KR102041664B1 - Method and apparatus for estimating localization of robot in wide range of indoor space using qr marker and laser scanner - Google Patents

Abstract

The apparatus for recognizing the position of the robot estimates the position of the robot by obtaining information about at least one QR marker disposed at a preset position and information sensed by at least one sensor installed in the robot.

Description

TECHNICAL AND APPARATUS FOR ESTIMATING LOCALIZATION OF ROBOT IN WIDE RANGE OF INDOOR SPACE USING QR MARKER AND LASER SCANNER}

The following embodiments relate to a method and apparatus for recognizing a robot's position in a large indoor space, and more particularly, to a method and apparatus for recognizing a robot's position in a large indoor space using a QR marker and a laser scanner. will be.

Industrial robots perform the same tasks repeatedly in fixed locations, and field robots perform various tasks in different environments. These field robots are emerging as robot technology has advanced recently.

Since field robots perform tasks in various environments, there are many difficulties for robot automation. One of the major issues with the automation of field robots is to solve the robot's self-awareness problem.

A technique using a laser scanner for location recognition has been proposed. However, in the case of acquiring scanned information about a wall including a reflective material such as glass, it is difficult to accurately obtain information about the target from the scanned information, and it is difficult to obtain the information of the reference data of the created map due to dynamic objects or changes in the environment. The change causes a decrease in location awareness reliability.

In addition, a QR code attached to a specific object or a specific location may provide various information about the object or the corresponding location. However, to properly recognize the QR code, the camera needs to be fixed. In addition, in order for the robot on which the camera is installed to properly recognize the QR code, the moving speed of the robot may be limited.

A method and apparatus for recognizing a location of a robot according to an embodiment of the present invention recognizes a location of a robot more accurately by using information on a QR marker disposed in an indoor space and information sensed by a sensor installed in the robot.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention, despite the robot traveling at a high speed by using information about the QR marker disposed in the indoor space and information sensed by the sensor installed in the robot , More accurately recognize the position of the robot.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention improves the position recognition of the robot by generating a graph using the information about the QR marker and the information sensed by the sensor installed in the robot as constraints. .

In addition, the robot's position recognition method and apparatus according to an embodiment of the present invention realizes more accurate position recognition by using a process of generating a map by optimizing the generated graph.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention more accurately estimates the position of the robot by matching the map scanned with the laser scanner.

According to one or more exemplary embodiments, a method of recognizing a location of a robot includes: obtaining information about at least one QR marker disposed at a preset location of an indoor space and information sensed by at least one sensor installed in the robot; And estimating the position of the robot using information on the QR marker and information sensed by the at least one sensor installed in the robot.

A plurality of bezels may be disposed at corners of the QR marker, and each of the plurality of bezels may have a different thickness.

The method of recognizing a position of the robot may include determining an orientation of the robot based on the thickness of each of the bezels recognized by the robot.

The estimating of the position of the robot may include generating a graph using information on the QR marker and information sensed by the at least one sensor installed in the robot as a constraint condition; And optimizing the graph to estimate the position of the robot.

The estimating of the position of the robot may include generating a map corresponding to the indoor space using the optimized graph; And matching the scanned data with the generated map using a laser scanner.

The location recognition method of the robot may further include filtering the estimated location of the robot using a particle filter.

The at least one sensor installed in the robot may include at least one of an odometry, a gyroscope, a light detection and ranging (LIDAR), or a camera.

An apparatus for recognizing a position of a robot according to an embodiment of the present invention includes an acquisition unit for obtaining information about at least one QR marker disposed at a preset position of an indoor space and information sensed by at least one sensor installed in the robot; And an estimator for estimating the position of the robot using the information on the QR marker and the information sensed by the at least one sensor installed in the robot.

A plurality of bezels may be disposed at corners of the QR marker, and each of the plurality of bezels may have a different thickness.

The position recognition apparatus of the robot may include an orientation determination unit that determines an orientation of the robot based on the thickness of each of the bezels recognized by the robot.

The estimator optimizes the graph to estimate the position of the graph generator and the robot to generate a graph using information on the QR marker and information sensed by the at least one sensor installed in the robot as constraint conditions. It may include an optimization unit.

