KR20130064636A - Gondola robot and method for controlling thereof - Google Patents

Abstract

PURPOSE: A gondola robot and a control method thereof are provided to facilitate the maintenance of the outer wall of a building and to monitor the outer wall of the building by recognizing windows forming the outer wall of the building. CONSTITUTION: A control method of a gondola robot is as follows: a step of obtaining images around a position where the gondola robot is position(S53); a step of matching the images(S55); a step of recognizing windows forming the outer wall of a building by analyzing the matched images(S57); and a step of grasping the current position of the gondola robot by counting a number of the windows(S59). [Reference numerals] (AA) Start; (BB) End; (S51) Gondola robot is on standby by hanging on a rope of a rope cart; (S533) If moving according to a length change of the rope, obtain images a building outer wall at a position where the gondola robot is located; (S55) Match the images; (S57) Recognize window glasses and window frames separately by analyzing the matched images; (S59) Grasp the current position of the gondola robot by counting the number of the recognized window glasses; (S61) Change a management tool according to the recognized window glasses and window frames; (S63) Establish a map of the building outer wall based on the recognized window glass and window frame information; (S65) Transmit the matched images to a remote controller of a manager

Description

Gondola robot and method for controlling thereof

The present invention relates to an intelligent robot system for maintenance of a building exterior wall and a control method thereof, and more particularly, to a gondola robot and a control method thereof for managing the exterior wall of a building by recognizing a glass window installed on the building exterior wall.

As urbanization proceeds rapidly with the development of industry, various buildings of large scale are entering the center of the city. In addition, with the urbanization, the problem of environmental pollution is also increasing.

For this reason, the exterior walls of buildings exposed to the external environment are easily contaminated by dust, rain, smoke, etc., and therefore need to be cleaned periodically. The cleaning of the exterior walls of these buildings is performed manually by the janitor descending from the top of the building in a rope or gondola, and there is always a risk of a safety accident such as a fall of the janitor during the cleaning.

In order to solve this problem, an intelligent robot system for maintaining the exterior wall of various kinds of buildings to replace the role of the cleaner has been introduced. Most of the intelligent robots for exterior wall maintenance introduced in the related art are attached to the outer wall of the building and move to perform cleaning on the outer wall of the building.

On the other hand, the conventional intelligent robot for exterior wall maintenance does not know the position of the robot when performing the management of the exterior wall of the building, for example, painting or cleaning. However, cleaning may be a problem with management tools that do not fit where to clean. For example, the conventional intelligent robot for maintaining the exterior wall does not distinguish between the glass windows installed on the exterior wall of the building and the parts other than the glass window, and thus does not perform the cleaning operation according to the characteristics of the material constituting the exterior wall of the building. Holding it. In particular, in the case of performing the painting work, a problem may occur in painting the window area of the outer wall of the building.

In addition, the conventional intelligent robot for exterior wall maintenance is always installed on the exterior wall of the building, so that the robot always carries the risk of falling from the exterior wall of the building.

In order to solve the problem of the existing intelligent robot for exterior wall maintenance, an object of the present invention is to prevent the robot from being damaged or a safety accident from falling off the building exterior wall. It is to provide a gondola robot and its control method.

It is another object of the present invention to provide a gondola robot and a control method thereof, which can perform maintenance on a building exterior wall more quickly through the glass window recognition of the building exterior wall.

Another object of the present invention is to provide a gondola robot and a control method thereof capable of identifying the current position of the gondola robot on the outer wall of the building through glass recognition.

It is still another object of the present invention to provide a gondola robot and a control method thereof capable of monitoring a building exterior wall of the gondola robot through glass window recognition.

In order to achieve the above object, the present invention is a gondola robot that performs the management of the outer wall of the building hanging on the rope of the rope cart, the obtaining step of obtaining a plurality of images of the building outer wall of the area adjacent to the point where the gondola robot is located; A matching step of the gondola robot matching the obtained plurality of images; Recognizing step of the gondola robot recognizes the glass window forming the outer wall of the building from the matched image; And a location determining step of counting the number of the recognized windows to determine the current position of the gondola robot on the outer wall of the building.

In the control method of the gondola robot according to the present invention, the recognition step may include: extracting a line from the matched image by the gondola robot; And distinguishing, by the gondola robot, a glass window and a window frame from a pattern formed by the extracted line.

