyeongju earthquake, there are major cities with populations exceeding 500,000, such as Ulsan, Daegu, Pohang, and Busan, as well as a dense concentration of large-scale industrial facilities including nuclear power plants. Consequently, there is growing concern regarding the expected scale and extent of damage in the event of a major earthquake like the Gyeongju one. The cause of this earthquake is presumed to be the movement of the Yangsan Fault, the largest active fault in the southeastern part of the Korean Peninsula, due to the influence of the Kumamoto (M7.3) earthquake that occurred in Japan on April 16, 2016, about three months before the Gyeongju earthquake. However, while it appears clear that the main shock of the Gyeongju earthquake occurred along the Yangsan Fault, the distribution of aftershocks was mainly on the western side of the Yangsan Fault, and the direction of the aftershocks showed a pattern of moving from the northeast to the southwest (Korea Institute of Geoscience and Mineral Resources, January 24, 2017, Yonhap News). Therefore, it is judged that such large-scale earthquakes and aftershocks are occurring in the Gyeongju region due to the movement of other faults in addition to the movement of the Yangsan Fault.

[0004] Meanwhile, although the intensity of the earthquake in Gyeongju was 5 to 10 times higher, the damage in the Pohang area was greater due to the weak ground. For the first time in the country, liquefaction occurred in about 200 locations, including a beach located 5.5 km from the epicenter.

[0005] Liquefaction is the subsidence of the ground or the soil turning into a liquid-like substance, which can cause damage such as buildings tilting or collapsing, and subsequently, ground subsidence occurred in various parts of the roads over the following years.

[0006] The research team that conducted a surface geological survey of the Pohang area at the time identified the mudstone layer formed from hardened mud as the cause of liquefaction. There is an analysis that Pohang is vulnerable to earthquakes because, having been submerged in water in the past, it has a large accumulation of sand and mud underground and a high volume of groundwater.

[0007] Judging from the aforementioned earthquake incidents, it has been demonstrated that earthquakes can occur in Korea at any time, and the importance of earthquake preparedness systems is being strongly emphasized.

[0008] To establish such an earthquake preparedness system, the most critical element is the essential collection of data to predict damage areas resulting from earthquakes, and data for predicting earthquakes is utilized in various fields such as oil, gas, construction, and environmental monitoring.

[0009] Meanwhile, existing earthquake prediction methods have a problem of low accuracy because they analyze one-dimensional ground waves using seismometers.

[0010] The matters described above as background technology are intended solely to enhance understanding of the background of the present invention and should not be construed as an acknowledgment that they constitute prior art already known to those skilled in the art. Prior art literature

[0011] Republic of Korea Registration No. 10-2170907 (Registered on Oct. 22, 2020) The problem to be solved

[0012] The present invention provides an earthquake prediction method using a drone capable of generating earthquake prediction information by collecting vibration signals, distance measurement signals, low-frequency sound wave signals, and geomagnetic values ​​through the autonomous flight of the drone, and then analyzing the collected data using artificial intelligence.

[0013] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems to be solved will be clearly understood by those skilled in the art to which the present invention pertains from the descriptions below. means of solving the problem

[0014] In order to achieve the above-mentioned problem to be solved, an earthquake prediction method using a drone according to an embodiment of the present invention comprises a flight controller equipped with a main processor for controlling flight, a GPS module, and a battery, and controls the drone to maintain a constant altitude by applying a PID control algorithm and measuring the distance from the ground using a LiDAR sensor, and performs hovering flight upon reaching a measurement point using GPS data and IMU data. The method further comprises the steps of: detecting a vibration signal at a measurement point using a seismometer mounted on the drone; measuring the distance from the ground at the measurement point using a distance measuring sensor mounted on the drone; generating an earthquake precursor signal by detecting low-frequency sound waves at the measurement point using a sound wave measuring sensor mounted on the drone; measuring a geomagnetic value using a geomagnetic sensor mounted on the drone; receiving data of the vibration signal, distance measuring signal, earthquake precursor signal, and geomagnetic value, and then performing a preprocessing process for filtering and noise removal by applying an AI-based signal processing algorithm to the data; and applying the data that has undergone the preprocessing process to an artificial intelligence-based earthquake prediction model to generate earthquake prediction information for predicting the probability of an earthquake. It can be included. Effects of the invention

