KR20240044611A - Object recognition apparatus for inventory management - Google Patents

Abstract

An object recognition device for inventory management is disclosed. The object recognition device for inventory management includes a rail installed at the top within a space where objects are loaded; A conveyor that moves along the rail and lowers and supports the wire below the rail; and a detector that is supported on the wire and moves up and down, and detects and recognizes the object. The transporter includes a transport body mounted on the rail and movable along the rail. and an arm module mounted on the lower part of the transport body, rotatable about the Z-axis in the vertical direction, and capable of adjusting the relative distance between the transport body and the detector in a direction perpendicular to the Z-axis.

Description

Object recognition apparatus for inventory management}

The present invention relates to an object recognition device for inventory management, and more specifically, to an object recognition device for inventory management that can obtain information such as the loading location, product information, loading quantity, and loading rate of objects loaded in a warehouse or store. It's about.

As vision-based recognition through cameras or wireless recognition technology using electronic tags such as RFID (Radio Frequency Identification) develops, automatic systems can be applied to collect information about displayed or stored items. Technologies are being developed.

Conventionally, in order to scan items on a shelf using a scanning device, a worker either directly scanned the items using a reader or scanned the items through a scanning device that randomly moves on the floor.

However, in spaces where multiple shelves are placed (stores, pharmacies, supermarkets, libraries, etc.), it is inefficient to scan each item displayed on the shelves manually.

In addition, devices that scan while moving on the floor using autonomous driving technology have frequent errors while driving, have poor reliability due to a low recognition rate for target items, and have problems with overlapping movement lines with customers or workers.

The present invention provides an object recognition device for inventory management that can obtain information on objects loaded on a shelf and on the floor.

Additionally, the present invention provides an object recognition device for inventory management that can minimize the length of the rail.

Additionally, the present invention provides an object recognition device for inventory management that can quickly obtain information on objects loaded in a wide area.

In addition, the present invention provides an object recognition device for inventory management that can obtain information such as the object's loading location, product information, loading quantity, and loading rate.

An object recognition device for inventory management according to the present invention includes a rail installed at the top of a space where objects are loaded; A conveyor that moves along the rail and lowers and supports the wire below the rail; and a detector that is supported on the wire and moves up and down, and detects and recognizes the object. The transporter includes a transport body mounted on the rail and movable along the rail. and an arm module mounted on the lower part of the transport body, rotatable about the Z-axis in the vertical direction, and capable of adjusting the relative distance between the transport body and the detector in a direction perpendicular to the Z-axis.

Additionally, the arm module is coupled to the transport body and may include at least two arms that are connected to each other and can rotate relative to each other.

In addition, the arm module is coupled to the ends of the arms and further includes a wire support portion for supporting the wire, and the wire support portion may be rotatable about the Z axis.

Additionally, a pair of the rails may be arranged side by side along a preset path.

In addition, it includes a plurality of position identification units installed on the rail at preset intervals along the rail, wherein each of the position identification units crosses the pair of rails in a direction perpendicular to the longitudinal direction of the rail. Support frame provided; and an electronic tag installed on an upper part of the support frame, wherein the transporter may further include a location recognition unit installed in the transport body and recognizing the electronic tag.

In addition, the rail is installed along the space between the first shelf and the second shelf on which the object is loaded, and the arm module is installed on one side and the other side of the first shelf, and one side and the other side of the second shelf. The detector can be placed on the side.

Additionally, the detector includes a first body connected to the wire; a second body connected to the first body and capable of rotating relative to the first body; a first camera provided on the second body and acquiring an image of the object in a first direction; a second camera provided on the second body and acquiring an image of the object in a second direction different from the first direction; A first LIDAR sensor (or first 3D camera) provided on the second body and acquiring spatial information on where the object is located in the first direction; And it is provided on the second body and may include a second LIDAR sensor (or a second 3D camera) that acquires spatial information on where the object is located in the second direction.

According to the present invention, the transporter moves along the rail, the detector moves up and down, and information on objects loaded on the shelf and on the warehouse floor can be obtained.