In the control method of the gondola robot according to the present invention, the positioning step may include: when only the nth window (n is a natural number) is recognized in the matched image, the gondola robot stops the window count; Preparing the window count by the gondola robot when the window frame is recognized by including the n-th window in the matched image; And when the (n + 1) th window adjacent to the nth window is recognized in the matched image, the gondola robot counts the (n + 1) th window to determine the current position of the gondola robot. +1) modifying the glass window; may include.

In the control method of the gondola robot according to the present invention, in the step of modifying, the gondola robot according to the position of the window frame connecting the n-th glass window and the (n + 1) glass window of the gondola robot up, down, left and right Recognize location movement.

The control method of the gondola robot according to the present invention is performed after the positioning step, the gondola robot is a step of changing the management tool according to the recognized window and window frame; The gondola robot constructing a building exterior wall map based on the recognized glass window and window frame; And transmitting, by the gondola robot, the matched image to a remote controller of a manager. At least one step may further include.

The present invention also provides a gondola robot including a robot body, an image information acquisition unit, and a controller. The robot body is suspended from the rope of the rope cart to perform management of the outer wall of the building. The image information acquisition unit is installed on the robot body facing the building exterior wall, and acquires a plurality of images of the building exterior wall in an area adjacent to the point where the robot body is located. The controller is further configured to match the obtained plurality of images, recognize glass windows forming a building exterior wall from the matched image, count the number of the recognized windows, and determine a current position of the robot body on the building exterior wall. do.

In the gondola robot according to the present invention, the image acquisition unit may include a wide-angle camera installed on at least one of the top, bottom, right and left of the robot body.

In the gondola robot according to the present invention, the control unit extracts a line from the matched image, distinguishes and recognizes a window and a window frame from a pattern formed by the extracted line, and manages the tool according to the recognized window and window frame. Change the structure, build a map of the exterior wall of the building based on the recognized windows and window frames, or transmit the matched image to the remote controller of the administrator.

The gondola robot according to the present invention can perform maintenance on the building exterior wall while grasping the location of the gondola robot in real time through registration of images of the building exterior wall. In other words, by taking images of the outer wall of the building around the gondola robot through a plurality of cameras installed on the gondola robot, the images are matched, and by analyzing the matched images, the glass window and the window frame are separated and recognized, thereby managing the window and the window frame. Can be performed separately.

In addition, since the gondola robot can determine the position of the gondola robot on the outer wall of the building through the count of the recognized windows, it is possible to improve the outside wall maintenance efficiency of the gondola robot. In other words, if the location of the gondola robot is not known correctly when the exterior wall management unit of the gondola robot is managed, for example, painting or cleaning, the cleaned area may be cleaned again or an area not cleaned may be generated. Can be cleaned with management tools that do not fit where you want to clean. However, by distinguishing and recognizing the glass window and the window frame as in the present invention, by accurately grasping the position of the gondola robot, this problem can be suppressed from occurring.

In addition, the gondola robot according to the present invention can build a building exterior wall map while grasping the location of the gondola robot in real time through the registration of the image of the building exterior wall.

In addition, the gondola robot according to the present invention provides an administrator with a matched image of the building exterior wall, the operator has the advantage of monitoring the state of the building exterior wall and the progress of work.

In addition, since the gondola robot according to the present invention basically hangs on a rope cart via a rope, and is fixed to the building exterior wall by an exterior wall adsorption unit, the gondola robot is damaged from a building exterior wall in the process of performing maintenance work on the building exterior wall. To prevent accidents or accidents.

1 is a view showing a gondola robot system according to an embodiment of the present invention.
2 is a view showing a state in which the gondola robot of FIG. 1 is installed on the outer wall of the building.
3 is a view showing a state in which the gondola robot performs management while moving along the outer wall of the building.
4 is a block diagram showing the configuration of the gondola robot of FIG.
5 is a flowchart illustrating a control method of a gondola robot according to an embodiment of the present invention.
FIG. 6 is a detailed flowchart according to the positioning step of the gondola robot of FIG. 5.

In the following description, only parts necessary for understanding the embodiments of the present invention will be described, and the description of other parts will be omitted so as not to obscure the gist of the present invention.

The terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary meanings and the inventor is not limited to the meaning of the terms in order to describe his invention in the best way. It should be interpreted as meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configuration shown in the drawings are only preferred embodiments of the present invention, and do not represent all of the technical idea of the present invention, and various equivalents may be substituted for them at the time of the present application. It should be understood that there may be variations and variations.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a view showing a gondola robot system according to an embodiment of the present invention. 2 is a view showing a state in which the gondola robot of FIG. 1 is installed on the outer wall of the building. 3 is a view showing a state in which the gondola robot performs management while moving along the outer wall of the building.

1 to 3, the gondola robot system 100 according to the present embodiment includes a rope cart 10 and a gondola robot 20, a rope cart 10, or a gondola robot 10 connected via a rope 18. And a remote controller 40 for performing wireless communication with 20). In this case, the gondola robot 20 includes an image information acquisition unit 28. The rope cart 10 is fixedly installed on the roof 81 of the building 80, adjusts the position of the gondola robot 20 through the length of the rope 18, and supports the gondola robot 20. The gondola robot 20 is connected to the end of the rope 18 to perform maintenance on the outer wall 83 of the building 80. The gondola robot 20 is fixed to the outer wall 83 of the building 80 by using the outer wall absorbing unit 23, and is fixed to the outer wall 83 of the building 80 by using the outer wall managing unit 25. Perform maintenance.

Here, the remote controller 40 remotely controls the driving of the rope cart 10 and the gondola robot 20 through wireless communication with the rope cart 10 or the gondola robot 20. The remote controller 40 may receive a matched image of the building exterior wall 83 from the gondola robot 20 and provide it to the manager. The manager may check the state of the building exterior wall 83 and the work progress through checking the matched image.

The rope cart 10 may include a cart body 12, a winch 14, a rope 18, and a transceiver 19. The transceiver 19 communicates with the communication unit 24 of the gondola robot 20 installed in the cart body 12. The cart body 12 is fixedly installed on the rooftop 81 of the building 80, and performs length reduction of the rope 18 at the request of the gondola robot 20 through the transceiver 19. The winch 14 is spaced apart from the cart body 12 via the winch support 16 and installed below the outer wall 83 of the building 80. The rope 18 connects the cart body 12 and the gondola robot 20 and is supported by the winch 14. At this time, the cart body 12 has a roll for winding or unwinding the rope 18, and a motor for providing a rotational force to the roll, and performs the length of the rope 18 through the communication with the gondola robot 20 do. Meanwhile, the rope cart 10 may directly perform wireless communication with the remote controller 40 of the manager, and may add or decrease the length of the rope 18 according to a control signal of the remote controller 40.

In addition, the gondola robot 20 maintains the outer wall 83 of the building 80 using the outer wall manager 25 while being fixed to the outer wall 83 of the building 80 by using the outer wall absorbing part 23. Perform management. In addition, the gondola robot 20 matches images of the outer wall 83 of the building 80, and distinguishes the window 85 and the window frame 87 of the outer wall 83 of the building by analyzing the matched images. Using the information on the glass window 85 and the window frame 87 recognized and separated, the management of the building exterior wall 83, the construction of the building exterior wall map, or the matched image is performed by the remote controller of the manager ( 40). The gondola robot 20 may build a building exterior wall map by performing SLAM (Simultaneous Localization And Map-Building). SLAM uses the sensors attached to the gondola robot 20 as the gondola robot 20 moves around the unknown environment (here, the outer wall 83 of the building 80) to create an accurate building exterior wall map of the environment without outside assistance. Work is one of the key technologies for autonomous driving.

The gondola robot 20 according to the present embodiment will be described with reference to FIGS. 1 to 4 as follows. 4 is a block diagram showing the configuration of the gondola robot 20 of FIG.

The gondola robot 20 according to the present embodiment includes a robot body 21, an image information acquisition unit 28, and a control unit 29. The robot body 21 is suspended from the rope 18 of the rope cart 10 to perform management of the outer wall 83 of the building. The image information acquisition unit 28 is installed on the robot body 21 facing the building exterior wall 83, and acquires a plurality of images of the building exterior wall 83 in a region adjacent to the point where the robot body 21 is located. The control unit 29 matches the acquired images, recognizes the glass windows 85 forming the building exterior walls 83 from the matched images, counts the number of the recognized glass windows 85, and then, at the building exterior walls 83. The current position of the robot body 21. When the control unit 29 recognizes the glass window 85 that forms the outer wall 83 of the building in the matched image, the control unit 29 may recognize the glass window 85 and the window frame 87 separately. In addition, the gondola robot 20 may further include a communication unit 24, a storage unit 26, an outer wall absorbing unit 23, and an outer wall managing unit 25.

