Wind turbine inspection method, unmanned aircraft, and program

Abstract

To provide an inspection method capable of automatically optimizing an inspection flight path of an unmanned aircraft when inspecting a wind turbine.SOLUTION: In a wind turbine inspection method, an unmanned aircraft flies from a takeoff point to a first spatial position from which an overhead photograph of a wind turbine to be inspected can be taken; the unmanned aircraft flies around the wind turbine from the first spatial position, and captures images of the wind turbine; the unmanned aircraft processes image data of the captured images of the wind turbine; the unmanned aircraft identifies the attitude of the wind turbine using the result of the image data processing; the unmanned aircraft creates an inspection flight path for the wind turbine based on the identified attitude of the wind turbine; the unmanned aircraft inspects the wind turbine while flying along the inspection flight path; and the unmanned aircraft lands at a landing point.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] Embodiments of the present invention relate to a method for inspecting a wind turbine, an unmanned aerial vehicle, and a program.

Background Art

[0002] A method for inspecting a wind turbine using an image captured by an unmanned aerial vehicle such as a drone is known. In the conventional inspection method, an operator turns the nacelle to a predetermined orientation at the site and stops the blade at a predetermined rotation angle. Then, the unmanned aerial vehicle flies along a pre-set inspection flight path.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a wind turbine, the orientation of the nacelle is indefinite depending on the wind direction, and the rotation angle of the blade is indefinite depending on the situation where the blade is stopped. For example, in a floating offshore wind turbine, the wind turbine floats and moves on the sea due to the wind direction, tidal current, and waves. Therefore, the attitude of the wind turbine at the time of inspection is indefinite. For this reason, in the conventional inspection method of flying an unmanned aerial vehicle along a pre-determined inspection flight path, the inspection flight path may not be appropriate, and as a result, it is assumed that an image necessary for inspecting the wind turbine cannot be obtained, or the unmanned aerial vehicle may collide with the wind turbine. In these cases, it takes time for the operator to adjust the orientation of the nacelle of the wind turbine, the rotation angle of the blade, or the inspection flight path of the unmanned aerial vehicle at the site.

[0005] The problem to be solved by the present invention is to provide an inspection method, an unmanned aerial vehicle, and a program capable of automatically optimizing the inspection flight path of the unmanned aerial vehicle at the time of inspecting a wind turbine.

Means for Solving the Problem

[0006] In a method for inspecting a wind turbine according to an embodiment, a drone flies from a takeoff point to a first spatial position where the wind turbine to be inspected can be photographed from above. The drone flies in a circular path around the wind turbine from the first spatial position while photographing the wind turbine. The drone processes the image data of the photographed wind turbine, identifies the attitude of the wind turbine using the processing result of the image data, creates an inspection flight path of the wind turbine based on the identified attitude of the wind turbine, inspects the wind turbine while flying along the inspection flight path, and lands at a landing point.

Advantages of the Invention

[0007] According to this embodiment, it is possible to automatically optimize the inspection flight path of the drone during the inspection of the wind turbine.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The following embodiments do not limit the present invention.

[0010] FIG. 1 is a block diagram showing a configuration example of an unmanned aerial vehicle according to an embodiment. The unmanned aerial vehicle 10 shown in FIG. 1 is, for example, a multi-copter type drone, and includes a positioning unit 11, a distance measuring unit 12, a wide-angle camera 13, an inspection camera 14, a drive unit 15, a processing unit 16, a flight controller 17, and a communication unit 18. In this embodiment, the use of the unmanned aerial vehicle 10 is to inspect wind turbines. The wind turbines to be inspected include floating offshore wind turbines, fixed-bottom offshore wind turbines, and onshore wind turbines.

[0011] The positioning unit 11 receives, for example, position data transmitted from GPS (Global Positioning System) or GNSS (Satellite positioning, navigation and timing system). Further, the positioning unit 11 periodically transmits the received position data to the flight controller 17.

[0012] The distance measuring unit 12 measures, for example, the distance to an object such as a blade or a tower in the wind turbine to be inspected. The distance measuring unit 12 is, for example, composed of a three-dimensional LiDAR (Light Detection And Ranging). The three-dimensional LiDAR measures the time from when a laser beam is emitted from a light source until the laser light reflected by an object is received by a light receiving unit, and calculates the distance to the object based on the measured time. The distance measuring unit 12 periodically transmits data indicating the calculation result to the processing unit 16. The distance measuring unit 12 is used for the unmanned aerial vehicle 10 to avoid collision with the wind turbine and maintain a certain distance. However, in this embodiment, since the flight path of the unmanned aerial vehicle 10 is set so as not to collide with the wind turbine as described later, the distance measuring unit 12 may not be provided on the unmanned aerial vehicle 10.