In addition, according to the present invention, since the detector can obtain information on objects loaded on multiple shelves by the arm module while the transport body is located at the same point, it is possible to quickly obtain information on objects loaded on a wide area. You can.

Additionally, according to the present invention, information on objects loaded on multiple shelves can be obtained with one rail, so the length of the rail can be minimized.

In addition, according to the present invention, the vertical/horizontal encoding unit mounted on the device, image information captured by the first and second cameras based on the vertical/horizontal and rotational movements of the scan panel, and the first and second LiDAR sensors ( By selectively applying the spatial information obtained from the first and second 3D cameras (omitted below) and the electronic tag identification information obtained from the antenna, information such as the object's loading location, product information, loading quantity, and loading rate can be obtained. there is.

Figure 1 is a perspective view showing an object recognition device for inventory management installed in a warehouse according to an embodiment of the present invention.
Figure 2 is a perspective view showing the object recognition device for inventory management of Figure 1.
Figure 3 is a front view showing the object recognition device for inventory management of Figure 2.
Figure 4 is a diagram showing one side of a detector according to an embodiment of the present invention.
Figure 5 is a view showing the other side of the detector of Figure 4.
Figure 6 is a diagram showing the shooting range of the first and second cameras and the measurement range of the first and second LiDAR sensors according to an embodiment of the present invention.
7 to 9 are diagrams showing a process in which an object recognition device for inventory management recognizes objects loaded on first and second shelves according to an embodiment of the present invention.
Figures 10 and 11 are diagrams showing a process in which an object recognition device for inventory management acquires object information in a Pyeongchi warehouse according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. However, the technical idea of the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed content will be thorough and complete and so that the spirit of the present invention can be sufficiently conveyed to those skilled in the art.

In this specification, when an element is referred to as being on another element, it means that it may be formed directly on the other element or that a third element may be interposed between them. Additionally, in the drawings, the thicknesses of films and regions are exaggerated for effective explanation of technical content.

Additionally, in various embodiments of the present specification, terms such as first, second, and third are used to describe various components, but these components should not be limited by these terms. These terms are merely used to distinguish one component from another. Accordingly, what is referred to as a first component in one embodiment may be referred to as a second component in another embodiment. Each embodiment described and illustrated herein also includes its complementary embodiment. Additionally, in this specification, 'and/or' is used to mean including at least one of the components listed before and after.

In the specification, singular expressions include plural expressions unless the context clearly dictates otherwise. In addition, terms such as "include" or "have" are intended to designate the presence of features, numbers, steps, components, or a combination thereof described in the specification, but are not intended to indicate the presence of one or more other features, numbers, steps, or components. It should not be understood as excluding the possibility of the presence or addition of elements or combinations thereof. Additionally, in this specification, “connection” is used to mean both indirectly connecting and directly connecting a plurality of components.

Additionally, in the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

Figure 1 is a perspective view showing an object recognition device for inventory management installed in a warehouse according to an embodiment of the present invention, Figure 2 is a perspective view showing the object recognition device for inventory management of Figure 1, and Figure 3 is an inventory of Figure 2 This is a front view showing the object recognition device for management.

Referring to FIGS. 1 to 3 , the object recognition device 10 for inventory management can obtain information on objects loaded in a warehouse or store. In this embodiment, the object recognition device 10 for inventory management acquires information on objects loaded in a warehouse as an example. According to one example, the object recognition device 10 may obtain information such as the object's loading location, product information, loading quantity, and loading rate. The object recognition device 10 can obtain information on objects loaded on the shelf 20. Additionally, the object recognition device 10 can obtain object information from a pyeongchi warehouse where objects are stacked on the floor. In this embodiment, a situation in which the object recognition device 10 acquires information about objects loaded on a shelf will be described as an example.

The object recognition device 10 includes a rail 100, a location identification unit 150, a transfer device 200, and a detector 300.

