KR102044912B1 - Unmanned aircraft for structural crack detection using machine learning - Google Patents

Abstract

The present invention relates to an unmanned aerial vehicle capable of detecting a crack in a structure from a photographed result by using a machine learning by improving a flying performance by mechanically connecting a plurality of rotorcrafts, the housing and the front of the housing It is installed on one side of the image pickup unit for acquiring the image data of the facing area, and connected to the shooting unit by wire or wireless, and detect the cracks of the structure included in the image data by analyzing the image data using machine learning It includes a crack analysis unit, spaced apart at equal intervals in the housing, a plurality of aircraft to generate a thrust to fly with the housing, and the main control unit for controlling the flight direction and flight speed by controlling the vehicle; .

Description

Unmanned aerial vehicle detecting machine cracks using machine learning {UNMANNED AIRCRAFT FOR STRUCTURAL CRACK DETECTION USING MACHINE LEARNING}

The present invention relates to an unmanned aerial vehicle for detecting a crack in a structure using machine learning, and more particularly, a plurality of rotorcrafts are mechanically connected to improve flight performance, and the structure of the structure in the result photographed using machine learning. It is a technology for unmanned aerial vehicles that can detect cracks.

In general, a drone is a general term for an unmanned aerial vehicle or uninhabited aerial vehicle (UAV) in the form of an airplane or helicopter that can be operated and controlled by radio wave guidance without a pilot. Drones are used for military purposes and are used as targets for practice shooting or for reconnaissance, surveillance and anti-submarine attacks. Around the 2010s, it has been used in various civilian fields besides military use. In the private sector, the field of application is being expanded by using drones for home delivery, broadcasting, hobby video shooting, and lifesaving.

Types of drones are categorized according to takeoff and landing modes, as well as a sliding takeoff and landing vehicle having a fixed wing and a lifting and landing vehicle with a rotorcraft.

Rotorcraft drones are being spread by hobbyists in watches for remote use, but recently, the concept of supplying (delivering) various cargoes, such as cargo, documents, books and emergency medicines, which does not have heavy take-off weight, has been realized.

In particular, efforts have been made to apply unmanned aerial vehicles to detect cracks in walls of structures such as buildings.

Korean Patent Publication No. 10-1796258 is a vision-based structure stability test method using a small unmanned aerial vehicle, and proposes a method of inspecting the stability of a structure using a small unmanned aerial vehicle and a camera. The patent acquires an image of a structure with an imaging device, compresses the image and wirelessly transmits the image, the image processing apparatus receives the wireless transmitted image and detects an outline, outputs the display and detects whether the user is cracked. It consists of steps. However, this crack detection method has a limitation that the user must check whether the crack manually. In addition, when the rotor blades collide with the wall, it is crashed, and when the wind is strong, it is difficult to be adjacent to the building due to the characteristics of the small drone vulnerable to disturbance, such as difficulty in flight control, making it difficult to detect a minute defect (cracks, etc.) of the building.

On the other hand, Patent Publication No. 10-1536574 proposed a vehicle for the structure inspection. Referring to FIG. 1, this patent addresses this problem by applying adsorption fixing means 10 adsorbed to a wall to prevent the aircraft from colliding with or vibrating the wall by disturbances such as side winds while inspecting the structure. Was to solve. However, when the aircraft is adsorbed on the wall, the aircraft moves and cannot take a picture, so it takes a considerable time to detect a crack of a large structure, and there is a problem that it may cause instability in the flight due to the change in dynamic characteristics. .

Patent Application Publication 10-1796258 Patent Publication 10-1536574

Therefore, the present invention has been proposed to solve the conventional problems as described above, an object of the present invention is a plurality of rotary wing aircraft is mechanically connected to improve the flight performance, the structure of the structure in the result taken using the machine learning It is to provide an unmanned aerial vehicle capable of detecting cracks.

In order to achieve the above object, the unmanned aerial vehicle which detects a crack of a structure by using the machine learning according to the technical idea of the present invention is installed on one side of the housing, facing the front, and an image of an area facing each other. A photographing unit for acquiring data, a cable analyzing unit connected to the photographing unit by wire or wirelessly, and analyzing the image data using machine learning to detect a crack of a structure included in the image data, and the housing It is spaced apart at equal intervals in the, characterized in that it comprises a plurality of aircraft to generate a thrust to fly with the housing, and the main controller for controlling the flight direction and flight speed by controlling the vehicle.

