KR101035058B1 - Moving Robot and Method Of Recognizing A Ceiling Region For Localization Thereof - Google Patents

Abstract

According to an aspect of the present invention, there is provided a ceiling area recognition method for recognizing a magnetic position of a mobile robot and a mobile robot equipped with an image device, the method comprising: calculating center coordinates of a mobile robot from a ceiling image obtained from an imaging device, Performing pattern division, extracting an image region including the center coordinates of the mobile robot among the ceiling image regions divided by the pattern division, as a ceiling candidate region, and an image region included inside the extracted ceiling candidate region Is recognized as a ceiling area, distinguishes the recognized ceiling areas, obtains the edges of the separated ceiling areas, sets the edges between the ceiling and the wall in the image area, and Disclosed are a mobile robot and a ceiling area recognition method for recognizing a magnetic position thereof, including extracting feature points.

According to the present invention, by efficiently recognizing the ceiling area to set the edge between the ceiling and the wall surface, and easily distinguish the feature points that are not affixed by distinguishing only the feature points in the ceiling area, the magnetic position and orientation of the mobile robot in the indoor space Can be recognized correctly.

Imaging Device, Pattern Segmenation, Ceiling, Wall, Feature

Description

Moving Robot and Method Of Recognizing A Ceiling Region For Localization Thereof}

The present invention relates to a mobile robot and a ceiling area recognition method for recognizing a magnetic position thereof, and more particularly, to efficiently recognize the ceiling area to accurately recognize the magnetic location and the orientation of the mobile robot in an indoor space. The present invention relates to a mobile robot and a ceiling area recognition method for magnetic position recognition thereof.

In general, the magnetic position recognition technology of the mobile robot uses a sensor or a camera to set the magnetic position and orientation of the mobile robot by recognizing or recognizing surrounding objects. However, such a recognition technology has a large variability in map composition due to change or movement of surrounding objects, and localization is difficult due to affine.

To overcome this problem, a mobile robot magnetic position recognition technology based on a ceiling vision camera has been proposed. That is, the conventional ceiling-oriented camera-based mobile robot magnetic position recognition technology extracts a feature from the acquired image, or extracts a line and uses a matching method to extract the magnetic position of the mobile robot. It was to recognize the direction.

However, the above-described conventional ceiling-oriented camera-based magnetic position recognition technology has an ineffective aspect because of a problem of affine and an unclear distinction between the ceiling and the wall. On the other hand, the root cause of this problem is that the mobile robot does not effectively recognize the ceiling area. For example, as shown in Figure 1, it can be seen that the position of the corner (corner) on the actual image (A) and the position of the corner on the image (B) recognized by the mobile robot clearly does not correspond (position uncertainty ). That is, in the conventional ceiling-oriented camera-based magnetic position recognition technology, there is a problem that the mobile robot does not clearly recognize the ceiling area and the feature points included in the ceiling area.

Therefore, there is an urgent need to efficiently distinguish and recognize ceiling areas, to easily obtain lines or edges between ceilings and walls, and to distinguish only feature points within ceiling areas.

Therefore, the present invention has been made to solve the above-mentioned conventional problems, and efficiently recognizes the ceiling area from the ceiling image obtained through the imaging device mounted on the mobile robot to distinguish the line or edge between the ceiling and the wall surface. By distinguishing only the feature points in the ceiling area and easily distinguishing feature points that are not affine, the mobile robot can recognize the magnetic position and orientation of the mobile robot accurately in the indoor space and the ceiling area recognition for its magnetic location recognition. The purpose is to provide a method.

In order to achieve the above object, according to a preferred embodiment of the present invention, in the ceiling area recognition method for recognizing the magnetic position of the mobile robot equipped with an imaging device, the mobile robot from the ceiling image obtained by the imaging device Calculating a center coordinate, performing a pattern division of the obtained ceiling image, and extracting an image region including the center coordinates of the robot from the ceiling image region divided by the pattern division as a ceiling candidate region; And recognizing the image area included inside the extracted ceiling candidate area as a ceiling area, separating the recognized ceiling area, obtaining edges of the separated ceiling area, and determining a ceiling from the image area. Setting the edges between the wall surfaces and extracting feature points of the ceiling area separated by the edges; The present invention provides a ceiling area recognition method for magnetic position recognition of a mobile robot.