The rail 100 is installed at a preset height and along a preset path inside the warehouse. A pair of rails 100 may be provided side by side along the same path. The rail 100 provides a movement path along the space between the shelves 10 arranged opposite each other at regular intervals.

A plurality of position identification units 150 are provided at preset intervals along the longitudinal direction of the rail 100. The position identification unit 150 is provided to identify the moving position of the transfer machine 200. The location identification unit 150 includes a support frame 151 and an electronic tag 155. The support frame 151 is provided with a predetermined length and is arranged in a direction perpendicular to the longitudinal direction of the rail 100. The support frame 151 connects a pair of rails 100. The electronic tag 155 is located on the upper part of the support frame 151. The electronic tag 155 provides a location identification signal. The plurality of location identification units 150 provide different location identification signals. According to the embodiment, the electronic tag 155 may use RFID.

The conveyor 200 can move along the rail 100, supports the wire 270, and can lower the wire 270 below the rail 100. The transfer machine 200 includes a transfer body 210, an arm module 220, and a CCTV camera 280.

The transfer body 210 has a predetermined shape. The transfer body 210 is mounted on the rail 100 and can move along the rail 100. For this purpose, a roller 211, a driver, and an encoder are provided within the transfer body 210. The roller 211 contacts the rail 100 and rotates about its central axis. As the roller 211 rotates, the transfer body 210 may move along the rail 100. According to an embodiment, at least two rollers 211 may be provided on each side of a pair of rails 100 arranged side by side. Since the rollers 211 contact the rail 100 at at least four points, the transfer body 210 can stably move along the rail 100 even if eccentricity occurs due to movement of the arm module 220.

The driver rotates the roller 211. A driving motor may be used as the actuator. The encoder measures the rotation amount of the roller 211. By measuring the rotation amount of the roller 211, the moving distance of the transfer body 210 can be calculated.

Additionally, a charging terminal 213 may be provided on the upper surface of the transfer body 210, and a battery may be provided inside. The charging terminal 213 receives power wired or wirelessly from a power supply module (not shown) installed at one point of the rail 100. The power supplied to the charging terminal 213 is charged to the battery. The battery may be located on the central axis of the transport body 210. The battery supplies power to various electronic components provided in the transporter 200 and the detector 300.

Additionally, a location recognition unit may be provided inside the transfer body 210. The location recognition unit recognizes the location of the transfer body 210 from the location information provided by the location identification unit 155. According to an embodiment, the location identification unit 155 may be provided as an RFID reader. When the position of the transfer body 210 is recognized based on the rotation amount of the roller 211 measured by the encoder, a position error may occur depending on the driving state of the roller 211. Since the position identification unit 155 can identify the exact position of the transfer body 210 through position information provided at each preset point of the rail 100, it is possible to correct the position error according to the driving state of the roller 211. You can.

The arm module 220 is located below the transfer body 210 and is mounted on the transfer body 210. The arm module 220 can rotate relative to the transfer body 210 about the Z-axis, which is the vertical direction, and moves the detector 300 on the XY plane perpendicular to the Z-axis. The arm module 200 may be provided as a multi-joint structure having at least two arms 230 and 240. In this embodiment, the structure in which the arm module 200 consists of two arms 230 and 240 is explained as an example, but the number of arms 230 and 240 may change depending on the distance to which the detector 300 must be moved. You can. The arm may also include a telescopic structure.

The arm module 220 includes a first arm 230, a second arm 240, and a wire support portion 250.

The first arm 230 is provided with a predetermined length, and its length direction is arranged perpendicular to the Z-axis. One end of the first arm 230 is coupled to the transfer body 210, and one end can be rotated relative to the transfer body 210 as an axis.

The second arm 240 is provided with a predetermined length, and its length direction is arranged perpendicular to the Z-axis. One end of the second arm 240 is coupled to the other end of the first arm 230, and can rotate relative to the first arm 230 about one end.

Although not shown in the drawing, the arm module 220 may further include a balancing arm. The balancing arm is located on the opposite side of the first and second arms 230 and 240 with respect to the transfer body 210 and is supported on the transfer body 210. The balancing arm has a weight corresponding to the first and second arms 230 and 240 and is provided to maintain the weight balance of the conveyor 200.

