KR102504743B1 - Position correction device and correction method of inspection drone based on the model of the facility - Google Patents

Abstract

The present invention relates to an autonomous flight of a drone inspecting a facility and, more specifically, to an apparatus and a method for correcting the position of an inspection drone based on modeling of a facility when a positioning algorithm of a drone does not operate smoothly due to a GPS shadow area during the inspection of the facility. To this end, provided is a position correction apparatus for an inspection drone based on modeling of a facility which includes a drone (100) and a drone control unit (200) which are wirelessly communicable with each other The drone control unit (200) includes: (i-1) a three-dimensional modeling unit (230) storing modeling (10a) data of the facility (10) to be inspected; (i-2) a path generation unit (260) that generates a flight path (30) of the drone (100) with respect to the modeling (10a); (i-3) a keyframe extraction unit (280) that extracts a frame image (55) for a camera pose (60) at a random location on the flight path (30); (i-4) a feature extraction unit (290) that extracts features of the facility (10) from the frame image (55); and (i-5) a control communication unit (220) that communicates wirelessly with the drone (100). The drone (100) includes: (ii-1) a wireless communication unit (140) that communicates wirelessly with the control communication unit (220); (ii-2) a path storage unit (190) storing the flight path (30); and (ii-3) a drone control unit (110) that corrects the predicted position (102) of the drone (100) based on the features received from the feature extraction unit (290).

Description

Position correction device and correction method of inspection drone based on the model of the facility {Position correction device and correction method of inspection drone based on the model of the facility}

The present invention relates to the autonomous flight of a drone inspecting a facility, and more particularly, a device for correcting the position of a drone inspecting a facility based on facility modeling when a positioning algorithm does not operate smoothly due to a GPS shaded area. and methods.

In general, for train tracks, bridges, etc., girders, substructures, abutments, etc. are photographed using a drone (or an unmanned aerial vehicle such as a multicopter) and then cracks, deformations, tilts, etc. are inspected. Such photographing and inspection are performed on a regular basis, and since time and cost are reduced, it has recently been greatly spotlighted and widely performed.

That is, after moving to the vicinity of the facility with a vehicle equipped with a drone, the drone is launched. The operator controls the drone from inside the vehicle and monitors the filming and inspection process.

However, during inspection of the lower structure of a railway bridge or a girder, there is a problem in that a reception rate is lowered due to a shadow area of the satellite navigation information (GPS signal) of the drone due to the facility. Due to such deterioration in the reception rate, it is difficult for the drone to accurately recognize the current location, and there have been cases where the drone collides with a facility or malfunctions. Conventionally, when this problem occurs, the operator has to turn off the automatic flight and manually adjust the drone. Therefore, the operator's flight control skill was required, and there were disadvantages in that automatic inspection and manual inspection were mixed, causing confusion and increasing inspection time.

1. Republic of Korea Patent Registration No. 10-1782039 (River and river facility surveying device and location measurement method using drone), 2. Republic of Korea Patent Registration No. 10-1881121 (drones measuring distances and drone control methods), 3. Republic of Korea Patent Publication No. 10-2019-0092789 (Drone position measurement method and drone position correction system using the same).

Therefore, the present invention has been devised to solve the above problems, and the problem to be solved by the present invention is that the reception of satellite navigation information is not smooth due to the shadow of the facility, and as a result, the positioning of the drone is inaccurate. To provide a position correction device and correction method for inspection drones based on modeling of a facility capable of correcting the location of a drone based on 3D modeling.

However, the technical problems to be achieved in the present invention are not limited to the above-mentioned technical problems, and other technical problems not mentioned will be clear to those skilled in the art from the description below. You will be able to understand.

In order to achieve the above technical task, in the drone 100 and the drone control unit 200 capable of wireless communication with each other, (i-1) a 3D modeling unit in which modeling (10a) data of the facility 10 to be inspected is stored (230); (i-2) a path generator 260 generating a flight path 30 of the drone 100 for the modeling 10a; (i-3) a key frame extractor 280 for extracting a frame image 55 for the camera pose 60 at an arbitrary position on the flight path 30; (i-4) a feature extraction unit 290 extracting features of the facility 10 from the frame image 55; and (i-5) a control communication unit 220 that wirelessly communicates with the drone 100; a drone control unit 200 including; and (ii-1) a wireless communication unit 140 that wirelessly communicates with the control communication unit 220; (ii-2) a path storage unit 190 in which the flight path 30 is stored; and (ii-3) a drone control unit 110 that corrects the predicted position 102 of the drone 100 based on the features received from the feature extraction unit 290. This can be achieved by a device for correcting the position of an inspection drone based on the modeling of a characteristic facility.

