Automatic inventory robot and obstacle map creation robot

Abstract

In step S3401 of the present invention, an automatic inventory robot moves a prescribed distance at a prescribed speed. In step S3402, the automatic inventory robot computes a duplicate reading rate, which is a rate of duplicate readings of product RFID tags. If the duplicate reading rate is at or above a threshold, the process moves to step S3401. However, if the duplicate reading rate is below the threshold, this indicates that there are still many product RFID tags being newly read, and an operation of stopping for a prescribed number of seconds is added in step S3403 in order to reliably read the position RFID tags.

Description

Automatic inventory robot and obstacle map creation robot

[0001] The present invention relates to a robot that automatically performs inventory.

[0002] Traditionally, stores have periodically conducted "inventory counts" to determine the number of products in the store. In Patent Document 1, the applicant discloses a technology in which item RFID tags (also referred to as "product RFID tags") are attached to products and location-indicating RFID tags are attached to fixtures such as shelves and floors, and store clerks read the RFID tags with a handheld reader while moving around the store. Product RFID tags and location RFID tags read at approximately the same time can be assumed to be in the same location, so information on the location of the products can be obtained. The technology of Patent Document 1 can be used for inventory counts. However, since products may be placed high on shelves, the technology of Patent Document 1 assumes that store clerks will read the RFID tags while changing the orientation of the handheld reader up and down. This results in significant fatigue for the store clerk's arms.

[0003] Patent No. 6803089

[0004] The applicant developed a robot that automatically reads product RFID tags and location RFID tags within a store while moving around the store. However, in experiments, while it was possible to achieve a 100% product RFID tag read rate by slowing the robot's movement speed sufficiently, even with a 100% read rate, the rate of incorrectly determining the location did not fall below a satisfactory level. Investigation into the cause revealed that the robot moved too quickly, resulting in failure to read the location RFID tags. Therefore, a solution was considered: controlling the robot's movement within the store based on the read status of the product RFID tags. Furthermore, the task of affixing location-indicating location RFID tags to fixtures such as shelves and floors requires a great deal of time and effort in large stores. Therefore, it would be desirable to realize a product management system that does not require the affixing of location RFID tags.

[0005] The present invention aims to provide an automated inventory robot that automatically controls the way it moves within a store. Another objective of the present invention is to provide a product management system that does not require affixing positional RFID tags to fixtures such as shelves and floors.

[0006] The inventors discovered that failure to read position RFID tags can be reduced by controlling the movement of an automated inventory robot within a store in accordance with the read status of product RFID tags, leading to the completion of the present invention. The inventors also discovered that automatic inventory can be performed even on floors where position RFID tags are not attached by creating an obstacle map using an obstacle map creating robot, leading to the completion of the present invention.

[0007] (1) A first automated inventory robot according to the present invention is an automated inventory robot that automatically travels within a store floor and reads product RFID tags and location RFID tags, and includes: a duplicate read rate calculation unit that calculates a duplicate read rate, which is the rate at which product RFID tags are read twice; a movement instruction creation unit that creates movement instructions to control movement of the automated inventory robot based on the duplicate read rate; and a travel control unit that controls movement of the automated inventory robot in accordance with the movement instructions. (2) The movement instruction creation unit may create movement instructions that are based on an operation of moving a predetermined distance at a predetermined speed and then stopping for a predetermined time, and that omit the operation of stopping for the predetermined time if the duplicate read rate is equal to or greater than a threshold. (3) The first automated inventory robot according to the present invention may include directional upper and lower antennas. (4) The obstacle map creation robot of the present invention includes a human-following travel control unit that causes the obstacle map creation robot to travel following a human, a travel history recording unit that records a travel history, a current position coordinate calculation unit that calculates the current position coordinates of the obstacle map creation robot based on the travel history, a LIDAR unit that detects the distance to an obstacle, an obstacle map creation unit that creates an obstacle map based on the distance to the obstacle and the current position coordinates of the obstacle map creation robot, and an overlay unit that overlays the obstacle map on a floor plan. (5) The obstacle map creation robot of the present invention may further include a section determination unit that divides a floor aisle into a plurality of sections corresponding to each shelf based on the result of overlaying the obstacle map on the floor plan. (6) The obstacle map creation robot of the present invention may further include a section / shelf number conversion table creation unit that creates coordinate information that defines the sections and creates a correspondence table between the coordinate information and shelf numbers. (7) A control method according to the present invention is a control method for an automatic searching robot that is capable of traveling and that can read RFID tags, and calculates a duplicate read rate, which is the rate at which RFID tags of search targets are read twice, and controls the movement of the automatic searching robot based on the duplicate read rate. (8) A control method according to the present invention may determine the self-position of the automatic searching robot by reading a position RFID tag affixed to a predetermined position in a planned travel area.(9) The control method according to the present invention may sequentially record a travel route of the automatic search robot and determine the self-position of the automatic search robot from the travel route. (10) The control method according to the present invention may record the self-position in association with the reading result of the RFID tag of the search target. (11) A second automatic inventory robot according to the present invention is an automatic inventory robot that travels within a predetermined area and reads product RFID tags affixed to products placed within the predetermined area, and includes a current position calculation unit that calculates current position information from which the current position of the automatic inventory robot in the predetermined area can be determined, a memory unit that stores information read from the product RFID tags in association with the current position information, a duplicate read rate calculation unit that calculates a duplicate read rate, which is the rate at which the product RFID tags are read twice, and a movement control unit that generates movement instructions to control movement of the automatic inventory robot based on the duplicate read rate. (12) A third automatic inventory robot according to the present invention is an automatic inventory robot that automatically travels around a store floor and reads product RFID tags, and is equipped with a control unit that controls a process of calculating the coordinates of the current position of the automatic inventory robot by adding the amount of movement of the automatic inventory robot to the coordinates of the starting point based on the orientation of the automatic inventory robot at the starting point.