Wire supports 250 are provided at the ends of the arms 230 and 240. According to the embodiment, the wire support part 250 is coupled to the end of the second arm 240 and is located at the lower part of the second arm 240. The wire support portion 250 can rotate relative to the second arm 240 about the Z-axis. The wire support portion 250 supports the wire 270 and can lower the wire 270 and pull it up. For this purpose, a reel and a drive motor may be provided inside the wire support portion 250. A wire is wound around the reel, and the reel rotates when driven by a drive motor. Depending on the direction of rotation of the reel, the wire 270 may move down or up as it is wound or unwound from the reel.

The CCTV camera 280 is installed at the bottom of the first arm 230 and captures surrounding images. The video captured by the CCTV camera 280 is transmitted to a control server (not shown). The user can monitor the situation inside the warehouse and the movement of the detector 300 through the video captured by the CCTV camera 280. When an error occurs in the object recognition device 10, the user can connect remotely and directly control the movement of the object recognition device 10. The CCTV camera 280 may be equipped with a storage device capable of storing captured images when an accident occurs. When an external impact is applied to the object recognition device 10, the captured image may be stored in the storage device for a preset time.

The detector 300 is supported by hanging on a wire 270 and can move up and down by the wire 270. The detector 300 has a rectangular shape with a predetermined thickness and has one side 301 facing the object and the other side 302 facing the user. The detector 300 detects product information and location information of the object, and spatial information where the object is located.

Figure 4 is a diagram showing one side of the detector according to an embodiment of the present invention, and Figure 5 is a diagram showing the other side of the detector of Figure 4.

4 and 5, the detector 300 includes a first body 310, a second body 320, a user camera 330, a display 350, a first camera 360, and a second camera ( 370), a first LiDAR sensor 380, a second LiDAR sensor 390, an antenna 400, and an information processing unit (not shown).

The first body 310 is suspended from a wire 270, and a user camera 330 and a display 350 are provided on the other side 302 facing the user. The user camera 330 photographs the user, and the captured image of the user is transmitted to the information processing unit and provided for user recognition. The display 350 displays various information to be provided to the user.

The second body 320 is located below the first body 310 and has a length corresponding to the width of the first body 310. The second body 320 is hinge-coupled with the first body 310 at both ends, and can rotate relative to the first bar 310 about the hinge as the axis 321. A first camera 360, a second camera 370, a first LiDAR sensor 380, a second LiDAR sensor 390, and an antenna 400 are provided on one side of the second body 320. .

Figure 6 is a diagram showing the shooting range of the first and second cameras and the measurement range of the first and second LiDAR sensors according to an embodiment of the present invention.

Referring to FIG. 6, the first camera 360 captures an image in a first direction, and the second camera 370 captures an image in a second direction. The first direction and the second direction form a preset angle. The area 362 captured by the first camera 360 and the area 371 captured by the second camera 370 partially overlap. Image information of a wider area can be obtained by merging the image captured by the first camera 360 and the image captured by the second camera 370 based on the overlapping area.

The first LiDAR sensor 380 acquires spatial information on the location of the object 30 in the first direction, and the second LiDAR sensor 390 acquires spatial information on the location of the object 30 in the second direction. do. The first direction and the second direction form a preset angle. The measurement area 381 of the first LiDAR sensor 380 and the measurement area 391 of the second LiDAR sensor 390 partially overlap. A wider range of spatial information can be obtained by merging the spatial information measured by the first LiDAR sensor 380 and the spatial information measured by the second LiDAR sensor 390 based on the overlapping area.

The antenna 400 can identify electronic tags. According to an embodiment, the antenna 400 can recognize the RFID attached to the object 30.