The estimator may include a map generator that generates a map corresponding to the indoor space by using the optimized graph, and a matching unit that matches the generated map with the scanned data using a laser scanner.

The location recognition apparatus of the robot may further include a filter for filtering the estimated location of the robot using a particle filter. At least one sensor installed in the robot may include at least one of an odometry, a gyroscope, a light detection and ranging (LIDAR), or a camera.

The method and apparatus for recognizing a location of a robot according to an embodiment of the present invention may recognize a location of a robot more accurately by using information about a QR marker disposed in an indoor space and information sensed by a sensor installed in the robot.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention, despite the robot traveling at a high speed by using information about the QR marker disposed in the indoor space and information sensed by the sensor installed in the robot Therefore, the position of the robot can be recognized more accurately.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention improve the method for recognizing the position of the robot by generating a graph using the information about the QR marker and the information sensed by the sensor installed in the robot as constraints. can do.

In addition, the robot location recognition method and apparatus according to an embodiment of the present invention can realize a more accurate location recognition by using a process of generating a map by optimizing the generated graph.

In addition, the method and apparatus for recognizing the position of the robot according to an embodiment of the present invention may more accurately estimate the position of the robot by matching the information scanned by the laser scanner with a map.

1 illustrates a position recognition system of a robot including a robot moving in an indoor space and a position recognition apparatus of the robot according to an embodiment of the present invention.
2 is a flowchart illustrating a method of recognizing a position of a robot according to an embodiment of the present invention.
3 is a diagram schematically illustrating a process of executing a mapping algorithm by using information of sensors installed in a robot according to an embodiment of the present invention.
4 shows an example of a QR marker disposed in an indoor space according to an embodiment of the present invention.
5 exemplarily illustrates an execution result of a marker-based map and mapping algorithm generated according to an embodiment of the present invention.
6 is a diagram schematically illustrating a process of executing a position recognition algorithm for estimating a position result of a robot according to an embodiment of the present invention.
7 is a view showing a laser scanner disposed in the indoor space according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Although limited embodiments are described below, these embodiments are examples of the invention, and those skilled in the art can easily change these embodiments.

1 illustrates a position recognition system of a robot including a robot moving in an indoor space and a position recognition apparatus of the robot according to an embodiment of the present invention.

Referring to FIG. 1, a plurality of QR markers 110 are disposed at preset locations in the indoor space 100. In addition, the robot 120 may move in the indoor space 100, and the position recognizing apparatus 101 estimates the position of the robot 120 in real time. Interior space 100 is based on a wide range.

The position recognizing apparatus 101 includes an acquirer 130, an orientation determiner 140, an estimator 150, a graph generator 151, an optimizer 152, a map generator 153, and a matcher 154. ) And a filter 160.

The acquirer 130 acquires information about the plurality of QR markers 110 disposed at a preset position of the indoor space 100 and information sensed by at least one sensor installed in the robot 120. In this case, the sensor may include at least one of an odometry, a gyroscope, a light detection and ranging (LIDAR), or a camera.

In this case, the information on the plurality of QR markers 110 may include location information of each of the QR markers 110, and the location recognizing apparatus 101 based on the location information of each of the QR markers 110. The position of the robot 120 can be estimated. However, when the robot 120 moves, the plurality of QR markers 110 may not be well recognized by the robot 120. Each of the plurality of QR markers 110 according to an embodiment of the present invention may be designed to increase the recognition rate of the robot 120, which will be described again with reference to FIG. 4.

In addition, the orientation determination unit 140 may determine the orientation of the robot using the information on the QR marker 110. Detailed description of the orientation determination will be described again with reference to FIG. 4.

The estimator 150 may estimate the position using information about the plurality of QR markers 110 and information sensed by at least one sensor installed in the robot 120. The estimator 150 includes a graph generator 151, an optimizer 152, a map generator 153, and a matcher 154.