[0008] According to the present invention, an automatic inventory robot can automatically perform inventory, thereby reducing the burden on store staff. Also, according to the present invention, an automatic inventory robot can perform inventory even on floors where position RFID tags are not attached, thereby eliminating the need to attach position RFID tags, further reducing the burden on store staff.

[0009] FIG. 1 is a diagram showing an automatic inventory robot according to a first embodiment of the present invention. FIG. 2 is a diagram showing an example of a floor layout of a store according to the first embodiment of the present invention. FIG. 3 is a diagram showing functional blocks of a tablet according to the first embodiment of the present invention. FIG. 4 is a diagram showing an example of calculation of a duplicate read rate according to the first embodiment of the present invention. FIG. 5 is a diagram showing an obstacle map creating robot according to a second embodiment of the present invention. FIG. 6 is a diagram showing functional blocks of a tablet according to the second embodiment of the present invention. FIG. 7 is an example of an obstacle map created by an obstacle map creating unit according to the second embodiment of the present invention. FIG. 8 is a diagram showing the result of an overlay unit according to the second embodiment of the present invention overlaying a rotated and enlarged / reduced obstacle map on a floor plan. FIG. 9 is a diagram showing the result of a section determination unit according to the second embodiment of the present invention dividing an aisle into a plurality of sections corresponding to each shelf. FIG. 10 is a diagram showing an example of a section / shelf number conversion table created by a section / shelf number conversion table creating unit according to the second embodiment of the present invention. FIG. 11 is a floor plan for explaining a third embodiment of the present invention.

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. [Embodiment 1] Fig. 1 is a diagram showing an automated inventory robot 3100 according to a first embodiment of the present invention. The automated inventory robot 3100 includes a reader 3030, an upper antenna 3010, a lower antenna 3020, a tablet 3040, a travel control unit 3060, and wheels 3050. The reader 3030 wirelessly communicates with product RFID tags and location RFID tags via the upper antenna 3010 and the lower antenna 3020, and reads these RFID tags.

[0011] The upper antenna 3010 has a directivity center at 45 degrees upward so as to read the product RFID tags of products placed in high places. The lower antenna 3020 has a directivity center in the horizontal direction so as to read the product RFID tags and location RFID tags of products placed in low places. The tablet 3040 receives information from the RFID tags read by the reader 3030 and creates instruction information for controlling the movement of the automated inventory robot based on that information. In FIG. 1 , the reader 3030 is partially hidden by the tablet 3040. The travel control unit 3060 controls the movement of the automated inventory robot 3100 in accordance with the instruction information from the tablet 3040.

[0012] A plurality of products 3300 are placed on the fixture 3200. A product RFID tag 3350 is attached to the product 3300. A position RFID tag 3450 is attached to the bottom of the fixture 3200. Number 3400 is the floor.

[0013] The upper antenna 3010 and the lower antenna 3020 are configured with an antenna array arranged in a matrix to receive reflected waves, with the center of directivity in the normal direction from the center of the plane on which the matrix is ​​formed, and an effective reception angle of approximately 45 degrees. Both the upper antenna 3010 and the lower antenna 3020 are arranged on the side of the direction in which the automated inventory robot 3100 travels during inventory, with the upper antenna 3010 being arranged so that its effective reception angle faces upward at 45 degrees to the plane of the floor 3400, and the lower antenna 3020 being arranged so that its effective reception angle faces parallel to the floor 3400.