[0013] The wide-angle camera 13 can photograph the windmill to be inspected over a wider range than the inspection camera 14. The wide-angle camera 13 periodically transmits image data indicating the photographed image to the processing unit 16. The photographed image of the wide-angle camera 13 is used to create the inspection flight path of the unmanned aircraft 10 during the inspection of the windmill.

[0014] The number of pixels of the inspection camera 14 is larger than that of the wide-angle camera 13. Therefore, the inspection camera 14 can photograph the windmill to be inspected with higher image quality than the wide-angle camera 13. The inspection camera 14 periodically transmits image data indicating the photographed image to the processing unit 16. The photographed image of the inspection camera 14 is used to photograph the windmill to be inspected when the unmanned aircraft 10 flies along the inspection flight path. In this embodiment, the wide-angle camera 13 is used for creating the inspection flight path, and the inspection camera 14 is used for inspecting the windmill. However, these two cameras can also be replaced by a single camera having the functions of each camera.

[0015] The drive unit 15 is composed of a plurality of parts necessary for driving the unmanned aircraft 10. The plurality of parts include, for example, a storage battery, a motor that rotates with the electric power supplied from the storage battery, an ESC (Electric Speed Controller) that controls the rotation speed of the motor, a propeller connected to the motor, and the like.

[0016] The processing unit 16 processes various data acquired from the distance measuring unit 12, the wide-angle camera 13, and the inspection camera 14. The processing unit 16 is composed of, for example, a computer that performs data processing according to a predetermined program.

[0017] The flight controller 17 identifies the current position of the unmanned aircraft 10 based on the position data from the positioning unit 11. Further, the flight controller 17 controls the drive unit 15 so that the unmanned aircraft 10 flies based on an instruction from the processing unit 16.

[0018] The communication unit 18 performs wireless communication with the port 20. The port 20 is installed at the takeoff and landing point of the unmanned aircraft 10. In this embodiment, the takeoff and landing point of the unmanned aircraft 10 is a windmill. However, the takeoff and landing point of the unmanned aircraft 10 is not limited to a windmill and may be, for example, a land-based point. Also, the takeoff point and the landing point do not have to be the same point and may be different points. For example, one of the takeoff point and the landing point may be a windmill and the other may be on land. Further, when a plurality of windmills are installed, each of the takeoff point and the landing point may be an adjacent or nearby windmill.

[0019] As shown in FIG. 1, the port 20 includes a processing unit 21, a communication unit 22, and a power supply unit 23. The processing unit 21 performs various data processes related to, for example, charging and flight of the unmanned aircraft 10. The communication unit 22 performs wireless communication with the communication unit 18 of the unmanned aircraft 10. Further, the communication unit 22 can also communicate with the terminal 30 used by the operator. The power supply unit 23 transmits power for charging the unmanned aircraft 10.

[0020] Next, a method for inspecting a windmill using the unmanned aircraft 10 according to this embodiment will be described.

[0021] FIG. 2 is a flowchart showing the procedure of the windmill inspection method. In this flowchart, first, the user operates the terminal 30 to instruct inspection of the windmill. The inspection instruction is received by the communication unit 18 of the unmanned aircraft 10 via the port 20. Note that the inspection instruction may be received directly from the terminal 30 to the communication unit 18. The received inspection instruction is transferred to the processing unit 16. Thereby, the processing unit 16 instructs the flight controller 17 to start takeoff of the unmanned aircraft 10 from the takeoff point P0, and the flight controller 17 controls the drive unit 15 (step S101).

[0022] Next, the flight controller 17 controls the drive unit 15 according to the instruction of the processing unit 16 and moves the unmanned aircraft 10 from the takeoff point P0 to the first spatial position P1 where the windmill 40 can be photographed from above (step S102). Here, with reference to FIG. 3, step S102 will be described.

