System, Location Information Management Device, Location Identification Method, and Program

Abstract

[Problem] To provide a system for correctly identifying a mobile object. [Solution] This system includes a mobile object and a position information management device. The mobile object is provided with a top board with a light emission unit laid thereon. The position information management device extracts, from an image in which the mobile object is captured, a first image including the light emission unit. The position information management device detects the presence of the mobile object by executing image processing on the first image, and identifies the position of the mobile object. The image processing may be processing for eliminating noise included in the region corresponding to the light emission unit in the first image.

Description

Technical Field

[0001] The present invention relates to a system, a position information management device, a position specifying method, and a program.

Background Art

[0002] In a production site such as a factory, the movement of articles such as parts and materials to be used is essential. Also, in a logistics warehouse, the movement of articles is required. For these article movements, a transport robot (AGV; Automated Guided Vehicle) is utilized.

[0003] Patent Document 1 describes providing a LIBS type object sorting device that performs LIBS (Laser-Induced Breakdown Spectroscopy) analysis while conveying an object to be sorted on a conveyor and performs sorting based on this. Patent Document 1 discloses a technique for sorting an object by irradiating a laser beam onto an object being conveyed on a conveyor and analyzing the wavelength of the reflected light. In the technique disclosed in Patent Document 1, the position of the object is specified using a camera, and the falling position of the object is adjusted by irradiating the object with a laser.

[0004] Patent Document 2 describes detecting the position and posture of a moving body moving on a moving surface with low contrast and the shape of the moving surface in an environment with low illuminance. In Patent Document 2, a dot-shaped light source is provided on the upper part of the moving body, and the light source is used as a marker for position and posture detection. In the technique disclosed in Patent Document 2, the moving body is imaged using a stereo camera, and unnecessary objects such as walls and cables are removed using a distance sensor.

[0005] Patent Document 3 describes realizing a movable object position detection system that can easily recognize the position and orientation of a movable object such as a robot. In Patent Document 3, the position of a light-emitting element captured by a camera is specified, the position of the light-emitting element is converted from the camera coordinate system to the absolute coordinate system, and the position of the object is specified. In Patent Document 3, a plurality of light-emitting elements are arranged on a movable object (mobile body), and based on the unique light-emitting pattern by the plurality of light-emitting elements, the coordinates of the object are specified and the object is identified.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0007] As described above, a robot (transport robot) may be used for transporting articles. Here, as forms of article transport by a transport robot, a type in which the transport robot autonomously moves along a transport path and a type in which a control device that communicates with the transport robot remotely controls the transport robot are conceivable.

[0008] In the latter case, it is necessary for the control device to grasp the position of the transport robot. At that time, as disclosed in Patent Documents 2 and 3, it is conceivable to use a light source (light-emitting element). Specifically, a light source is provided on the top plate or the like of the transport robot, and a camera device attached to the ceiling images the transport robot. The control device acquires image data from the camera device and calculates the position of the transport robot by analyzing the image data.

[0009] For example, the control device needs to identify and locate the transport robot that is holding an article in the image data. In this case, depending on the positional relationship between the article to be transported, the cart on which the article is placed, and the camera, there is a possibility that the control device cannot identify (extract) the transport robot from the image data.

[0010] For example, as shown in FIG. 21, if the height of the article to be transported (or the cart on which the article is placed) is not very high, the top plate of the transport robot can be included in the camera's field of view. On the other hand, as shown in FIG. 22, when the height of the article to be transported is high, the field of view of the camera may be blocked by the article, and only a part of the transport robot may be captured in the image data.

[0011] Specifically, when an article is placed on a cage cart and the transport robot transports the entire cage cart, the frame of the cage cart may hide a part of the top plate of the transport robot. In the image data captured in such a situation, the top plate of the robot is divided into a plurality of regions, and the control device cannot correctly recognize the transport robot.

[0012] The main object of the present invention is to provide a system, a position information management device, a position identification method, and a program that contribute to accurately identifying a moving body.

Means for Solving the Problem

[0013] According to a first aspect of the present invention, there is provided a system including: a moving body having a top plate on which a light emitting unit is laid; a position information management device that extracts a first image including the light emitting unit from an image of the moving body and detects the presence of the moving body by performing image processing on the first image and identifies the position of the moving body.

[0014] According to a second aspect of the present invention, from an image of a moving body provided with a top plate on which a light emitting part is laid, a first image including the light emitting part is extracted, and the presence of the moving body is detected by performing image processing on the first image, and a position information management device for specifying the position of the moving body is provided.

[0015] According to a third aspect of the present invention, in a position information management device, a step of extracting a first image including the light emitting part from an image of a moving body provided with a top plate on which the light emitting part is laid, a step of detecting the presence of the moving body by performing image processing on the first image, and a step of specifying the position of the moving body are included, and a position specifying method is provided.

[0016] According to a fourth aspect of the present invention, a program is provided that causes a computer mounted on a position information management device to execute a process of extracting a first image including the light emitting part from an image of a moving body provided with a top plate on which the light emitting part is laid, a process of detecting the presence of the moving body by performing image processing on the first image, and a process of specifying the position of the moving body.

[0017] According to a fifth aspect of the present invention, a system includes a moving body provided with a light emitting part, a camera device that images a field including the moving body, and a position information management device that calculates the position of the moving body in the field from an image acquired from the camera device. The position information management device extracts a high-luminance image including at least one high-luminance region that is a set of pixels having a luminance value equal to or higher than a predetermined value among a plurality of pixels forming the image acquired from the camera device, performs image processing on the high-luminance image, and determines whether or not the moving body is included in the image acquired from the camera device according to the area of the high-luminance region included in the image after the image processing is performed.

Effects of the Invention

[0018] According to each aspect of the present invention, a system, a position information management device, a position identification method, and a program are provided that contribute to accurately identifying a moving object. Note that according to the present invention, other effects may be achieved instead of or together with the above effects.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10A

Figure 10B

Figure 10C

Figure 11A

Figure 11B

Figure 11C

Figure 12

Figure 13

Figure 14A

Figure 14B

Figure 14C

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Embodiments for Carrying Out the Invention

[0020] First, an overview of an embodiment will be described. Note that the reference numerals in the drawings appended to this overview are for convenience and are appended to each element as an example to assist understanding, and the description of this overview is not intended to be limiting in any way. In this specification and the drawings, for elements that can be similarly described, duplicate description may be omitted by assigning the same reference numerals.

[0021] The system according to one embodiment includes a mobile body 101 and a position information management device 102 (see FIG. 1). The mobile body 101 includes a top plate 112 on which a light emitting unit 111 is laid. The position information management device 102 extracts a first image including the light emitting unit 111 from an image of the mobile body 101. The position information management device 102 detects the presence of the mobile body 101 and identifies the position of the mobile body 101 by performing image processing on the first image.

[0022] The operations of the position information management device 102 according to the above embodiment are summarized as shown in FIG. 2. The position information management device 102 extracts a first image including the light emitting unit 111 from an image of the mobile body 101 (step S1). The position information management device 102 detects the presence of the mobile body 101 and identifies the position of the mobile body 101 by performing image processing on the first image (step S2).

[0023] The position information management device 30 sharpens the region corresponding to the light emitting unit 111 of the image data by performing image processing, particularly noise removal processing such as closing processing, on the image data including the mobile body 101 to be detected. The position information management device 30 can accurately identify a mobile body such as a transport robot by performing detection processing of the mobile body 101 using the image after the image processing is performed.

[0024] Specific embodiments will be described in more detail below with reference to the drawings.