Images acquired from the first and second cameras 370 and 370, spatial information obtained from the first and second LIDAR sensors 380 and 390, and information recognized from the antenna 400 are transmitted to the information processing unit. . The information processing unit may analyze the received information and calculate product information, location information, loading quantity, loading ratio, etc. of the object 30 loaded in the corresponding space. The calculated information may be displayed on the display 350 at the user's request.

7 to 9 are diagrams showing a process in which an object recognition device for inventory management recognizes objects loaded on first and second shelves according to an embodiment of the present invention.

First, referring to FIG. 7, the object recognition device 10 is positioned at a point between the first shelf 21 and the second shelf 22 as the transfer body 210 moves along the rail 100. Then, the first arm 230 and the second arm 240 rotate so that the detector 300 moves in the XY plane. With the detector 300 located on one side of the first shelf 21, the wire support part 250 is rotated so that one side of the detector 300 faces the first shelf 21. In this state, the wire 270 is lowered and the detector 300 is positioned in front of the object 31. The first and second cameras 360 photograph the object, and the first and second LIDAR sensors 370 and 380 obtain spatial information where the object is located. Then, the antenna 400 recognizes the RFID information attached to the object 30. The acquired information is transmitted to the information processing unit, and the information processing unit calculates product information, location information, loading quantity, loading ratio, etc. of the object 30.

Referring to FIG. 8, while the transport body 210 is located at the same point as in FIG. 7, the first arm 230 and the second arm 240 rotate so that the detector 300 is positioned on the first shelf 21. It is located on the other side. Then, the wire support part 250 rotates so that one side of the detector 300 faces the first shelf 21 and the wire 270 comes down. The detector 300 acquires product information, location information, loading quantity, loading ratio, etc. of the object 32 loaded on the other side of the first shelf 31.

Referring to FIG. 9, while the transport body 210 is located at the same point as in FIG. 7, the first arm 230 and the second arm 240 rotate so that the detector 300 is positioned on the second shelf 22. It is located on one side. Then, the wire support 250 rotates so that one side of the detector 300 is positioned toward the second shelf 22 and the wire 270 comes down. The detector 300 obtains product information, location information, loading quantity, loading ratio, etc. of the object 33 loaded on the other side of the second shelf 32.

Through the same process, the object recognition device 10 for inventory management can obtain product information, location information, loading quantity, loading ratio, etc. of the object 34 loaded on the other side of the second shelf 22.

Figures 10 and 11 are diagrams showing a process in which an object recognition device for inventory management acquires object information in a Pyeongchi warehouse according to an embodiment of the present invention.

First, referring to FIG. 10 , the object recognition device 10 scans the inside of a warehouse to generate object loading maps 30a and 30b. The object loading maps 30a and 30b refer to three-dimensional spatial information that allows you to know the loading positions and loading heights of the objects 31 to 33 loaded inside the warehouse.

According to one embodiment, the object recognition device 100 may generate an object loading map with the detector 300 located on one side of the warehouse. According to another embodiment, the object recognition device 100 may generate an object loading map after the transport body 210 moves along the rail 100 and the detector 300 scans the inside of the warehouse.

The first body 310 of the object recognition device 100 rotates about the Z-axis with respect to the wire support part 250, and the second body 320 rotates in the vertical direction with respect to the first body 310, And images inside the warehouse are acquired from the second cameras 360 and 370, and information on space inside the warehouse is acquired from the first and second LiDAR sensors 380 and 390. The information processing unit uses the acquired image information and spatial information to create object loading maps 30a and 30b including the positions and heights of the objects 31 to 33 loaded inside the warehouse. In the drawing, the objects (31 to 33) inside the warehouse are expressed as being loaded in two groups, and for convenience of explanation, the loading groups of the objects (31 to 33) are referred to as a first loading group (30a) and a second loading group ( It is expressed as 30b). The information processing unit uses the object loading map to calculate a movement path for obtaining information on individual objects.

According to the calculated movement path, the transfer body 210 moves along the rail 100 and is located at a point between the first loading group 30a and the second loading group 30b. The first arm 230 and the second arm 240 rotate so that the detector 300 is located on one side of the first loading group 30a, and the wire support part 250 rotates to expose one side of the detector 300. It is positioned to face the first loading group 30a. In this state, the wire 270 comes down to obtain product information and location information of the object 31 loaded on one side of the first loading group 30a.