In addition, the feature extraction unit 290 includes a camera pose extraction unit 292 that extracts information about the camera pose 60; an imaging plane extraction unit 294 for extracting the imaging planes 80 and 82 from the frame image; and an outline extraction unit 296 for extracting an outline 84 from the photographing planes 80 and 82 .

In addition, the drone 100 further includes at least one of a camera 120, a GPS 130, and an IMU 160 to calculate the predicted location 102.

In addition, the drone controller 110 further includes a depth sensor 150 to correct the predicted position 102 .

In addition, in order to correct the predicted position 102, the drone control unit 110 includes relative distances (Dx, Dy, Dz) from the predicted position 102 to the corrected position 104; and the nose rotation angle (θyaw) of the drone 100.

Another object of the present invention as described above is another category, and in the correction method using the above-described position correction device, the drone control unit 110 receives the flight path 30 from the drone control unit 200, and the path storage unit ( 190) pre-step (S90) to store; The step of the drone controller 110 calculating the predicted position 102 based on the output signal of at least one of the camera 120, the GPS 130, and the IMU 160, and extracting the pose of the camera 120 ( S100); When the pose of the camera 120 is upward shooting, the Z-axis relative distance from the predicted position 102 to the corrected position 104 based on the modeling features received from the feature extraction unit 290 of the drone controller 200 Correcting (Dz) (S130); And correcting the X-axis relative distance (Dx), Y-axis relative distance (Dy) and nose rotation angle (θyaw) of the drone 100 from the predicted position 102 to the corrected position 104 based on the characteristics of the modeling Step (S140); including, when the camera 120 pose is forward shooting, the X-axis relative distance (Dx) and nose rotation angle from the predicted position 102 to the corrected position 104 based on the characteristics of the modeling Correcting (θyaw) (S160); And correcting the Y-axis relative distance (Dy) and the Z-axis relative distance (Dz) from the predicted position 102 to the corrected position 104 based on the characteristics of modeling (S170); including, a drone control unit ( 110) flying to the correction position 104 (S150); it can also be achieved by a method for correcting the position of an inspection drone based on modeling of a facility, characterized in that it includes.

Also, in the correction steps S130 and S160, the modeling features include data of the imaging planes 80 and 82 extracted from the modeling 10a by the imaging plane extraction unit 294 of the feature extraction unit 290.

Further, in the correction steps S140 and S170, the modeling features include outline 84 data extracted from the modeling 10a by the outline extractor 296 of the feature extractor 290.

Also, when the pose of the camera 120 is upward shooting, the drone 100 flies below the facility 10 .

Also, in the calibration steps (S130, S140, S160, and S170), the drone control unit 110 may perform correction based on the output signal of the depth sensor 150.

Also, modeling is 3-D modeling of facilities.

According to one embodiment of the present invention, even if the reception rate of satellite navigation information (GPS signal) decreases during facility inspection or flies in a shaded area, the drone corrects the location based on facility modeling to calculate an accurate location.

Through this, facilities can be inspected smoothly and quickly, and the operator can focus more on inspecting or determining the facilities rather than flying the drone.

However, the effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

The following drawings attached to this specification illustrate preferred embodiments of the present invention, and together with the detailed description of the present invention serve to further understand the technical idea of the present invention, the present invention is the details described in such drawings should not be construed as limited to
1 is an operating state diagram showing a state in which a drone inspects a facility according to the present invention;
2 is a perspective view showing some of the 3D modeling 50 of the facility used in the present invention;
3 is a plan view conceptually showing a predicted position 102 and a corrected position 104 of a drone 100 in position correction of an inspection drone based on facility modeling according to the present invention;
4 is a schematic block diagram of an inspection drone 100 and a drone control unit 200 based on facility modeling according to an embodiment of the present invention;
5 is a detailed block diagram of the feature extraction unit 290 in FIG. 4;
6 is a front view showing an example in which the capturing plane extraction unit 294 and the capturing plane outline extracting unit 296 extract the capturing plane and outline from the frame image 55;
7 is a flowchart illustrating a method for calibrating the position of an inspection drone based on facility modeling according to an embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily carry out the present invention. However, since the description of the present invention is only an embodiment for structural or functional description, the scope of the present invention should not be construed as being limited by the embodiments described in the text. That is, since the embodiment can be changed in various ways and can have various forms, it should be understood that the scope of the present invention includes equivalents capable of realizing the technical idea. In addition, since the object or effect presented in the present invention does not mean that a specific embodiment should include all of them or only such effects, the scope of the present invention should not be construed as being limited thereto.