[0025] [First Embodiment] The first embodiment will be described in more detail with reference to the drawings.

[0026] FIG. 3 is a diagram showing an example of the schematic configuration of the transport system according to the first embodiment. Referring to FIG. 3, the transport system includes a plurality of transport robots 10-1 and 10-2, a camera device 20, a position information management device 30, a terminal 40, and a control device 50.

[0027] In the following description, when there is no particular reason to distinguish between the transport robots 10-1 and 10-2, they are simply referred to as "transport robot 10". The same notation is used for other components. The configuration shown in FIG. 2 is an example, and is not intended to limit the number of components such as the camera device 20 included in the transport system. For example, a plurality of camera devices 20 may be included in the system. For example, the entire field may be covered by the image data captured by each of the plurality of camera devices 20.

[0028] The transport robot 10 is a robot that transports the article 60. In the first embodiment, the transport robot 10 is a cooperative transport robot that transports the article 60 in cooperation with other robots. The two transport robots 10 sandwich the article 60 from opposite directions and move while holding the article 60 in the sandwiched state, thereby transporting the article 60. The transport robot 10 is configured to be communicable with the control device 50 and moves based on a control command (control information) from the control device 50.

[0029] Note that a light source such as an LED (Light Emitting Diode) is attached to the top plate of the transport robot 10. Identification and position calculation of the transport robot 10 are performed using a region (high-brightness region) illuminated by the light source (light-emitting portion) laid on the top plate of the transport robot 10. For example, by making the area of the top plate of the transport robot 10-1 and the area of the top plate of the transport robot 10-2 different, with each having a light source (light-emitting portion), the two transport robots 10 can be identified.

[0030] Note that the article 60 is fixed to a cart with wheels. Therefore, when two transfer robots 10 lightly sandwich the article 60 and move, the article 60 also moves. In the following description, a pair consisting of two transfer robots 10 is referred to as a transfer robot pair.

[0031] The camera device 20 is a device that images the inside of the field. For example, the camera device 20 includes a camera capable of calculating the distance to an object, such as a stereo camera. The camera device 20 is installed on the ceiling, pillars, etc. The camera device 20 is connected to the position information management device 30. The camera device 20 images the inside of the field at a predetermined interval (predetermined sampling period) and transmits the image data to the position information management device 30. The camera device 20 images the situation inside the field in real time and transmits the image data including the situation inside the field to the position information management device 30.

[0032] The position information management device 30 is a device that manages the positions of objects inside a field (for example, a factory or a logistics warehouse). The position information management device 30 is a device that extracts a first image including a light emitting part (light source) from an image of a moving body (transfer robot 10), and detects the presence of the moving body and identifies the position of the moving body by performing image processing on the first image.

[0033] The position information management device 30 identifies a moving body (transfer robot 10) located inside the field based on the image data received from the camera device 20, and generates position information of the moving body. For example, in the example of FIG. 3, the position information management device 30 generates the position information of the transfer robot 10-1 and the position information of the transfer robot 10-2.

[0034] The position information management device 30 calculates the position (absolute position) of the transfer robot 10 in a coordinate system (X-axis, Y-axis) with an arbitrary point (for example, the entrance and exit) inside the field as the origin. The position information management device 30 transmits the calculated position information of the transfer robot 10 (hereinafter referred to as robot position information) to the control device 50.

[0035] The terminal 40 is a terminal used by an operator. Examples of the terminal 40 include portable terminal devices such as smartphones, mobile phones, game machines, tablets, etc., and computers (personal computers, notebook computers), etc. However, the terminal 40 is not intended to be limited to these examples. The terminal 40 inputs information regarding the conveyance of the article 60 from the operator. Specifically, the terminal 40 displays an operation screen (GUI; Graphical User Interface) for inputting the source and destination of the article 60 to be conveyed by the pair of conveyance robots. Based on the information input by the operator, the terminal 40 generates article conveyance plan information including information regarding the article to be conveyed, the source of the article to be conveyed, and the destination. The terminal 40 transmits the generated article conveyance plan information to the control device 50.

[0036] The control device 50 is a device that remotely controls the conveyance robot 10. Specifically, the control device 50 controls the conveyance robot 10 using the robot position information acquired from the position information management device 30 and the article conveyance plan information acquired from the terminal 40.

[0037] The control device 50 transmits a control command to each of the two conveyance robots 10 and remotely controls the pair of conveyance robots to move to the destination of the article 60. At that time, the control device 50 performs the above remote control so that the pair of conveyance robots moves to the destination while sandwiching the article 60. For example, the control device 50 transmits a control command (control information) so that the two opposing conveyance robots 10 move while maintaining the distance between them.

[0038] Subsequently, details of each device included in the conveyance system will be described.

[0039] FIG. 4 is a diagram showing an example of the processing configuration (processing modules) of the conveyance robot 10 according to the first embodiment. Referring to FIG. 4, the conveyance robot 10 includes a communication control unit 201 and an actuator control unit 202.

[0040] The communication control unit 201 is a means for controlling communication with the control device 50. The communication control unit 201 communicates with the control device 50 using wireless communication means such as a wireless LAN (Local Area Network), LTE (Long Term Evolution), or a network used in a specific area such as local 5G.

[0041] The actuator control unit 202 is a means for controlling an actuator composed of a motor or the like based on a control command (control information) received from the control device 50. For example, the control device 50 transmits a control command including the start of motor rotation, the rotation speed of the motor, the stop of motor rotation, etc. to the transport robot 10. The actuator control unit 202 controls the motor or the like according to the control command.

[0042] FIG. 5 is a diagram showing an example of the processing configuration (processing module) of the position information management device 30 according to the first embodiment. Referring to FIG. 5, the position information management device 30 includes a communication control unit 301, a position information generation unit 302, and a storage unit 303.

[0043] The communication control unit 301 is a means for controlling communication with other devices (e.g., the camera device 20, the control device 50) connected by wire (e.g., LAN, optical fiber, etc.) or wirelessly.

[0044] The position information generation unit 302 is a means for generating the above-described robot position information. The position information generation unit 302 generates robot position information based on the image data acquired from the camera device 20.

[0045] FIG. 6 is a diagram showing an example of the processing configuration of the position information generation unit 302. As shown in FIG. 6, the position information generation unit 302 includes sub-modules composed of a robot detection unit 311 and a robot position information generation unit 312.

[0046] The robot detection unit 311 is a means for detecting the transport robot 10 from the image data acquired from the camera device 20.

[0047] The robot detection unit 311 extracts an image including an area of pixels having a luminance higher than a predetermined value among the pixels constituting the image data acquired from the camera device 20. In the following description, a pixel having a luminance higher than a predetermined threshold value is referred to as a "high-luminance pixel". Further, an area composed of the high-luminance pixels is referred to as a "high-luminance area". An image including at least one or more high-luminance areas is referred to as a "high-luminance image".

[0048] For example, consider the case where the position information generation unit 302 acquires image data as shown in FIG. 7. As described above, a light source is attached to the top plate of the transport robot 10. When the light source attached to the top plate of the transport robot 10 emits light, the luminance of the area corresponding to the top plate becomes higher than a predetermined threshold value. As a result, the area corresponding to the top plate of the transport robot 10 is extracted by the robot detection unit 311.