[0015] According to the aforementioned embodiment of the present invention, by generating earthquake-related data through the flight of a drone and then generating earthquake prediction information based thereon, it is possible to collect various data necessary for earthquake prediction, thereby increasing the accuracy of earthquake prediction. Brief explanation of the drawing

[0016] FIG. 1 is a configuration diagram illustrating an earthquake prediction system using a drone according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating the structure of a drone according to an embodiment of the present invention. FIG. 3 is a flowchart illustrating an earthquake prediction process using a drone according to an embodiment of the present invention. Specific details for implementing the invention

[0017] Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the methods, apparatuses, and / or systems described herein. However, this is merely illustrative and the present invention is not limited thereto.

[0018] In describing the embodiments of the present invention, if it is determined that a detailed description of known technology related to the present invention may unnecessarily obscure the essence of the present invention, such detailed description will be omitted. Furthermore, the terms described below are defined considering their functions in the present invention, and these may vary depending on the intentions or practices of the user or operator. Therefore, such definitions should be based on the content throughout this specification. The terms used in the detailed description are merely for describing the embodiments of the present invention and should not be limiting in any way.

[0019] Referring to the attached drawings, an environmental measurement drone and a method for collecting environmental data using it will be described below.

[0020] FIG. 1 is a configuration diagram illustrating an earthquake prediction system using a drone according to an embodiment of the present invention, and FIG. 2 is a block diagram illustrating the structure of a drone according to an embodiment of the present invention.

[0021] As illustrated in FIG. 1, the earthquake prediction system can collect and generate data through the autonomous flight of a drone (100) and then transmit it to a cloud server (200) via a network, such as a wired or wireless communication network. Accordingly, the cloud server (200) can generate prediction information that can predict the location of an earthquake using an artificial intelligence-based earthquake prediction model (310).

[0022] As illustrated in FIG. 2, the drone (100) may include a communication circuit (102) for connecting to a wireless communication network, a sensor (104) for sensing earthquake-related data, a flight controller (109) for controlling the flight of the drone (100), a location information acquisition unit (110) for acquiring location information of the drone (100), a data generation unit (114) for generating matching data in which earthquake-related data and location information are matched and storing it in a memory (112), a battery (not shown), etc.

[0023] The detector (104) can collect earthquake-related data by being equipped with a seismometer (105) that senses vibration signals by detecting vibration data of the ground surface, a distance measuring sensor (106) for measuring the distance from the ground during flight of the drone (100), a sound wave sensing sensor (107) that detects low-frequency sound waves during flight to generate earthquake precursor signals, and a geomagnetic sensor (108) for sensing geomagnetic values.

[0024] The seismometer (105) is designed to detect vibrations occurring on the surface of the earth and to analyze the frequency and amplitude to sense vibration signals, and may be a MEMS (Micro-Electro-Mechanical Systems) based seismometer.

[0025] In particular, the seismometer (105) can be attached to the bottom of the drone (100) so as to be grounded upon landing, and preferably, the sensor can be embedded in the landing gear of the drone (100) so as to be in contact with the ground only at the time of measurement. Infrasound microphone: Detects low-frequency sound waves to analyze earthquake precursor signals.

[0026] The distance measuring sensor (106) can be installed at the bottom of the drone (100) to measure the distance to the ground surface in real time, and as an example, it may be an ultrasonic sensor.

[0027] The sound wave measurement sensor (107) detects low-frequency pressure changes in the atmosphere during flight and generates an earthquake precursor signal, and may be an infrasound detection sensor or a high-sensitivity microphone.

[0028] In particular, the sound wave measuring sensor (107) can be attached to the side or top of the drone (100) to detect changes in air pressure.

[0029] The geomagnetic sensor (108) is capable of sensing geomagnetic values, and an example of such a 3-axis geomagnetic sensor can be cited.

[0030] In particular, the geomagnetic sensor (108) can be placed on the top or side of the drone (100).

[0031] The flight controller (108) may be composed of a main processor that performs control functions for the operation of the drone (100), such as flight attitude control of the drone (100).