The vehicle may be coupled to a link extending from the housing, and the vehicle and the link may be coupled to each other through a universal joint.

In addition, the vehicle may be characterized in that it comprises a plurality of rotor blades for generating a thrust.

In addition, the main control unit may control the thrust of the rotor blades of the position opposite to the flight direction more strongly than the rotary blades located in the flight direction of the plurality of the rotor blades of the rotor, when the flying direction control in the front and rear or left and right directions. .

The main controller may control the thrust of the aircraft disposed behind the housing more strongly than the aircraft disposed in front of the housing when the flight attitude is controlled to increase the rear of the housing. You can do

The main controller may control the thrust of the aircraft disposed on the right side of the housing more strongly than the aircraft corresponding to the left side of the housing when the flight attitude is controlled to raise the right side of the housing from the left side. It may be characterized by.

In addition, the main control unit controls the thrust force of the front of the vehicle disposed on the left side of the housing among the plurality of the aircraft more strongly than the rotary blades located in the rear when the flight attitude control of the front of the housing is turned in the left direction. Or, it may be characterized in that the thrust is more strongly controlled than the rotary blades located in front of the rotary blades located in the rear of the vehicle disposed on the right side of the housing.

The apparatus may further include a space keeping member coupled to the front of the housing such that the structure positioned at the front and the housing may fly at a distance spaced apart from each other.

In addition, the space keeping member is formed to extend in the front from the housing long, it may be characterized in that the extended end is coupled to the wheel rolling in the left and right directions.

The main controller may control a flight direction in the direction of the structure when the image data of the structure is acquired.

According to the unmanned aerial vehicle which detects the crack of a structure using the machine learning by this invention,

First, as the plurality of vehicles are coupled to the housing, the payload that can fly is increased.

Second, as the housing and the vehicle are coupled through the universal joint, the gimbal for maintaining the posture of the photographing unit can be omitted.

Third, the payload is increased by the plurality of vehicles, and as the plurality of vehicles are all connected to the housing, it is possible to apply a battery having a larger capacity than the conventional one, thereby allowing a longer flight.

Fourth, as the unmanned aerial vehicle moves in one direction to the structure and performs shooting and analysis in real time, it is possible to perform crack detection of the structure more quickly than before.

Fifth, the unmanned aerial vehicle is contacted with the structure by using the space keeping member to reduce the vibration caused by the internal or external causes, thereby improving the quality of the image data.

Sixth, as the wheels of the space keeping member is rotated only in the left and right directions, the unmanned aerial vehicle is restricted from moving up and down, and thus stable left or right movement is possible.

1 is a side view of one embodiment of Patent Publication No. 10-1536574.
2 is a perspective view of an unmanned aerial vehicle for detecting cracks in a structure using machine learning according to an embodiment of the present invention.
Figure 3 is a block diagram of the electrical components of the unmanned aerial vehicle for detecting cracks of the structure using the machine learning according to an embodiment of the present invention.
4 is a side view of a state in which the front of the unmanned aerial vehicle for detecting the crack of the structure is turned downward by using the machine learning according to an embodiment of the present invention.
5 is a plan view of an unmanned aerial vehicle for detecting cracks in a structure using machine learning according to an embodiment of the present invention.
FIG. 6 is a side view illustrating an example of controlling a thrust to maintain a state where an unmanned aerial vehicle detecting a crack of a structure is in contact with the structure by using machine learning according to an embodiment of the present invention. FIG.
7 is a view showing that the unmanned aerial vehicle is restricted in movement in the vertical direction by the wheels coupled to the space keeping member, and is moved to the right or the left.
FIG. 8 is a view illustrating a drone that detects a crack in a structure by using machine learning according to an embodiment of the present invention and moves in a right direction along a preset shooting path.

With reference to the accompanying drawings will be described in detail with respect to the unmanned aerial vehicle for detecting the crack of the structure by using the machine learning according to embodiments of the present invention. As the inventive concept allows for various changes and numerous modifications, particular embodiments will be illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to the specific form disclosed, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the drawings, similar reference numerals are used for similar elements.

Also, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art and shall not be construed in ideal or excessively formal meanings unless expressly defined in this application. Do not.