[0014] 1, the automated inventory robot 3100 moves toward the back of the drawing, with its antenna facing to the right in the direction of travel. Therefore, the lower receiving end of the upper antenna 3010 and the upper receiving end of the lower antenna 3020 overlap, providing an effective receiving angle from the floor surface 3400 on one side of the automated inventory robot 3100 to almost directly above. By traveling parallel to the shelves at a distance of 30 cm to 90 cm from the fixtures 3200, it is possible to detect all product RFID tags installed on the shelves.

[0015] In the first embodiment, the upper and lower antennas are arranged only on the right side in the direction of travel, and no antenna with an effective receiving angle is arranged on the left side in the direction of travel. The travel control unit 3060 outputs a control signal for controlling the output of a motor (not shown) that drives the wheels 3050, and causes the automated inventory robot 3100 to travel parallel to the fixtures 3200 at a speed of, for example, 500 m to 1 km per hour.

[0016] The tablet 3040 displays a UI (user interface) that allows a person to check the behavior of the automated inventory robot 3100 and the RFID tag reading status, and to give instructions for behavior as needed. Although the same functions can be achieved with a set of a PC (personal computer), monitor, mouse, and keyboard, the tablet 3040 is lighter in weight and is fixed into one, making it easier to use. The display screen is positioned facing backward in the direction of travel so that a person can follow the autonomously traveling automated inventory robot 3100 and operate it while walking.

[0017] 2 is a diagram showing an example of a store floor layout according to the first embodiment of the present invention. A floor plan 3700 includes wall fixtures 3210 and 3230 that are adjacent to the wall, and a central fixture 3220 that looks like an island. "A001," "B001," etc. are shelf numbers. However, the floor plan 3700 does not have coordinate information with a certain point on the floor as the origin.

[0018] 3 is a diagram showing functional blocks of the tablet 3040 according to the first embodiment of the present invention. The tablet 3040 includes a control unit 3041, a storage unit 3042, and an input / output unit 3043. The control unit 3041 controls the operation of the tablet 3040 and also creates instruction information for controlling the movement of the automated inventory robot 3100. The control unit 3041 includes a duplicate reading rate calculation unit 3046 and a movement instruction creation unit 3047.

[0019] The reader 3030 reads RFID tags multiple times per second. Therefore, some of the RFID tags read this time are the same as the RFID tags read up to the previous reading. The proportion of RFID tags that have already been read up to the previous reading among the RFID tags read this time is defined as the "duplicate read rate." The duplicate read rate calculation unit 3046 calculates the duplicate read rate.

[0020] The reader 3030 sequentially stores the IDs read from the RFID tags in the storage unit 3042. The IDs read from the position RFID tags 3450 are linked to the IDs of the product RFID tags 3350 read at the same time, and the product is recognized as being placed on the fixture 3200 to which the position RFID tag 3450 is attached. In other words, reading the position RFID tag determines the current location, and in the first embodiment, the reader 3030 can be said to also function as a current location calculation unit, and the ID of the position RFID tag has the meaning of current location information indicating the current location.

[0021] The movement instruction creation unit 3047 creates instruction information to control the movement of the automated inventory robot 3100 according to the overlap read rate. In the first embodiment, the basic operation is to move a predetermined distance of 1 m to 1.5 m at a predetermined speed of 500 m to 1 km per hour, and then stop for a predetermined rest time of 1 second. If the distance is 1 m to 1.5 m, even if the automated inventory robot 3100 passes an RFID tag leaving an unreadable RFID tag behind, it can still read it at the stationary position. If the predetermined movement distance is set longer than this, it becomes difficult to read RFID tags that have passed by without being read, even if the robot stops. On the other hand, if the predetermined movement distance is too short, the robot will stop frequently, reducing inventory efficiency. Therefore, in the first embodiment, in which the travel path of the automated inventory robot 3100 is set 30 cm to 90 cm from the fixture 3200, a predetermined movement distance of 1 m to 1.5 m was optimal.

[0022] Here, if the duplicate read rate is equal to or greater than the threshold, it is determined that "the duplicate read rate is high. → New RFID tags are no longer being read. → All RFID tags in this location have already been read." The robot may be controlled to omit the operation of stopping for a predetermined rest time, or to resume running without waiting for the predetermined rest time of one second to elapse. In the first embodiment, if the duplicate read rate exceeded 65%, the read rate achieved approximately 100%. Therefore, the threshold for the duplicate read rate is set to 65%, and in the first embodiment, if the duplicate read rate during running is less than the threshold, the robot performs a rest operation for a predetermined rest time, and resumes running when either the duplicate read rate exceeds 65% or one second has elapsed.

