Method, system and computer readable recording medium for generating a map for a robot

Abstract

According to one embodiment of the present invention, a method is provided, which is a method for generating a map for a robot, the method comprising: a step of obtaining a basic map related to a task of the robot; a step of identifying pixels inferred to be mobile obstacles in the basic map based on at least one of the color of a specific pixel in the basic map and the size of an area related to the pixel; and a step of performing a dilation operation and an erosion operation on the pixels inferred to be the mobile obstacles as objects, and determining the boundary of the mobile obstacle based on a polygon.

Description

Technical Field

[0001] The present invention relates to a method, a system and a non-transitory computer-readable recording medium for generating a map for a robot. Background Art

[0002] Robots can automatically perform difficult tasks or repetitive operations, and are therefore used in a variety of fields to replace or assist humans. In order for robots to move smoothly, maps that accurately reflect information about the surrounding environment are required. In recent years, various studies have been conducted on such robot map generation technology.

[0003] As an example of the related prior art, the technology disclosed in Korean Patent Gazette No. 10-2020-0085142 can be cited. According to this technology, a robot cleaning device that generates map data is introduced. The robot cleaning device includes a communication interface, a memory that stores one or more instructions, and a processor that controls the robot cleaning device by executing the one or more instructions. The processor generates basic map data related to the cleaning space by executing the one or more instructions, and generates object information about at least one object in the cleaning space, divides the area in the basic map data into multiple areas according to the object information, and determines the identification values ​​of the differentiated multiple areas according to the object information.

[0004] However, according to the technologies introduced so far, represented by the above-mentioned prior art, a basic map generated initially based on the robot's sensor information (e.g., information obtained by an image sensor, a laser sensor, a laser radar (LIDAR) sensor, etc.) or provided from an external system is directly applied to the robot's task execution (e.g., serving food), or such a basic map is corrected by manual operation (e.g., noise removal, setting of annotations about obstacle boundaries, etc.) and then applied. When the basic map is directly applied, the map contains considerable noise, the obstacle boundaries are inaccurate, and dynamic obstacles such as people may be recognized as static obstacles such as walls and remain in the map as they are, which in many cases leads to inefficient walking of the robot. In addition, when the basic map is corrected by manual operation, it is necessary to find and remove fine noise in the basic map with the naked eye to correct the fuzzy boundaries of the obstacles, and set annotations about the boundary range of the obstacles one by one, so there is a problem of requiring considerable time, cost and labor.

[0005] To this end, the present inventors have proposed a novel and creative technology that can rapidly and accurately automate the process of processing a basic map related to a robot's task. Summary of the invention

[0006] Problem that the invention aims to solve

[0007] The object of the present invention is to solve all the problems of the above-mentioned prior art.

[0008] Furthermore, another object of the present invention is to reduce the time, cost, and labor required to generate a map for a robot to perform a task.

[0009] Furthermore, another object of the present invention is to accurately remove even minute noises that are difficult to detect with the naked eye of a human being so as to enable the robot to walk stably and efficiently.

[0010] Furthermore, another object of the present invention is to be able to apply dilation and erosion operations to pixels inferred as obstacles to specify the obstacle boundary in a manner that can include all subsidiary obstacles related to the obstacle (and further, provide annotations about the obstacle boundary), and to be able to specify the obstacle boundary based on a polygon, thereby being able to set the obstacle boundary more accurately and allowing the user to easily edit the obstacle boundary afterwards.

[0011] Means used to solve problems

[0012] A representative structure of the present invention for achieving the above-mentioned object is as follows.

[0013] According to one embodiment of the present invention, a method is provided, which is a method for generating a map for a robot, the method comprising: a step of obtaining a basic map related to a task of the robot; a step of identifying pixels inferred to be mobile obstacles in the basic map based on at least one of the color of a specific pixel in the basic map and the size of an area related to the pixel; and a step of performing a dilation operation and an erosion operation on the pixels inferred to be the mobile obstacles as objects, and determining the boundary of the mobile obstacle based on a polygon.