[0049] For example, in the example shown in FIG. 7, an image (high-luminance image) as shown in FIG. 8 is extracted. Note that the upper limit of the area (size) of the high-luminance area cut out by the robot detection unit 311 from the image data is determined in advance. Specifically, the upper limit is determined in consideration of the size of the top plate of the transport robot 10 and the like. Therefore, in the example of FIG. 7, an image including the areas corresponding to the top plates of the transport robots 10-1 and 10-2 is not extracted as a high-luminance image. The robot detection unit 311 extracts an image including the areas corresponding to the top plates of the transport robots 10-1 and 10-2 as a "high-luminance image".

[0050] Here, as described above, if the height of a cage cart or the like carrying the article 60 is low, the frame of the cage cart does not appear in the top plate of the transport robot 10. However, if the height of a cage cart or the like carrying the article 60 is high, the frame thereof may appear in the top plate of the transport robot 10-1. In this case, the robot detection unit 311 extracts a high-luminance image as shown in FIG. 9.

[0051] As described above, the upper limit of the area of the high-brightness regions that can be included in the high-brightness image is predetermined. In other words, when the upper limit is not reached, the robot detection unit 311 extracts a single high-brightness image including a plurality of high-brightness regions. That is, as shown in FIG. 9, when the area of the top plate of the transfer robot 10 is separated by a frame or the like, an image including the two separated high-brightness regions is extracted.

[0052] Alternatively, instead of determining the upper limit of the area of the high-brightness regions that can be included in the high-brightness image, an upper limit may be provided for the distance from one high-brightness region to another high-brightness region. For example, as shown in FIG. 9, even when the top plate of the transfer robot 10 is separated by the frame of the carriage or the like, if the distance between the region 401 and the region 402 is short, a single high-brightness image including the two high-brightness regions (region 401, region 402) is extracted.

[0053] As described above, the robot detection unit 311 extracts a high-brightness image including at least one high-brightness region, which is a set of pixels having a luminance value equal to or greater than a predetermined value among the plurality of pixels forming the acquired image.

[0054] Note that the robot detection unit 311 also calculates the position of the high-brightness image extracted from the image data. For example, the robot detection unit 311 uses a specific location (for example, a point in the lower left) in the image data acquired from the camera device 20 as a reference point, and calculates the coordinates of the four points (the number of pixels with respect to the reference point) forming the high-brightness image with respect to the reference point. In the example of FIG. 7, the coordinates P1 to P4 are calculated.

[0055] The robot detection unit 311 calculates the area of the high-brightness region in the high-brightness image. Specifically, the robot detection unit 311 counts the number of high-brightness pixels that make up each high-brightness region. Next, the robot detection unit 311 calculates the distance from the camera device 20 to the object (the top plate of the transfer robot 10). Specifically, the robot detection unit 311 calculates the above distance using information such as the lens interval distance and focal length of the stereo camera that makes up the camera device 20. Note that the distance calculation using the image data captured by the stereo camera is obvious to those skilled in the art, so the detailed description thereof is omitted.

[0056] Based on the calculated distance, the robot detection unit 311 converts the number of high-brightness pixels forming the high-brightness region into the area of the high-brightness region. When the transfer robot 10 is imaged at a position far from the camera device 20, the number of high-brightness pixels is small, so the area of each pixel is largely converted and the area of the high-brightness region is calculated.

[0057] On the other hand, when the transfer robot 10 is imaged near the camera device 20, the number of high-brightness pixels is large, so the area of each pixel is small and the area of the high-brightness region is calculated. Note that the conversion formula between the number of pixels and the area can be determined in advance according to the distance between the transfer robot 10 and the camera device 20, the measured value of the number of pixels, the size of the top plate of the transfer robot 10, etc.

[0058] In the example of FIG. 9, the robot detection unit 311 calculates the area of each of the regions 401 and 402, and calculates the sum as the area of the high-brightness region.

[0059] Next, the robot detection unit 311 determines whether or not the calculated area is within a predetermined range. For example, let the lower limit of the top plate area when identifying as the transfer robot 10-1 be Amin1 and the upper limit of the top plate area be Amax1. In this case, the robot determination unit 3012 determines whether or not the calculated area A satisfies the relationship of "Amin1 ≦ A ≦ Amax1".

[0060] If the calculated area A satisfies the above relational expression, the robot detection unit 311 determines that the extracted high-luminance image corresponds to the top plate of the transfer robot 10-1. That is, the robot detection unit 311 detects the presence of the transfer robot 10-1.

[0061] On the other hand, when the calculated area A does not satisfy the above relational expression, the robot detection unit 311 performs predetermined image processing on the extracted image. In particular, when the calculated area A is smaller than the lower limit Amin1 of a predetermined range, the robot detection unit 311 performs the above image processing.

[0062] When the calculated area A is larger than the upper limit Amax1 of the predetermined range, the robot detection unit 311 determines that the extracted high-luminance image does not correspond to the top plate of the transfer robot 10-1.

[0063] In the following description, the predetermined range for the above robot determination is referred to as the "robot determination range".

[0064] When the calculated area is smaller than the lower limit of the robot determination range, the robot detection unit 311 performs predetermined image processing on the extracted high-luminance image. The above predetermined image processing is a process for removing noise included in an area corresponding to a light source (emission unit) of the high-luminance image. In the first embodiment, the case of using image processing called "closing" will be described.

[0065] The closing process is an image process in which, after performing a dilation process of replacing the luminance value of a target pixel with the luminance values of the neighboring pixels of the target pixel, a contraction process of replacing the luminance value of the target pixel with the luminance values of the neighboring pixels of the target pixel is performed.

[0066] For example, when dilation processing is performed on the image shown in FIG. 10A, the images shown in FIGS. 10B and 10C can be obtained. Note that in FIGS. 10A to 10C, FIGS. 11A to 11C, and FIG. 12, one cell in the figure represents one pixel. Also, in FIGS. 10A to 10C, FIGS. 11A to 11C, and FIG. 12 for explaining the closing processing, the images are shown in binary form for ease of understanding.

[0067] Note that in the dilation processing and erosion processing that constitute the closing processing, the number of bits to be dilated (hereinafter referred to as the dilation bit number) and the number of bits to be eroded (hereinafter referred to as the erosion bit number) can be input as parameters.

[0068] For example, when "1" is set for the dilation bit number, the luminance values of the pixels located above, below, to the left, and to the right of the target pixel are replaced with the luminance value of the target pixel (see FIG. 10B). Similarly, when "2" is set for the dilation bit number, the luminance values of the pixels located two pixels away from the target pixel are replaced with the luminance value of the target pixel (see FIG. 10C).

[0069] Also, when erosion processing is performed on the image shown in FIG. 11A, the images shown in FIGS. 11B and 11C can be obtained. For example, as shown in FIG. 11B, when "1" is set for the erosion bit number, the luminance value of the target pixel is replaced by the luminance values of the pixels located above, below, to the left, and to the right of the target pixel. Similarly, when "2" is set for the erosion bit number, the luminance value of the target pixel is rewritten by the luminance values of the pixels located within a range two pixels away from the target pixel (see FIG. 11C).

[0070] In the closing processing, by performing erosion processing on the target image after performing dilation processing, it is possible to remove noise existing in the original image or connect broken figures.

[0071] Also, in the closing process, it is possible not only to perform one dilation process and one contraction process, but also to perform the same number of contraction processes after multiple dilation processes. For example, it is also possible to continuously perform the dilation process twice on the original image and then perform the same number of contraction processes on the resulting image. By performing such multiple dilation processes and contraction processes, the noise removal ability and the like can be enhanced.