A ceiling area recognition method for recognizing a magnetic position of a mobile robot according to the present invention, comprising: adjusting a focal center of the imaging apparatus performed before calculating a center coordinate of the mobile robot, and obtaining a ceiling from the imaging apparatus A calibration step of planarizing the image may be further included.

In the ceiling area recognition method for magnetic position recognition of the mobile robot according to the present invention, in the step of calculating the center coordinates of the mobile robot, if the image device is mounted on the rotation center of the mobile robot, from the image device The center coordinates of the acquired image may be calculated as the center coordinates of the mobile robot.

In the ceiling area recognition method for magnetic position recognition of the mobile robot according to the present invention, in the step of calculating the center coordinates of the mobile robot, if the imaging device is not mounted on the rotation center of the mobile robot, A center coordinate of the mobile robot is calculated by using an equation, that is, h = f / d * H, where h is a center coordinate of the mobile robot, and f is a focal length at which the imaging device acquires the ceiling image. d may be a distance from the imaging apparatus to the ceiling, and H may be a distance coordinate from the imaging apparatus to the rotation center of the mobile robot.

The present invention also provides a mobile robot configured to include an imaging device and a control unit. The imaging apparatus acquires a ceiling image by photographing a ceiling or a wall surface. The controller calculates a center coordinate of the mobile robot from the ceiling image acquired by the imaging apparatus. The controller performs pattern division of the obtained ceiling image, and extracts an image area including the center coordinates of the robot from the ceiling image area divided by the pattern division as a ceiling candidate area. The controller recognizes the image area included inside the extracted ceiling candidate area as a ceiling area and classifies the recognized ceiling area. The control unit obtains an edge of the divided ceiling area, sets the edge between the ceiling and the wall in the image area, and extracts feature points of the ceiling area divided by the edge.

According to the present invention, by effectively recognizing the ceiling area to distinguish the edge between the ceiling and the wall, and by distinguishing only the feature points within the ceiling area to easily distinguish the feature points that are not affixed, the magnetic position and orientation of the mobile robot in the indoor space There is an effect that can be recognized correctly.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, only parts necessary for understanding the operation according to the embodiment of the present invention will be described, and the description of other parts will be omitted so as not to disturb the gist of the present invention.

The terms or words used in the specification and claims described below should not be construed as being limited to the ordinary or dictionary meanings, and the inventors may appropriately use the concept of terms to describe their own invention in the best way. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configuration shown in the drawings are only exemplary embodiments of the present invention, and do not represent all of the technical ideas 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 water and variations.

Figure 2 is a block diagram showing the configuration of a mobile robot according to an embodiment of the present invention.

Referring to FIG. 2, the mobile robot 100 according to the present embodiment efficiently recognizes the ceiling area, accurately recognizes and sets the magnetic position and the orientation in the indoor space, and stores the storage unit 10, the driving unit 20, and the image. It comprises a device 30 and the control unit 40.

In this case, the mobile robot 100 according to the present embodiment may be a cleaning robot used in a home or a building, but is not limited thereto.

The storage unit 10 stores a program necessary for controlling the operation of the mobile robot 100 and data generated while the program is executed, and includes one or more volatile memory devices and nonvolatile memory devices. The storage unit 10 stores an execution program for recognizing the ceiling area for recognizing the magnetic position of the mobile robot 100.

The driver 20 moves the mobile robot 100 under the control of the controller 40.

The imaging apparatus 30 performs a photographing function of a subject and acquires surrounding image information necessary for recognizing and setting the magnetic position and orientation of the mobile robot 30 that is moving. The imaging apparatus 30 may include an image sensor, a signal processor, and an image processor. The image sensor converts the optical signal of the captured image into an analog signal. The signal processor converts the analog signal into a digital signal. The image processing unit obtains image data by processing the image signal input through the signal processing unit, and transfers the acquired image data to the control unit 70 or stores it in the storage unit 10.