2 and 3, the unmanned aerial vehicle 100 that detects a crack in a structure using machine learning according to an embodiment of the present invention has a housing 130 and a front side of the housing 130. Is connected to the photographing unit 120 and wired or wirelessly connected to the photographing unit 120 to acquire image data of an area facing each other. The image data is analyzed by using machine learning. Crack analysis unit 102 for detecting a crack of the structure, a plurality of spaced to be spaced apart at equal intervals in the housing 130, generating a thrust 170 flying with the housing 130, and the vehicle It is characterized in that it comprises a main control unit 101 for controlling the flight direction and flight speed by controlling the 170.

In this case, the front is a direction that the imaging unit 120 gazes at, and the rear is a direction opposite to the direction that the imaging unit 120 gazes at, and becomes an X-axis. The lower side is the direction of gravity, and the upper side is the direction opposite to gravity, which is Z-axis. The right direction and the left direction are the right and left sides of the direction in which the photographing unit 120 prepares, and are Y-axis.

The structure includes a large building such as an apartment, a building, a bridge, and a tunnel. The unmanned aerial vehicle 100 of this embodiment detects cracks formed on the outer wall and the inner wall of the structure.

In one embodiment, the crack analysis unit 102 and the main control unit 101 may be accommodated in a case 150 separately provided in the middle of the housing 130. In the case 150, the GPS sensor 104 for receiving a GPS signal, the main gyro sensor 106 for acquiring a posture value of the housing 130, and a separate control system provided on the ground are wirelessly communicated with each other. The power supply unit 190 for supplying power to the configuration of the communication module 180, the main control unit 101, the crack analysis unit 102, the plurality of aircraft 170, and the like to transmit and receive the photographing position, image data, etc. may be further accommodated. Can be.

In addition, the proximity of the case 150 or the housing 130 may include a proximity sensor 103 for detecting a distance away from the adjacent structure. The main controller 101 controls the flight direction and flight attitude so that the unmanned aerial vehicle 100 of this embodiment is not in close contact with the structure using the spaced distance values detected by the proximity sensor 103. Proximity sensor 103 is preferably disposed in the front, rear, left, right, up and down directions of the case 150 or the housing 130 to prevent a collision in the entire direction. The proximity sensor 103 may be configured as an optical sensor using a laser, a sound wave sensor using a sound wave, or the like.

The photographing unit 120 may be a camera that acquires an image image in the direction of attention, or a photographing apparatus that acquires a stereoscopic image as a signal received after scanning sound waves, ultrasound waves, lasers, and the like. The photographing unit 120 may also include illumination (not shown) that illuminates light such as infrared rays and visible light in a region to be photographed so that photographing may be performed even in a dark environment.

The crack analyzing unit 102 receives the image data acquired by the photographing unit 120 and detects the crack by inputting the image data into a preloaded machine learning algorithm. The machine learning algorithm is based on the assumption that learning for crack detection is performed by inputting image data prepared in advance.

The crack analyzing unit 102 may be implemented as mounted in the housing 130, and the crack analyzing unit 102 may be provided only on the ground to receive image data through the communication module 180.

The crack analysis unit 102 is preferably a computing device that is executed by including a machine learning algorithm.

Aircraft 170 is characterized in that it comprises a plurality of rotor blades for generating thrust. The aircraft 170 flying with the thrust generated from the plurality of rotor blades is useful when shooting the structure can stop freely at one position in the flight state, or can freely move forward and backward, left and right, up and down.

The vehicle 170 uses an electric motor 175 for driving a rotor blade, a drone gyro sensor 176 for obtaining a posture value of the aircraft 170, and a posture value obtained through the drone gyro sensor 176. The drone controller 173 may control the attitude of the vehicle 170 by controlling the driving of the vehicle 175.

The plurality of vehicles 170 may be lighter by using the GPS sensor 104, the proximity sensor 103, and the power supply unit 190 coupled to the housing 130.

As shown, the aircraft 170 is preferably spaced apart from each other at a uniform interval on the edge portion of the housing 130. The number of aircraft 170 may be at least two, and may be configured in a larger number than this embodiment of four. As the number of vehicles 170 increases, the payload that can fly in the unmanned aerial vehicle 100 of this embodiment increases.

As a method for increasing the payload, there is a method of applying a larger rotor blade of the aircraft 170, but since the payload is not increased in proportion to the size of the rotor blade, a rotor blade of a certain size is more effective for effective flight. Inappropriate.