[0023] For example, in industries such as apparel and drugstores where a large number of items are displayed on fixtures, a 1% reading loss may not be a problem. In such cases, the threshold for the duplicate reading rate can be set low to reduce stationary time, or the travel speed can be set fast to shorten the time required for inventory. Conversely, in stores that handle brand-name goods or high-priced products, where even a 1% reading loss is unacceptable, it is advisable to set the threshold high or the travel speed slow. Such thresholds, stationary time, and travel speed can be set from the tablet 3040.

[0024] In addition, the basic operation for all or part of the stationary operation may be to move a predetermined distance at a predetermined speed, and then slow the moving speed for a predetermined period of time.In this case, if the overlapping reading rate is above a threshold, the operation of slowing the moving speed for a predetermined period of time may be omitted, or the time for slowing down may be controlled to be shorter.

[0025] The read RFID information is stored in the storage unit 3042. The input / output unit 3043 is a unit for inputting and outputting to and from the tablet 3040, and is a touch panel or the like.

[0026] 4 is a diagram showing the processing flow on the tablet 3040. Before START, the automated inventory robot 3100 is positioned at a predetermined start position within the store. Default values ​​and values ​​previously specified by the user for the travel speed, specified rest time, and specified travel distance are stored in the memory unit 3042. If a user-specified value exists, that set value is displayed on the display screen; if not, the default value is displayed. The START button is displayed on the screen, and the tablet is in a standby state waiting for the user to press it.

[0027] START clears the previous inventory data remaining in the memory unit 3042, and the automated inventory robot 3100 begins moving in a predetermined direction while activating the reader 3030 to start reading RFID tags. The RFID tags receive radio waves emitted from the upper antenna 3010 and lower antenna 3020 via the location RFID tag 3450 and product RFID tag 3350. The generated electromotive force transmits the specified ID stored in the RFID tag back to the reader 3030. The reader 3030 sequentially stores the RFIDs it reads in the memory unit 3042. In the first embodiment, the reader 3030 transmits radio waves 100 to 400 times per second and receives corresponding replies 100 to 400 times per second. Because each RFID tag replies with only a specific ID, the same ID is received each time a reply is received from the same RFID tag. Consequently, the same ID is stored multiple times in the memory unit 3042.

[0028] Here, if the read ID of the position RFID tag is different from the expected RFID tag ID, it can be assumed that START was pressed at a location other than the expected starting point, and the vehicle can stop traveling and enter a standby mode to wait for user input, or the vehicle can recognize its own location from the read RFID tag and the furniture layout map stored in memory unit 3042 and perform control to correct the route.

[0029] In step S3401, the movement instruction creation unit 3047 creates a movement instruction to move a predetermined distance at a predetermined speed and sends it to the travel control unit 3060. The movement instruction creation unit 3047 reads the ID of the last read position RFID tag from the memory unit 3042 as needed, and creates a travel history of the area within the store traveled by the automated inventory robot 3100 while comparing it with the fixture layout map. In step S3402, the duplicate read rate calculation unit 3046 calculates the duplicate read rate. It then determines whether the duplicate read rate is equal to or greater than a threshold. If the duplicate read rate is equal to or greater than the threshold, the process proceeds to step S3401. If the duplicate read rate is not equal to or greater than the threshold, the process proceeds to step S3403.

[0030] In step S3403, the automated inventory robot 3100 is stopped for the number of seconds set as the predetermined rest time. In step S3404, the travel history is used to determine whether the robot has reached the end point of its journey around the store floor. If so, the reader 3030 is stopped, and the automated inventory robot 3100 is caused to travel to the predetermined end position, ending its operation. If not, the process returns to step S3401. In this way, the movement of the automated inventory robot 3100 is controlled according to the duplicate read rate, reducing the chance of incorrectly determining location information without reading the location RFID tag.

[0031] FIG. 5 is a diagram showing an example of calculating the duplicate read rate. Item P1 indicates the number of times the reading has been performed. Item Q1 indicates the IDs of the product RFID tags that have been read. In FIG. 5, the number of IDs in item Q1 is four each time, but the number of IDs in item Q1 may vary depending on the time. Item R1 is the number of IDs of the product RFID tags that have been read.

[0032] The item S1 is the ID of the product RFID tag that has already been read from the item Q1. The item T1 is the number of IDs in the item S1. The item U1 is the overlap rate calculated by T1 / R1.

[0033] As described above, in the first embodiment of the present invention, the basic operation is to move a predetermined distance at a predetermined speed and then stop for a predetermined time, and if the duplicate reading rate is equal to or greater than a threshold, the operation of stopping for the predetermined time is controlled to be omitted. As a result, sufficient time can be secured to read the position RFID tag, and the occurrence of incorrect operation such as incorrectly determining the position of a product is reduced.

[0034] 6 is a diagram showing an obstacle map creation robot 3121 according to a second embodiment of the present invention. The reader 3030, upper antenna 3010, lower antenna 3020, and wheels 3050 of the obstacle map creation robot 3121 are the same as those of the automated inventory robot 3100 according to the first embodiment.