The controller 40 is a microprocessor that performs the overall control operation of the mobile robot 100. The controller 40 recognizes the ceiling area for recognizing the magnetic position of the mobile robot 100 using the image information received from the imaging device 30.

In this case, the controller 40 may recognize the ceiling area for recognizing the magnetic position of the mobile robot 100 as follows.

First, the controller 40 adjusts the focal center of the imaging apparatus 30 and flattens the ceiling image acquired from the imaging apparatus 30. The controller 40 calculates the center coordinates of the mobile robot 100 from the flattened ceiling image. The controller 40 performs pattern division of the obtained ceiling image, and extracts an image area including the center coordinates of the mobile robot 100 from the ceiling image area divided by the pattern division as the ceiling candidate area. The controller 40 recognizes the image area included inside the extracted ceiling candidate area as the ceiling area. The controller 40 classifies the recognized ceiling area, obtains an edge of the separated ceiling area, and sets the edge between the ceiling and the wall in the image area. In addition, the controller 40 can accurately recognize the magnetic position and the orientation of the mobile robot 100 in the indoor space by extracting the feature points of the ceiling areas separated by the edges.

At this time, when calculating the center coordinates of the mobile robot 100, if the imaging device 30 is mounted to the rotation center of the mobile robot 100, the controller 40 is the center coordinates of the image obtained from the imaging device 30 May be calculated as the center coordinates of the mobile robot 100.

 On the other hand, when calculating the center coordinates of the mobile robot 100, if the imaging device 100 is not mounted to the rotation center of the mobile robot 100, the control unit 40 uses the following equation (1) the mobile robot ( The center coordinate of 100) can be calculated.

h = f / d * H

Here, h is the center coordinate of the mobile robot 100, f is the focal length for the imaging device 30 to obtain a ceiling image, d is the distance from the imaging device 30 to the ceiling on the physical indoor space H may be a distance coordinate from the imaging apparatus 30 to the rotation center of the mobile robot 100. In addition, h and H are vectors as coordinates of a two-dimensional phase, and f and d are scalars as distances of a two-dimensional phase.

As described above, the mobile robot 100 equipped with the imaging apparatus 30 according to an exemplary embodiment of the present invention calculates the center coordinates of the mobile robot 100 from the ceiling image acquired by the imaging apparatus 30 to recognize the ceiling region. By easily distinguishing the edge between the ceiling and the wall surface, and by distinguishing only the feature points within the ceiling area to easily distinguish the feature points that are not affixed, it is possible to accurately recognize the magnetic position and orientation of the mobile robot 100 in the indoor space. .

The ceiling area recognition method for recognizing the magnetic position of the mobile robot 100 according to the embodiment of the present invention will be described with reference to FIGS. 2 to 4G as follows. 3 is a schematic flowchart of a method for recognizing a ceiling area for recognizing a magnetic position of a mobile robot 100 according to a preferred embodiment of the present invention, and FIGS. 4A to 4G are diagrams for recognizing a magnetic position of a mobile robot according to FIG. 3. Each step of the ceiling area recognition method is illustrated.

2 to 4G, the ceiling area recognition method for magnetic position recognition of the mobile robot 100 according to the present invention, the center adjustment and ceiling image flattening step (S210), the center coordinate calculation step of the imaging device 30 (S220), pattern division performing step (S230), ceiling candidate area extraction step (S240), ceiling area recognition step (S250), ceiling area classification step (S260), edge setting step between ceiling and wall surface (S270), and Ceiling area feature point extraction step (S280).

In the method of recognizing the ceiling region for recognizing the magnetic position of the imaging device 30, for example, the mobile robot 100 equipped with the camera, first, the controller 40 calibrates the imaging device 30. That is, the controller 40 adjusts the focal center of the imaging device 30, obtains a ceiling image from the imaging device 30, and flattens it (S210). Here, the controller 40 refers to the focusing center of the imaging apparatus 30 that acquires the ceiling image. Since the controller 40 may proceed in the same manner as before, the controller 40 will not be described in detail.