[0072] That is, in the closing process, it is also possible to input the number of times of repeating the dilation process and the contraction process as a parameter. For example, if the dilation bit number is fixed to "1" and the original image shown in the upper left of FIG. 12 is subjected to the dilation process twice, the image shown in the upper right of the same figure is obtained. When the contraction bit number is fixed to "1" and the contraction process is performed twice on the obtained image, the image shown in the lower left of FIG. 12 is obtained. As shown in FIG. 12, it can be seen that the high-brightness regions are connected by the two dilation processes and contraction processes. If the region (black region) sandwiched between the two high-brightness regions is regarded as noise, it can be said that the noise is removed by the closing process.

[0073] Note that in FIG. 12, the case where the dilation bit number and the contraction bit number are set to "1" is illustrated, but even when the dilation bit number and the contraction bit number are set to "2", the image shown in the lower left of FIG. 12 can be obtained by one dilation process and one contraction process. That is, in the closing process, the dilation bit number and the contraction bit number can be treated as equivalent to the number of repetitions of the dilation process and the contraction process.

[0074] Also, as is clear from FIG. 12, when the bit number or the number of repetitions is changed, the noise removal ability changes. In one dilation / contraction process (dilation bit number and contraction bit number are "1"), the noise (black region) sandwiched between the two high Hui brightness regions cannot be removed. However, the noise can be removed by two dilation / contraction processes. Therefore, the dilation bit number (contraction bit number) and the number of repetitions function as parameters for determining the intensity of the closing process. In the following description, the dilation bit number and the number of repetitions may also be referred to as "intensity parameters".

[0075] The robot detection unit 311 calculates the area of a high-luminance region (region corresponding to a light source) included in a high-luminance image for each of the variable parameters while varying the parameters that determine the noise removal ability by image processing. The robot detection unit 311 detects the presence of the transfer robot 10 based on the calculated area.

[0076] Specifically, the robot detection unit 311 performs a closing process on the extracted image (high-luminance image including a high-luminance region) while changing an intensity parameter that determines the intensity of the closing process. The robot detection unit 311 calculates the area of the high-luminance region included in the high-luminance image obtained for each closing process, and determines whether or not the calculated area is included in the robot determination range.

[0077] FIG. 13 is a flowchart showing an example of the operation of the robot detection unit 311.

[0078] The robot detection unit 311 sets an initial value of the intensity parameter (step S101). Here, a case where the number of dilation bits and the number of erosion bits are treated as the intensity parameters will be described. The robot detection unit 311 sets initial values (for example, "1") of the number of dilation bits and the number of erosion bits. Note that the determination of the parameter (determination of the initial value of the intensity parameter) may be made by the administrator, or may be made by the position information management device 30. Alternatively, the initial value may be calculated based on the accuracy of the camera, or a value detected by the transfer robot 10 in the past may be stored, and the value stored in the past (a value with a record of detecting the transfer robot 10 in the past) may be used as the initial value of the intensity parameter.

[0079] The robot detection unit 311 performs a closing process on the extracted high-luminance image (step S102).

[0080] The robot detection unit 311 calculates the area of the high-luminance region in the image after the closing process (step S103).

[0081] The robot detection unit 311 determines whether or not the area of the calculated high-brightness region is included in the robot determination range (step S104).

[0082] If the area of the high-brightness region is included in the robot determination range (step S104, Yes branch), the robot detection unit 311 determines that the high-brightness image is the top plate of the transfer robot 10 (determined as the transfer robot 10; step S105). When it is determined that the high-brightness image is the top plate of the transfer robot 10, the robot detection unit 311 executes the process of step S109.

[0083] If the area of the high-brightness region is not included in the robot determination range (step S104, No branch), the robot detection unit 311 increases the intensity parameter (step S106). For example, the robot detection unit 311 increments the intensity parameter (number of dilation bits, number of erosion bits) to increase the noise removal ability of the closing process by one level.

[0084] The robot detection unit 311 determines whether or not the intensity parameter has reached a predetermined upper limit (step S107).

[0085] If the intensity parameter has not reached the upper limit (step S107, No branch), the robot detection unit 311 returns to step S102 and continues the process. Note that the image to be subjected to the closing process implemented for the second and subsequent times is the image (high-brightness image) initially extracted by the robot detection unit 311. That is, the closing process is not executed repeatedly on the image after the closing process has ended. However, if the process of step S106 shown in FIG. 13 is not executed, the closing process may be executed repeatedly on the image after the closing process has ended. This is because the loop process from steps S102 to S107 shown in FIG. 13 without step S106 is substantially the same as repeatedly executing the closing process with the initial value of the intensity parameter (for example, the number of dilation bits and the number of erosion bits are 1).

[0086] If the intensity parameter has reached the upper limit (branch at step S107, Yes), the robot detection unit 311 does not determine that the high-intensity image is the top plate of the transfer robot 10 (non-determination as the transfer robot; step S108).

[0087] The robot detection unit 311 notifies the robot position information generation unit 312 of the determination result (whether the high-intensity image corresponds to the top plate of the transfer robot 10 or not) (step S109). Specifically, the robot detection unit 311 notifies the robot position information generation unit 312 of the image data acquired from the camera device 20, the identifier (ID; Identifier) of the transfer robot 10, and its position information (for example, coordinates P1 to P4 in the example of FIG. 7).

[0088] When a plurality of high-intensity images are included in a single image data acquired from the camera device 20, the robot detection unit 311 performs the above determination process for each high-intensity image. In the example of FIG. 7, since the high-intensity image by the transfer robot 10-1 and the high-intensity image by the transfer robot 10-2 are included in the image data, the robot detection unit 311 executes the robot determination process for each high-intensity image.

[0089] Hereinafter, the effect of the closing process executed by the robot detection unit 311 will be specifically described.

[0090] For example, consider the case where the robot detection unit 311 extracts an image as shown in FIG. 14A. In this case, the robot detection unit 311 calculates the areas of the region 401 and the region 402 (the total value of the areas of the two regions) and determines whether the area is included in the robot determination range.

[0091] If the calculated area is smaller than the lower limit of the robot determination range, the robot detection unit 311 executes the closing process on the image shown in FIG. 14A. As a result, an image as shown in FIG. 14B is obtained. The robot detection unit 311 calculates the areas of the region 401a and the region 402a in FIG. 14B and determines whether the area is included in the robot determination range.

[0092] If the calculated area is not included in the robot determination range, the robot detection unit 311 increases the intensity parameter by one step and executes the closing process on the image shown in FIG. 14A again. As a result, an image as shown in FIG. 14C is obtained. The robot detection unit 311 calculates the areas of the regions 401b and 402b in FIG. 14C and determines whether or not the areas are included in the robot determination range.

[0093] The robot detection unit 311 repeats the above-described process up to the upper limit of the intensity parameter and determines whether or not the extracted high-intensity image corresponds to the top plate of the transfer robot 10. By repeating the closing process while the robot detection unit 311 increases the intensity parameter, finally, the width of the black line (the black line that divides the high-intensity region) in the image illustrated in FIG. 14A can be narrowed (the region of the black line can be reduced). As a result, the robot detection unit 311 can accurately determine whether or not the high-intensity image corresponds to the top plate of the transfer robot 10.

[0094] Note that for the identification (specification) of the transfer robot 10 by the robot detection unit 311, the fact that the sizes (the areas of the high-intensity regions) of the light sources attached to the top plates of the respective transfer robots 10 are different may be utilized. For example, in the example of FIG. 7, the robot detection unit 311 may identify the two transfer robots 10 according to whether the area of the high-intensity region is included in the robot determination range of the transfer robot 10-1 or the robot determination range of the transfer robot 10-2.