When the photographing unit 120 of the present embodiment includes a separate light or the computing device equipped with the crack analysis unit 102 is included in the housing 130, the load of this embodiment may be greatly increased. The larger the number of h), the higher the payload, making it easier to fly large loads.

Referring to FIG. 4, in this embodiment, the link 140 extends at each vertex portion of the rectangular housing 130, and the vehicle 170 is coupled to the end of the link 140. In particular, the vehicle 170 is coupled to the link 140 via the universal joint 142 in order to take a free flight posture even when coupled to the link 140. The universal joint 142 has a rotation axis in the front and rear direction of the portion coupled to the drone frame 171 constituting the body of the aircraft 170, and the portion coupled to the link 140 has a left and right rotation shaft orthogonal to the rotation axis. By being implemented to have, it is preferable that the vehicle 170 is configured to be freely tilted in the front and rear and left and right directions on the link 140.

Only the attitude of the aircraft 170 is inclined when the flight direction and the flight attitude are controlled, and as the housing 130 can maintain the attitude, the photographing unit 120 has a configuration such as a conventional gimbal for maintaining a constant posture. Can be omitted.

The main controller 101 controls the flight of the unmanned aerial vehicle 100 of this embodiment. A method of controlling the flight direction and attitude by controlling the plurality of vehicles 170 by the main controller 101 will be described.

The main controller 101 controls the thrust of the rotor blade at a position opposite to the flight direction of the rotor blades in the flight direction among the rotor blades of the plurality of vehicles 170 when controlling the flight direction in the front-rear (X-axis) or left-right (Y-axis) directions. More control. For example, when the flying direction is to be forward, as shown in FIG. 2, the main controller 101 allows the rotary blades (motors 175) located at the rear of the plurality of rotary blades included in the vehicle 170 to be driven more strongly. , So that the rotary blade (motor 175) located in the front is driven to be weaker, so that the thrust of the rotary blade located in the rear is relatively stronger. When the thrust of the rotor blades located at the rear of the aircraft 170 becomes relatively stronger, the aircraft 170 is in a forward tilted downward position, but the aircraft 170 and the link 140 are coupled to the universal joint 142 Since the housing 130 is in an inclined state, the housing 130 also does not tilt.

In addition, the main controller 101 can control the flight attitude so that the pitch axis of the unmanned aerial vehicle 100 of this embodiment is rotated. For example, as shown in FIG. 4, the main controller 101 has a flying posture that causes the housing 130 viewed from the side direction to be turned counterclockwise, that is, the front of the housing 130 is lowered and the rear is raised. During flight attitude control, the thrust of the vehicle 170 disposed at the rear of the housing 130 among the plurality of vehicles 170 is more strongly controlled than the aircraft 170 disposed at the front of the housing 130. The pitch axis rotation of the unmanned aerial vehicle 100 is used to maintain the level of the housing 130 or to control the photographing angle of the photographing unit 120.

In addition, the main controller 101 can control the flight attitude so that the roll axis of the unmanned aerial vehicle 100 of this embodiment is rotated. For example, the main controller 101 may control a plurality of aircrafts in a flight attitude that causes the housing 130 to be turned counterclockwise, that is, when the right side of the housing 130 is raised from the left side. The thrust of the aircraft 170 disposed on the right side of the housing 130 among the 170 is controlled to be stronger than the aircraft 170 disposed on the left side of the housing 130. The roll shaft rotation of the unmanned aerial vehicle 100 is used to keep the housing 130 in a horizontal position even in a disturbance such as wind.

In addition, the main controller 101 can control the flight attitude so that the yaw axis of the unmanned aerial vehicle 100 of this embodiment is rotated. For example, referring to FIG. 5, the main controller 101 is a flight attitude that causes the housing 130 to be rotated counterclockwise in a planar direction, that is, a flight in which the front of the housing 130 is turned to the left direction. In the posture control, the thrust of the front rotor of the aircraft 170 disposed on the left side of the housing 130 among the plurality of vehicles 170 is controlled to be stronger than the rotor blades located at the rear, or the right side of the housing 130. The rotary blades located at the rear of the vehicle 170 disposed in the control the thrust relatively stronger than the rotary blades located in the front.

On the other hand, the unmanned aerial vehicle 100 that detects the crack of the structure by using the machine learning according to an embodiment of the present invention so that the structure and the housing 130 located in front of the space can be maintained at a distance spaced apart It further includes a space keeping member 135 coupled to the front of the housing 130.