The meaning of terms described in the present invention should be understood as follows.

Terms such as "first" and "second" are used to distinguish one component from another, and the scope of rights should not be limited by these terms. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element. It should be understood that when an element is referred to as “connected” to another element, it may be directly connected to the other element, but other elements may exist in the middle. On the other hand, when an element is referred to as being “directly connected” to another element, it should be understood that no intervening elements exist. Meanwhile, other expressions describing the relationship between components, such as “between” and “immediately between” or “adjacent to” and “directly adjacent to” should be interpreted similarly.

Singular expressions should be understood to include plural expressions unless the context clearly dictates otherwise, and terms such as “comprise” or “having” refer to a described feature, number, step, operation, component, part, or It should be understood that it is intended to indicate that a combination exists, and does not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs, unless defined otherwise. Terms defined in commonly used dictionaries should be interpreted as consistent with meanings in the context of the related art, and cannot be interpreted as having ideal or excessively formal meanings unless explicitly defined in the present invention.

Configuration of the embodiment

Hereinafter, the configuration of a preferred embodiment will be described in detail with reference to the accompanying drawings. 1 is an operational state diagram showing a state in which a drone inspects a facility according to the present invention. As shown in FIG. 1 , the facilities 10 may be train tracks, bridges, piers 20, girders, substructures, abutments, and the like. The drone 100 is a small unmanned aerial vehicle, and may be an unmanned helicopter or a multicopter. These drones 50 are moved, collected, controlled, and communicated by the vehicle-type drone control unit 200 parked around the facility 10 . As shown in FIG. 1, the drone 100 photographs the side and bottom of the facility 10 while moving along a predetermined flight path 30, and cracks, tilts, deformations, foreign substances, peeling, Check for leaks, etc.

Figure 2 is a perspective view showing a part of the 3D modeling 50 of the facility used in the present invention. As shown in FIG. 2 , in the 3D modeling unit 230 of the drone control unit 200, the facility modeling 10a of the corresponding facility 10 to be inspected is stored in 3D form.

In the 3D modeling 50, the drone position 70 is movable along the flight path 30, and the camera pose 60 represents the shooting direction of the camera 120 of the drone 100. In the 3D modeling unit 230, coordinate data for the drone position 70 and direction data for the camera pose 60 are generated.

3 is a plan view conceptually showing a predicted position 102 and a corrected position 104 of a drone 100 in position correction of an inspection drone based on facility modeling according to the present invention. As shown in FIG. 3, the predicted position 102 is a position measured by the drone 100 based on output signals of the camera 120, the GPS 130, and the IMU 160. This predicted position 102 differs from the actual position of the drone 100 due to a drop in GPS reception or a shadow area. This difference in location may gradually widen as the inspection progresses.

The correction position 104 is the position of the drone 100 corrected according to the correction method of the present invention, and more closely matches the actual position of the drone 100. The relative distances (Dx, Dy, Dz) are relative distances from the predicted position 102 to the corrected position 104 in the space of the X-Y-Z axes.

The nose rotation angle (θyaw) of the drone 100 is the direction in which the front end (nose) of the drone 100 at the predicted position 102 faces on the X-Y plane and the front end of the drone 100 at the corrected position 104 faces. is the angle between the directions. Since the drone 100 maintains a horizontal state by itself during autonomous flight, correction of the nose rotation angle θyaw, which is a rotation in the Z-axis direction, may be sufficient. When the corrected position is calculated, the drone 100 flies to the corrected position 104 .

4 is a schematic block diagram of an inspection drone 100 and a drone control unit 200 based on facility modeling according to an embodiment of the present invention. As shown in FIG. 4, the present invention is largely composed of a drone 100 and a drone control unit 200.

The drone control unit 110 of the drone 100 may be a microprocessor or a CPU, and programs necessary for inspection, filming, and flight are loaded and executed.

The camera 120 photographs the facility 10, and the drone control unit 110 processes the photographed image to detect cracks, tilts, deformations, foreign substances, peeling, water leakage, and the like. Optionally, the captured image may be transmitted to the drone control unit 200 for detection. In particular, pose (rotation information and zoom information in the X, Y, and Z axis directions) data of the camera 120 is transmitted to the drone controller 110 .