[0035] The obstacle map creating robot 3121 includes a tablet 3040, a human following travel control unit 3090, and a lidar unit 3080. The tablet 3040 creates a map of obstacles on the floor based on information acquired by the lidar unit 3080. The tablet 3040 also creates a conversion table from floor position coordinates to shelf numbers by overlaying the created obstacle map on a floor plan 3700.

[0036] The LIDAR unit 3080 has a LIDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging) function. LIDAR is a type of remote sensing technology that uses light and measures scattered light in response to pulsed laser irradiation, allowing it to measure the distance to a distant object. The LIDAR unit 3080 is oriented in the same direction as the upper antenna 3010 and the lower antenna 3020, and in the second embodiment, has directivity to the right of the direction of travel of the obstacle map creation robot 3121, which travels in the depth direction of the drawing. The LIDAR unit 3080 records the distance measurement results in the memory unit 3042 as needed, linking them to time information from the start of operation. The human-following travel control unit 3090 controls the obstacle map creation robot 3121 to move behind a person walking around the floor. The lidar unit 3080 also constantly measures the distance to a person walking in front of the obstacle map creation robot 3121 and records the distance from the obstacle map creation robot 3121 to the person in the memory unit 3042.

[0037] The human-following running control unit 3090 outputs a drive command to the motor (not shown) to drive the wheels 3050 so as to follow the person within a range that does not exceed a pre-set maximum speed. At this time, the human-following running control unit 3090 uses the start point as the origin and calculates the coordinates of its own position from the history of drive commands to the motor, i.e., the running speed and time information, as needed, and stores this in the memory unit 3042. In this way, a running history is created from the history of the coordinates of its own position. In addition, the coordinates of obstacles can be calculated and created as a map from the distance measurement results of the lidar unit 3080 and its own position.

[0038] Here, among the objects for which measurement data from the lidar unit 3080 exists, those whose coordinate positions do not change over time are reflected as obstacles in the obstacle map, while those whose coordinate positions change over time, such as people, are not reflected in the obstacle map. Therefore, during inventory work, an item that has fallen on the floor 3400 is initially detected by the lidar unit 3080, but once it is no longer detected because the person puts the item away, it is not reflected in the obstacle map, and even if it was once reflected, it is removed from the map and updated. The created map is updated as needed and displayed on the tablet 3040.

[0039] In this way, the obstacle map creating robot 3121 runs while creating a map in accordance with its own running, and is able to recognize its own position as coordinates on that map. As in the first embodiment, the reader 3030 reads the ID of the commodity RFID tag as needed through the upper antenna 3010 and the lower antenna 3020, and records the read commodity RFID tag ID in the memory unit 3042 in association with the coordinates of the obstacle (i.e., presumed to be the fixture 3200) detected at that time.

[0040] A plurality of products 3300 are placed on the fixture 3200, and product RFID tags 3350 are attached to the products 3300, as in the first embodiment. However, in the second embodiment, the position RFID tags 3450 are not attached to the fixture 3200.

[0041] 7 is a diagram showing functional blocks of the tablet 3040 according to the second embodiment of the present invention. The tablet 3040 includes a control unit 3041, a storage unit 3042, and an input / output unit 3043. The control unit 3041 creates an obstacle map of the floor. In addition, by overlaying the created obstacle map on a floor plan 3700, a conversion table from floor position coordinates to shelf numbers is created.

[0042] The control unit 3041 includes a travel history recording unit 4101, a current position coordinate calculation unit 4102, an obstacle map creation unit 4103, an overlay unit 4104, a section determination unit 4105, and a section / shelf number conversion table creation unit 4106. The travel history recording unit 4101 keeps a record of the movement of the obstacle map creation robot 3121, including the direction and distance it has moved. This movement record is stored in the memory unit 3042.

[0043] When the obstacle map creating robot 3121 moves within a floor, the current position coordinate calculation unit 4102 calculates the coordinates of the current position when a certain point within the floor is set as the origin, based on the travel history recorded by the travel history recording unit 4101. The obstacle map creating unit 4103 creates a map (hereinafter referred to as an "obstacle map") showing the positions of obstacles such as walls or furniture, based on the information created by the rider unit 3080. The superimposition unit 4104 performs processing such as rotation and enlargement / reduction on the obstacle map created by the obstacle map creating unit 4103, and then superimposes it on the floor plan 3700.

[0044] The section determination unit 4105 divides the aisles of the floor into a plurality of sections corresponding to each shelf based on the result of overlaying the obstacle map on the floor plan 3700. The section / shelf number conversion table creation unit 4106 creates coordinate information that defines the sections, and creates a correspondence table between the coordinate information and the shelf number.