[0032] That is, the flight controller (108) according to an embodiment of the present invention can be linked with an IMU (Inertial Measurement Unit) sensor for maintaining balance by detecting the tilt and movement of the drone (100) using a main processor, a GPS (Global Positioning System) for providing position information of the drone (100), and a plurality of motors for flight control of the drone (100).

[0033] Additionally, the flight controller (108) can control the flight of the drone (100) by executing an autonomous driving program, that is, an autonomous driving program stored in a form executable in memory (112), to drive the drone (100) autonomously through the main processor.

[0034] The autonomous driving program is designed to drive the drone (100) autonomously in a predetermined area (hereinafter referred to as the "earthquake detection area") so that the drone (100) does not go outside a preset radius and does not collide with various obstacles, and may be equipped with functions such as detection, perception, thinking, and action.

[0035] These functions are composed of at least one command and can be executed using sensor values ​​output from a sensor equipped in the drone (100), such as an IMU sensor.

[0036] First, the detection function can be performed through earthquake-related data output from an IMU sensor capable of measuring environmental data necessary for autonomous driving of the drone (100) in an earthquake detection area.

[0037] The perception function may mean recognizing objects, recommending objects, and perceiving states and situations within an environment sensing area based on environmental data collected by the sensing function and location information of the drone (100).

[0038] Thinking function may refer to real-time decision-making, action planning, movement path setting, situational reasoning and prediction, and big data-based judgment to respond appropriately to states and situations identified through cognitive functions.

[0039] The action function can perform the function necessary for the flight of the drone (100) within the flight controller (108) according to the judgment determined by the thinking function.

[0040] That is, the flight controller (108) according to the embodiment of the present invention can control parts within the drone (100), such as a transmission, to enable the drone (100) to fly autonomously within an earthquake detection area through the position information of the drone (100), path information obtained through a judgment function, and flight control information obtained through the execution of an autonomous driving program.

[0041] That is, the flight controller (108) according to the embodiment of the present invention measures the distance from the ground using a LiDAR sensor and applies a PID control algorithm to maintain a constant altitude in the range of 1 to 2 m, and can also control the flight of the drone (100) by combining it with an IMU (Inertial Measurement Unit) sensor to respond to wind or obstacles.

[0042] Additionally, the flight controller (108) can correct the position of the drone (100) by combining GPS and IMU data for stationary flight, i.e., hovering of the drone (100), and accurately measure as it reaches a measurement point in automatic flight mode.

[0043] Meanwhile, the flight data generation unit (114) of the drone (100) according to an embodiment of the present invention can generate matching data by matching the flight location information of the drone (100) provided by the location information acquisition unit (110) with earthquake-related data, and then store the data in the memory (112).

[0044] In particular, the data generation unit (114) applies data detected by the detector (104), namely, vibration signals, distance signals, sound waves, and geomagnetic values, to an edge AI-based signal processing algorithm to generate earthquake-related data through preprocessing processes such as filtering and noise removal, and can store matching data that matches the earthquake-related data with location information in memory (112).

[0045] The data generation unit (114) according to an embodiment of the present invention may generate earthquake occurrence prediction information by applying time series analysis, spectrum analysis, an artificial intelligence-based earthquake prediction model, etc. to the matching data.

[0046] Specifically, the data generation unit (114) can analyze the vibration signal that has undergone a preprocessing process to distinguish between the P-wave (Primary wave) and S-wave (Secondary wave) of the seismic wave, convert the ground change into time-series data through the distance signal that has undergone a preprocessing process, and detect abnormal changes.

[0047] In addition, the data generation unit (114) can detect earthquake precursors by analyzing the preprocessed sound wave signal, i.e., low-frequency data, into abnormal patterns by applying an artificial intelligence model, and can detect geomagnetic anomalies by continuously analyzing geomagnetic changes by processing the preprocessed geomagnetic values ​​with a deep learning-based time series analysis model (e.g., RNN, LSTM, etc.).

[0048] In addition, the data generation unit (114) can generate earthquake prediction information that can predict the likelihood of an earthquake through multi-variable analysis by applying earthquake-related data to an artificial intelligence-based earthquake prediction model, such as a CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network) based earthquake prediction model.