The robot body 21 is connected to the rope 18 of the rope cart 10 installed in the building 80 to move. The upper portion of the robot body 21 is provided with a fixed base 27 to which the rope 18 is connected and fixed. The example of the fixing stand 27 is installed on both sides of the upper portion of the robot body 21, but only one may be installed at the center of the upper portion of the robot body 21, two or more may be provided at the upper portion. In this case, the robot body 21 may have a communication unit 24, a storage unit 26, and a control unit 29. The communication unit 24, the storage unit 26, and the control unit 29 may be embedded in the robot body 21 in a form mounted on a printed circuit board. In addition, the communication unit 24 may be installed in the robot body 21 in a form exposed to the outside of the robot body 21.

In addition, the robot body 21 may further include a management material supply unit capable of supplying a management material to the outer wall management unit 25. At this time, the management material supply unit may store the management material or receive the management material from the outside and deliver the management material to the outer wall management unit 25. The management material may include a cleaning agent for cleaning the outer wall 83 of the building 80, a paint for painting, and the like.

The communication unit 24 communicates with the transmission / reception unit 19 of the rope cart 10 under the control of the control unit 29. The communication unit 24 transmits information for requesting the length of the rope 18 required for the vertical movement of the robot body 21 and the position control in the fixed state to the transceiver 19 of the rope cart 10. The communication unit 24 communicates with the remote controller 40 of the manager, and receives the remote control signal of the manager from the remote controller 40 and transmits the signal to the controller 31. The remote control signal of the manager may include various control signals required for remote control of the gondola robot 20, including a start signal and an end signal of the gondola robot 20. The communication unit 24 may transmit the matched image of the building exterior wall 83 to the remote controller 40. At this time, the communication unit 24 may communicate with the transceiver 19 and the remote controller 40 of the rope cart 10 through a communication network. For example, the communication network may be one of a wired communication network, a mobile communication network, a WiBro (Wireless Broadband) network, a high-speed downlink packet access (HSDPA) network, a satellite communication network, and a local area network. It is preferable to use a local area network, and the local area network may be Bluetooth, infrared, Zigbee, UWB, NFC, or Wi-Fi.

The storage unit 26 stores a program necessary for controlling the operation of the gondola robot 20 and data generated while executing the program. The storage unit 26 stores an execution program for performing a function according to a remote control signal of the remote controller 40. The storage unit 26 matches images of the outer wall 83 of the building 80, and distinguishes and recognizes the glass window 85 and the window frame 87 of the outer wall 83 of the building by analyzing the matched image. In addition, the management of the outer wall 83 of the building using the information on the glass window 85 and the window frame 87 recognized and divided, building the outer wall map of the building, or match the image of the remote controller 400 of the manager Save the executable program for sending to). In this case, the storage unit 26 may store the constructed outer wall map by matching the identification information of the corresponding building 80.

The outer wall suction unit 23 is installed on the robot body 21 and performs a function of fixing the robot body 21 to the outer wall 83 of the building 80. The outer wall adsorption unit 23 is installed on one surface of the robot body 21 facing the outer wall 83 of the building 80 so as to be fixed to the outer wall 83 of the building 80 by a vacuum adsorption method. At this time, the outer wall suction unit 23 may directly fix the robot main body 21 to the outer wall 83 of the building 80 by a vacuum suction method using a fan. Alternatively, the outer wall adsorption unit 23 may fix the robot body 21 to the outer wall of the building 80 by using the adsorption robot arm.

The outer wall manager 25 is installed in the robot body 21, and manages the outer wall 83 of the building 80 under the control of the controller 29. The outer wall manager 25 is installed on one surface of the robot body 21 facing the outer wall 83 of the building 80. Here, the outer wall management unit 25 includes a management tool for performing the outer wall 83 management, and the management tool may be a cleaning device or a painting device. The outer wall management unit 25 may be installed in the robot body 21 in the form of a robot arm, but is not limited thereto.