The GPS 130 receives satellite navigation information and outputs location information.

The wireless communication unit 140 can perform two-way wireless data communication with the control communication unit 220 of the drone control unit 200. To this end, the wireless communication unit 140 includes one of a Wi-Fi module, a Bluetooth module, a 4G communication module, a wireless Internet module, a wireless-LAN module, and a ZigBee communication module.

The depth sensor 150 measures the depth of a video or image using a time-of-flight (TOF) measurement method. That is, the depth sensor 150 calculates the distance between the depth sensor 150 and the facility by measuring the delay time until the signal emitted from the source is reflected by the facility and returns. The output signal of the depth sensor 150 is input to the drone control unit 110 and used for location measurement along with data transmitted from the feature extraction unit 290.

The IMU (Inertial Measurement Unit, inertial measurement unit, 160) has a built-in 3-axis accelerometer and a 3-axis angular accelerometer of the drone 100 to detect the flight, rotation, turning, acceleration and deceleration of the drone 100, and the drone control unit 110 ) is sent to An example of the IMU 160 may be a gyro sensor.

The blade motor 170 provides rotation to blades for flight of the drone 100 . Two to eight blade motors 170 may be provided according to the number of blades.

The power supply unit 180 provides power for flight, communication, control, filming, and storage of the drone 100 . The power supply unit 180 may be a removable secondary battery.

The route storage unit 190 stores a route on which the drone 100 should fly to inspect the facility 10 . The route storage unit 190 stores the location, speed, direction, camera pose, and the like of the route. The route is received and stored from the drone control unit 200 before flight.

The control unit 210 of the drone control unit 200 may be a microprocessor or a CPU, and programs necessary for operation of the drone control unit 200 and control and monitoring of the drone 100 are loaded and executed.

The control communication unit 220 is capable of two-way wireless data communication with the wireless communication unit 140 . To this end, the control communication unit 220 includes one of a Wi-Fi module, a Bluetooth module, a 4G communication module, a wireless Internet module, a wireless-LAN module, and a ZigBee communication module.

The 3D modeling unit 230 stores 3D modeling information of the facility 10 as shown in FIG. 2 . That is, the 3D modeling unit 230 has 3D modeling information such as train tracks, bridges, abutments, piers 20, girders, lower structures, and towers.

The display 240 displays an image captured by the drone 100, an inspection process of the facility 10, 3D modeling information, and the like. The display 240 may be a monitor or LCD panel having a touch screen function.

The input unit 250 is for inputting necessary information by the operator, and may be a mouse, keyboard, USB port, touch screen, drone remote controller for manual flight, and the like.

The route generator 260 designates or creates a route for a drone flying around the facility 10 and stores the route. Route information generated by the route generator 260 is transmitted to the route storage unit 190 of the drone 100.

The virtual depth image generation unit 270 generates (eg, captures and converts) 3D images of a specific part, a specific zoom-in, and a specific camera pose from the corresponding facility modeling 10a of the 3D modeling unit 230. . The created virtual depth image may be a JPG file. In this case, the specific part may be a part (eg, a protruding part, a corner, a wall, an inclined surface, a pillar, etc.) that is easily identified by the camera among facilities in the GPS shaded area.

The key frame extractor 280 extracts (eg, cropping) a specific part as a frame from the generated virtual depth image.

The feature extraction unit 290 extracts a feature (eg, a photographing plane or an outline) from the extracted frame image.

5 is a detailed block diagram of the feature extraction unit 290 in FIG. 4, and FIG. 6 shows the capturing plane extracting unit 294 and the capturing plane outline extracting unit 296 capturing the capturing plane and outlines from the frame image 55. It is a front view showing an example of extracting. As shown in FIGS. 5 and 6 , the camera pose extractor 292 of the feature extractor 290 extracts the specific camera pose 60 generated by the virtual depth image generator 270 . The camera pose 60 includes X, Y, and Z-axis camera rotation information and zoom information in a direction in which the camera looks from a specific position (eg, drone position 70).

The planar feature extraction unit 294 extracts a capturing plane from the frame image 55 . 6 is an enlarged frame image 55 of a railroad bridge and a pier, and the imaging plane extractor 294 extracts a first imaging plane 80 and a second imaging plane 82 from these frame images 55, respectively. do.