The graph generator 151 generates a graph to estimate the position of the robot 120. For example, the position of the robot 120 may be estimated based on the position information of each of the QR markers 110 and the information sensed by an odometry sensor installed in the robot 120. Here, the Odometry estimates the displacement of the robot 120 using the data obtained by the motion sensor. The graph generator 151 generates a graph using the displacement information of the robot 120 and the position information of each of the plurality of QR markers 110 as constraint conditions. Constraints prevent the unrealistic values from being obtained due to numerical problems such as lack of coefficients or poor conditions. A detailed description of the process of generating the graph is described with reference to FIG. 3.

In addition, the optimizer 152 performs an optimization process on the generated graph. The map generator 153 generates a map using the optimized graph. The map generator 153 registers a few initial QR markers 110 using the optimized graph and automatically registers the remaining markers to generate a map quickly and efficiently. A detailed description of the process of generating the map is described with reference to FIG. 5.

The matching unit 154 matches the generated map with the data scanned by the laser scanner. The position of the robot 120 may be estimated based on the matched scanned data and map. The data scanned in the laser scanner may be provided from the robot 120. A detailed description of the laser scanner is given in FIG. 7. A detailed description of the process of matching the two data is described in detail with reference to FIG. 6.

In addition, the filter 160 filters the position result using the particle filter. Resampling of the particle filter occurs at the same time as obtaining information about the plurality of QR markers (110).

2 is a flowchart illustrating a method of recognizing a position of a robot according to an embodiment of the present invention.

Referring to FIG. 2, the method for recognizing a location according to an embodiment of the present invention includes information on a plurality of QR markers 110 disposed at a preset location of an indoor space 100 and at least one installed in the robot 120. Information sensed by the sensor is obtained (S210).

At this time, the acquisition of information on the plurality of QR markers 110 and the information sensed by at least one sensor installed in the robot 120 may be performed sequentially but may be performed in parallel.

In addition, the orientation determination of the robot 120 uses information on the plurality of QR markers 110 (S221). In this case, the orientation determination of the robot 120 may be performed in parallel with or sequentially obtained with information about the plurality of QR markers 110 and information sensed by at least one sensor installed in the robot 120. have.

The location recognizing apparatus 101 generates a graph based on information about the plurality of QR markers 110 and information sensed by at least one sensor installed in the robot 120, and optimizes the generated graph (S222). , S230). Step S241 generates a map using the optimized graph. In operation S250, the generated map may be matched with the data scanned by the laser scanner to estimate a location. Data scanned by the laser scanner may be received from the robot 120 (S242).

For example, the coordinates of the robot 120 having the map and the coordinates of the data scanned by the laser scanner may compensate for the error of the respective data. The error of the position information of the robot 120 having the map may occur when the light environment is vulnerable in the QR code recognition and camera recognition process due to the fast moving speed of the robot 120. Errors in the data scanned with the laser scanner may be due to environmental changes. Therefore, matching of the generated map and the data scanned by the laser scanner is made in a direction to compensate for each error.

In operation S270, the estimated location using the matched data may be filtered using a particle filter.

3 is a diagram schematically illustrating a process of executing a mapping algorithm by using information of sensors installed in a robot according to an embodiment of the present invention.

The location recognition apparatus 101 of the robot 120 obtains information about the QR marker 110 and information sensed by the sensor of the robot 120 to derive a mapping result 340 through the mapping algorithm 330. . The mapping algorithm 330 may include a graph generator 151, an optimizer 152, and a map generator 153. The graph generator 151 generates a graph using information about the QR marker 110 and information sensed by the sensor of the robot 120.

For example, when the robot 120 approaches a plurality of QR markers 110 above a predetermined level, information about the robot 120 may be obtained. The graph generator 151 may coordinate the location information of each of the QR markers 110 with (x1, y1), (x2, y2), (x3, y3), and the like. In addition, the graph generator 151 may coordinate the later position with respect to the initial position of the robot 120 using the displacement information of the robot 120 obtained by the odometry. The graph generator 151 may generate a graph by comparing the position information of each of the coordinated QR markers 110 and the coordinate values of the initial position and the later position of the robot 120. For example, the graph generator 151 may generate coordinates by comparing the position of the robot 120 with the QR marker 110 having (x1, y1) as a reference point as a reference point. The graph generator 151 may express the coordinates of the graph as (x1, y1), (x1 + X1, y1 + Y1), or the like. The graph generator 151 may repeat this example with markers having different reference coordinates. The graph generator 151 may generate a graph by identifying the relative positions of the robots 120 with respect to the plurality of QR markers 110 by using repetitive coordinates.