Thereafter, as shown in FIG. 4A, the controller 40 calculates the center coordinates of the mobile robot 100 from the flattened ceiling image acquired by the imaging apparatus 30 (S220).

Here, when the imaging apparatus 30 is mounted at the center of the mobile robot 100, that is, the rotation center, the controller 40 adjusts the center coordinates of the ceiling image obtained from the imaging apparatus 30 to the center of the mobile robot 100. It calculates by the coordinate h. For example, when the lens center of the camera, which is the imaging device 30, is positioned at the rotation center of the mobile robot 100, the center coordinate of the ceiling image becomes the center coordinate h of the mobile robot 100.

On the other hand, if the lens center of the imaging device 30 is not mounted to the rotation center of the mobile robot 100, the controller 40 may calculate the center coordinate h of the mobile robot 100 by using Equation (1). Can be.

Thereafter, as illustrated in FIG. 4B, the controller 40 performs pattern segmentation of the obtained ceiling image (S230). Here, the ceiling image may be divided into at least one image area by pattern division. Referring to FIG. 4B, it can be seen that, for example, the image area divided by the pattern division is clearly distinguished by colors.

Thereafter, as shown in FIG. 4C, the control unit 40 calculates the mobile robot calculated in the step S220 of calculating the center coordinate h of the above-described mobile robot 100 among the ceiling image areas divided by the pattern division. The image region including the center coordinate h of 100 is extracted as the ceiling candidate region (S240). On the other hand, it is well known that the image area including the center coordinate h of the mobile robot 100 may be a ceiling candidate area due to the structure of the imaging device 30 and the mobile robot 100.

Thereafter, as shown in FIG. 4D, the controller 40 recognizes the image area included inside the extracted ceiling candidate area as the ceiling area (S250). That is, the image areas r1, r2, and r3 inherent in the ceiling candidate area are comprehensively recognized as the ceiling area.

In operation S250, the controller 40 may recognize a ceiling area by recognizing an area not recognized as a ceiling area as a wall area.

Thereafter, as shown in FIG. 4E, the controller 40 distinguishes the recognized ceiling area R from the image area r4 that is not recognized as the ceiling area R (S260). That is, since the ceiling area R is clearly distinguished from the image area r4 that is not recognized as the ceiling area R by color, the controller 40 recognizes the recognized ceiling area R as the ceiling area R. It may be distinguished from the non-image area r4.

Thereafter, as illustrated in FIG. 4F, the controller 40 sets an edge between the ceiling and the wall by obtaining an edge from the ceiling region image (S270). That is, in step S270, the controller 40 obtains an edge of the ceiling area R that is separated from the image area that is not recognized as the ceiling area R in step S260. In addition, the controller 40 sets a line or an edge E between the ceiling R and the wall surface W in the image area.

Thereafter, as shown in FIG. 4G, the controller 40 extracts only a feature F included in the ceiling areas divided by the edges (S280).

Therefore, as described above, according to the present invention, in the mobile robot 100 equipped with the imaging device 30, by calculating the center coordinates of the mobile robot 100 from the acquired ceiling image to recognize the ceiling area to distinguish the ceiling By easily distinguishing the edge between the wall and the wall, and easily distinguishing the feature points that are not affixed by distinguishing only the feature points in the ceiling area, the magnetic position and the orientation of the mobile robot 100 can be accurately recognized in the indoor space.

In the above description of the exemplary embodiment of the present invention, a ceiling area recognition method for recognizing a magnetic position of a mobile robot having a specific shape and structure has been described with reference to the accompanying drawings. It is to be understood that such modifications and variations are intended to be included within the protection scope of the present invention.

1 illustrates an example of image analysis for recognizing a ceiling region by a magnetic position recognition method of a mobile robot according to the related art.

Figure 2 is a block diagram showing the configuration of a mobile robot according to a preferred embodiment of the present invention.