[0014] According to another embodiment of the present invention, a system is provided, which is a system for generating a map for a robot, the system comprising: a basic map acquisition unit, which acquires a basic map related to the robot's task; a pixel recognition unit, which identifies pixels inferred to be moving obstacles in the basic map based on at least one of the color of a specific pixel in the basic map and the size of an area related to the pixel; and a boundary determination unit, which performs dilation and erosion operations on the pixels inferred to be the moving obstacles as objects, and determines the boundaries of the moving obstacles based on polygons.

[0015] In addition, other methods and other systems for implementing the present invention and a non-transitory computer-readable recording medium having recorded thereon a computer program for executing the methods are also provided.

[0016] Effects of the Invention

[0017] According to the present invention, it is possible to reduce the time, cost, and labor required for generating a map used for a robot to perform a task.

[0018] Furthermore, according to the present invention, even minute noises that are difficult to detect with the naked eye of a human being can be accurately removed, thereby enabling the robot to walk stably and efficiently.

[0019] Furthermore, according to the present invention, it is possible to apply dilation and erosion operations to pixels inferred as obstacles as objects to identify the obstacle boundary in a manner that includes all subsidiary obstacles related to the obstacle (and further, provide annotations about the obstacle boundary), and it is possible to identify the obstacle boundary based on polygons, thereby enabling the obstacle boundary to be set more accurately and allowing the user to easily edit the obstacle boundary later. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a diagram schematically showing the structure of an overall system for generating a map for a robot according to one embodiment of the present invention.

[0021] Figure 2 This is a diagram that exemplarily shows the internal structure of a map management system according to an embodiment of the present invention.

[0022] Figure 3A and Figure 3B The following are diagrams showing by way of example a basic map according to an embodiment of the present invention and a conventional map processed by manual operation.

[0023] Figures 4 to 6 FIG. 1 is a diagram showing an example of a robot map generated according to an embodiment of the present invention.

[0024] Figure 7 and Figure 8 This is a diagram exemplarily showing the structure of a robot according to an embodiment of the present invention.

[0025] Description of Reference Numerals

[0026] 100-Communication Network

[0027] 200-Map Management System

[0028] 210-Basic map acquisition department

[0029] 220-Pixel recognition unit

[0030] 230-Boundary Determination Department

[0031] 240-Noise Removal Unit

[0032] 250-Boundary Correction Department

[0033] 260-Ministry of Communications

[0034] 270-Control Department

[0035] 300-Robot DETAILED DESCRIPTION

[0036] The detailed description of the present invention described below refers to the accompanying drawings shown as examples of specific embodiments that can implement the present invention. Such embodiments will be described in sufficient detail so that those skilled in the art can implement the present invention. The various embodiments of the present invention are different from each other, but it must be understood that the various embodiments do not need to be mutually exclusive. For example, as long as it does not depart from the spirit and scope of the present invention, the specific shapes, structures and characteristics recorded in this specification can be changed from one embodiment to other embodiments to achieve. In addition, it must be understood that as long as it does not depart from the spirit and scope of the present invention, the position or configuration of individual structural elements in each embodiment can also be changed. Therefore, the detailed description described below should not be understood as having a limiting meaning, and the scope of the present invention must be understood to include the scope requested by the claims of the claims and all scopes equivalent thereto. In the drawings, similar reference symbols represent the same or similar structural elements from all sides.

[0037] Hereinafter, in order to enable persons having common knowledge in the technical field to which the present invention belongs to easily implement the present invention, various preferred embodiments of the present invention are described in detail with reference to the accompanying drawings.

[0038] Overall system structure

[0039] Figure 1 This is a diagram schematically showing the structure of an overall system for generating a map for a robot according to one embodiment of the present invention.

[0040] like Figure 1 As shown, the overall system based on an embodiment of the present invention may include a communication network 100, a map management system 200 and a robot 300.

[0041] First, according to an embodiment of the present invention, the communication network 100 can be constructed regardless of the communication mode such as wired communication or wireless communication, and can be composed of a variety of communication networks such as a local area network (LAN), a metropolitan area network (MAN), and a wide area network (WAN). Preferably, the communication network 100 mentioned in the present invention can be the well-known Internet or the World Wide Web (WWW). However, the communication network 100 does not need to be limited to this, and can also include a well-known wired / wireless data communication network, a well-known telephone network, or a well-known wired / wireless television communication network in at least a part thereof.