The image information acquisition unit 28 obtains image information of the outer wall 83 of the building where the gondola robot 20 is located and transmits the image information to the control unit 29. The image information acquisition unit 28 performs a photographing function of a subject, and in particular, acquires image information by photographing a building outer wall 83. In this case, the image information acquisition unit 28 may include an image sensor, a signal processor, and an image processor. The image sensor converts the optical signal of the photographed image into an analog signal. The signal processor converts an analog signal into a digital signal. The image processor may obtain image data by processing an image signal input through the signal processor, and output the obtained image data to the controller 29 or store the image data in the storage unit 26.

The image information acquisition unit 28 includes a camera array in which a plurality of cameras are arranged, a camera module in which a pan-tilt module is coupled to the camera, a wide angle camera, and the like. In the present embodiment, as the image information acquisition unit 28, as shown in FIG. 3, a plurality of cameras may be installed in the direction toward the outer wall 83 of the building on the top, bottom, left, and right sides of the robot body 21.

 On the other hand, the image information acquisition unit 28 may be installed on the outer surface of the robot body 21 that can face the outer wall 83 of the building (80). In the present embodiment, the camera is disclosed in the upper, lower, left, and right sides of the robot body 21, but may be installed in at least one of the upper, lower, left, and right, it is preferably installed in two or more places. For example, the camera may be installed above, below, left, or right of the robot body 21.

And the control unit 29 is built in the robot body 21, and performs a function as a microprocessor for performing the overall control operation of the gondola robot (20). The controller 29 matches the obtained image, recognizes the window 85 and the window frame 87 forming the building exterior wall 83 in the matched image, counts the number of recognized windows 85 The current position of the robot body 21 on the building outer wall 83 is grasped. In addition, the control unit 29 changes the management tool according to the recognized glass window 85 and the window frame 87, constructs a building exterior wall map based on the recognized glass window 85 and the window frame 87, or displays a matched image. Transmission can be transmitted to the remote controller 40 of the manager via the communication unit 24.

In this case, the controller 29 may perform image registration in such a manner as to partially overlap the boundary portions of the plurality of images photographed by the image information acquirer 28. When the camera is installed on the top, bottom, left, and right sides of the robot body 21 as in the present embodiment, the controller 29 may obtain a matched image by matching the images photographed by the top, bottom, left, and right cameras. Alternatively, the image registration may be performed by overlapping the boundary portions of the plurality of images captured by the image information acquisition unit 28 so that they overlap in part. Since a difference may occur in the position of the gondola robot 20 with time, the image may be slightly distorted according to the position of the gondola robot 20 at the time of shooting between the images before registration. Therefore, the obtained image may be registered by correcting the degree of distortion between the acquired images based on the position of the gondola robot 20 determined at the time of image registration. Alternatively, the image information acquiring unit 28 may overlap the boundary portions of the plurality of images captured by time, so as to overlap each other, and then perform image registration by aligning a plurality of feature points of an object included in the captured image. have. The position of the gondola robot 20 may be grasped from the distorted information of the second image that is matched with respect to the first image that is a reference when the image is matched.

In addition, the controller 29 may apply an AVM (around view monitoring) to match the image to enhance the monitoring effect. That is, the control unit 29 provides the matched image to the remote controller 40 of the manager through the communication unit 24. At this time, the matched image to which the manager visually sees 360 degrees outside the gondola robot 20 is applied. Can be provided. In order to provide an AVM-matched image, an image information acquisition unit 28 capable of capturing 360 degrees outside the gondola robot 20 is required. Accordingly, in the present embodiment, the image information acquisition unit 28 including the cameras installed on the top, bottom, left, and right sides of the gondola body 21 is disclosed.

The controller 29 may recognize the window 85 and the window frame 87 by using various image analysis techniques for the matched image. For example, the glass window 85 and the window frame 87 constituting the building exterior wall 83 may be divided by lines formed by the window frame 87. Therefore, the controller 29 extracts a line from the matched image. In addition, the controller 29 recognizes the window 85 and the window frame 87 separately from the pattern formed by the extracted line. In addition, the glass window 85 and the window frame 87 may be distinguished and recognized using various conventional image analysis techniques.

