Goods picking method and apparatus, and warehousing system

Abstract

The present disclosure relates to the technical field of warehousing logistics, and discloses a goods picking method and apparatus, and a warehousing system. The method comprises: acquiring an order to be processed corresponding to a first workstation, and determining a target container for placing at least one item to be picked corresponding to the order to be processed; when a handling device transports a target container to a first workstation, controlling an image acquisition apparatus to acquire a first image corresponding to the target container; determining, according to the first image, a first item from the at least one item to be picked, and when a type of the first item is a first item type, determining first target picking information of the first item; and according to the first target picking information, controlling a picking and placing apparatus to pick the first item from the target container, and transferring the first item to a target location.

Description

Goods picking method, device and warehouse system

[0001] This application claims priority to Chinese Patent Application No. 202410985263.6, filed on July 22, 2024, and Chinese Patent Application No. 202510976042.7, filed on July 15, 2025, the contents of which are incorporated herein by reference in their entirety. TECHNICAL FIELD

[0002] The present disclosure relates to the technical field of warehouse logistics, and in particular to a goods picking method, device and warehouse system. BACKGROUND

[0003] In a "goods-to-person" warehouse system, a robot can carry a carton or a goods shelf from a storage area to a work station, and then a picker or a picking device at the work station picks the goods required by an order from the carton and sows the goods into a corresponding order box on a sowing wall. When the picker performs the picking operation, if the number of orders in the warehouse system is large, the required labor cost is high, and the picking efficiency is low. When the picking device performs the picking operation, picking exceptions may occur, thereby affecting the picking efficiency. SUMMARY

[0004] Embodiments of the present disclosure provide a goods picking method, device and warehouse system.

[0005] The first aspect of the embodiments of the present disclosure provides a goods picking method applied to a control device in a warehouse system, which comprises the following steps. First, obtaining a to-be-processed order corresponding to a first work station, and determining a target container in which at least one to-be-picked goods corresponding to the to-be-processed order is placed. Second, in the case that a carrying device carries the target container to the first work station, controlling an image acquisition device to acquire a first image corresponding to the target container. Third, determining a first goods according to the first image, and in the case that the type of the first goods is a first goods type, determining first target picking information of the first goods; wherein the first target picking information comprises a picking position and a picking direction. Finally, according to the first target picking information, controlling a taking and placing device to pick the first goods from the target container and transfer the first goods to a target position.

[0006] In a second aspect, the embodiments of the present disclosure provide a warehouse system, comprising a first workstation, a control device, a carrying device and an image acquisition device, the first workstation comprising a taking and placing device. The control device is configured to: acquire a to-be-processed order corresponding to the first workstation, determine a target container in which at least one to-be-picked goods corresponding to the to-be-processed order is placed, and generate an acquisition instruction according to the target container. The image acquisition device is configured to: in a case where the carrying device carries the target container to the first workstation, acquire a first image corresponding to the target container according to the acquisition instruction. The control device is configured to: determine a first goods from the at least one to-be-picked goods according to the first image, and determine first target picking information of the first goods in a case where a type of the first goods is a first goods type; generate a picking instruction according to the first target picking information; wherein the target picking information comprises a picking position and a picking direction. The taking and placing device is configured to: pick the first goods from the target container according to the picking instruction, and transfer the first goods to a target position.

[0007] In a third aspect, the embodiments of the present disclosure provide a goods picking device, comprising an acquisition module, a control module and a determination module. The acquisition module is configured to: acquire a to-be-processed order corresponding to a first workstation, and determine a target container in which at least one to-be-picked goods corresponding to the to-be-processed order is placed. The control module is configured to: in a case where a carrying device carries the target container to the first workstation, control an image acquisition device to acquire a first image corresponding to the target container. The determination module is configured to: determine a first goods from the at least one to-be-picked goods according to the first image, and determine first target picking information of the first goods in a case where a type of the first goods is a first goods type; wherein the first target picking information comprises a picking position and a picking direction. The control module is configured to: control a taking and placing device of the first workstation to pick the first goods from the target container according to the first target picking information, and transfer the first goods to a target position.