[0042] For example, the communication network 100 may be a wireless data communication network, and in at least a portion thereof may implement existing communication methods such as WiFi communication, WiFi-Direct communication, Long Term Evolution (LTE) communication, Bluetooth communication (more specifically, Bluetooth Low Energy (BLE) communication), infrared communication, and ultrasonic communication.

[0043] Next, the map management system 200 based on an embodiment of the present invention is able to communicate with the robot 300 described later through the communication network 100, and is able to perform the following functions: obtain a basic map related to the task of the robot 300, identify pixels inferred to be mobile obstacles in the basic map based on at least one of the color of a specific pixel in the basic map and the size of an area related to the pixel, perform dilation operations and erosion operations on the pixels inferred to be mobile obstacles as objects, and determine the boundaries of the mobile obstacles based on polygons.

[0044] On the other hand, the map management system 200 is described as above, but this description is illustrative, and it is self-evident to those skilled in the art that at least a portion of the functions or structural elements required for the map management system 200 can be implemented in the robot 300 or an external system (not shown) described later, or can be included in the robot 300 or the external system, as required. In addition, all functions and all structural elements of the map management system 200 can be implemented in the robot 300, or can be included in the robot 300, as appropriate.

[0045] Next, the robot 300 according to an embodiment of the present invention is a device that can communicate with the map management system 200 through the communication network 100, and can autonomously perform a predetermined function or assigned task (e.g., serving food, recycling containers, etc.) without the user's operation, and can include at least one of a module (e.g., a grab, a manipulator module, etc.) for loading and unloading an object (e.g., a food tray), an image module (e.g., a visible light camera, an infrared camera, etc.) for obtaining surrounding images, and a driving module (e.g., a motor, etc.) for moving the robot 300. For example, such a robot 300 may be a robot having characteristics or functions similar to at least one of a guide robot, a transport robot, a cleaning robot, a medical robot, an entertainment robot, a pet robot, and a flying remote-controlled robot.

[0046] On the other hand, according to one embodiment of the present invention, the robot 300 may include an application for controlling the destination of the robot 300. Such an application may be downloaded from the map management system 200 or an external application distribution server (not shown).

[0047] Structure of map management system

[0048] The following describes the internal structure of the map management system 200 that performs important functions to realize the present invention and the functions of each component.

[0049] Figure 2 FIG. 2 is a diagram exemplarily showing the internal structure of a map management system 200 according to an embodiment of the present invention.

[0050] refer to Figure 2, the map management system 200 according to an embodiment of the present invention may include a basic map acquisition unit 210, a pixel recognition unit 220, a boundary determination unit 230, a noise removal unit 240, a boundary correction unit 250, a communication unit 260 and a control unit 270. According to an embodiment of the present invention, at least a part of the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240, the boundary correction unit 250, the communication unit 260 and the control unit 270 may be a program module for communicating with an external system (not shown). Such a program module may be included in the map management system 200 in the form of an operating system, an application module and other program modules, and may be physically stored in various well-known storage devices. In addition, such a program module may also be stored in a remote storage device capable of communicating with the map management system 200. On the other hand, such a program module includes a routine, a subroutine, a program, an object, a component, a data structure, etc. that executes a specific business described later or executes a specific virtual data type according to the present invention, but is not limited thereto.

[0051] First, the basic map acquisition unit 210 according to an embodiment of the present invention can perform the function of obtaining a basic map related to the task of the robot 300. According to an embodiment of the present invention, the basic map acquisition unit 210 can obtain the basic map as described above based on sensor modules such as image sensors, radar sensors, and lidar sensors that can be included in the robot 300 or can be interconnected, or obtain the basic map from an external system. A task according to an embodiment of the present invention can refer to a specific work, behavior, or action assigned to the robot 300. For example, the robot 300 can be assigned to move to a specific place, serve food, or recycle containers as a task.