And the transfer body 210 is located at the same point, and the first arm 230 and the second arm 240 move from one side to the other across the upper area of the first loading group 30a, thereby forming the first loading group 30a. Product information and location information of the object 31 loaded on the top of (30a) can be obtained.

In addition, the first arm 230 and the second arm 240 rotate so that the detector 300 is located on the other side of the first loading group 30a, and one side of the detector 300 is located on the first loading group 30a. ), the wire comes down and product information and location information of the object 32 loaded on the other side of the first loading group 30a can be obtained.

The object recognition device 10 may obtain product information and location information of the object 33 in the same manner for the second loading group 33.

The government processing unit maps the product information and location information of the objects 31 to 33 obtained through the above-described process with the object loading maps 30a and 30b. Through the mapped information, the user can check the 3D location information of the objects 31 to 33 loaded in the warehouse and check the product information loaded at each point.

Above, the present invention has been described in detail using preferred embodiments, but the scope of the present invention is not limited to the specific embodiments and should be interpreted in accordance with the appended claims. Additionally, those skilled in the art should understand that many modifications and variations are possible without departing from the scope of the present invention.

10: Object recognition device
100: rail
200: transfer machine
210: transfer body
220: arm module
230: first arm
240: Second arm
250: wire support
280: CCTV camera
300: detector
310: first body
320: second body
330: User camera
350: display
360: first camera
370: Second camera
380: First LiDAR sensor (or first 3D camera)
390: Second LiDAR sensor (or first 3D camera)
400: Antenna

Claims (8)

A rail installed at the top within the space where the object is loaded;
A conveyor that moves along the rail and lowers and supports the wire below the rail; and
It is supported on the wire and moves up and down, and includes a detector that detects and recognizes the object,
The transfer machine
a transfer body mounted on the rail and movable along the rail; and
Object recognition for inventory management, including an arm module mounted on the lower part of the transport body, rotatable about the Z-axis in the vertical direction, and capable of adjusting the relative distance between the transport body and the detector in a direction perpendicular to the Z-axis. Device. According to claim 1,
The arm module is,
An object recognition device for inventory management, which is coupled to the transfer body and includes at least two arms that are connected to each other and can rotate relative to each other. According to claim 2,
The arm module is,
Object recognition device for inventory management, further comprising a balancing arm located on an opposite side of the two or more arms around the transfer body and having a weight corresponding to the two or more arms. According to claim 2,
The arm module is
It further includes a wire support part coupled to the ends of the arms and supporting the wire,
The wire support part is an object recognition device for inventory management that is rotatable around the Z axis. According to claim 1,
The rail is an object recognition device for inventory management in which a pair of rails are arranged side by side along a preset path. According to claim 5,
It includes a plurality of position identification units installed on the rail at preset intervals along the rail,
Each of the location identification units,
a support frame provided across the pair of rails in a direction perpendicular to the longitudinal direction of the rails; and
It includes an electronic tag installed on the upper part of the support frame,
The conveyor,
An object recognition device for inventory management further comprising a location recognition unit installed in the transfer body and recognizing the electronic tag. According to claim 1,
The rail is installed along the space between the first shelf and the second shelf on which the object is loaded,
The arm module is an object recognition device for inventory management capable of positioning the detector on one side and the other side of the first shelf and on one side and the other side of the second shelf. According to claim 1,
The detector is
a first body connected to the wire;
a second body connected to the first body and capable of rotating relative to the first body;
a first camera provided on the second body and acquiring an image of the object in a first direction;
a second camera provided on the second body and acquiring an image of the object in a second direction different from the first direction;
A first LIDAR sensor provided on the second body and acquiring spatial information on the location of the object in the first direction; and
An object recognition device for inventory management including a second LIDAR sensor provided on the second body and acquiring spatial information on where the object is located in the second direction.