The optimizer 152 enables the realization of more accurate information about the position of the robot 120. For example, the optimizer 152 may improve the position estimation of the robot 120 by calculating the maximum and minimum values that the graph generated in the above-described constraint condition may have.

The optimized graph is used to generate a map in the map generator 153. A detailed description of the process of generating the map is described with reference to FIG. 5.

4 shows an example of a QR marker disposed in an indoor space according to an embodiment of the present invention.

The QR code 420 is disposed at the center of the QR marker 110, and a plurality of bezels 420 are disposed at corners of the QR marker 110. QR code 420 may be a minimum size code that can be recognized when the robot moves at a speed of 1m / s. The bezel 420 refers to an edge disposed at each corner of each of the plurality of QR markers 110. Each of the bezels 420 may have a different thickness.

The position recognition device 101 of the robot 120 may determine the orientation of the robot 120 based on the thickness of each of the bezels 420. For example, assuming that the upper bezel 420 is 2CM and the lower bezel 420 is 1CM, when the robot 120 recognizes that the lower bezel 420 is closer to the upper bezel 420 than the lower bezel 420. That is, when the robot 120 recognizes that it is closer to the thick bezel 420 than the thin bezel 420, it may be estimated that the orientation of the robot 120 is located at the top. For example, assuming that the bezel 420 on the left side is 2CM and the bezel 420 on the right side is 1CM, the robot 120 recognizes that the bezel 420 is closer to the left side than the bezel 420 on the right side. If the robot 120 recognizes that the robot 120 is closer to the thick bezel 420 than the thin bezel 420, it may be estimated that the orientation of the robot 120 is located on the left side.

5 exemplarily illustrates an execution result of a marker-based map and mapping algorithm generated according to an embodiment of the present invention.

5 illustrates a result generated when the location recognizing apparatus 101 executes the mapping algorithm 330. The mapping algorithm 330 uses the SLAM technique. The simultaneous localization and mapping (SLAM) technique is a technique in which mapping occurs while simultaneously obtaining location information of each of the plurality of QR markers 110. The SLAM technique is understood similarly to the chicken and egg problem. For example, the mapping algorithm 330 uses location information of each of the plurality of QR markers 110 to generate a graph, and simultaneously maps location information of each of the plurality of QR markers 110 using the graph. In addition, it may be used to estimate the position of each of the plurality of QR markers 110. The figure on the left illustrates the marker-based map generated at this time. The figure on the right illustrates the optimized graph produced at this time.

6 is a diagram schematically illustrating a process of executing a position recognition algorithm 610 for estimating the position result of the robot 120 according to an embodiment of the present invention.

6 illustrates a process of generating a position recognition algorithm 610 using a matching unit 154 that matches the information scanned by a laser scanner with a mapping result 340 generated using the mapping algorithm 330. . The location recognition algorithm 610 may estimate the location result 620. The matching unit 154 may match the feature points of the mapping result 340 generated using the information scanned by the laser scanner and the mapping algorithm 330. The matching unit 154 extracts the feature points of the information scanned by the laser scanner and extracts the feature points of the mapping result 340. The matching unit 154 may match the feature points of the information scanned by the laser scanner with the feature points of the mapping result 340. The location recognition algorithm 610 may estimate the location with the matched result.

7 is a view showing a laser scanner disposed in an indoor space according to an embodiment of the present invention.

The laser scanner is a device that informs the absolute distance between the robot 120 and the surrounding object to estimate the position of the robot 120. Laser scanning is used mainly by two definitions in modern engineering. The first definition, which is considered more general, uses the principle of returning with limited laser beam reflection. The position of the object may be estimated by determining whether the laser beam output from the laser is reflected from the object and returned to the laser to be recognized or impossible. The second definition allows for more detailed object recognition compared to the first definition. It measures the distance to the object by outputting the laser in all directions. At this time, the laser scanner uses the principle that the flight time between the reference signal and the laser light signal reflected back to the object changes with distance. The above invention uses the definition of the second laser scanning. The location recognizing apparatus 101 may coordinate the data scanned by the laser scanner. The data scanned with the laser scanner may refer to the relative position of the robot 120 relative to the object. The data scanned with the laser scanner is used to estimate the position of the robot 120. Laser scanning can be applied in various ways such as autonomous vehicle driving, shipyard loading station automation, civil engineering / building environment, distance / area, digital 3D shape for restoration of cultural property, and virtual game.