[0045] Figure 8 is an example of an obstacle map 3800 created by the obstacle map creation unit 4103. The obstacle map 3800 in Figure 8 is an example of a map created by the obstacle map creation robot 3121 following a human as it moves around the floor in Figure 2. Number 3840 corresponds to the wall portion of the floor in Figure 2. Number 3830 is the portion recessed from the wall by the wall fixture 3210. Number 3810 corresponds to the passage portion. Number 3820 is the portion corresponding to the central fixture 3220.

[0046] 9 is a diagram showing the result of the overlay unit 4104 overlaying the obstacle map 3900, which is obtained by rotating and scaling the obstacle map 3800, onto the floor plan 3700. As described above, the floor plan 3700 does not necessarily have coordinate information with a certain point on the floor as its origin. In contrast, the obstacle map 3900 has coordinate information with a certain point on the floor (typically the starting point where the obstacle map creation robot 3121 starts its operation) as its origin. Therefore, by overlaying the obstacle map 3800 created by the obstacle map creation robot 3121 on the floor plan 3700, it is possible to obtain the coordinate information of each piece of furniture on the obstacle map 3900.

[0047] Figure 10 is a diagram showing the state in which the section determination unit 4105 has divided the aisle into multiple sections A through C corresponding to each piece of furniture. By creating the correspondence relationship shown in Figure 10, when an automated inventory robot receives radio waves from a product RFID tag on a floor where no positional RFID tags are attached, as shown in Figure 6, the robot can identify the section based on the coordinate position of the automated inventory robot and narrow down the shelves on which the product is located to two. For example, if the automated inventory robot is determined to be in section C based on the coordinates of its current location when radio waves are received from a product RFID tag, the robot can estimate that the product is on shelf A003 or B003.

[0048] Figure 11 is a diagram showing an example of a section / shelf number conversion table 5000 created by the section / shelf number conversion table creation unit 4106. Item V1 is a section. Item W1 is coordinate information that defines the section. In Figure 11, the coordinate information that defines the section is expressed as a round number, but in reality, it would be an odd number. Item X1 is the shelf number of the shelf where the product with the product RFID tag is presumed to be located when the automated inventory robot that received the radio waves from the product RFID tag determines that it is in the section of item V1.

[0049] When an automated inventory robot receives radio waves from a product RFID tag, the first thing that is determined is the coordinates of the automated inventory robot's current location. Based on the coordinates of the automated inventory robot's current location, the shelf where the product is likely to be located is narrowed down by referencing the section / shelf number conversion table 5000. In the case of the table in Figure 11, this narrows down the list to two shelves. Which of the two candidate shelves is the shelf can be determined by the orientation of the upper antenna 3010 and lower antenna 3020 when the automated inventory robot receives the radio waves from the product RFID tag.

[0050] In this way, if the obstacle map creation robot 3121 creates an obstacle map and then creates a section / shelf number conversion table based on that obstacle map, inventory can be performed by an automatic inventory robot even on floors where positional RFID tags are not affixed.

[0051] [Embodiment 3] In embodiment 2, the overlay unit 4104 performs a process of overlaying an obstacle map 3900, which is obtained by rotating and scaling the obstacle map 3800, onto a floor plan 3700, as shown in Fig. 9. However, because the obstacle map 3800 is a map created by the obstacle map creation robot 3121 through actual measurements, it is possible to predict a situation in which the obstacle map 3800 will not be properly overlaid on the floor plan 3700 no matter how much it is rotated and scaled. Therefore, it would be preferable if the process of the overlay unit 4104 to overlay the obstacle map 3900, which is obtained by rotating and scaling the obstacle map 3800, onto the floor plan 3700 could be omitted.

[0052] The current position of the automated inventory robot 3100 or the obstacle map creation robot 3121, based on its travel history, is determined with the robot's starting point as the origin. Therefore, if the coordinates of the robot's starting point on the floor plan 3700 are known, it is possible to determine the robot's current position on the floor plan. However, the robot can only record the amount of movement in the x-axis direction and the y-axis direction for the robot. What the robot records is not necessarily the amount of movement in the x-axis direction and the y-axis direction on the floor plan.

[0053] FIG. 12 is a floor plan for explaining the third embodiment. Point A in FIG. 12 is the origin of the floor plan. Point B is the starting point of the robot. Point C is the current position of the robot. The coordinates of point B in a coordinate system with point A as the origin can be obtained using a ruler tool in drawing software. It is assumed that the y-axis direction for the robot is the direction obtained by rotating the x-axis direction for the robot by 90 degrees counterclockwise. Whether the x-axis direction for the robot coincides with the x-axis direction of the coordinate system with point A as the origin and whether the y-axis direction for the robot coincides with the y-axis direction of the coordinate system with point A as the origin depends on the orientation of the robot at point B, which is the starting point.