[0008] In a fourth aspect, the embodiments of the present disclosure provide an electronic device, comprising: a processor and a memory, the memory being configured to store computer executable instructions; and the processor being configured to read the instructions from the memory and execute the instructions to implement the goods picking method of the first aspect.

[0009] In a fifth aspect, the embodiments of the present disclosure provide a computer readable storage medium, the storage medium storing computer program instructions, when a computer reads the instructions, the goods picking method of the first aspect is executed.

[0010] The sixth aspect of the embodiments of the present disclosure provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the goods picking method of the first aspect.

[0011] The seventh aspect of the embodiments of the present disclosure provides a computer program which, when executed by a processor, can implement the goods picking method of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0012] FIG. 1 is a schematic diagram of a warehouse system according to an embodiment of the present disclosure;

[0013] FIG. 2A is a schematic diagram of a taking and placing device according to an embodiment of the present disclosure;

[0014] FIG. 2B is a schematic diagram of a workstation according to an embodiment of the present disclosure;

[0015] FIG. 3 is a schematic diagram of a goods picking method according to an embodiment of the present disclosure;

[0016] FIG. 4 is a schematic diagram of another goods picking method according to an embodiment of the present disclosure;

[0017] FIG. 5 is a schematic diagram of a second identification model according to an embodiment of the present disclosure;

[0018] FIG. 6 is a schematic diagram of yet another goods picking method according to an embodiment of the present disclosure;

[0019] FIG. 7 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0020] FIG. 8 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0021] FIG. 9 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0022] FIG. 10 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0023] FIG. 11 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0024] FIG. 12A is a schematic diagram of another warehouse system according to an embodiment of the present disclosure;

[0025] FIG. 12B is a schematic diagram of yet another warehouse system according to an embodiment of the present disclosure;

[0026] FIG. 13 is a schematic diagram of still another goods picking method according to an embodiment of the present disclosure;

[0027] FIG. 14 is a schematic diagram of another goods picking method according to an embodiment of the present disclosure;

[0028] FIG. 15 is a schematic diagram of another warehouse system according to an embodiment of the present disclosure;

[0029] FIG. 16 is a schematic diagram of a goods picking device according to an embodiment of the present disclosure;

[0030] FIG. 17 is a schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present disclosure, and to make the above-mentioned purposes, features and advantages of the embodiments of the present disclosure more apparent and easy to understand, the technical solutions in the embodiments of the present disclosure will be further described in detail below with reference to the drawings.

[0032] In the warehouse system, the carrying robot can carry the goods box to the work station, and the operator of the work station picks the goods required by the order from the goods box and sows the goods into the order box corresponding to the order, thereby completing the picking of the goods. Generally, the picking process accounts for a high proportion of work in the warehouse system, for example, the working time of picking accounts for about 40% to 60% of the working time of the entire warehouse system. Therefore, improving the working efficiency of picking and sowing can greatly improve the working efficiency of the warehouse system. The cost of manual picking is relatively high, and when the number of orders is large, the picking efficiency is low, which cannot meet the demand.

[0033] FIG. 1 is a schematic diagram of a warehouse system according to an embodiment of the present disclosure.

[0034] As shown in FIG. 1, the warehouse system 100 includes a plurality of carriers 10, a carrying device 20, a work station 30, a taking and placing device 40, and a control device (not shown in FIG. 1).

[0035] In some examples, the carrier 10 can be a shelf, a tray, a trolley, a cage, etc. The carrier 10 includes a plurality of storage positions (which can also be referred to as storage sites) thereon. The storage position can be a cuboid-shaped accommodation space, and the plurality of storage positions on the carrier 10 can be arranged in order along the length direction, the width direction and the height direction of the carrier 10.