The space keeping member 135 of this embodiment was implemented to be formed to extend in the front from the housing 130 long.

The space keeping member 135 maintains a constant distance between the photographing unit 120 and the structure so that the photographed image data has a certain quality. For example, when the photographing unit 120 is configured as a camera, if the distance between the photographing unit 120 and the structure is kept constant, the focus of the photographing unit 120 does not always need to be controlled. It is possible to omit the lens so that a clearer image can be obtained.

In addition, when the space keeping member 135 is in contact with the structure, the vibration generated by the disturbance such as the electric motor 175 or the surrounding wind of the aircraft 170 is reduced to the quality of the image data of the photographing unit 120 This can be further improved.

In addition, the space keeping member 135 to prevent the aircraft 170 from colliding with the structure by maintaining the distance between the structure and the housing 130 spaced apart.

Since the unmanned aerial vehicle 100 of this embodiment performs imaging in a state adjacent to the structure, the range that can be photographed can be greatly narrowed compared to the size of the structure. That is, when the unmanned aerial vehicle 100 is photographed while the unmanned aerial vehicle 100 is adsorbed to the structure, as in Patent Publication No. 10-1536574, it takes a very long time to photograph the entire large structure such as a bridge.

6 to 8, wheels rolling in the left and right directions are coupled to the extended ends of the space keeping member 135. When the image data of the structure is acquired, the main control unit 101 controls the flight direction in the direction of the structure so that the wheels of the space keeping member 135 closely adhere to the structure. When the space keeping member 135 is in close contact with the structure, the main control unit 101 moves the flight direction to the left or right direction. At this time, the flying speed moving in the left direction or the right direction is preferably kept constant. The photographing unit 120 and the crack analyzing unit 102 acquire image data in real time while the unmanned aerial vehicle 100 moves in a left direction or a right direction, and detects a crack in the image data. As the unmanned aerial vehicle 100 constantly moves and cracks are detected in real time, the time for photographing the entire structure is greatly increased.

In the drawings of this embodiment, the space keeping member 135 and the wheel is largely applied for emphasis, but in the embodiment to be manufactured, it is preferable that the thickness and size are smaller.

In addition, since the image data photographed while the unmanned aerial vehicle 100 is stopped is a two-dimensional image, the performance of crack detection is degraded due to the color of the structure, interference of ambient light, and the like. On the other hand, the unmanned aerial vehicle 100 according to an embodiment of the present invention uses at least a certain distance from the structure and moves in one direction, at least two or more image data photographed at different positions of the same portion of the structure Detect cracks by comparing For example, when a crack is generated in any part of the structure and the unmanned aerial vehicle 100 moves to the right and performs a crack inspection, the first image data and the crack photographed from the left side are taken from the right side. Combining one second image data makes it possible to obtain a three-dimensional image. The crack analysis unit 102 generates a three-dimensional image by combining a plurality of image data and then detects a crack, thereby improving the detection performance of the crack and detecting even finer cracks.

In addition, this embodiment may further include an imprint (not shown) coupled to the circumference of the wheel to leave an indication on the structure when the wheel is clouded in contact with the structure. In one embodiment, the engraving portion is discharged ink when the wheel is clouded, it can leave the ink in the structure in contact with the wheel when the unmanned aerial vehicle 100 is moved in the left and right direction in close contact with the structure. The ink marks left on the structure are included in the image data captured by the photographing unit 120 in real time. The ink marks included in the image data are transmitted to the main controller 101, and the main controller 101 analyzes the direction in which the ink marks are formed, and then horizontally and uniformly when the unmanned aerial vehicle 100 moves to the left or right direction. You can feedback whether it is moving.

The marking may be a fine line that is not visually identified at a distance, or may be applied with a special ink that can be identified by illuminating ultraviolet rays, and may be a fluid such as water or alcohol which naturally evaporates over time.

Next, an example of use of the unmanned aerial vehicle 100 for detecting a crack of a structure according to an embodiment of the present invention will be described.

First, the position or shape information of the structure to be crack detection object is input to the main control unit 101. The location information may be information such as a GPS value, and the shape information may be such as a drawing of a structure or a three-dimensional image.