[0054] At point B, which is the starting point, if the x-axis direction for the robot coincides with the x-axis direction of a coordinate system with point A as the origin, and the y-axis direction for the robot coincides with the y-axis direction of the coordinate system with point A as the origin, then the coordinate of point C in the coordinate system with point A as the origin can be determined simply by adding the amount of movement in the x-axis direction for the robot to the x-coordinate of point B, and adding the amount of movement in the y-axis direction for the robot to the y-coordinate of point B.

[0055] However, if the orientation of the robot at point B, its starting point, is rotated 90 degrees clockwise, the x-axis direction for the robot becomes the negative y-axis direction of the coordinate system whose origin is point A. The y-axis direction for the robot becomes the x-axis direction of the coordinate system whose origin is point A. Therefore, in this case, the coordinate of point C in the coordinate system whose origin is point A can be determined by adding the amount of movement in the y-axis direction for the robot to the x-coordinate of point B and subtracting the amount of movement in the x-axis direction for the robot from the y-coordinate of point B.

[0056] If the orientation of the robot at point B, its starting point, is rotated another 90 degrees clockwise, the x-axis direction for the robot becomes the negative x-axis direction of the coordinate system whose origin is point A. The y-axis direction for the robot becomes the negative y-axis direction of the coordinate system whose origin is point A. Therefore, in this case, the coordinate of point C in the coordinate system whose origin is point A can be determined by subtracting the amount of movement in the x-axis direction for the robot from the x-coordinate of point B, and subtracting the amount of movement in the y-axis direction for the robot from the y-coordinate of point B.

[0057] If the orientation of the robot at point B, its starting point, is rotated another 90 degrees clockwise, the x-axis direction for the robot becomes the y-axis direction in a coordinate system with point A as the origin. The y-axis direction for the robot becomes the minus x-axis direction in the coordinate system with point A as the origin. Therefore, in this case, the coordinate of point C in the coordinate system with point A as the origin can be determined by subtracting the amount of movement in the y-axis direction for the robot from the x-coordinate of point B and adding the amount of movement in the x-axis direction for the robot to the y-coordinate of point B.

[0058] As described above, the robot can record only the amount of movement in the x-axis direction and the amount of movement in the y-axis direction. Therefore, by inputting the orientation of the robot at point B, which is the starting point, into the robot, the amount of movement of the robot can be appropriately added to the coordinate value of point B, which is the starting point, to determine the coordinates of point C, which is the robot's current position with point A as the origin.

[0059] Once the coordinates of point C, which is the robot's current position with point A as the origin, are determined, the positions of all fixtures are set on floor plan 3700, so based on the orientation of the robot's antenna when it receives radio waves from the product RFID tag, it can be determined that "shelf D003 contains the product with the product RFID tag that received the radio waves," or "shelf C003 contains the product with the product RFID tag that received the radio waves."

[0060] In the first embodiment, the upper antenna 3010, the lower antenna 3020, and the lidar unit 3080 are arranged with their directivity facing to the right of the direction of travel, but they may of course be arranged to the left, in another direction, on both sides, or in all directions. In this case, antennas facing different directions will simultaneously detect the same RFID tag, so an algorithm is required to identify in which direction the RFID tag is located from the automated inventory robot. While such identification is possible by measuring changes in the RFID tag read rate due to changes in position or by measuring radio wave intensity, in the first embodiment, the direction of the RFID tag is identified in a simpler manner by arranging the antennas with directivity in only one direction.

[0061] Furthermore, in many stores, the distance between fixtures is set wide enough for shopping carts used in the store to pass each other. In the first embodiment, the automated inventory robot 3100 travels along a path 30 to 90 cm away from the fixture 3200. However, in physical stores, it is often not possible to provide 30 to 90 cm on either side of the automated inventory robot 3100. The directivity of the upper antenna 3010 and the lower antenna 3020 is not very wide when the antennas are in close proximity to each other. By providing a distance of 30 to 90 cm, the two antennas can cover the entire area from the floor 3400 to the ceiling. Therefore, even if antennas are placed on both sides of the automated inventory robot 3100, it is difficult to ensure a distance of 30 to 90 cm from both fixtures. Therefore, unless a larger number of antennas are installed, it will be impossible to read the product RFID tags 3350.

[0062] Therefore, arranging the upper antenna 3010 and the lower antenna 3020 on one side provides better cost performance. There is no problem with arranging the lidar unit 3080 on both sides, but if the direction for reading RFID tags is one-sided, even if only the lidar unit 3080 is oriented on both sides, it will be necessary to pass through the same passage again to orient the antenna, and so although the accuracy of the obstacle map may improve, the benefit is not enough to justify the increased cost. Therefore, it is sufficient to arrange the lidar unit 3080 only in the same direction as the antenna that reads RFID.