[0052] For example, the basic map acquisition unit 210 can generate a basic map based on sensor information about at least one object (e.g., all types of objects such as people, animals, and objects) disposed outside the robot 300. More specifically, the basic map acquisition unit 210 can generate a two-dimensional or three-dimensional basic map centered on the robot based on image data obtained by photographing at least one object disposed outside using an image sensor, distance data obtained by measuring using at least one of a lidar sensor, a radar sensor, a laser sensor, and an ultrasonic sensor, and the like.

[0053] As another example, the basic map acquisition unit 210 may acquire a basic map related to a task of the robot 300 from an external system related to the basic map or another robot 300 .

[0054] Next, the pixel recognition unit 220 according to an embodiment of the present invention can recognize at least one of a pixel inferred as a mobile obstacle, a pixel inferred as noise, and a pixel inferred as a fixed obstacle in the basic map based on at least one of the color of a specific pixel in the basic map and the size of an area associated with the pixel. Here, a fixed obstacle may refer to an obstacle that cannot be moved to another place or is difficult to move to another place above a predetermined level, such as a building structure (e.g., a wall), and a mobile obstacle may refer to an obstacle that is easy to move to another place above a predetermined level, such as a table, a chair, and a dining table.

[0055] For example, according to an embodiment of the present invention, the color of a specific pixel in the basic map can be represented by grayscale according to the sensor information (specifically, the intensity of the sensed signal) obtained during the basic map generation process. According to this grayscale, the color of a specific pixel in the basic map can be roughly divided into white, gray and black. White pixels can represent an area where no object exists (i.e., an empty area), black pixels can represent an area where an object exists in reality, and gray pixels can represent an area where an object in reality may exist. In this case, when the size of an area (e.g., a closed area) formed by including (or connecting) adjacent gray pixels and black pixels (e.g., close to or existing within a predetermined level) in the basic map is less than the first critical size, the pixel identification unit 220 can identify the pixels in the area as pixels inferred to be noise. Here, this first critical size can be set, or can be determined by probability statistical analysis based on the number of areas formed by including adjacent gray pixels and black pixels in the basic map and the area of ​​the area. Furthermore, when the size of an area (e.g., a closed area) formed by including (or connecting) adjacent black pixels in the basic map is greater than the first critical size and less than the second critical size, the pixel recognition unit 220 can recognize the pixels in the area as pixels inferred to be mobile obstacles. Here, this second critical size can be set, or can be determined by probabilistic statistical analysis based on the number of areas formed by including adjacent black pixels in the basic map and the area of ​​the area. Furthermore, when the size of an area (e.g., a closed area) formed by including (or connecting) adjacent black pixels in the basic map is greater than the second critical size, the pixel recognition unit 220 can recognize the pixels in the area as pixels inferred to be fixed obstacles.

[0056] Next, the boundary determination unit 230 according to an embodiment of the present invention can perform a function of applying dilation and erosion operations to the pixels inferred as moving obstacles by the pixel recognition unit 220 and determining the boundary of the moving obstacle based on polygons.

[0057] For example, the boundary determination unit 230 can perform a dilation operation on pixels inferred to be the first moving obstacle and then perform an erosion operation to group at least one second moving obstacle that is within a predetermined distance from the first moving obstacle together with the first moving obstacle, and can identify the convex hull corresponding to the boundaries of the above-mentioned grouped multiple moving obstacles (i.e., the first moving obstacle and at least one second moving obstacle) based on a convex hull algorithm, and determine the boundaries of the above-mentioned grouped multiple moving obstacles based on the identified convex hull (for example, connecting the convex hulls to each other).

[0058] Next, the noise removal unit 240 according to an embodiment of the present invention can perform a function of removing pixels inferred as noise in the basic map by the pixel identification unit 220 .

[0059] For example, as described above, when the color of a specific pixel in the basic map is expressed in grayscale based on the sensor information (specifically, the intensity of the sensed signal) obtained during the basic map generation process, the noise removal unit 240 corrects the pixels inferred as noise by the pixel recognition unit 220 (for example, pixels within an area formed by containing (or connecting) adjacent gray pixels and black pixels in the basic map when the size of the area is less than the first critical size) to white pixels, thereby removing the noise.