[0023] FIG. 3 is a schematic diagram showing a state in which the unmanned aircraft 10 flies from the takeoff point P0 to the first spatial position P1. The takeoff point P0 is preset around the lower end of the tower 41 of the wind turbine 40. However, the takeoff point P0 is not limited to the lower end of the tower 41. For example, it may be above the nacelle 44, inside the nacelle 44, or on the floating structure when the wind turbine 40 is a floating offshore wind turbine. The first spatial position P1 is preset at a distance at which the wind turbine 40 can be photographed from above by the wide-angle camera 13. As shown in FIG. 3, it is desirable that the first spatial position P1 is a spatial position where the entire wind turbine 40 to be inspected can be photographed. However, the first spatial position P1 may be a spatial position where the attitude can be estimated by capturing features such as the orientation of the nacelle 44 and the rotation angle of the blade 43 in the wind turbine 40, rather than photographing the entire wind turbine 40, as long as it can be photographed from above to some extent. In FIG. 3, the first spatial position P1 is in front of the wind turbine 40, but it does not have to be in front.

[0024] In step S102, the unmanned aircraft 10 flies from the takeoff point P0 to the first spatial position P1 along the flight path f1 shown in FIG. 3. The flight path f1 is preset as a flight path that does not contact the wind turbine 40 regardless of the orientation and angle of the wind turbine 40.

[0025] In the wind turbine 40, the nacelle 44 is supported by the tower 41. A hub 42 is installed on the front surface of the nacelle 44. Blades 43 are attached to the hub 42. In the present embodiment, a contact space 50 that may contact the wind turbine 40 is preset regardless of the orientation of the nacelle 44 and the rotation angle (azimuth angle) of the blade 43. The flight path f1 is preset in the space outside the contact space 50. As shown in FIG. 3, the unmanned aircraft 10 ascends vertically from the takeoff point P0, and then ascends obliquely upward toward the first spatial position P1 while keeping away from the wind turbine 40 outside the contact space 50.

[0026] When the unmanned aircraft 10 flies to the first spatial position P1 along the flight path f1, next, the processing unit 16 instructs the flight controller 17 so that the unmanned aircraft 10 makes a turning flight from the first spatial position P1 to the windmill 40, and the flight controller 17 controls the drive unit 15 (step S103). Also, in step S103, during the turning flight of the unmanned aircraft 10, the wide-angle camera 13 periodically photographs the windmill 40.

[0027] Note that in step S103, the distance measuring unit 12 may measure the distance from the unmanned aircraft 10 to the windmill 40. When the measured distance of the distance measuring unit 12 becomes shorter than the reference value, the possibility that the unmanned aircraft 10 collides with the windmill 40 increases. Therefore, in such a case, the processing unit 16 instructs the flight controller 17 so that the unmanned aircraft 10 turns away from the windmill 40 and makes a turning flight, and the flight controller 17 controls the drive unit 15. Thereby, a collision between the unmanned aircraft 10 and the windmill 40 can be avoided.

[0028] FIG. 4 is a schematic diagram showing a state in which the unmanned aircraft 10 makes a turning flight around the windmill 40. In step S103, the unmanned aircraft 10 makes a turning flight around the windmill 40 along a preset photographing flight path f2 from the first spatial position P1.

[0029] During the turning flight of the unmanned aircraft 10, the wide-angle camera 13 periodically photographs the windmill 40, thereby generating image data of the windmill 40 at different photographing locations. The generated image data is transmitted to the processing unit 16.

[0030] The processing unit 16 processes the image data of the windmill 40 (step S104). Here, the image processing of the processing unit 16 will be described with reference to FIG. 5.

[0031] FIG. 5 is a schematic diagram for explaining the image processing of the processing unit 16. In step S104, the processing unit 16 extracts a feature point G related to the posture of the windmill 40 for each piece of image data captured by the wide-angle camera 13 as shown in FIG. 5. The feature point G corresponds to a plurality of characteristic locations of the windmill 40, such as the tip of each blade 43, the hub 42 that is the rotation center location of the windmill 40, and the corner of the nacelle 44.

[0032] Based on the difference (displacement amount) between the windmill model obtained by connecting the extracted plurality of feature points G with lines as shown in FIG. 5 and the standard windmill model stored in the processing unit 16 in advance, the posture of the windmill 40 such as the orientation of the nacelle 44 and the rotation angle of the blade can be roughly grasped. However, the number of extractable feature points G varies depending on the position where the wide-angle camera 13 photographs the windmill 40. Therefore, if the number of extracted feature points G is small, the posture of the windmill 40 cannot be grasped. Also, there is a possibility of grasping an incorrect posture. Further, even if there is no problem with the number of feature points G, there is a possibility of grasping an incorrect posture.