The capturing plane outline extraction unit 296 extracts an outline (Edge) from the capturing plane. In FIG. 6 , the outline extraction unit 296 recognizes and extracts a plurality of outlines 84 from the second imaging plane 82 . The outline 84 may be composed of a plurality of straight lines.

The imaging plane extractor 294 uses the output signal of the depth sensor 150 for precise correction, and may include depth data in the first and second imaging planes 80 and 85 .

Operation of the embodiment

Hereinafter, the operation of the present invention having the above configuration will be described in detail. 7 is a flowchart illustrating a method for calibrating the position of an inspection drone based on facility modeling according to an embodiment of the present invention.

As shown in FIG. 7, first, the drone control unit 110 receives the flight path 30 from the drone control unit 200 and stores it in the path storage unit 190 (S90).

Next, the drone control unit 110 calculates the predicted position 102 based on the route information, the position signal of the camera 120 and the GPS 130, and the output signal of the IMU 160, and the pose of the camera 120 Extract (S100). The pose of the camera includes the current shooting direction of the camera.

If the pose of the camera 120 is upward shooting, the drone 100 photographs the lower surface of the bridge upward while flying the lower part of the facility 10 . At this time, the Z-axis relative distance ( Dz) is corrected (S130). The Z-axis relative distance (Dz) is the difference between the predicted height of the drone 100 and the actual flight height with respect to the height direction (Z-axis direction) between the drone 100 and the bottom of the bridge. The Z-axis relative distance (Dz) is caused by GPS shadowing or overlapping of other errors as the drone 100 enters the lower part of the bridge.

Then, the X-axis relative distance (Dx), Y-axis relative distance (Dy), and nose rotation of the drone 100 from the predicted position 102 to the corrected position 104 based on the characteristics of modeling (eg, outline) The angle (θyaw) is corrected (S140). At this stage, the drone precisely corrects its position and heading direction on the horizontal plane (X-Y plane). By correcting the nose rotation angle θyaw, the pose of the camera 120 is also corrected.

If the pose of the camera 120 is forward shooting, the drone 100 photographs the side of the bridge and the side of the pier 20 while flying on the side of the facility 10 . At this time, the flight direction may be a horizontal direction, a vertical direction, or a circular path. Based on the characteristics of the modeling and the output signal of the depth sensor 150, the X-axis relative distance (Dx) and nose rotation angle (θyaw) from the predicted position 102 to the corrected position 104 are corrected (S160). The X-axis relative distance (Dx) is the difference between the predicted distance of the drone 100 and the actual flight distance with respect to the horizontal (X-axis direction) distance between the drone 100 and the side of the bridge.

Then, the Y-axis relative distance (Dy) and Z-axis relative distance (Dz) from the predicted position 102 to the corrected position 104 are corrected based on the characteristics of the modeling (S170).

In this way, when the relative distance of each axis and the nose rotation angle (θyaw) are corrected, this corrected position is updated to the current predicted position. Then, by correcting the drone to the corrected position (S150), the drone is placed on an accurate path.

Such a process may be performed at regular intervals, may be performed according to a GPS reception rate, or may be performed on a specific route.

Detailed descriptions of the preferred embodiments of the present invention disclosed as described above are provided to enable those skilled in the art to implement and practice the present invention. Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the scope of the present invention. For example, those skilled in the art can use each configuration described in the above-described embodiments in a manner of combining with each other. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The present invention may be embodied in other specific forms without departing from the spirit and essential characteristics of the present invention. Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered as illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention. The invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. In addition, claims that do not have an explicit citation relationship in the claims may be combined to form an embodiment or may be included as new claims by amendment after filing.

10: facilities,
10a: facility modeling,
20: piers,
30: flight path,
50: 3D modeling,
55: frame image,
60: camera pose,
70: drone location,
80: 1st shooting plane,
82: second shooting plane,
84: outline,
100: drone,
102: predicted position,
104: correction position,
110: drone control unit,
120: camera,
130: GPS,
140: wireless communication unit,
150: depth sensor,
160: IMU,
170: blade motor,
180: power supply,
190: path storage unit,
200: drone control department,
210: control unit,
220: control and communication department,
230: 3D modeling unit,
240: display,
250: input unit,
260: path creation unit,
270: virtual depth image generation unit,
280: key frame extraction unit,
290: feature extraction unit,
292: camera pose extraction unit,
294: shooting plane extraction unit,
296: outline extraction unit of the shooting plane,
Dx, Dy, Dz: Compensation distance on X-axis, Y-axis, Z-axis,
θyaw: The angle of rotation of the nose of the drone.