[0060] Next, the boundary correction unit 250 according to an embodiment of the present invention can perform a function of correcting at least one of color and thickness of the boundary of the pixel inferred as a fixed obstacle in the basic map by the pixel recognition unit 220 .

[0061] For example, as described above, when the color of a specific pixel in the basic map is represented in grayscale based on the sensor information (specifically, the intensity of the sensed signal) obtained during the basic map generation process, the boundary correction unit 250 can correct the boundary of the pixel inferred as a fixed obstacle by the pixel recognition unit 220 (for example, the pixels within the area when the size of the area formed by including (or connecting) adjacent black pixels in the basic map is greater than the second critical size) to be black and have a thickness above a predetermined level.

[0062] Furthermore, the boundary correction unit 250 can dynamically determine the thickness of the boundary by referring to the size of the region specified by the pixels estimated to be the fixed obstacle.

[0063] For example, the boundary correction unit 250 can determine that the thickness of the boundary is thicker as the size of the region specified by the pixel estimated to be a fixed obstacle is larger, and can determine that the thickness of the boundary is thinner as the size of the region is smaller.

[0064] On the other hand, the boundary correction unit 250 according to an embodiment of the present invention can suppress pixels estimated as moving obstacles and selectively identify pixels estimated as fixed obstacles to perform correction.

[0065] For example, when the size of an area formed by including (or connecting) adjacent black pixels in the basic map is less than the second critical size, the boundary correction unit 250 suppresses the black pixels contained in the area (for example, disabling the layer related to the black pixels contained in the area or temporarily or temporarily removing them). When the size of an area formed by including (or connecting) adjacent black pixels in the basic map is greater than the second critical size, the above-mentioned correction can be performed after only selecting and specifically including the black pixels in the area.

[0066] Next, according to one embodiment of the present invention, the communication unit 260 is capable of executing the function of transmitting and receiving data from the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240 and the boundary correction unit 250 or to the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240 and the boundary correction unit 250.

[0067] Finally, according to an embodiment of the present invention, the control unit 270 can perform a function of controlling the data flow between the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240, the boundary correction unit 250, and the communication unit 260. That is, the control unit 270 according to an embodiment of the present invention can control the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240, the boundary correction unit 250, and the communication unit 260 to perform their own functions respectively by controlling the data flow from or to the outside of the map management system 200 or the data flow between the structural elements of the map management system 200.

[0068] Figure 3A and Figure 3B The following are diagrams showing by way of example a basic map according to an embodiment of the present invention and a conventional map processed by manual operation.

[0069] Figures 4 to 6 FIG. 1 is a diagram showing an example of a robot map generated according to an embodiment of the present invention.

[0070] refer to Figure 3A , Figure 3B and Figure 4 , if a basic map of the area around the robot 300 is obtained according to an embodiment of the present invention (reference Figure 3A), in the past, the only way was to directly apply the basic map, or manually remove the noise one by one and correct the boundaries of obstacles by eye (refer to Figure 3B ), so there are limitations in terms of speed or accuracy. However, through the map management system 200 based on the present invention, noise removal and boundary correction can be performed quickly, accurately and finely (refer to Figure 4 ).

[0071] First, a basic map related to the task of the robot 300 can be obtained according to an embodiment of the present invention.

[0072] Next, according to one embodiment of the present invention, pixels inferred to be noise in the basic map can be identified based on at least one of the color of a specific pixel in the basic map and the size of an area associated with the pixel, and the pixels inferred to be noise can be removed from the basic map.

[0073] Next, refer to Figures 4 to 6 According to one embodiment of the present invention, it is possible to identify pixels inferred to be fixed obstacles 420, 520, 620 based on at least one of the color of specific pixels and the size of an area associated with the pixels in a basic map from which pixels inferred to be noise are removed, and to use the boundary pixels of the inferred fixed obstacles 420, 520, 620 as objects to correct at least one of color and thickness.