The main controller 101 sets a position at which image pickup is to be started, a photographing path, and a position at which image pickup ends in an area of the structure where crack detection is to be performed. The photographing path is set to one line which is moved in the left direction or the right direction, and when the photographing of one line is completed, the unmanned aerial vehicle 100 is moved to the next adjacent line.

The main controller 101 controls the aircraft 170 to allow the unmanned aerial vehicle 100 to fly toward the input structure.

After reaching the structure, the unmanned aerial vehicle 100 approaches the structure such that the space keeping member 135 contacts after moving to the position where the shooting starts.

When the space keeping member 135 is in contact with the structure, the unmanned aerial vehicle 100 is moved in the left direction or the right direction along the shooting path, while the photographing unit 120 photographs the structure in real time. While the photographing unit 120 acquires the image data in real time, the crack analysis unit 102 also analyzes the image data in real time to detect cracks in the structure.

Information about the crack of the detected structure is transmitted to a separate system provided on the ground through the communication module 180, or stored in the crack analysis unit 102.

The unmanned aerial vehicle 100 returns to the preset return point when the shooting end position is reached after passing through the preset shooting path.

If the power of the power supply unit 190 is depleted during the imaging of the structure, the unmanned aerial vehicle 100 may return to a moment, charge the power, and move to the stopped shooting path to continue crack detection. In this case, since the mark remains at the position where the image is taken through the engraving portion, the path and the area where the image is not detected may be easily identified.

Although the preferred embodiment of the present invention has been described above, the present invention may use various changes, modifications, and equivalents. It is clear that the present invention can be applied in the same manner by appropriately modifying the above embodiments. Therefore, the above description does not limit the scope of the present invention as defined by the limitations of the following claims.

100: unmanned aerial vehicle 101: main control unit
102: crack analysis unit 103: proximity sensor
104: GPS sensor 106: main gyro sensor
120: photographing unit 130: housing
135: space keeping member 140: link
142: universal joint 150: case
170: aircraft 171: drone frame
173: drone control unit 175: electric motor
176: drone gyro sensor 180: communication module
190: power supply

Claims (10)

A housing;
A photographing unit installed at one side of the housing while looking forward and acquiring image data of an area facing each other;
A crack analyzing unit connected to the photographing unit by wire or wirelessly and detecting a crack of a structure included in the image data by analyzing the image data using machine learning;
A plurality of vehicles disposed at equal intervals in the housing and generating thrust and flying together with the housing;
It includes a main control unit for controlling the flight direction and flight speed by controlling the vehicle,
The vehicle is coupled to a link extending from the housing,
And the vehicle and the link are coupled to each other via a universal joint. delete The method of claim 1,
The aircraft detects a crack of the structure, characterized in that it comprises a plurality of rotor blades for generating a thrust. The method of claim 3,
The main control unit controls the thrust of the rotor blades at positions opposite to the flight direction than the rotor blades located in the flight direction among the rotor blades of the plurality of flying bodies when controlling the flight direction in the front, rear, left and right directions. Unmanned drone detecting. The method of claim 3,
The main controller may control the thrust of the vehicle disposed behind the housing among the plurality of the vehicles more strongly than the vehicle disposed at the front of the housing when the flight attitude of the rear of the housing is raised. Unmanned aerial vehicle that detects cracks in structures. The method of claim 3,
The main controller may control the thrust of the aircraft disposed on the right side of the housing more strongly than the aircraft corresponding to the left side of the housing when the flight attitude is controlled to raise the right side of the housing than the left side. Unmanned aerial vehicle for detecting cracks in structures. The method of claim 3,
The main controller may control the thrust of the front rotor of the vehicle disposed on the left side of the housing more strongly than the rotor of the rear, when the flight attitude control of the front of the housing is turned in the left direction, Unmanned aerial vehicle for detecting cracks of the structure, characterized in that the thrust control stronger than the rotary blades located in front of the rotor positioned in the rear of the vehicle disposed on the right side of the housing. The method of claim 1,
And a space keeping member coupled to the front of the housing so that the structure positioned in front of the housing and the housing can be spaced at a spaced distance apart from each other. The method of claim 8,
The space keeping member is formed to extend in a long direction from the housing forward, unmanned aerial vehicle for detecting cracks of the structure, characterized in that coupled to the wheel wheels rolling in the left and right directions to the extended end. The method of claim 8,
The main control unit detects a crack of the structure, characterized in that for controlling the flight direction toward the structure when the image data of the structure is acquired.

Images