Mobile object and control method thereof

Abstract

A moving body according to the present disclosure: includes an image capture device that captures an image of a recognition target, and a recognition unit that recognizes the recognition target on the basis of image data captured by the image capture device; and performs a predetermined operation on the basis of the recognition target recognized by the recognition unit. The moving body further includes an image capture control unit that, when the recognition unit does not recognize the recognition target, changes the image capture conditions of the image capture device in stages so that the recognition unit recognizes the recognition target, and causes a storage device to store the image capture conditions when the recognition unit recognizes the recognition target, wherein, under the same or similar circumstances, the image capture control unit applies the image capture conditions stored in the storage device when the image capture device captures an image of the recognition target or another recognition target. This makes it possible to ensure smooth operation of the moving body by allowing the moving body to recognize a recognition target on the basis of image data captured by the image capture device even if the ambient environment or the like changes.

Description

[Technical Field]

[0001] The present disclosure relates to a moving object that recognizes a recognition target based on image data captured by an imaging device, and a control method thereof. [Background technology]

[0002] Conventionally, a robot device mounted on an automated guided vehicle is known that includes a camera attached to a hand unit and that captures images of a workpiece set on a workbench (see, for example, Patent Document 1). To recognize the shape and position of a workpiece using the camera, this robot device sequentially adjusts the aperture value of an aperture adjustment device attached to the camera while capturing an image of a master workpiece, learning the optimal aperture value corresponding to the illuminance. Furthermore, the robot device changes parameter values ​​while the optimal aperture value is set, thereby determining the optimal parameter value corresponding to the illuminance when the feature values ​​of the master workpiece are closest to the target value, and storing the parameter value in memory. When recognizing a workpiece, the robot device adjusts the aperture adjustment device to the optimal aperture value corresponding to the current illuminance, sets the optimal parameter value corresponding to the current illuminance as a parameter for feature extraction, and recognizes the workpiece based on the image signal from the camera. This allows the workpiece to be recognized even if the brightness of the work position where the workpiece is set fluctuates. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 11-272845 Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, the efficiency of transporting goods can be improved by having a mobile object, such as an AMR (Autonomous Mobile Robot) or an AGV (Automatic Guided Vehicle), recognize a recognition target, such as a marker or a transport target, based on image data (image data) captured by an imaging device. However, the surrounding environment of the mobile object changes depending on the time of day and the location of the mobile object, and if the mobile object is unable to recognize the recognition target due to changes in the surrounding environment, this can cause disruption to work. Furthermore, the locations where mobile objects are used and the ranges of movement are diverse, making it virtually impossible to fully learn the imaging conditions of the imaging device in advance.

[0005] Therefore, a main object of the present disclosure is to ensure smooth operation of a moving object by having the moving object recognize a recognition target based on imaging data from an imaging device, even when the surrounding environment or the like changes. [Means for solving the problem]

[0006] The mobile body of the present disclosure includes an imaging device that captures an image of a recognition target, and a recognition unit that recognizes the recognition target based on image data from the imaging device, and the mobile body performs a predetermined operation based on the recognition target recognized by the recognition unit.When the recognition unit does not recognize the recognition target, the mobile body includes an imaging control unit that gradually changes the imaging conditions of the imaging device so that the recognition unit recognizes the recognition target, stores the imaging conditions when the recognition unit recognizes the recognition target in a storage device, and applies the imaging conditions stored in the storage device when the imaging device captures an image of the recognition target or another recognition target under the same or similar circumstances.

[0007] In the mobile body disclosed herein, when the recognition unit does not recognize the recognition target, the imaging conditions of the imaging device are gradually changed so that the recognition unit can recognize the recognition target, and the imaging conditions when the recognition unit recognizes the recognition target are stored in the storage device. Furthermore, when the imaging device captures an image of the recognition target or another recognition target under the same or similar conditions, the imaging conditions stored in the storage device are applied. This allows the mobile body to recognize the recognition target based on the imaging data of the imaging device even if the surrounding environment changes, for example, and therefore makes it possible to ensure smooth operation of the mobile body without having to fully learn the imaging conditions of the imaging device in advance.