[0074] Next, according to an embodiment of the present invention, pixels inferred to be mobile obstacles can be identified based on at least one of the color of specific pixels and the size of an area associated with the pixels in a basic map in which the boundary pixels of the fixed obstacles 420, 520, 620 are corrected, and dilation and erosion operations are applied to the pixels inferred to be mobile obstacles as objects, and the boundaries 410, 510, 610 of the mobile obstacles are determined based on polygons 401, 402, 403, 404, 405, 406, 501, 502, 503, 504, 505, 506, 507, 508, 601, 602, 603, 604.

[0075] Next, according to an embodiment of the present invention, at least one of the boundaries 410, 510, 610 of the moving obstacle and the coordinate data related to the boundaries 410, 510, 610 (for example, the reference coordinate system related to the above-mentioned basic map, the reference coordinate system of the robot 300 or the coordinate data in the absolute coordinate system) can be generated as annotation information of the above-mentioned basic map.

[0076] Next, according to an embodiment of the present invention, the robot 300 can perform tasks according to the corrected basic map and the annotation information of the basic map.

[0077] On the other hand, when it is necessary to correct the boundaries 410, 510, 610 of the moving obstacles afterwards, they can be easily adjusted based on the polygons 401, 402, 403, 404, 405, 406, 501, 502, 503, 504, 505, 506, 507, 508, 601, 602, 603, 604.

[0078] On the other hand, in order to generate a map for a robot according to the present invention, various computer programs or tools such as the OpenCV library functions of the Python programming language can be applied. For example, functions such as threshold(), connectedcomponentswithstats(), canny(), houghlinesp(), erode(), dilate(), findcontours(), convexhull(), and approxpolydp() of the Python programming language can be utilized.

[0079] Furthermore, the above mainly describes an embodiment of sequentially performing noise removal, boundary correction of fixed obstacles, and boundary determination of moving obstacles, but the present invention is not limited to the above sequence. It should be noted that the above sequence can be arbitrarily changed to other sequences within the scope of achieving the purpose of the present invention.

[0080] The structure of the robot

[0081] The robot 300 based on one embodiment of the present invention can be a robot that performs work similar to that performed by at least one of a guide robot, a serving robot, a transport robot, a cleaning robot, a medical robot, an entertainment robot, a pet robot and a flying remote-controlled robot. To this end, it can be implemented in a variety of shapes that suit each task.

[0082] refer to Figure 7 The robot 300 may be configured to include a main body 710 , a driving unit 720 , and a processor 730 .

[0083] For example, the main body 710 according to an embodiment of the present invention may include at least one loading space for loading a transport object or a recycling object. Such transport objects and recycling objects according to an embodiment of the present invention are concepts that collectively refer to all movable tangible objects, such as objects, animals, and people. For example, the transport object may be food, and the recycling object may be a container containing the food.

[0084] refer to Figure 8For example, when the robot 300 is a serving robot, it may include a first space 810 for providing a transport object and a second space 820 for recovering a target object. In addition, the robot 300 may also include an expansion space, i.e., a third space 830, provided by a detachable support. Further expansion spaces may be added as needed, thereby providing more loading space. In addition, the robot 300 may also include a tray 840 dedicated to transporting a target object or recovering a target object. For example, when viewed from above, the tray 840 may have a structure with a plurality of circular grooves formed on its upper surface. Each circular groove may be formed so that the lower part of a cup containing a beverage is stably placed and easily fixed to a certain extent. The size of such a circular groove may be varied. In addition, the lower part of the first space 810 of the robot 300 may also include a fourth space 850 that can be taken out through the side of the robot 300. The fourth space 850 based on an embodiment of the present invention may have a basket-like shape with an empty space formed inside, a blocked side, an open upper surface, and a blocked bottom surface. However, the loading space of the robot 300 according to the present invention is not limited to the above-mentioned contents, and can be variously realized in the form of loading spaces of different forms within the scope that can achieve the purpose of the present invention.

[0085] On the other hand, refer again Figure 7 The main body 710 may also include a photographic module (eg, a visible light camera, an infrared camera, etc.) (not shown) and a scanner module (eg, a lidar sensor, etc.) for obtaining surrounding images, object and obstacle information.