[0049] In the embodiment of the present invention, an example of generating earthquake prediction information through the data generation unit (114) of the drone (100) has been described, but earthquake prediction information may also be generated through the earthquake prediction model (210) of an external cloud server (200). In this case, the drone (100) may transmit matching data to the cloud server (200) at preset time intervals through the communication circuit (102), or transmit matching data to the cloud server (200) as the drone (100) lands.

[0050] Accordingly, the cloud server (200) can generate earthquake prediction information by applying the matching data received from the drone (100) to the earthquake prediction model (210).

[0051] Hereinafter, the earthquake prediction process using a drone according to an embodiment of the present invention will be described with reference to FIG. 3.

[0052] FIG. 3 is a flowchart illustrating an earthquake prediction process using a drone according to an embodiment of the present invention.

[0053] As illustrated in FIG. 3, the drone (100) receives flight path data from a cloud server (200) connected to a communication network via a communication circuit (102) and sets it in memory (112) (S300). At this time, the flight path data may have a GPS coordinate-based ROI (Region Of Interest) (hereinafter referred to as 'measurement point') set.

[0054] Afterwards, the flight controller (108) of the drone (100) makes the drone (100) fly autonomously based on flight path data (S310). That is, the flight controller (108) can make the drone (100) fly autonomously based on the location information and flight path data obtained from the location information acquisition unit (110).

[0055] During autonomous flight, if the location information acquired by the location information acquisition unit (110) corresponds to the GPS coordinates of the measurement point (S320), the flight controller (108) causes the flight to stop for a preset time and also activates the detector (104) (S330).

[0056] During stationary flight, the detector (104) can detect vibration signals at the measurement point using a seismometer (105), measure the distance from the ground using a distance measuring sensor (106), and generate earthquake precursor signals by detecting low-frequency sound waves using a sound wave measuring sensor (107) mounted on the drone (100), and measure geomagnetic values ​​(S340) using a geomagnetic sensor (108) mounted on the drone (100).

[0057] Afterwards, the data generation unit (114) applies data measured by the detector (104), namely, vibration signals, distance measurement signals, low-frequency sound wave signals, and geomagnetic values, to an edge AI-based signal processing algorithm to generate earthquake-related data (S350) through preprocessing processes such as filtering and noise removal, and can store matching data that matches the earthquake-related data with location information in memory (112).

[0058] Afterwards, the data generation unit (114) can generate earthquake occurrence prediction information (S360) by applying time series analysis, spectrum analysis, and an artificial intelligence-based earthquake prediction model to the matching data.

[0059] According to the embodiment of the present invention as described above, although it has been explained as an example that earthquake occurrence prediction information is generated by a data generation unit (114) within a drone (100), earthquake occurrence prediction information may also be generated by transmitting earthquake-related data generated through a communication circuit (102) to a cloud server (200) and then using an earthquake prediction model (210) within the cloud server (200).

[0060] The above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains may make various modifications and variations within the scope of the essential characteristics of the present invention. Accordingly, the embodiments expressed in the present invention are intended to explain, not limit, the technical concept of the present invention, and the scope of rights of the present invention is not limited by such embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts equivalent to or within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0061] 100 : Drone

Claims

Claim 1 A method for generating earthquake prediction information using a drone comprising a flight controller equipped with a main processor for controlling flight, a GPS module, and a battery, controlling to maintain a constant altitude by applying a PID control algorithm and measuring the distance from the ground using a LiDAR sensor, and performing hover flight upon reaching a measurement point using GPS data and IMU data, the method comprising: a step of detecting a vibration signal at a measurement point using a seismometer mounted on the drone; a step of measuring the distance from the ground at the measurement point using a distance measuring sensor mounted on the drone; a step of generating an earthquake precursor signal by detecting a low-frequency sound wave at the measurement point using a sound wave measuring sensor mounted on the drone; a step of measuring a geomagnetic value using a geomagnetic sensor mounted on the drone; a step of receiving data of the vibration signal, distance measuring signal, earthquake precursor signal, and geomagnetic value, and then performing a preprocessing process for filtering and noise removal by applying an AI-based signal processing algorithm to the data; and a step of generating earthquake prediction information to predict the possibility of an earthquake by applying the data that has undergone the preprocessing process to an artificial intelligence-based earthquake prediction model.

Images