[0063] Although the above embodiments 1 to 3 have been described using a store as an example, this is not limited to this. RFID tags may be attached to tools in a factory, or to medical instruments in a medical institution, and various other applications are possible in which an RFID tag is attached to an object to be searched and the position of the RFID tag is identified using a self-propelled robot.

[0064] The above embodiment has been described using location RFID tags and product RFID tags, but wireless communication tags such as short-range Bluetooth (registered trademark) or optically readable tags such as QR codes (registered trademark) or barcodes may also be used. RFID tags do not require a power source and can be read even when products are stacked, making them more suitable than wireless communication or optically readable tags.

[0065] 3010 Upper antenna 3020 Lower antenna 3030 Reader 3040 Tablet 3041 Control unit 3042 Memory unit 3046 Duplicate reading rate calculation unit 3047 Movement instruction creation unit 3050 Wheels 3060 Travel control unit 3080 Lidar unit 3090 Human following travel control unit 3100 Automatic inventory robot 3121 Obstacle map creation robot 3200 Fixtures 3350 Product RFID tag 3450 Position RFID tag 3700 Floor plan 3800 Obstacle map 4101 Travel history recording unit 4102 Current position coordinate calculation unit 4103 Obstacle map creation unit 4104 Superposition unit 4105 Section determination unit 4106 Section / shelf number conversion table creation unit 5000 Section / shelf number conversion table

Claims

1. An automated inventory robot that automatically travels around a store floor and reads product RFID tags and location RFID tags, comprising: a duplicate read rate calculation unit that calculates a duplicate read rate, which is the percentage of duplicate reads of product RFID tags; a movement instruction creation unit that creates movement instructions to control the movement of the automated inventory robot based on the duplicate read rate; and a travel control unit that controls the movement of the automated inventory robot in accordance with the movement instructions.

2. The automatic inventory robot of claim 1, wherein the movement instruction creation unit creates movement instructions that are based on the operation of moving a predetermined distance at a predetermined speed and then stopping for a predetermined period of time, and if the duplicate reading rate is above a threshold, omits the operation of stopping for a predetermined period of time.

3. The automated inventory robot according to claim 1, which is equipped with directional upper and lower antennas.

4. An obstacle map creating robot comprising: a human-following driving control unit that causes the obstacle map creating robot to drive while following a human; a driving history recording unit that records the driving history; a current position coordinate calculation unit that calculates the current position coordinates of the obstacle map creating robot based on the driving history; a lidar unit that detects the distance to an obstacle; an obstacle map creating unit that creates an obstacle map based on the distance to the obstacle and the current position coordinates of the obstacle map creating robot; and an overlay unit that overlays the obstacle map on a floor plan.

5. An obstacle map creation robot as described in claim 4, further comprising a section determination unit that divides the aisles of the floor into a plurality of sections corresponding to each shelf based on the result of overlaying the obstacle map and the floor plan.

6. An obstacle map creating robot according to claim 5, further comprising a section / shelf number conversion table creating unit that creates coordinate information defining the section and creates a correspondence table between the coordinate information and shelf numbers.

7. A control method for an automatic search robot that is capable of moving and reading RFID tags, the control method comprising: calculating a duplicate read rate, which is the percentage of RFID tags being searched that are read twice; and controlling the movement of the automatic search robot based on the duplicate read rate.

8. A control method according to claim 7, wherein the self-position of the automatic search robot is determined by reading a positional RFID tag attached to a predetermined position in a planned travel area.

9. A control method according to claim 7, wherein the travel route of the automatic search robot is recorded sequentially, and the self-position of the automatic search robot is determined from the travel route.

10. A control method according to claim 8 or 9, wherein the self-location and the read result of the RFID tag of the search target are recorded in association with each other.

11. An automated inventory robot that travels within a specified area and reads product RFID tags affixed to products placed within the specified area, comprising: a current position calculation unit that calculates current position information that enables determination of the current position of the automated inventory robot within the specified area; a memory unit that stores information read from the product RFID tags in association with the current position information; a duplicate read rate calculation unit that calculates a duplicate read rate, which is the rate at which the product RFID tags are read twice; and a movement control unit that creates movement instructions to control the movement of the automated inventory robot based on the duplicate read rate.

12. An automated inventory robot that automatically travels around a store floor and reads product RFID tags, and is equipped with a control unit that controls the process of calculating the coordinates of the current position of the automated inventory robot by adding the amount of movement of the automated inventory robot to the coordinates of the starting point based on the orientation of the automated inventory robot at the starting point.

Images