[0095] Note that the method of identifying the transfer robot by the area of the high-intensity region is an example, and other MethodIt can also be used. For example, a marker with an identification function such as a QR code (registered trademark) or an AR (Augmented Reality) marker may be attached to the transport robot 10, and the robot detection unit 311 may identify the transport robot 10 by reading the marker. Alternatively, the robot detection unit 311 may transmit a specific signal or message to the transport robot 10, and the transport robot 10 that has received the signal or the like may respond with an identification number or the like to identify the transport robot 10. That is, even if identification information (for example, characters or patterns) is not given to the outside of the transport robot 10, the robot detection unit 311 can identify the transport robot 10 based on a signal or the like from the transport robot 10.

[0096] The robot position information generation unit 312 shown in FIG. 6 calculates the absolute position (position within the field) of the transport robot 10 and notifies the control device 50 of the absolute position as the robot position information. Specifically, the robot position information generation unit 312 converts the position (number of pixels from the reference point) in the image data of the transport robot 10 into an absolute position within the field based on the information of the camera device 20 (resolution of the imaging element, etc.).

[0097] The robot position information generation unit 312 converts the position (number of pixels) from the reference point (for example, the lower left of the image) in the image data of the transport robot 10 into the position (relative position) with respect to the reference point of the image data within the field. Since the absolute position of the reference point in the image data within the field is known in advance, the robot position information generation unit 312 calculates the absolute position of the transport robot 10 by adding the converted relative position to the absolute position of the reference point.

[0098] The robot position information generation unit 312 transmits the identifier of the detected transport robot 10 and its absolute position to the control device 50. The absolute position of the object may be the absolute coordinates of four points forming the transport robot 10, or the absolute coordinates of one point (for example, the center of the transport robot 10) representing the transport robot 10.

[0099] FIG. 15 is a diagram showing an example of the robot position information transmitted from the position information management device 30. Note that, as the identifier of the transfer robot 10, the IP (Internet Protocol) address or the like of each transfer robot 10 can be used.

[0100] The terminal 40 generates the above-described article transfer plan information. The terminal 40 displays a GUI for inputting the transfer source and the transfer destination of the article 60 on a liquid crystal display or the like. For example, the terminal 40 generates a GUI for inputting (designating) the transfer source and the transfer destination of the article 60 as shown in FIG. 16 and provides it to the operator. The terminal 40 transmits the information input by the operator according to the GUI to the control device 50. Specifically, the terminal 40 transmits the transfer source and the transfer destination of the article 60 to the control device 50 as "article transfer plan information".

[0101] FIG. 17 is a diagram showing an example of the processing configuration (processing module) of the control device 50 according to the first embodiment. Referring to FIG. 17, the control device 50 includes a communication control unit 501, a route calculation unit 502, a robot control unit 503, and a storage unit 504.

[0102] Similar to the communication control unit 301 of the position information management device 30, the communication control unit 501 controls communication with other devices. When the communication control unit 501 acquires the robot position information from the position information management device 30 and the article transfer plan information from the terminal 40, it stores these pieces of information in the storage unit 504.

[0103] The storage unit 504 stores field configuration information indicating the configuration of the field and robot management information for managing information of the transfer robot 10. For example, the position information (absolute position within the field) of the transfer source and the transfer destination indicated in the article transfer plan information is described in the field configuration information.

[0104] The route calculation unit 502 is a means for calculating a route for the transfer robot pair to transfer the article 60 from the transfer source to the transfer destination based on the article transfer plan information generated by the terminal 40.

[0105] The path calculation unit 502 calculates a path for transporting the article 60 from the source to the destination, for example, using a path search algorithm such as Dijkstra's algorithm or Bellman - Ford algorithm. Since the path search algorithms such as Dijkstra's algorithm are obvious to those skilled in the art, detailed descriptions thereof are omitted.

[0106] The robot control unit 503 is a means for controlling the transport robot 10. The robot control unit 503 transmits control information for transporting the article 60 with a pair of transport robots to each transport robot 10 based on the position information of the transport robot 10 and the position information of another transport robot 10 that forms a pair with the transport robot 10. The robot control unit 503 controls the transport robot 10 by transmitting a control command (control information) to the transport robot 10.

[0107] The robot control unit 503 grasps the absolute position of the transport robot 10 within the field based on the robot position information notified from the position information management device 30. Also, when controlling the transport robot 10, the robot control unit 503 needs information regarding the orientation of the transport robot 10. In this case, a gyro sensor or the like is attached to the transport robot 10, and the robot control unit 503 may acquire information regarding its orientation from the transport robot 10. Alternatively, the orientation when the transport robot 10 is first placed in the field may be determined in advance, and the orientation of the transport robot 10 may be estimated based on the control command transmitted from the robot control unit 503 to the transport robot 10.

[0108] The robot control unit 503 controls the two transport robots 10 to sandwich the article 60 placed at the source by transmitting a control command to the transport robots 10. Specifically, the robot control unit 503 moves these robots so that the two transport robots 10 face each other across the article 60 and moves them so that the distance between the robots becomes narrower.

[0109] Thereafter, the robot control unit 503 generates a control command so that the pair of transfer robots sandwiching the article 60 moves along the path calculated as the transfer path of the pair of transfer robots, and transmits the command to each transfer robot 10.

[0110] The robot control unit 503 treats one of the two transfer robots 10 as the "leading transfer robot" and the other as the "following transfer robot". Then, the robot control unit 503 acquires the current position of the leading transfer robot 10 among the transfer robots 10 described in the robot management information. Next, the robot control unit 503 determines the position to which the leading transfer robot 10 will arrive based on the transfer path calculated by the path calculation unit 502.

[0111] When making the pair of transfer robots move straight, the robot control unit 503 calculates the time and speed for rotating the motors of each transfer robot 10 according to the distance between the current position of the leading transfer robot 10 and the calculated arrival position. At this time, the robot control unit 503 generates a control command so that the motor rotation speeds of each transfer robot 10 are the same.

[0112] When making the pair of transfer robots rotate, the robot control unit 503 uses a model of circular motion in which a curve is drawn by the speed difference between the left and right wheels. Specifically, the robot control unit 503 calculates the input speeds to the left and right wheels for reaching the target position from the current position on a circular orbit based on the target position, the position, and the orientation of the robot. For the leading transfer robot 10, the robot control unit 503 uses the calculated input speed as it is and generates a control command to be transmitted to the leading transfer robot 10 based on the calculated input speed. On the other hand, for the following transfer robot 10, the robot control unit 503 calculates a front-rear direction speed correction value based on the distance between the robots (the distance between the plates by which each transfer robot sandwiches the article 60) and an offset correction value for the left and right wheels based on the rotation angle. The robot control unit 503 generates a control command to be transmitted to the following transfer robot 10 based on these correction values (speed correction value, offset correction value).

[0113] When the transport robot pair arrives at the destination, the robot control unit 503 controls the transport robot pair to place the article 60 at the destination. Specifically, the robot control unit 503 controls to increase the distance between the two transport robots 10 to complete the transport of the article 60.

[0114] As described above, in the transport system according to the first embodiment, the position information management device 30 calculates the position of the moving body (transport robot 10) in the field from the image acquired from the camera device 20. The position information management device 30 executes image processing on the high-intensity image and determines whether the moving body is included in the image acquired from the camera device 20 according to the area of the high-intensity region included in the image after the image processing is executed. More specifically, the position information management device 30 executes a closing process on the high-intensity image (first image), and detects the presence of the moving body based on the processed image (second image) obtained as a result of the closing process.