[0008] The control method for a moving body of the present disclosure is a control method for a moving body that includes an imaging device that captures an image of a recognition target and a recognition unit that recognizes the recognition target based on image data of the imaging device, and when the recognition unit does not recognize the recognition target, the imaging conditions of the imaging device are gradually changed so that the recognition unit can recognize the recognition target, the imaging conditions when the recognition unit recognizes the recognition target are stored in a storage device, and the imaging conditions stored in the storage device are applied when the imaging device captures an image of the recognition target or another recognition target under the same or similar conditions.

[0009] According to this method, even if the surrounding environment changes, the moving body can recognize the recognition target based on the imaging data of the imaging device, so it is possible to ensure smooth operation of the moving body without having to fully learn the imaging conditions of the imaging device in advance. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a schematic configuration diagram showing a logistics center to which a mobile body according to the present disclosure is applied; [Figure 2] FIG. 2 is a block diagram showing a logistics management system for the logistics center of FIG. 1. [Figure 3] FIG. 2 is a perspective view showing a cart used in the logistics center of FIG. 1. [Figure 4]FIG. 1 is a perspective view showing a moving body according to the present disclosure. [Figure 5] FIG. 2 is a control block diagram of a moving body according to the present disclosure. [Figure 6] 10 is a flowchart illustrating a procedure for recognizing a recognition target in a moving body according to the present disclosure. [Figure 7] 10A and 10B are schematic diagrams showing modified aspects of the moving body of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Next, embodiments of the present disclosure will be described with reference to the drawings.

[0012] FIG. 1 is a schematic diagram illustrating a logistics center 1 to which an autonomous mobile robot (AMR) 50, which is a mobile object according to the present disclosure, is applied. FIG. 2 is a block diagram illustrating a logistics management system 10 of the logistics center 1. The logistics center 1 illustrated in FIG. 1 is a facility where storage, transportation, loading and unloading, packaging, distribution processing, and the like of packages (goods) are performed. As illustrated, the logistics center 1 includes, in addition to multiple transport robots 50, an entrance 2 for delivery vehicles T such as trucks, a shipping area 3 having multiple shipping gates 3a, 3b, 3c, 3d, 3e, 3f, 3g, 3h, 3i, 3j, 3k, and 3l (hereinafter, the shipping gates will be collectively referred to as "shipping gate 3x" as appropriate), a conveyor 4 for transporting a large number of packaged packages P from a warehouse (not shown) to the shipping area 3, and multiple carts 5. In the logistics center 1, the packages P are transported by the conveyor 4 to the shipping area 3 and transferred onto the carts 5. Furthermore, the cart 5 as a loading and unloading tool carrying the package P is transported to the corresponding shipping gate 3x by the transport robot 50. Then, the cart 5 carrying the package P is loaded onto the corresponding delivery vehicle T at the shipping gate 3x, and is transported to the shipping destination by the delivery vehicle T.

[0013] At the entrance 2 of the logistics center 1, an entrance camera 20 is installed to capture images of delivery vehicles T arriving at the entrance 2, a monitor 21 for visually providing various information to the driver of the delivery vehicle T, and an entrance management device 25 (see FIG. 2 ). The entrance management device 25 is a computer including a CPU, ROM, RAM, a storage device, a communication module, etc., and acquires the ID of the delivery vehicle T arriving at the entrance 2 based on the image data (image data) captured by the entrance camera 20. In other words, the entrance camera 20 and the entrance management device 25 function as an ID acquisition device for the entrance 2. In this embodiment, the ID of the delivery vehicle T is, for example, a number assigned to the delivery vehicle T in advance by the logistics center 1, and a marker (not shown) indicating the ID is affixed to the delivery vehicle T so that it can be recognized by the entrance camera 20. However, the ID of the delivery vehicle T may be the vehicle registration number of the delivery vehicle T or the number of an on-board device such as an ETC (Electronic Toll Collection) acquired via a communication device (not shown). Furthermore, the entrance management device 25 displays information to be provided to the driver on the monitor 21.