[0033] Therefore, in the present embodiment, the processing unit 16 counts the case where the posture of the windmill 40 can be grasped from the image data of the wide-angle camera 13 (step S105). When the number of times the posture of the windmill 40 can be grasped is less than or equal to the threshold value (step S105: NO), since the posture of the windmill 40 cannot be correctly grasped, the processing unit 16 continues the turning flight of the unmanned aircraft 10 and the photographing operation of the wide-angle camera 13. As a result, a plurality of grasped postures are accumulated.

[0034] For example, when the processing unit 16 processes the image data captured by the wide-angle camera 13 while the unmanned aircraft 10 is flying at the second spatial position P2 (see FIG. 4), if the number of counts when the attitude can be grasped is greater than a threshold value (step S105: YES), the attitude of the windmill 40 can be grasped more accurately from the grasped attitudes. Therefore, the processing unit 16 stops the turning flight of the unmanned aircraft 10 and the shooting operation of the wide-angle camera 13 at the second spatial position P2. In this way, it is possible to more accurately grasp the attitude of the windmill 40 in a short time. Thereby, the inspection preparation time of the windmill 40 can be shortened. When the windmill 40 is a fixed-type offshore windmill or an onshore windmill, the processing unit 16 processes the image data of the wide-angle camera 13 to specify only the attitude of the windmill 40. On the other hand, when the windmill 40 is a floating-type offshore windmill, the windmill 40 can move offshore due to wind direction, tidal current, and waves. Therefore, in this case, the processing unit 16 processes the image data of the wide-angle camera 13 to specify not only the attitude of the windmill 40 but also the position offshore. For example, similar to the attitude, the processing unit 16 calculates the difference between the windmill model created based on the image data captured by the wide-angle camera 13 and the pre-stored standard windmill model with respect to the standard windmill model, and calculates the relative position with the unmanned aircraft 10 in combination with the data of the positioning unit 11, thereby specifying the position of the windmill 40.

[0035] Next, the processing unit 16 creates an inspection flight route for the unmanned aircraft 10 (step S106). Here, step S106 will be described with reference to FIGS. 6 and 7.

[0036] FIG. 6 is a schematic diagram for explaining the standard inspection flight route of the unmanned aircraft 10. The processing unit 16 stores in advance the standard inspection flight route f0 shown in FIG. 6. The standard inspection flight route f0 is set for the windmill 40 in a standard state in which one of the three blades 43 extends along the vertical direction and the nacelle 44 faces a predetermined azimuth.

[0037] FIG. 7 is a schematic diagram for explaining the inspection flight path of the unmanned aircraft 10. In step S106, the processing unit 16 first identifies the rotation angle of the blade 43 and the orientation of the nacelle 44 based on the result of the image processing. Subsequently, the processing unit 16 calculates the difference between the identified rotation angle of the blade 43 and the orientation of the nacelle 44 and the rotation angle of the blade 43 and the orientation of the nacelle 44 in the wind turbine 40 in the standard state shown in FIG. 6. Finally, the processing unit 16 creates an inspection flight path f3 (see FIG. 7) based on the calculated difference.

[0038] For example, assume that the rotation angle of the blade 43 in the standard state is 0° and the rotation angle based on the result of the image processing is 30°. In this case, the processing unit 16 generates the inspection flight path f3 by rotating the standard inspection flight path f0 by 30° in the rotation direction of the blade 43.

[0039] When the inspection flight path f3 is generated, next, the processing unit 16 instructs the flight controller 17 to fly the unmanned aircraft 10 to the third spatial position P3, which is the inspection start position on the inspection flight path f3, and the flight controller 17 controls the drive unit 15 (step S107). In step S107, the third spatial position P3 is the spatial position closest to the second spatial position P2, which is the end position of the turning flight, on the inspection flight path f3. That is, the processing unit 16 instructs the flight controller 17 to fly the unmanned aircraft 10 from the second spatial position P2 to the third spatial position P3 at the shortest distance, and the flight controller controls the drive unit 15.

[0040] When the unmanned aircraft 10 reaches the third spatial position P3, next, the processing unit 16 instructs the flight controller 17 to fly the unmanned aircraft 10 along the inspection flight path f3, and the flight controller 17 controls the drive unit 15 (step S108). Also, in step S108, during the flight of the unmanned aircraft 10, the inspection camera 14 periodically photographs the wind turbine 40. As a result, inspection image data of the wind turbine 40 taken from the inspection flight path f3 is generated. The generated inspection image data is transmitted to the processing unit 16.