[0115] As described with reference to FIG. 12, the noise of the image is removed by executing the closing process. In the example shown in FIG. 14A, the black line that divides the two high-intensity regions corresponds to noise, and the black line is removed. The position information management device 30 can accurately identify (detect) the transport robot 10 that transports the article 60 placed on a tall cage cart or the like by determining whether the image after the noise (black line) is removed corresponds to the top plate of the transport robot 10.

[0116] [Second Embodiment] Subsequently, the second embodiment will be described in detail with reference to the drawings.

[0117] In the first embodiment, the closing process is executed while sequentially increasing the intensity parameter. In the second embodiment, a description will be given of the case where an intensity parameter suitable for the extracted high-intensity image is calculated in advance and the closing process is executed with the intensity parameter.

[0118] Note that since the configuration of the position information management device 30 according to the second embodiment can be the same as that of the first embodiment, the description corresponding to FIG. 5 is omitted. Hereinafter, the description will focus on the differences from the first embodiment.

[0119] If a plurality of high-brightness regions are included in a single high-brightness image, the robot detection unit 311 according to the second embodiment calculates the shortest distance between the plurality of high-brightness regions and determines an intensity parameter according to the shortest distance. For example, in the example of FIG. 9, the shortest distance between region 401 and region 402 is calculated, and an intensity parameter is determined according to the distance.

[0120] For example, the robot detection unit 311 extracts the edges of each high-brightness region. The robot detection unit 311 calculates the distance between a pixel of one high-brightness region (a pixel on the extracted edge) and a pixel of the other high-brightness region (a pixel on the extracted edge). For example, the robot detection unit 311 fixes a pixel of one high-brightness region and calculates the distance between the fixed pixel and each pixel on the edge corresponding to the other high-brightness region. Next, the robot detection unit 311 moves the fixed pixel to another pixel on the edge, and in the same manner as above, calculates the distance between each pixel on the edge corresponding to the other high-brightness region. The robot detection unit 311 repeats the above-described process until the fixed pixel makes a full circle on the edge, and selects the minimum value from the calculated distances to calculate the shortest distance between the high-brightness regions.

[0121] For example, as shown in FIG. 18, consider the case where two regions 403 and 404 are extracted as high-brightness regions. In this case, the robot detection unit 311 fixes a pixel on the edge of region 403 and calculates the distance between the fixed pixel and each pixel on the edge of region 404. Next, the robot detection unit 311 moves the calculation target to another pixel on the edge of region 403 for the fixed pixel, and again calculates the distance between each pixel on the edge of region 404. The robot detection unit 311 calculates the minimum value among the distances calculated by such processing as the shortest distance between the high-brightness regions.

[0122] Similarly, when a high - brightness image includes three or more high - brightness regions, the shortest distance between the high - brightness regions can be calculated in the same manner.

[0123] The robot detection unit 311 determines the intensity parameter used for the closing process according to the calculated shortest distance. For example, the robot detection unit 311 sets the number of pixels of the shortest distance as the number of dilation bits (number of dilated pixels) and the number of erosion bits (number of eroded pixels). Alternatively, the robot detection unit 311 may set the number of pixels of the shortest distance as the number of repetition times.

[0124] FIG. 19 is a flowchart showing an example of the operation of the robot detection unit 311 according to the second embodiment. In the flowcharts shown in FIGS. 13 and 19, the same reference numerals (step names) are given to the processes that can have the same content, and detailed descriptions are omitted.

[0125] The robot detection unit 311 calculates the shortest distance between the high - brightness regions (step S201).

[0126] The robot detection unit 311 sets the intensity parameter according to the calculated shortest distance (step S202).

[0127] Thereafter, the robot detection unit 311 determines whether the area of the high - brightness region is included in the robot determination range by one closing process. That is, compared with the first embodiment, the robot detection unit 311 according to the second embodiment determines whether the high - brightness image corresponds to the image of the top plate of the transfer robot 10, and can reduce the number of repetitions of changing the intensity parameter and executing the closing process.

[0128] As described above, when a plurality of high - brightness regions are included in one high - brightness image, the position information management device 30 according to the second embodiment calculates the shortest distance between the plurality of high - brightness regions and determines the intensity parameter according to the shortest distance. As a result, the number of times of changing the intensity parameter and repeating the closing process, which were necessary in the first embodiment, is reduced, and the load on the position information management device 30 can be reduced.

[0129] Next, the hardware of each device constituting the transport system will be described. FIG. 20 is a diagram showing an example of the hardware configuration of the position information management device 30.

[0130] The position information management device 30 can be configured by an information processing device (so-called computer) and has the configuration exemplified in FIG. 20. For example, the position information management device 30 includes a processor 321, a memory 322, an input / output interface 323, a communication interface 324, and the like. The components such as the processor 321 are connected by an internal bus or the like and are configured to communicate with each other.

[0131] However, the configuration shown in FIG. 20 is not intended to limit the hardware configuration of the position information management device 30. The position information management device 30 may include hardware not shown, or may not include the input / output interface 323 if necessary. Also, the number of components such as the processor 321 included in the position information management device 30 is not intended to be limited to the example shown in FIG. 20. For example, a plurality of processors 321 may be included in the position information management device 30.

[0132] The processor 321 is, for example, a programmable device such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or a DSP (Digital Signal Processor). Alternatively, the processor 321 may be a device such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). The processor 321 executes various programs including an operating system (OS; Operating System).

[0133] The memory 322 is a RAM (Random Access Memory), ROM (Read Only Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), etc. The memory 322 stores an OS program, an application program, and various data.

[0134] The input / output interface 323 is an interface for a display device and an input device (not shown). The display device is, for example, a liquid crystal display or the like. The input device is a device that receives user operations such as a keyboard and a mouse.

[0135] The communication interface 324 is a circuit, a module, etc. that communicate with other devices. For example, the communication interface 324 includes a NIC (Network Interface Card), a wireless communication circuit, etc.

[0136] The functions of the position information management device 30 are realized by various processing modules. The processing modules are realized, for example, by the processor 321 executing a program stored in the memory 322. Further, the program can be recorded on a computer-readable storage medium. The storage medium can be a non-transitory one such as a semiconductor memory, a hard disk, a magnetic recording medium, an optical recording medium, etc. That is, the present invention can also be embodied as a computer program product. Also, the above program can be downloaded via a network or updated using a storage medium storing the program. Furthermore, the above processing module may be realized by a semiconductor chip.

[0137] Note that the terminal 40, the control device 50, etc. can also be configured by an information processing device in the same manner as the position information management device 30, and the basic hardware configuration is the same as that of the position information management device 30, so the description thereof is omitted.

[0138] [Modification Example] Note that the configuration, operation, etc. of the transport system described in the above embodiments are examples and are not intended to limit the configuration of the system.

[0139] In the above embodiment, the transport robot 10 is used as an example of the moving body, but the moving body to which the present disclosure can be applied is not limited to the transport robot 10. For example, by applying the present disclosure, the position of a worker or the like working in the field may be specified.