[0014] Each shipping gate 3a-3l is equipped with a camera 30 that captures images of delivery vehicles T arriving at the shipping gate 3a-3l, a trolley detector 31 that can capture images of trolleys 5 passing through when loading onto the delivery vehicles T, and a shipping gate management device 35 (see FIG. 2). The shipping gate management device 35 is a computer including a CPU, ROM, RAM, storage device, communication module, etc., and acquires the ID of the delivery vehicle T arriving at the shipping gate 3a, 3b, ..., or 3l based on the image data (image data) captured by the camera 30. That is, the camera 30 and shipping gate management device 35 function as a delivery vehicle ID acquisition device at each shipping gate 3a-3l. The shipping gate management device 35 also acquires the ID of the trolley 5 to be loaded onto the delivery vehicle T based on the image data (image data) captured by the trolley detector 31. That is, the trolley detector 31 and shipping gate management device 35 function as a trolley ID acquisition device at each shipping gate 3a-3l.

[0015] The conveyor 4 includes a main conveyor line 40 and multiple shipping lines 4a, 4b, 4c, 4d, 4e, 4f, 4g, 4h, 4i, 4j, 4k, and 4l branching off from the main conveyor line 40 so as to face a corresponding one of the shipping gates 3a-3l (hereinafter, the shipping lines will be collectively referred to as "shipping lines 4x" as appropriate). The conveyor 4 is controlled by a conveyor control device 14 (see FIG. 2), which is a computer including a CPU, ROM, RAM, storage device, and communication module. Each package P handled at the logistics center 1 has a barcode sticker affixed thereto indicating its ID. The conveyor control device 14 controls the conveyor 4 based on the package P's ID. Specifically, the conveyor control device 14 obtains the ID of each package P on the main conveyor line 40 via a barcode reader (not shown), and controls the conveyor 4 to sort each package P transported to the shipping area 3 by the main conveyor line 40 into the corresponding shipping line 4x based on the obtained ID.

[0016] The cart 5 is a so-called basket cart as shown in FIG. 3, and includes a marker 5m as a recognition target, such as an AR marker, a QR code, or a barcode, which indicates the ID of the cart 5. The marker 5m is provided in at least one location (two locations in the example of FIG. 3) on the cart 5 so that it can be recognized by the transport robot 50 and the cart detector 31. In addition, in the logistics center 1, as shown in FIG. 1, a storage location (distribution location) 6 is secured within the shipping area 3. Unused carts 5 are aligned and stored in the storage location 6, and carts 5 to be used are identified by the transport robot 50 and transported from the storage location 6 to a designated location. Furthermore, a plurality of markers 3m as recognition targets, such as AR markers and QR codes, which allow each transport robot 50 to travel autonomously, are attached to the floor, pillars, etc. of the shipping area 3 within the building of the logistics center 1.

[0017] As shown in FIG. 2, the logistics management system 10 includes a management server 11 as a management device and a storage device 12. The management server 11 is a computer having a CPU, ROM, RAM, a communication module, etc. The management server 11 acquires various information from the entrance management device 25 and the shipping gate management devices 35 of each shipping gate 3a-3l, and transmits requested information and command signals to the entrance management device 25 and each shipping gate management device 35. Furthermore, the management server 11 transmits information necessary for sorting the packages P, such as the ID of the packages P, to the conveyor control device 14, and exchanges information with the AMR management device 15. The AMR management device 15 is a computer having a CPU, ROM, RAM, a storage device, a communication module, etc., and manages the multiple transport robots 50 by exchanging information with the multiple transport robots 50 via wireless communication.

[0018] The storage device 12 of the logistics management system 10 also stores a delivery database in which information regarding the delivery of packages P is stored for each of multiple delivery vehicles T that have been assigned an ID by the logistics center 1. The delivery database stores the ID of a delivery vehicle T, in association with the ID of the package P to be loaded onto the delivery vehicle T, the shipping destination of the package P, and the ID of a shipping gate 3x (hereinafter referred to as "loading gate 3z" as appropriate) where the package P is loaded onto the delivery vehicle T. Furthermore, the storage device 12 stores area information in the shipping area 3 (logistics center 1), including areas where the transport robot 50 is allowed to move and areas where movement is prohibited, identification sign information for each area, position information of the marker 3m, time periods when movement is prohibited, etc.