The controller 29 may determine the position of the gondola robot 20 as follows. That is, when only the nth window (n is a natural number) is recognized in the matched image, the controller 29 stops the window count. When the window frame 87 is recognized by including the n-th window in the matched image, the controller 29 prepares the window count. When the n-th glass window and the (n + 1) th glass window that are adjacent to the matched image are recognized, the controller 29 counts the (n + 1) th glass window to determine the current position of the gondola robot 20 (n). +1) Modify it with a glass window. In this case, the controller 29 may recognize whether the gondola robot 20 is moved up, down, left, or right according to the position of the window frame 87 connecting the nth glass window and the (n + 1) th glass window.

After determining the position of the gondola robot 20, the control unit 29 may perform the following functions.

The control unit 29 may change the management tool according to the recognized glass window 85 and the window frame 87. For example, when cleaning the window 85 and the window frame 87, the control unit 29 may use different management tools according to the window and the window frame. In this case, the control unit 29 may perform cleaning by changing the management tool according to the recognized glass window 85 and the window frame 87. Alternatively, when performing a painting operation, the control unit 29 may prevent the glass window 85 from being painted. As such, when the location of the gondola robot 20 generated when the management, for example, painting or cleaning of the building exterior wall 83 is performed by using the exterior wall management unit 25 of the gondola robot 20, cleaning is performed. The cleaning can be performed again with one place, a non-cleaning area is created, or with a management tool that is not suitable for the place to be cleaned. However, by distinguishing and recognizing the glass window 85 and the window frame 87 as in the present embodiment, by accurately identifying the position of the gondola robot 20, it can be suppressed that such a problem occurs.

In addition, the controller 29 may construct a building exterior wall map using information on the recognized glass window 85 and the window frame 87. In this case, the controller 29 may build a building exterior wall map by reflecting the image information matched with the position of the gondola robot 20.

In addition, the control unit 29 may transmit the matched image through the communication unit 24 to the remote controller 40 of the manager.

The control method of the gondola robot 20 according to the present embodiment will be described with reference to FIGS. 1 to 6 as follows. 5 is a flow chart according to the control method of the gondola robot 20 according to an embodiment of the present invention. 6 is a detailed flowchart according to the positioning step of the gondola robot 20 of FIG.

First, when the gondola robot 20 moves along the length of the rope 18 while the gondola robot 20 is suspended in the rope 18 in step S51, the gondola robot 20 in step S53 is a rope. According to the change in the length of 18, an image of the outer wall 83 of the building where the gondola robot 20 is located is obtained through the image information acquisition unit 28. For example, the gondola robot 20 captures and acquires an image of the building exterior wall 83 through a camera of the image information acquisition unit 28 installed on the top, bottom, left, and right sides of the robot body 21.

Next, in step S55, the gondola robot 20 matches the acquired image. The gondola robot 20 may perform image registration in such a manner as to partially overlap the boundary portions of the plurality of images photographed by the image information acquisition unit 28.

Next, in step S57, the gondola robot 20 analyzes the matched image and recognizes the window and the window frame separately. That is, the gondola robot 20 may recognize the window 85 and the window frame 87 by using various image analysis techniques for the matched image. For example, the glass window 85 and the window frame 87 constituting the building exterior wall 83 may be divided by lines formed by the window frame 87. Therefore, the gondola robot 20 extracts a line from the matched image. In addition, the gondola robot 20 recognizes the window 85 and the window frame 87 separately from the pattern formed by the extracted line. In addition, the glass window 85 and the window frame 87 may be distinguished and recognized using various conventional image analysis techniques.

Subsequently, in step S59, the gondola robot 20 determines the position of the gondola robot by counting the number of recognized windows. At this time, the gondola robot 20 may determine the position of the gondola robot 20 as follows.

First, in step S591, the gondola robot 20 determines whether only the nth glass window (n is a natural number) is recognized in the matched image.

When only the nth window is recognized as a result of the determination in step S591, the gondola robot 20 stops the window count in step S593.

If only the n-th window is not recognized as a result of the determination in step S591, the gondola robot recognizes the window frame 87 by including the n-th window in the matched image in step S595. In operation S597, the gondola robot 20 prepares a window count.