[0140] For example, in the above embodiment, the case where a pair of transport robots consisting of two transport robots 10 transports the article 60 has been described, but the number of transport robots used may be one. That is, a conventional transport robot (for example, a robot of a type that places the article 60 on itself or a robot of a type that uses a towing device to tow the article 60 and the robot) may be used to transport the article 60. In this case, the control device 50 may control one transport robot based on the article transport plan information and the like acquired from the terminal 40, so that the article 60 can be transported by simpler control. Alternatively, the number of transport robots 10 to be controlled by the control device 50 may be three or more. By increasing the number of transport robots 10, it becomes possible to transport heavier articles 60 and the like using smaller (less expensive) transport robots 10. to tow In the above embodiment, the use of closing processing as image processing for noise removal has been described, but other processing may be used. For example, a Gaussian filter called "Gaussian blur" may be used. In this case, the position information management device 30 applies a Gaussian filter to the high-brightness image and detects the presence of the transport robot 10 based on the image obtained by applying the Gaussian filter. At that time, the position information management device 30 sequentially increases the parameter that determines the intensity of the Gaussian filter, calculates the area of the high-brightness region, and attempts to detect the transport robot 10.

[0141] In the above embodiment, the use of closing processing as image processing for noise removal has been described, but other processing may be used. For example, a Gaussian filter called "Gaussian blur" may be used. In this case, the position information management device 30 applies a Gaussian filter to the high-brightness image and detects the presence of the transport robot 10 based on the image obtained by applying the Gaussian filter. At that time, the position information management device 30 sequentially increases the parameter that determines the intensity of the Gaussian filter, calculates the area of the high-brightness region, and attempts to detect the transport robot 10.

[0142] Alternatively, the position information management device 30 may apply a low-pass filter to the high-brightness image and detect the transport robot 10 based on the resulting image. As shown in FIG. 7, if the area of the light source laid on the top plate of the transport robot 10 is large, fine noise can be removed by applying a low-pass filter to the high-brightness image.

[0143] The position information management device 30 may execute predetermined image processing before image processing such as closing processing. For example, the position information management device 30 may execute geometric transformation such as affine transformation or density transformation for converting contrast as necessary.

[0144] In the above embodiment, the number of dilation bits and the number of shrinkage bits are taken as examples of the parameters mainly determining the intensity of the closing process, but the parameters may be the number of repetitions of the dilation process and the shrinkage process respectively.

[0145] The position information management device 30 may change the intensity parameter linearly (in a straight line) or non-linearly like an exponential function. That is, when the position information management device 30 repeatedly executes closing processing or the like, the parameter may be changed so that the noise removal ability is rapidly increased as the number of times increases.

[0146] In the closing process of the above embodiment, the case of expanding crosswise from the target pixel (target pixel) has been described. However, in the expansion process, it is not limited to crosswise, and it is also possible to expand so as to change the luminance values of all or some of the pixels located around the target pixel. For example, it is also possible to convert the luminance values of the pixels including the upper left diagonal of the target pixel. Alternatively, in the expansion process and the shrinkage process, the number of pixels to be set may be changed in the horizontal direction and the vertical direction. For example, the number of pixels may be set asymmetrically in the vertical and horizontal directions such that it expands by 2 pixels vertically and 1 pixel horizontally.

[0147] In the above-described embodiment, the case of identifying the transfer robot 10 based on the area of the light source has been described. However, the transfer robot 10 may be identified according to the intensity of the light source (brightness on the image) laid on the transfer robot 10. For example, in the example of FIG. 7, the transfer robot 10 may be distinguished by changing the intensities of the light sources of the transfer robot 10-1 and the transfer robot 10-2. That is, even if the sizes of the light sources (the areas of the high-brightness regions) laid on each of the two transfer robots 10 are the same, the two transfer robots 10 can be distinguished by changing the intensities of the respective light sources. Alternatively, the two transfer robots 10 may be distinguished using the difference in the colors of the light sources.

[0148] Depending on the conditions in the field, by making the top plate of the transfer robot 10 white or the like, the light source may become unnecessary.

[0149] In the above-described embodiment, the position information management device 30 and the control device 50 have been described as different devices. However, the functions of the position information management device 30 may be realized by the control device 50. Alternatively, the position information management device 30 may be installed inside the field, and the control device 50 may be implemented on a server on the network. That is, the transfer system disclosed in the present application may be realized as an edge cloud system.

[0150] In the above-described embodiment, the case of using a camera (for example, a stereo camera) capable of detecting the distance between the ceiling and the transfer robot 10 has been described. However, a normal camera and a sensor (for example, an infrared sensor, a distance sensor) for measuring the distance between the camera and the transfer robot 10 may be used in combination.

[0151] By installing a position identification program in the storage unit of the computer, the computer can function as the position information management device 30. Further, by causing the computer to execute the position identification program, the position identification method can be executed by the computer.

[0152] In addition, in the plurality of flowcharts used in the above description, a plurality of steps (processes) are described in order, but the execution order of the steps executed in each embodiment is not limited to the described order. In each embodiment, for example, the order of the illustrated steps can be changed within a range that does not substantially affect the content, such as executing each process in parallel.

[0153] The above-described embodiments can be combined within a range where the contents do not conflict. For example, the number of bits of the shortest distance calculated in the second embodiment may be set as the initial value of the expansion bit number and the contraction bit number.

[0154] From the above description, the industrial applicability of the present invention is clear, and the present invention is preferably applicable to the conveyance of articles in factories, logistics warehouses, etc.