[0019] FIG. 4 is a perspective view showing a transport robot 50, and FIG. 5 is a control block diagram of the transport robot 50. As shown in FIGS. 4 and 5, the transport robot 50 includes a plurality of (four) Mecanum wheels 51, a plurality of (four) electric motors 52 that rotate and drive the corresponding Mecanum wheels 51, and a lifting unit 53 that raises and lowers the carriage 5. The transport robot 50 is configured to be movable forward and backward below the carriage 5. Each Mecanum wheel 51 includes a pair of support plates 51p and a plurality of rollers 51r that are supported by the pair of support plates 51p in a circular arrangement and can rotate around an axis inclined at 45° relative to the axle. As a result, by controlling the rotation direction and rotation speed of the plurality of electric motors 52, the transport robot 50 can be moved in all directions and can be made to make super-pivot turns, pivot turns, or gentle turns. The lifting unit 53 includes a support member that supports the cart 5 and a drive device that raises and lowers the support member relative to the main body of the transport robot 50 so that each wheel of the cart 5 moves away from the traveling road surface or makes contact with the traveling road surface.

[0020] Furthermore, the transport robot 50 includes a camera 54 as an imaging device, a control device 55, a communication module 56 used for communication (wireless communication) with the AMR management device 15 and other transport robots 50, and a storage device 57. The camera 54 captures images of the surroundings of the transport robot 50 and transmits the captured image data (image data) to the control device 55. The control device 55 is a computer including a CPU, ROM, RAM, etc. As shown in FIG. 5, the control device 55 includes an arithmetic processing unit 55G, a travel control unit 55D, and an elevation control unit 55L, which are configured through cooperation between hardware such as the CPU, ROM, and RAM and various pre-installed programs.

[0021] The arithmetic processing unit 55G of the control device 55 utilizes VSLAM technology (self-position estimation technology) and acquires the self-position (3D coordinates) of the transfer robot 50 in the shipping area 3 (logistics center 1) and an environmental map based on the image data of the camera 54 at predetermined time intervals (short time intervals) while the transfer robot 50 is traveling, and stores these in the storage device 57. The arithmetic processing unit 55G can recognize (identify) the markers 3m in the shipping area 3 and the markers 5m (IDs of the carts 5) on the dollies 5 from the image data of the camera 54, as well as recognize (identify) the objects themselves, such as the dollies 5. Furthermore, the arithmetic processing unit 55G recognizes the markers 3m arranged in the shipping area 3 while the transfer robot 50 is traveling, and corrects the estimated self-position and environmental map based on the previously determined positions of the markers 3m. The arithmetic processing unit 55G can also adjust the sensitivity (gain of the image sensor) and exposure time (shutter speed) of the camera 54. Furthermore, the calculation processing unit 55G stores in the memory device 57 the imaging conditions of the camera 54, i.e., the sensitivity and exposure time, when the recognition target such as markers 3m and 5m was successfully recognized from the imaging data of the camera 54, in association with situational information, i.e., the location (ID) of the marker 3m, etc., the time period and season when the marker 3m, etc. was imaged.

[0022] The travel control unit 55D of the control device 55 controls the multiple electric motors 52 based on the self-position (current position) and the environmental map acquired by the arithmetic processing unit 55G. Furthermore, the lifting control unit 55L controls the lifting unit 53 to raise or lower the cart 5 when the transport robot 50 is positioned below the corresponding cart 5. The arithmetic processing unit 55G of the control device 55 may be configured to acquire the self-position and the environmental map using SLAM technology using a 2D or 3D LiDAR (laser sensor) or the like, may be configured to estimate the self-position based only on the position of a marker, or may be configured to acquire the self-position and the environmental map using indoor positioning technology using a beacon or the like. Furthermore, the transport robot 50 may include wheels other than Mecanum wheels, such as wheels including general rubber tires.

[0023] The shipping area 3 of the logistics center 1 where the transport robot 50 is used has a large area, and the amount of sunlight varies depending on the location. Therefore, the surrounding environment of the transport robot 50 changes depending on the time of day and the location of the transport robot 50 in the shipping area 3. This change in the surrounding environment can cause the transport robot 50 to be unable to recognize markers 3m, 5m, etc., as recognition targets, which can interfere with tasks such as estimating its own position and transporting the cart 5 using the estimated self-position. Meanwhile, the transport robot 50 has a wide range of movement, making it virtually impossible to fully learn the imaging conditions of the camera 54 as an imaging device in advance. Taking these factors into consideration, the transport robot 50 recognizes markers 3m, 5m, etc. according to the procedure shown in FIG. 6 to prevent misrecognition of markers 3m, 5m, etc. Below, the series of processes shown in FIG. 6 will be described using an example case in which the transport robot 50 recognizes a marker 3m, etc., affixed to the floor or pillar of the shipping area 3, as a recognition target, when transporting the cart 5 from one of the shipping lines 4x to the corresponding shipping gate 3x.