Next, it is determined whether the nth glass window and the (n + 1) th glass window that are adjacent to the image matched in step S599 are recognized.

In operation S599, when the first (n + 1) window is recognized, the gondola robot 20 counts the (n + 1) window to determine the current position of the gondola robot 20 in step S601. 1) Modify with glass window. In this case, the gondola robot 20 may recognize whether the gondola robot 20 has moved up, down, left, or right according to the position of the window frame 87 connecting the nth glass window and the (n + 1) th glass window.

In operation S61, S63, and S65, the gondola robot 20 performs management on the outer wall 83 of the building by using information about the glass window 85 and the window frame 87 that are separately recognized (S63). To construct a building exterior wall map (S63), or transmits the matched image to the remote controller 40 of the manager (S65). At this time, in step S61, the gondola robot 20 may change the management tool of the outer wall management unit 25 according to the recognized glass window 85 and the window frame 87.

On the other hand, the embodiments disclosed in the specification and drawings are merely presented specific examples to aid understanding, and are not intended to limit the scope of the present invention. It will be apparent to those skilled in the art that other modifications based on the technical idea of the present invention are possible in addition to the embodiments disclosed herein.

10: rope cart 12: cart body
14 winch 16 winch support
18: rope 19: transceiver
20: gondola robot 21: robot body
23: outer wall adsorption unit 24: communication unit
25: outer wall management unit 26: storage unit
28: image information acquisition unit 29: control unit
40: remote controller 80: building
81: roof 83: outer wall
85: glass window 87: window frame
100: Gondola Robot System

Claims (8)

The gondola robot which manages the building exterior wall by hanging on the rope of a rope cart, The acquisition step of acquiring a plurality of images of the building exterior wall of the area adjacent to the point where the gondola robot is located;
A matching step of the gondola robot matching the obtained plurality of images;
Recognizing step of the gondola robot recognizes the glass window forming the outer wall of the building from the matched image;
A location determining step of determining, by the gondola robot, the current position of the gondola robot on the outer wall of the building by counting the number of the recognized windows;
Control method of the gondola robot comprising a. The method of claim 1, wherein the recognition step,
The gondola robot extracting a line from the matched image;
The gondola robot distinguishes the window and the window frame from the pattern formed by the extracted line;
Control method of the gondola robot comprising a. The method of claim 2, wherein the positioning step,
If only the nth window (n is a natural number) is recognized in the matched image, the gondola robot stopping the window count;
Preparing the window count by the gondola robot when the window frame is recognized by including the n-th window in the matched image;
When the (n + 1) th window adjacent to the nth window is recognized in the matched image, the gondola robot counts the (n + 1) th window to determine the current position of the gondola robot (n +). 1) modifying with a glass window;
Control method of the gondola robot comprising a. The method of claim 3, wherein in the modifying step,
The gondola robot is a control method of the gondola robot, characterized in that the gondola robot recognizes the position movement up, down, left and right according to the position of the window frame connecting the n-th glass window and the (n + 1) th glass window. The method of claim 2, wherein the step is performed after the positioning step,
Changing the management tool according to the recognized windows and window frames by the gondola robot;
The gondola robot constructing a building exterior wall map based on the recognized glass window and window frame; And
The gondola robot transmitting the matched image to a remote controller of a manager;
The control method of the gondola robot, characterized in that it further comprises at least one step. A robot body hanging on a rope of a rope cart to perform management of a building exterior wall;
An image acquisition unit installed on the robot body facing the building exterior wall and acquiring a plurality of images of the building exterior wall in an area adjacent to the point where the robot body is located;
A controller for matching the obtained plurality of images, recognizing a glass window forming a building outer wall from the matched image, counting the number of the recognized glass windows, and determining a current position of the robot body on the building outer wall;
Gondola robot comprising a. The method of claim 6, wherein the image acquisition unit,
Gondola robot comprising a wide-angle camera installed on at least one of the top, bottom, right and left of the robot body. 7. The apparatus of claim 6,
Extract lines from the matched image, classify and recognize glass windows and window frames from patterns formed by the extracted lines, change management tools according to the recognized glass windows and window frames, or based on the recognized windows and window frames Gondola robot, characterized in that the building of the building outer wall map or transmits the matched image to the remote controller of the manager.

Images