[0086] Next, the driving unit 720 according to an embodiment of the present invention may be configured to include a module for moving the main body 710 to another location or a module for loading and unloading a transport target object and a recovery target object.

[0087] For example, the driving unit 720 may include modules related to wheels, propellers, etc. driven electrically, mechanically or hydraulically as modules for moving the main body 710 to other locations, and may include a manipulator module for installing and moving transport objects and recycling objects as modules for loading and unloading transport objects and recycling objects.

[0088] Next, the processor 730 based on an embodiment of the present invention can be electrically connected to the driving unit 720 to perform the function of controlling the driving unit 720 (it can also include a communication module for communicating with an external system), for example, it can refer to a data processing device built into the hardware, which has a physically structured circuit for executing the function represented by the code or command contained in the program. For example, the data processing device built into the hardware can include a microprocessor, a central processing unit, a processor core, a multiprocessor, an application-specific integrated circuit (ASIC (application-specific integrated circuit)), a field programmable gate array (FPGA (field programmable gate array)) and other processing devices.

[0089] Furthermore, the processor 730 is capable of executing at least one function of the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240 and the boundary correction unit 250 of the map management system 200 based on the present invention (for example, the function can be modularized and included in the processor 730), and is also capable of executing the function of controlling the driving unit 720 by communicating with an external system (not shown) that executes at least one function of the basic map acquisition unit 210, the pixel recognition unit 220, the boundary determination unit 230, the noise removal unit 240 and the boundary correction unit 250.

[0090] Specifically, the processor 730 is capable of performing the following functions: obtaining a basic map related to the task of the robot 300, identifying pixels inferred to be mobile obstacles in the basic map based on the color of a specific pixel in the basic map and at least one of the size of an area related to the pixel, applying dilation and erosion operations to the pixels inferred to be mobile obstacles as objects, and determining the boundaries of the mobile obstacles based on polygons.

[0091] The above-described embodiments based on the present invention can be implemented and recorded in a computer-readable recording medium in the form of program instructions that can be executed by a variety of computer structural elements. The computer-readable recording medium may include a separate program instruction, a data file, a data structure, etc., or a combination of program instructions, data files, data structures, etc. The program instructions recorded in the computer-readable recording medium may be program instructions specially designed and constructed for the present invention, or may be program instructions known and usable by technicians in the field of computer software. Examples of computer-readable recording media include hardware devices specially constructed to store and execute program instructions, such as: magnetic media such as hard disks, floppy disks and tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and ROMs, RAMs, flash memories, etc. Examples of program instructions include, in addition to machine language codes made by compilers, high-level language codes that can be executed by computers using interpreters, etc. Hardware devices can be changed to more than one software module in order to perform processing based on the present invention, and vice versa.

[0092] The present invention has been described above using specific matters such as specific structural elements and limited embodiments and drawings, but this is only provided to help a more comprehensive understanding of the present invention and does not limit the present invention to the above-mentioned embodiments. Anyone with common knowledge in the technical field to which the present invention belongs can make various modifications and changes based on this description.

[0093] Therefore, the concept of the present invention should not be limited to the above-described embodiments, and it can be said that not only the claims described below, but also all scopes equivalent to the claims or equivalently modified therefrom belong to the scope of the concept of the present invention.

Claims

1. A method for generating a map for a robot, in, include: Steps to obtain a basic map relevant to the robot's task; A step of identifying pixels inferred as moving obstacles, pixels inferred as noise, and pixels inferred as fixed obstacles in the basic map based on the color of a specific pixel in the basic map and the size of an area associated with the pixel; and The step of performing dilation and erosion operations on pixels inferred as the moving obstacle and determining the boundary of the moving obstacle based on a polygon, In the obtaining step, the basic map is generated based on sensing information obtained by sensing at least one object disposed outside the robot. The color of a specific pixel in the basic map is represented in grayscale according to the sensor information, so that white pixels are associated with areas where no object exists, black pixels are associated with areas where objects exist in reality, and gray pixels are associated with areas where objects may exist in reality, In the identification step, pixels within an area formed by connecting adjacent gray pixels and black pixels and having a size less than a first critical size are identified as pixels inferred to be noise, pixels within an area formed by connecting adjacent black pixels and having a size greater than the first critical size and less than a second critical size are identified as pixels inferred to be mobile obstacles, and pixels within an area formed by connecting adjacent black pixels and having a size greater than the second critical size are identified as pixels inferred to be fixed obstacles.