[0024] In the logistics center 1, in response to the arrival of the delivery vehicle T at the entrance 2, the management server 11 determines the shipping gate 3x that will become the loading gate 3z and transmits necessary information such as the ID of the shipping gate 3x to the AMR management device 15. The AMR management device 15 transmits the necessary information, including the ID of the shipping gate 3x that will become the loading gate 3z, to the corresponding transport robot 50, and reflects the shipping gate 3x that will become the target point in the travel route of the transport robot 50. The transport robot 50 that receives the information from the AMR management device 15 recognizes (identifies) the cart 5 with the corresponding ID from the image data captured by the camera 54, and travels based on its own position estimated by the calculation processing unit 55G to transport the cart 5 to the shipping gate 3x.

[0025] When the transfer robot 50 travels within the shipping area 3 in this manner, the arithmetic processing unit 55G of the control device 55 of the transfer robot 50 determines that the self-position of the transfer robot 50 estimated based on the imaging data of the camera 54 is within a range where any one of the markers 3m in the shipping area 3 can be imaged, the arithmetic processing unit 55G reads out the imaging conditions (sensitivity and exposure time) of the camera 54 stored in the storage device 57 (step S100). Furthermore, the arithmetic processing unit 55G determines whether the read-out imaging conditions include imaging conditions under which the marker 3m could be normally recognized from the imaging data of the camera 54 in the same time period as the current time (hereinafter referred to as "suitable imaging conditions") (step S110).

[0026] If the processing unit 55G determines that the imaging conditions stored in the storage device 57 do not include suitable imaging conditions (step S110: YES), it requests the other transport robots 50 and the AMR management device 15 (other devices) to transmit suitable imaging conditions (step S120), and determines whether suitable imaging conditions have been received from the other transport robots 50 or the like (step S130). Furthermore, if the processing unit 55G determines that suitable imaging conditions have not been received from the other transport robots 50 or the like (step S130: YES), it sets predetermined initial conditions as imaging conditions for the camera 54 (step S140). In this embodiment, the initial conditions set in step S140 are, for example, minimum values ​​for the sensitivity and exposure time of the camera 54. Furthermore, the processing unit 55G causes the camera 54 to capture an image of the marker 3m (step S150), acquires imaging data from the camera 54, and determines whether the marker 3m has been correctly recognized from the imaging data (step S160).

[0027] If the calculation processing unit 55G determines that the marker 3m was successfully recognized from the image data captured by the camera 54 (step S160: NO), it links the image capturing conditions set in step S140 to the situation information (location, time period, and season) and stores them in the storage device 57, and transmits the image capturing conditions (linked to the location, time period, etc.) to the AMR management device 15 (step S165), temporarily terminating the series of processes shown in Fig. 6. The AMR management device 15 stores the image capturing conditions received from the transport robot 50 in the storage device of the AMR management device 15, and if the image capturing conditions stored in the storage device include suitable image capturing conditions that meet the request in step S120, it transmits the suitable image capturing conditions to the corresponding transport robot 50.

[0028] Furthermore, if the processing unit 55G determines that the marker 3m could not be properly recognized from the imaging data of the camera 54 (step S160: YES), it changes the imaging conditions of the camera 54 (step S170). In step S170 of this embodiment, if the processing unit 55G set the imaging conditions to the above-mentioned initial conditions in step S140, it increases the sensitivity and exposure time by predetermined values. After the processing of step S170, the processing unit 55G causes the camera 54 to image the marker 3m again (step S180), acquires imaging data of the camera 54, and determines whether the marker 3m could be properly recognized from the imaging data (step S190).