Claims (11)

In the drone 100 and the drone control unit 200 capable of mutual wireless communication,
(i-1) a 3D modeling unit 230 in which modeling 10a data of the facility 10 to be inspected is stored;
(i-2) a path generator 260 generating a flight path 30 of the drone 100 for the modeling 10a;
(i-3) a key frame extractor 280 for extracting a frame image 55 for a camera pose 60 at an arbitrary position on the flight path 30;
(i-4) a feature extraction unit 290 extracting features of the facility 10 from the frame image 55; and
(i-5) the drone control unit 200 including; a control communication unit 220 wirelessly communicating with the drone 100; and
(ii-1) a wireless communication unit 140 wirelessly communicating with the control communication unit 220;
(ii-2) a path storage unit 190 for storing the flight path 30; and
(ii-3) The drone 100 including; a drone controller 110 correcting the predicted position 102 of the drone 100 based on the feature received from the feature extractor 290; constituted,
The feature extraction unit 290,
a camera pose extractor 292 extracting information about the camera pose 60;
an imaging plane extractor 294 for extracting the imaging planes 80 and 82 from the frame image; and
An apparatus for calibrating the position of an inspection drone based on modeling of a facility, characterized in that it includes; delete According to claim 1,
The drone 100, in order to calculate the predicted position 102,
A position correction device for an inspection drone based on modeling of a facility, further comprising at least one of a camera 120, a GPS 130, and an IMU 160. According to claim 1,
The drone control unit 110 further includes a depth sensor 150 to correct the predicted position 102. Position correction device for an inspection drone based on modeling of a facility. According to claim 1,
The drone control unit 110, in order to correct the predicted position 102,
Relative distances (Dx, Dy, Dz) from the predicted position 102 to the corrected position 104; and
A position correction device for an inspection drone based on modeling of a facility, characterized in that for calculating the nose rotation angle (θyaw) of the drone 100. In the correction method using the position correction device according to claim 5,
A preliminary step (S90) in which the drone control unit 110 receives the flight path 30 from the drone control unit 200 and stores it in the path storage unit 190;
The drone controller 110 calculates the predicted position 102 based on an output signal of at least one of the camera 120, the GPS 130, and the IMU 160, and extracts the pose of the camera 120 Step (S100);
When the pose of the camera 120 is upward shooting,
Correcting the Z-axis relative distance (Dz) from the predicted position 102 to the corrected position 104 based on the modeling features received from the feature extraction unit 290 of the drone control unit 200 (S130 ); and
X-axis relative distance (Dx), Y-axis relative distance (Dy) and nose rotation angle (θyaw) of the drone 100 from the predicted position 102 to the corrected position 104 based on the characteristics of the modeling Including; step of correcting (S140);
When the pose of the camera 120 is forward shooting,
Correcting the X-axis relative distance (Dx) and the nose rotation angle (θyaw) from the predicted position 102 to the corrected position 104 based on the characteristics of the modeling (S160); and
Correcting the Y-axis relative distance (Dy) and the Z-axis relative distance (Dz) from the predicted position 102 to the corrected position 104 based on the characteristics of the modeling (S170);
The drone control unit 110 flies to the correction position 104 (S150); a position correction method of an inspection drone based on facility modeling, characterized in that it includes. According to claim 6,
The characteristics of the modeling in the correction steps (S130 and S160) are,
Position correction of the inspection drone based on the modeling of the facility, characterized in that the capturing plane extracting unit 294 of the feature extracting unit 290 includes the capturing planes 80 and 82 data extracted from the modeling 10a. method. According to claim 6,
The characteristics of the modeling in the correction steps (S140, S170) are,
Position correction method of an inspection drone based on modeling of a facility, characterized in that the outline extractor 296 of the shooting plane among the feature extractors 290 includes the outline 84 data extracted from the modeling 10a. . According to claim 6,
The position correction method of the inspection drone based on the modeling of the facility, characterized in that when the camera 120 pose is upward shooting, the drone 100 flies below the facility 10. According to claim 6,
In the correction steps (S130, S140, S160, S170),
The drone control unit 110 corrects the position of the inspection drone based on the modeling of the facility, characterized in that the correction is further based on the output signal of the depth sensor 150. According to claim 6,
The method of correcting the position of an inspection drone based on modeling of a facility, wherein the modeling is a 3-D modeling of the facility.

Images