[0155] Some or all of the above embodiments can be described as follows in the following supplementary notes, but are not limited thereto. [Supplementary Note 1] A moving body (10, 101) including a top plate (112) on which a light emitting unit (111) is laid, A position information management device (30, 102) that extracts a first image including the light emitting unit (111) from an image of the moving body (10, 101), detects the presence of the moving body (10, 101) by performing image processing on the first image, and specifies the position of the moving body (10, 101), A system including. [Supplementary Note 2] The system according to Supplementary Note 1, wherein the image processing is a process for removing noise included in a region corresponding to the light emitting unit (111) in the first image. [Supplementary Note 3] The system according to Supplementary Note 1 or 2, wherein the position information management device (30, 102) calculates the area of the light emitting unit (111) included in the first image for each of the variable parameters while varying a parameter that determines the noise removal ability by the image processing, and detects the presence of the moving body (10, 101) based on the calculated area. [Supplementary Note 4] When a plurality of regions corresponding to the light emitting unit (111) are included in the first image, the position information management device (30, 102) determines a parameter for determining the noise removal ability by the image processing based on the distances between the plurality of regions, and executes the image processing using the determined parameter. The system according to Appendix 1 or 2. [Appendix 5] The position information management device (30, 102) Performs a closing process on the first image, and detects the presence of the moving body (10, 101) based on a second image obtained as a result of the closing process. The system according to any one of Appendices 1 to 4. [Appendix 6] The position information management device (30, 102) Applies a Gaussian filter to the first image, and detects the presence of the moving body (10, 101) based on a second image obtained by applying the Gaussian filter. The system according to any one of Appendices 1 to 4. [Appendix 7] Extracts a first image including the light emitting unit (111) from an image of a moving body (10, 101) provided with a top plate (112) on which the light emitting unit (111) is laid, and executes image processing on the first image to detect the presence of the moving body (10, 101), and a position information management device (30, 102) for specifying the position of the moving body (10, 101). [Appendix 8] The image processing is a process for removing noise included in a region corresponding to the light emitting unit (111) of the first image. The position information management device (30, 102) according to Appendix 7. [Appendix 9] While varying a parameter for determining the noise removal ability by the image processing, calculates the area of the light emitting unit (111) included in the first image for each of the varied parameters, and detects the presence of the moving body (10, 101) based on the calculated area. The position information management device (30, 102) according to Appendix 7 or 8. [Appendix 10] When a plurality of regions corresponding to the light emitting unit (111) are included in the first image, a parameter for determining the noise removal ability by the image processing is determined based on the distance between the plurality of regions, and the image processing is executed using the determined parameter. The position information management device (30, 102) according to appended claim 7 or 8. [Appended Note 11] Performing a closing process on the first image, and detecting the presence of the moving body (10, 101) based on a second image obtained as a result of the closing process. The position information management device (30, 102) according to any one of appended claims 7 to 10. [Appended Note 12] Applying a Gaussian filter to the first image, and detecting the presence of the moving body (10, 101) based on a second image obtained by applying the Gaussian filter. The position information management device (30, 102) according to any one of appended claims 7 to 10. [Appended Note 13] In the position information management device (30, 102), extracting a first image including the light emitting unit (111) from an image of a moving body (10, 101) provided with a top plate (112) on which the light emitting unit (111) is laid; detecting the presence of the moving body (10, 101) by performing image processing on the first image; identifying the position of the moving body (10, 101); A position identification method including. [Appended Note 14] The image processing is a process for removing noise included in a region corresponding to the light emitting unit (111) of the first image. The position identification method according to appended note 13. [Appended Note 15] The step of detecting the presence of the moving body (10, 101) is while varying a parameter for determining the noise removal ability by the image processing, calculating the area of the light emitting unit (111) included in the first image for each of the varied parameters, and detecting the presence of the moving body (10, 101) based on the calculated area. The position identification method according to appended note 13 or 14. [Appendix 16] The step of detecting the presence of the moving body (10, 101) is When a plurality of regions corresponding to the light emitting unit (111) are included in the first image, a parameter for determining the noise removal ability by the image processing is determined based on the distances between the plurality of regions, and the image processing is executed using the determined parameter. The position specifying method according to Appendix 13 or 14. [Appendix 17] The step of detecting the presence of the moving body (10, 101) is Performing a closing process on the first image, and detecting the presence of the moving body (10, 101) based on a second image obtained as a result of the closing process. The position specifying method according to any one of Appendices 13 to 16. [Appendix 18] The step of detecting the presence of the moving body (10, 101) is Applying a Gaussian filter to the first image, and detecting the presence of the moving body (10, 101) based on a second image obtained by the application of the Gaussian filter. The position specifying method according to any one of Appendices 13 to 16. [Appendix 19] To the computer (321) mounted on the position information management device (30, 102), A process of extracting a first image including the light emitting unit (111) from an image captured by a moving body (10, 101) including a top plate (112) on which the light emitting unit (111) is laid, A process of detecting the presence of the moving body (10, 101) by performing image processing on the first image, A process of specifying the position of the moving body (10, 101), A program for executing the above. [Appendix 20] A moving body (10, 101) including a light emitting unit (111), A camera device (20) that images a field including the moving body (10, 101), A position information management device (30, 102) that calculates the position of the moving body (10, 101) in the field from an image acquired from the camera device (20). including The position information management device (30, 102) extracts a high-brightness image including at least one high-brightness region, which is a set of pixels having a luminance value equal to or higher than a predetermined value among a plurality of pixels forming the image acquired from the camera device (20), executes image processing on the high-brightness image, and determines whether the moving body (10, 101) is included in the image acquired from the camera device (20) according to the area of the high-brightness region included in the image after the image processing is executed. A system. [Appendix 21] The system according to Appendix 20, wherein the position information management device (30, 102) executes a closing process as the image processing. [Appendix 22] The system according to Appendix 21, wherein the position information management device (30, 102) executes the closing process while varying a parameter that determines the intensity of the closing process. [Appendix 23] The system according to Appendix 22, wherein when the high-brightness image includes a plurality of the high-brightness regions, the position information management device (30, 102) calculates the shortest distance between the plurality of high-brightness regions and determines the parameter according to the shortest distance.

[0156] It should be noted that the disclosures of the above-cited prior art documents are incorporated herein by reference. Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments. It will be understood by those skilled in the art that these embodiments are merely illustrative and that various modifications can be made without departing from the scope and spirit of the present invention.

[0157] This application claims priority based on Japanese Patent Application No. 2019-153966 filed on August 26, 2019, and incorporates all of its disclosures herein.

Explanation of Reference Numerals

[0158] 10, 10-1, 10-2 Transfer Robots 20 Camera Device 30, 102 Location Information Management Device 40 Terminal 50 Control Device 60 Articles 101 Moving Body 111 Light Emitting Unit 112 Top Plate 201, 301, 501 Communication Control Unit 202 Actuator Control Unit 302 Location Information Generation Unit 303, 504 Memory Unit 311 Robot Detection Unit 312 Robot Location Information Generation Unit 321 Processor 322 Memory 323 Input / Output Interface 324 Communication Interface 401, 401a, 401b, 402, 402a, 402b, 403, 404 Areas 502 Route Calculation Unit 503 Robot Control Unit

Claims

1. A system including a moving body, wherein the moving body includes a top plate on which a light emitting portion is laid, and the system extracts a first image including the light emitting portion from an image of the moving body, specifies a parameter of noise removal ability based on distances between a plurality of high luminance regions in the light emitting portion included in the first image, detects the presence of the moving body by performing image processing on the first image based on the parameter of the noise removal ability, and specifies the position of the moving body. A system.

2. The system according to claim 1, wherein the image processing is a process for removing noise included in a region corresponding to the light emitting portion of the first image.

3. The system according to claim 1 or 2, wherein a closing process is performed on the first image, and the presence of the moving body is detected based on a second image obtained as a result of the closing process.

4. The system according to claim 1 or 2, wherein a Gaussian filter is applied to the first image, and the presence of the moving body is detected based on a second image obtained by applying the Gaussian filter.

5. extracts a first image including the light emitting portion from an image of a moving body including a top plate on which a light emitting portion is laid, specifies a parameter of noise removal ability based on distances between a plurality of high luminance regions in the light emitting portion included in the first image, detects the presence of the moving body by performing image processing on the first image based on the parameter of the noise removal ability, and specifies the position of the moving body. A position information management device.

6. In a position information management device, a step of extracting a first image including the light emitting portion from an image of a moving body including a top plate on which a light emitting portion is laid, A step of specifying a parameter of noise removal ability based on the distances between a plurality of high-brightness regions in the light-emitting unit included in the first image; A step of detecting the presence of the moving body by performing image processing on the first image based on the parameter of the noise removal ability; A step of specifying the position of the moving body; A position specifying method including the above.

7. On a computer mounted on a position information management device, A process of extracting a first image including the light-emitting unit from an image captured by a moving body including a top plate on which the light-emitting unit is laid; A process of specifying a parameter of noise removal ability based on the distances between a plurality of high-brightness regions in the light-emitting unit included in the first image; A process of detecting the presence of the moving body by performing image processing on the first image based on the parameter of the noise removal ability; A process of specifying the position of the moving body; A program for executing the above.

8. A moving body including a light-emitting unit; A camera device that captures a field including the moving body; A position information management device that calculates the position of the moving body in the field from an image acquired from the camera device; including The position information management device extracts a high-brightness image including at least one high-brightness region that is a set of pixels having a luminance value equal to or higher than a predetermined value among a plurality of pixels forming the image acquired from the camera device, and based on the distances between the plurality of high-brightness regions in the light-emitting unit included in the high-brightness image, specifies a parameter of noise removal ability, performs image processing on the high-brightness image based on the parameter of the noise removal ability, and determines whether or not the moving body is included in the image acquired from the camera device according to the area of the high-brightness region included in the image after the image processing is performed. A system.

Images