[0029] If it is determined that the marker 3m could not be properly recognized from the image data captured by the camera 54 despite the change in the image capturing conditions in step S170 (step S190: YES), the calculation processing unit 55G increments the counter C (step S200) and then determines whether the counter C is equal to or greater than a predetermined threshold value Cref (a relatively large positive value) (step S210). If it is determined that the counter C is less than the threshold value Cref (step S210: NO), the calculation processing unit 55G repeatedly executes the processes of steps S170 to S210. As a result, the image capturing conditions of the camera 54 are changed in stages until it is determined in step S190 that the marker 3m could be properly recognized from the image data captured by the camera 54.

[0030] Furthermore, if it is determined that the marker 3m has been correctly recognized from the image data captured by the camera 54 (step S190: NO), the processing unit 55G stores the image capturing conditions set in the most recent step S170 in the storage device 57 and transmits them to the AMR management device 15 (step S165), and temporarily ends the series of processes in Fig. 6. Note that, although it is an extremely rare case, if it is determined in step S210 that the counter C is equal to or greater than the threshold value Cref, the processing unit 55G transmits a signal indicating that an error has occurred to the AMR management device 15 (step S220), temporarily ends the series of processes in Fig. 6, and stops the travel of the transport robot 50 (transport of the cart 5).

[0031] On the other hand, if it is determined in step S110 that the imaging conditions stored in the storage device 57 include suitable imaging conditions (step S110: NO), the processing unit 55G sets the suitable imaging conditions as the imaging conditions of the camera 54 (step S115) and executes the processes of step S150 and later. If it is determined in step S130 that suitable imaging conditions have been received from another transport robot 50 or the like (step S130: NO), the processing unit 55G sets the suitable imaging conditions as the imaging conditions of the camera 54 (step S115) and executes the processes of step S150 and later. If it is determined after the process of step S115 that the marker 3m has been correctly recognized from the imaging data of the camera 54 (step S160: NO), the processing unit 55G associates the imaging conditions set in step S115 with the situation information (location, time zone, and season), stores them in the storage device 57, and transmits them to the AMR management device 15 (step S165), and temporarily ends the series of processes of FIG. 6.

[0032] Furthermore, if it is determined after the processing of step S115 that the marker 3m could not be correctly recognized from the imaging data of the camera 54 (step S160: YES), the calculation processing unit 55G changes the imaging conditions of the camera 54 (step S170). If the calculation processing unit 55G sets the above-mentioned preferred imaging conditions as the imaging conditions in step S115, then in step S170, for example, the calculation processing unit 55G increases the sensitivity and exposure time from the values ​​in the preferred imaging conditions by predetermined values, and if the marker 3m cannot still be recognized, decreases the sensitivity and exposure time from the values ​​in the preferred imaging conditions by predetermined values. Then, if it is determined after the processing of step S115 that the marker 3m could be correctly recognized from the imaging data of the camera 54 (step S190: NO), the calculation processing unit 55G stores the imaging conditions set in the most recent step S170 in the storage device 57 and transmits them to the AMR management device 15 (step S165), and temporarily ends the series of processes shown in FIG. 6.

[0033] As described above, in the transfer robot 50, when the processing unit 55G as a recognition unit fails to recognize the marker 3m or the like as a recognition target, the processing unit 55G as an imaging control unit gradually changes the imaging conditions of the camera 54 as an imaging device so that the marker 3m or the like is recognized (steps S150-S210). In addition, the processing unit 55G stores the imaging conditions used when the marker 3m or the like is recognized in the storage device 57 (step S165). Furthermore, in the transfer robot 50, when the same marker 3m or the like is imaged by the camera 54 at the same location and at the same time, i.e., under the same or similar circumstances, the imaging conditions (preferred imaging conditions) stored in the storage device 57 are applied (steps S100-S115).

[0034] As a result, even if the surrounding environment changes, the arithmetic processing unit 55G of the transport robot 50 can recognize the marker 3m or the like as a recognition target based on the imaging data (image data) of the camera 54. As a result, it is possible to ensure smooth operation of the transport robot 50 without having to fully learn the imaging conditions of the camera 54 in advance. It goes without saying that the above-mentioned preferable imaging conditions may also be applied when imaging another recognition target (e.g., a structure with a marker 3m) different from the previous recognition target (e.g., marker 3m) by the camera 54 at the same place and at the same time.