[0041] Note that in step S108, the distance measuring unit 12 may measure the distance from the unmanned aircraft 10 to the windmill 40. When the measured distance of the distance measuring unit 12 becomes shorter than the reference value, the processing unit 16 issues an instruction to the flight controller 17, and the flight controller 17 controls the drive unit 15 so that the unmanned aircraft 10 flies away from the windmill 40. As a result, the inspection flight path f3 is corrected, so that a collision between the unmanned aircraft 10 and the windmill 40 can be avoided.

[0042] The processing unit 16 processes the inspection image data of the windmill 40 (step S109). In step S109, for example, the processing unit 16 extracts a characteristic region indicating the damaged portion of the blade 43 from the inspection image data.

[0043] When the flight along the inspection flight path f3 ends, finally, the processing unit 16 issues an instruction to the flight controller 17 so that the unmanned aircraft 10 lands at the landing point, and the flight controller 17 controls the drive unit 15 (step S110). In the present embodiment, the landing point is the port 20. When the unmanned aircraft 10 lands at the port 20, the processing unit 16 transmits various data such as the inspection image data obtained by image processing to the communication unit 22 of the port 20 via the communication unit 18. The communication unit 22 transfers the received data to the terminal 30 (step S111). Thereby, the inspection operation of the windmill 40 by the unmanned aircraft 10 ends. Thereafter, the power supply unit 23 of the port 20 charges the unmanned aircraft 10. Note that the inspection image data processing in step S109 described above may be performed by another device such as the terminal 30 after step S111 instead of the processing unit 16.

[0044] According to the above-described embodiment, based on the image data captured by the unmanned aircraft 10 while orbiting and flying around the wind turbine 40, the processing unit 16 identifies the posture of the wind turbine 40, such as the rotation angle of the blade 43 and the orientation of the nacelle 44. Further, the processing unit 16 creates an inspection flight path f3 based on the identified posture of the wind turbine 40. Therefore, even if the posture of the wind turbine 40 is unknown before inspection, the inspection flight path f3 is automatically formed according to the posture of the wind turbine 40. Thus, it is possible to automatically optimize the inspection flight path f3 of the unmanned aircraft 10 during the inspection of the wind turbine 40.

[0045] In this embodiment, when the processing unit 16 acquires image data capable of grasping the posture of the wind turbine 40 up to a threshold value, the flight of the unmanned aircraft 10 is stopped. Subsequently, the processing unit 16 statistically analyzes the distribution of the rotation angle of the blade 43 and the orientation of the nacelle 44, which are indicators representing the postures of the plurality of identified wind turbines 40. The processing unit 16 creates the inspection flight path f3 using the results of the statistical analysis. For example, the processing unit 16 creates the inspection flight path f3 using the average value of the distribution as the result of the statistical analysis.

[0046] Depending on the positional relationship between the wind turbine 40 and the unmanned aircraft 10, the background environment such as a forest or sea of the wind turbine 40, and the positional relationship with the sun (such as backlighting), the processing unit 16 may create an incorrect inspection flight path f3. Therefore, when performing the statistical analysis, the processing unit 16 may exclude values that deviate significantly from the above distribution. Thereby, the inspection flight path f3 can be created with higher accuracy. Here, with reference to FIG. 8, an example of a case where the processing unit 16 statistically obtains the orientation and the blade angle will be described.

[0047] FIG. 8 is a diagram for explaining an example in which the processing unit 16 statistically obtains the azimuth and the blade rating. In FIG. 8, the azimuth and the blade angle obtained by performing image processing eight times are plotted on a graph. Referring to FIG. 8, in the seven image processings, a distribution 45 of substantially the same numerical values is formed, whereas the distribution 46 has numerical values clearly different from those of the distribution 45. Therefore, the processing unit 16 excludes the values belonging to the distribution 46 with a high possibility of error. Further, the processing unit 16 calculates the average value of the values belonging to the distribution 45 as the final detected azimuth and the blade angle. Thereby, a highly accurate detected value can be calculated. When the wind turbine 40 is a floating offshore wind turbine, the processing unit 16 excludes, regarding the position of the wind turbine 40, the positions belonging to the distribution with a high possibility of error, similarly to the above-described azimuth and blade angle. Then, the processing unit 16 specifies the average of the positions belonging to the remaining distribution as the final position. Thereby, the position of the wind turbine 40 with high accuracy can be specified.