3 is a schematic flowchart of a ceiling area recognition method for recognizing a magnetic position of a mobile robot according to a preferred embodiment of the present invention.

4A illustrates a step of calculating the center coordinates of the mobile robot from the flattened ceiling image of FIG. 3.

FIG. 4B illustrates a step of performing pattern division of the ceiling image according to FIG. 3.

4C illustrates a step of extracting a ceiling candidate region of FIG. 3.

4D illustrates a step of recognizing the ceiling area of FIG. 3.

FIG. 4E illustrates the step of classifying the recognized ceiling area according to FIG. 3.

4F illustrates the step of setting the edge between the ceiling and the wall in the image area shown in FIG. 3.

FIG. 4G illustrates a step of extracting feature points of the ceiling area of FIG. 3.

Claims (6)

In the ceiling area recognition method for magnetic position recognition of a mobile robot equipped with an image device, Calculating a center coordinate of the mobile robot from the ceiling image obtained from the imaging apparatus; Performing pattern division of the obtained ceiling image; Extracting an image region including the center coordinates of the robot among the ceiling image regions divided by the pattern division as a ceiling candidate region; Recognizing an image area included inside the extracted ceiling candidate area as a ceiling area; Classifying the recognized ceiling area; Obtaining an edge of the divided ceiling area and setting the edge between the ceiling and the wall in the image area including the center coordinates of the robot; Extracting feature points of the ceiling region separated by the edges; Including, In calculating the center coordinates of the mobile robot, If the imaging device is not mounted at the rotation center of the mobile robot, the following equation, h = f / d * H Compute the center coordinates of the mobile robot by, where h is the center coordinates of the mobile robot, f is the focal length for the image device to obtain the ceiling image, d is from the image device to the ceiling Distance, H is a distance coordinate from the imaging device to the center of rotation of the mobile robot, h and H is a vector as a coordinate on a two-dimensional, f and d is a scalar as a distance on a two-dimensional Ceiling area recognition method for magnetic position recognition of a mobile robot, characterized in that. The method of claim 1, Before the step of calculating the center coordinates of the mobile robot, A calibration step of adjusting a focal center of the imaging device and flattening a ceiling image to be obtained from the imaging device; Ceiling area recognition method for magnetic position recognition of the mobile robot, characterized in that it further comprises. The method of claim 1, In calculating the center coordinates of the mobile robot, When the imaging apparatus is mounted at the rotation center of the mobile robot, the center coordinates of the mobile robot may be calculated by calculating the center coordinates of the image acquired from the imaging apparatus as the center coordinates of the mobile robot. Way. delete In a mobile robot, Imaging apparatus for acquiring the ceiling image by photographing the ceiling or wall surface; The center coordinates of the mobile robot are calculated from the ceiling image acquired by the imaging apparatus, the pattern division of the obtained ceiling image is performed, and the center coordinates of the robot are included among the ceiling image regions divided by the pattern division. Extracts the extracted image area as the ceiling candidate area, recognizes the image area included inside the extracted ceiling candidate area as the ceiling area, classifies the recognized ceiling area, and obtains an edge of the divided ceiling area to obtain the robot. A controller configured to set an edge between the ceiling and the wall in the image area including the center coordinates of the extracted area, and extract feature points of the ceiling area separated by the edges; Including, When the control unit calculates the center coordinates of the mobile robot, If the imaging device is not mounted at the rotation center of the mobile robot, the following equation, h = f / d * H Compute the center coordinates of the mobile robot by, where h is the center coordinates of the mobile robot, f is the focal length for the image device to obtain the ceiling image, d is from the image device to the ceiling Distance, H is a distance coordinate from the imaging device to the center of rotation of the mobile robot, h and H is a vector as a coordinate on a two-dimensional, f and d is a scalar as a distance on a two-dimensional Mobile robot, characterized in that. The method of claim 5, And the control unit adjusts a focal center of the imaging apparatus and flattens the ceiling image to be obtained from the imaging apparatus.

Images