The system or apparatus described above may be implemented with hardware components, software components, and / or combinations of hardware components and software components. For example, the systems, devices, and components described in the embodiments may include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable arrays (FPAs). ), A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of explanation, one processing device may be described as being used, but one of ordinary skill in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and configure the processing device to operate as desired, or process it independently or collectively. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. Or may be permanently or temporarily embodied in a signal wave to be transmitted. The software may be distributed over networked computer systems so that they may be stored or executed in a distributed manner. Software and data may be stored on one or more computer readable recording media.

The method according to the embodiments may be embodied in the form of program instructions that may be executed by various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described by the limited embodiments and the drawings as described above, various modifications and variations are possible to those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components. Or even if replaced or substituted by equivalents, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the claims that follow.

Claims (14)

Obtaining information about at least one QR marker disposed at a preset position of the indoor space and information sensed by at least one sensor installed in the robot; And
Estimating a position of the robot using information on the QR marker and information sensed by the at least one sensor installed in the robot,
A plurality of bezels are arranged at the corners of the QR marker, and further comprising the step of determining the orientation of the robot based on each of the plurality of bezels having different thicknesses,
Determining the orientation of the robot
Characteristic of the plurality of QR markers and the information sensed by at least one sensor installed in the robot characterized in that to perform in parallel,
Estimating the position of the robot
Optimizing the graph to estimate the position of the robot;
Generating a map corresponding to the indoor space using the optimized graph; And
Matching the scanned data with the generated map using a laser scanner,
Matching the generated map is
The position of the robot is estimated by matching the generated map with the scanned data by compensating an error between the coordinates of the robot included in the generated map and the coordinates included in the scanned data.
Extracting the feature points of the scanned information by extracting feature points of the scanned information, and matching the feature points of the scanned information with the feature points of the mapping result to estimate the position of the robot. How to recognize your location. delete delete delete delete The method of claim 1,
Filtering the estimated position of the robot using a particle filter
How to recognize the position of the robot in the indoor space further comprising. The method of claim 1,
At least one sensor installed in the robot
A method of recognizing a robot's position in an indoor space including at least one of an odometry, a gyroscope, a light detection and ranging (LIDAR), or a camera. An acquisition unit for obtaining information about at least one QR marker disposed at a preset position of the indoor space and information sensed by at least one sensor installed in the robot; And
Including an estimation unit for estimating the position of the robot using the information on the QR marker and the information sensed by the at least one sensor installed in the robot,
A plurality of bezels are disposed at the corners of the QR marker, and further includes an orientation determination unit for determining the orientation of the robot based on each of the plurality of bezels having different thicknesses.
The orientation determination unit
Characteristic of the plurality of QR markers and the information sensed by at least one sensor installed in the robot characterized in that to perform in parallel,
The estimating unit
An optimizer for optimizing the graph to estimate the position of the robot;
A map generator for generating a map corresponding to the indoor space by using the optimized graph; And
A matching unit matching the scanned data with the generated map using a laser scanner,
The matching part
The position of the robot is estimated by matching the generated map with the scanned data by compensating an error between the coordinates of the robot included in the generated map and the coordinates included in the scanned data.
Extracting the feature points of the scanned information by extracting feature points of the scanned information, and matching the feature points of the scanned information with the feature points of the mapping result to estimate the position of the robot. Location-aware device. delete delete delete delete The method of claim 8,
Filter for filtering the estimated position of the robot using a particle filter
Apparatus for recognizing the position of the robot in the indoor space further comprising. The method of claim 8,
At least one sensor installed in the robot
A device for recognizing a robot's position in an indoor space including at least one of an odometry, a gyroscope, a light detection and ranging (LIDAR), or a camera.

Images