[0048] Also, in the present embodiment, since the inspection targets of the wind turbine 40 are mainly the blade 43 and the nacelle 44, the standard inspection flight path f0 and the inspection flight path f3 are set so that the blade 43 and the nacelle 44 can be imaged by the inspection camera 14. However, the inspection targets may include, for example, the tower 41 in addition to the blade 43 and the nacelle 44. In this case, the standard inspection flight path f0 and the inspection flight path f3 are set so that not only the blade 43 and the nacelle 44 but also the tower 41 can be photographed.

[0049] As described above, several embodiments have been described, but these embodiments are presented only as examples and are not intended to limit the scope of the invention. The novel system described in this specification can be implemented in various other forms. Also, various omissions, substitutions, and changes can be made to the form of the system described in this specification without departing from the gist of the invention. The appended claims and equivalents thereof are intended to include such forms and modifications within the scope and gist of the invention.

Description of Reference Numerals

[0050] 10: Unmanned aircraft 13: Wide-angle camera 16: Processing unit 40: Windmill 41: Tower 43: Blade 44: Nacelle

Claims

1. The unmanned aircraft flies from the takeoff point to a first spatial position where it can take an aerial photograph of the wind turbine to be inspected, the unmanned aircraft orbits the wind turbine from the first spatial position and takes an image of the wind turbine, the unmanned aircraft processes the image data of the image taken of the wind turbine, the unmanned aircraft identifies the attitude of the wind turbine using the processing result of the image data, the unmanned aircraft creates an inspection flight path of the wind turbine based on the identified attitude of the wind turbine, the unmanned aircraft inspects the wind turbine while flying along the inspection flight path, the unmanned aircraft lands at the landing point, A method for inspecting a wind turbine.

2. The flight path from the takeoff point to the first spatial position is preset in a space outside the contact space that may come into contact with the wind turbine. The method for inspecting a wind turbine according to claim 1.

3. The unmanned aircraft ascends directly above the takeoff point, and then ascends obliquely upward toward the first spatial position while keeping away from the wind turbine outside the contact space. The method for inspecting a wind turbine according to claim 2.

4. The unmanned aircraft extracts feature points related to the attitude of the wind turbine from the image data, The unmanned aircraft orbits the wind turbine until the number of the feature points is more than a threshold value. The method for inspecting a wind turbine according to claim 1.

5. The unmanned aircraft acquires a plurality of image data during the orbiting flight of the wind turbine, The unmanned aircraft identifies the attitude of the wind turbine from the plurality of image data. The method for inspecting a wind turbine according to claim 1.

6. When the orbiting flight of the wind turbine ends, the unmanned aircraft flies from the second spatial position where the orbiting flight ends to a third spatial position where the inspection in the inspection flight path starts at the shortest distance. The method for inspecting a wind turbine according to claim 1.

7. The landing point is different from the takeoff point. The method for inspecting a wind turbine according to claim 1.

8. The inspection flight path is a flight path along which the unmanned aircraft can photograph the blades of the wind turbine. The method for inspecting a wind turbine according to claim 1.

9. The first spatial position is a position where the entire wind turbine can be photographed. The method for inspecting a wind turbine according to claim 1.

10. The first spatial position is not to photograph the entire wind turbine, but is a position where it can take an aerial photograph to capture the characteristics of the wind turbine and estimate the attitude to a certain extent. The method for inspecting a wind turbine according to claim 1.

11. The method for inspecting a wind turbine according to claim 1, wherein the unmanned aircraft also identifies the position of the wind turbine using the processing result of the image data.

12. The inspection flight path according to claim 8, wherein the inspection flight path is a flight path along which the unmanned aircraft can photograph the nacelle and tower of the wind turbine.

13. An unmanned aircraft for inspecting a wind turbine, a camera that takes an image of the wind turbine when the unmanned aircraft makes a turning flight of the wind turbine from a first spatial position where the wind turbine can be photographed from above; a processing unit that processes the image data captured by the camera, identifies the attitude of the wind turbine using the processing result of the image data, and creates an inspection flight path of the wind turbine based on the identified attitude of the wind turbine; An unmanned aircraft comprising the above.

14. Processing of image data obtained by photographing a wind turbine when an unmanned aircraft makes a turning flight of the wind turbine from a first spatial position where the wind turbine can be photographed from above, processing of identifying the attitude of the wind turbine using the processing result of the image data, processing of creating an inspection flight path of the wind turbine based on the identified attitude of the wind turbine, A program for causing a computer to execute the above.

Images