2. The method for generating a robot map according to claim 1, wherein: Also included is the step of removing pixels inferred as said noise in said base map.

3. The method for generating a robot map according to claim 1, wherein: The method further includes the step of correcting at least one of the color and thickness of the boundary of the pixel inferred as the fixed obstacle in the basic map.

4. The method for generating a robot map according to claim 3, wherein: In the correction step, pixels estimated to be the moving obstacle are suppressed for the correction.

5. The method for generating a robot map according to claim 1, wherein: In the determination step, the dilation operation and the erosion operation are sequentially performed on the pixels estimated to be the moving obstacles.

6. The method for generating a robot map according to claim 1, wherein: In the determining step, at least one other moving obstacle existing within a predetermined distance from the moving obstacle is grouped together with the moving obstacle, and a boundary of the grouped moving obstacles is determined, thereby determining the boundary of the moving obstacle.

7. The method for generating a robot map according to claim 1, wherein: In the determining step, a convex hull associated with the moving obstacle is identified based on a convex hull algorithm, and a boundary of the moving obstacle is determined based on the identified convex hull.

8. A non-transitory computer-readable recording medium, wherein: A computer program for executing the method according to claim 1 is recorded.

9. A system for generating a map for a robot, in, include: A basic map acquisition unit, which acquires a basic map related to the robot's task; a pixel identifying unit that identifies pixels inferred to be moving obstacles, pixels inferred to be noise, and pixels inferred to be fixed obstacles in the basic map based on the colors of specific pixels in the basic map and the size of an area associated with the pixels; and The boundary determination unit performs a dilation operation and an erosion operation on the pixels estimated to be the moving obstacle, and determines the boundary of the moving obstacle based on the polygon. The basic map acquisition unit is configured to generate the basic map based on sensing information obtained by sensing at least one object disposed outside the robot. The color of a specific pixel in the basic map is represented in grayscale according to the sensor information, so that white pixels are associated with areas where no object exists, black pixels are associated with areas where objects exist in reality, and gray pixels are associated with areas where objects may exist in reality, The pixel identification unit is configured to identify pixels within an area formed by connecting adjacent gray pixels and black pixels and having a size less than a first critical size as pixels inferred to be noise, identify pixels within an area formed by connecting adjacent black pixels and having a size greater than the first critical size and less than a second critical size as pixels inferred to be mobile obstacles, and identify pixels within an area formed by connecting adjacent black pixels and having a size greater than the second critical size as pixels inferred to be fixed obstacles.

10. The system for generating a robot map according to claim 9, wherein: The method further includes a noise removing unit that removes pixels inferred as the noise in the basic map.

11. The system for generating a robot map according to claim 9, wherein: The system further includes a boundary correction unit configured to correct at least one of a color and a thickness of a boundary of a pixel inferred as the fixed obstacle in the basic map.

12. The system for generating a robot map according to claim 11, wherein: The boundary correction unit suppresses pixels estimated to be the moving obstacle for the correction.

13. The system for generating a robot map according to claim 9, wherein: The boundary identification unit sequentially performs the dilation operation and the erosion operation on pixels estimated as the moving obstacle.

14. The system for generating a robot map according to claim 9, wherein: The boundary determination unit groups at least one other moving obstacle existing within a predetermined distance from the moving obstacle together with the moving obstacle and determines a boundary of the grouped moving obstacles, thereby determining the boundary of the moving obstacle.

15. The system for generating a robot map according to claim 9, wherein: The boundary determination unit identifies a convex hull associated with the moving obstacle based on a convex hull algorithm, and determines a boundary of the moving obstacle based on the identified convex hull.

Images