[0035] Furthermore, when the marker 3m or the like as the recognition target is not recognized even when the imaging conditions (preferred imaging conditions) stored in the storage device 57 are applied, the calculation processing unit 55G as the imaging control unit changes the imaging conditions of the camera 54 so that the marker 3m or the like is recognized (steps S115, S150-S210), and stores the imaging conditions when the marker 3m or the like is recognized in the storage device 57 (step S165). This makes it possible to better respond to changes in the surrounding environment of the transport robot 50.

[0036] Furthermore, the arithmetic processing unit 55G of the control device 55 estimates the self-position of the transport robot 50 based on the recognized marker 3m. By enabling the transport robot 50 including such a control device 55 to recognize the recognition target even when the surrounding environment changes, for example, it becomes possible to smoothly move the transport robot 50 from any shipping line 4x (starting point) to any shipping data 3x (destination point). However, it goes without saying that the recognition target of the transport robot 50 is not limited to the marker 3m arranged in the shipping area 3 (the area where the transport robot 50 can move), but may also be a structure (stationary object) arranged in the shipping area 3.

[0037] Furthermore, in the above embodiment, the imaging conditions of the camera 54 include at least the sensitivity and exposure time of the camera 54. This makes it possible to gradually change the imaging conditions to cause the processing unit 55G to recognize the marker 3m, etc., when the processing unit 55G is unable to recognize the marker 3m, etc. However, if the camera 54 includes a light emitting device, the imaging conditions of the camera 54 may include the settings of the light emitting device (on / off, light emission timing, etc.).

[0038] Furthermore, in the above embodiment, the imaging conditions (preferred imaging conditions) when the marker 3m as the recognition target is recognized by the processing unit 55G as the recognition unit of any one of the transport robots 50 are shared by the other transport robots 50 (steps S120-S130, S115). This makes it possible to efficiently perform work using a plurality of transport robots 50.

[0039] The transport robot 50 recognizes the marker 5m attached to the cart 5 as a recognition target, and supports and transports the cart 5. Therefore, the transport robot 50 may be used not only in the logistics center 1, but also in the back yard of a retail store such as a shopping center to transport empty carts or carts loaded with cargo, or may be used to transport the cargo itself.

[0040] Furthermore, the transport robot 50 travels along a route to a target point while recognizing a plurality of markers 3m arranged inside the building of the logistics center 1. Therefore, the transport robot 50 may be used in places other than the logistics center 1, such as a shopping center. In a shopping center, for example, the transport robot 50 carrying featured products, bargain items, etc. may be autonomously driven at a low speed along a predetermined route while recognizing a plurality of markers 3m arranged inside the store, or the transport robot 50 carrying bargain items, etc. may be autonomously driven to, for example, a cash register in response to the arrival of a time sale. Note that the plurality of markers 3m are not limited to those arranged inside the building of the logistics center 1, but may also be arranged outside the building of the logistics center 1 (or at the boundary between the inside and outside).

[0041] In addition, as shown in Figure 7, the transport robot 50 may be configured to recognize a marker M as a recognition target attached to the worker's clothing, etc. or another transport robot 50, and move by following the worker or the other transport robot 50.

[0042] Furthermore, the transport robot 50 includes a plurality of Mecanum wheels 51 each rotated by a corresponding electric motor 52. This allows for a higher degree of freedom of movement for the transport robot 50. However, as mentioned above, the transport robot 50 may include wheels other than Mecanum wheels, such as wheels including general rubber tires.

[0043] The present invention is not limited to the above-described embodiment, and various modifications can be made within the scope of the present disclosure. Furthermore, the above-described embodiment is merely a specific form of the invention described in the Summary of the Invention, and does not limit the elements of the invention described in the Summary of the Invention. [Industrial Applicability]

[0044] The presently disclosed invention can be used in industries such as the manufacturing of moving objects, which recognize targets based on image data captured by an imaging device. [Explanation of symbols]

[0045] 1 logistics center, 2 entrance, 3 shipping area, 3a, 3b, 3c, 3d, 3e, 3f, 3g, 3h, 3i, 3j, 3k, 3l, 3x shipping gate, 3m marker, 4 conveyor, 4a, 4b, 4c, 4d, 4e, 4f, 4g, 4h, 4i, 4j, 4k, 4l, 4x shipping line, 40 main transport line, 5 trolley, 5m marker, 6 storage location, 10 logistics management system, 11 management server, 12 storage device, 14 conveyor control device, 15 AMR management device, 20 entrance camera, 21 monitor, 25 entrance management device, 30 camera, 31 trolley detector, 35 shipping gate management device, 50 transport robot, 51 Mecanum wheel, 51p support plate, 51r roller, 52 electric motor, 53 lifting unit, 54 Camera, 55 control device, 55D driving control unit, 55G calculation processing unit, 55L lift control unit, 56 communication module, 57 storage device, M marker, P luggage, T delivery vehicle.

Claims

1. A mobile body includes an imaging device that images a recognition target, and a recognition unit that recognizes the recognition target based on image data of the imaging device, and moves within a predetermined area by performing a predetermined operation based on the recognition target recognized by the recognition unit, a control unit that gradually changes the imaging conditions of the imaging device so that the recognition unit can recognize the recognition target when the recognition unit does not recognize the recognition target, stores the imaging conditions when the recognition unit recognizes the recognition target in a storage device by linking them to situational information including the location, time of day, and season within the specified area, and applies the imaging conditions stored in the storage device when the recognition target or another recognition target is imaged by the imaging device under the same or similar circumstances including the location, time of day, and season within the specified area.

2. 2. The moving body according to claim 1, When the recognition unit does not recognize the recognition target even when the imaging conditions stored in the storage device are applied, the imaging control unit changes the imaging conditions of the imaging device so that the recognition unit can recognize the recognition target, and links the imaging conditions when the recognition unit recognizes the recognition target to the situation information and stores them in the storage device.

3. In the mobile body described in claim 1 or 2, when the imaging conditions stored in the storage device do not include preferred imaging conditions that would have allowed the recognition target to be successfully recognized at the same location, time of day, and season as the current time, the imaging control unit requests the transmission of the preferred imaging conditions from other mobile bodies and / or a management device that manages the imaging conditions, and when the preferred imaging conditions are received, the imaging control unit recognizes the recognition target using the received preferred imaging conditions.

4. In the mobile body described in claim 3, when the imaging control unit is unable to receive the preferred imaging conditions from other mobile bodies and / or the management device, and when the recognition unit does not recognize the recognition target, the imaging control unit gradually changes the imaging conditions of the imaging device so that the recognition unit can recognize the recognition target, and the imaging conditions when the recognition unit recognizes the recognition target are linked to the situation information and stored in a storage device.

5. 5. The moving body according to claim 1, wherein the imaging conditions include at least a sensitivity and an exposure time of the imaging device.

6. 6. The mobile body according to claim 1, wherein the imaging conditions used when the recognition unit recognizes the recognition target are shared by other mobile bodies.

7. 7. A mobile body according to any one of claims 1 to 6, wherein the mobile body estimates its own position based on the recognition object recognized by the recognition unit, or recognizes the recognition object attached to a cart by the recognition unit and supports and transports the cart, or travels along a route to a target point while recognizing a plurality of recognition objects arranged inside and / or outside a building.

8. The moving body according to any one of claims 1 to 6, A mobile object that recognizes the recognition target attached to a worker or another mobile object and moves following the worker or the other mobile object.

9. The moving body according to any one of claims 1 to 8, A moving body including a plurality of Mecanum wheels, each of which is driven to rotate by a corresponding electric motor.

10. A method for controlling a moving object including an imaging device that captures an image of a recognition target and a recognition unit that recognizes the recognition target based on image data captured by the imaging device, the moving object moving in a predetermined area by performing a predetermined operation based on the recognition target recognized by the recognition unit, A method for controlling a moving body, comprising: when the recognition unit does not recognize the recognition target, gradually changing the imaging conditions of the imaging device so that the recognition unit can recognize the recognition target; storing the imaging conditions when the recognition unit recognizes the recognition target in a storage device while linking them to situational information including the location, time of day, and season within the specified area; and applying the imaging conditions stored in the storage device when the recognition target or another recognition target is imaged by the imaging device under the same or similar circumstances including the location, time of day, and season within the specified area.

Images