[0036] For example, the storage position on the carrier 10 is used to carry an inventory object such as a container or a cargo box. The container can be a matching product specially designed for the carrier 10, or can be a common cargo box (also referred to as a material box), or can be a package of the cargo (also referred to as a primary box), and the present disclosure does not limit this. The present disclosure takes the container placed on the carrier 10 as an example for illustrative description.

[0037] In some examples, the area for storing and parking the carriers 10 is a storage area 101, which can also be referred to as a shelf area. The storage area 101 can be a dense storage area or a non-dense storage area. The storage area 101 can include a plurality of storage locations, which can be arranged in rows and columns. Each storage location can be used to place a carrier 10.

[0038] For example, the warehouse system 100 includes a plurality of handling devices 20, which can be used to handle goods, containers, or carriers. For example, the handling devices 20 can be automated guided vehicles (AGVs). The handling devices 20 can carry inventory objects (e.g., goods, containers, or carriers) to corresponding destinations.

[0039] In some examples, the handling devices 20 can carry inventory objects to the workstations 30 for order picking, or the handling devices 20 can carry inventory objects to corresponding destinations for order picking or sorting operations.

[0040] For example, the warehouse system 100 can include one or more workstations 30. The workstations 30 can perform order picking operations on inventory objects carried to the workstations 30 by the handling devices 20. For example, the workstations 30 can pick goods in containers to order containers.

[0041] In some embodiments, the warehouse system 100 further includes a plurality of picking devices 40.

[0042] In some examples, the picking devices 40 can be disposed in correspondence with the workstations 30. Each workstation 30 can be disposed in correspondence with at least one picking device 40, and the number of picking devices 40 corresponding to each workstation 30 is not limited in the embodiments of the present disclosure. For example, the picking devices 40 can perform order picking operations on inventory containers carried to the workstations 30. For example, the picking devices 40 can pick goods in inventory containers to order containers at the workstations 30.

[0043] In other examples, the picking devices 40 can also be disposed in areas other than the workstations 30. For example, as shown in FIG. 1, the picking devices 40 can also be disposed in the storage area 101. The number of picking devices 40 in the storage area 101 can be one or more, and the number of picking devices 40 in the storage area 101 is not limited in the embodiments of the present disclosure.

[0044] For example, an area in the storage area 101 in which the picking devices 40 are disposed can be referred to as a work area 102. The picking devices 40 in the work area 102 can pick goods in to-be-picked containers to another inventory container, thereby facilitating subsequent warehousing operations on the inventory container.

[0045] In some examples, the control device can be coupled with the carrying device 20 and the taking and placing device 40 respectively, for controlling the work of the carrying device 20 and the taking and placing device 40. For example, the control device can control the carrying device 20 to carry the container to the work station 30 or the work area 102, and control the taking and placing device 40 to perform the picking operation on the goods in the container.

[0046] In some examples, the control device can be a server or a terminal device. The terminal device can include at least one of a personal computer, a notebook computer, a smart phone, a tablet computer and a portable wearable device; the server can include a stand-alone server or a server cluster composed of multiple servers, and the embodiments of the present disclosure do not limit this.

[0047] In some examples, the control device is in communication connection with the carrying device 20 and the taking and placing device 40 for data communication. For example, the control device can be in communication connection with the carrying device 20 and the taking and placing device 40 through a local area network (LAN), a wireless local area network (WLAN) and other networks.

[0048] It should be noted that the taking and placing device 40 of the work area 102 performs a picking operation similar to the picking operation performed by the taking and placing device 40 corresponding to the work station 30, and the following embodiments can be exemplarily described taking the picking process of the taking and placing device 40 of the work station 30 as an example.

[0049] In some examples, the taking and placing device 40 can take the target goods required by the order from the inventory container according to the order demand, and transfer the target goods to the order container corresponding to the order in the work station 30. When the taking and placing device 40 picks the target goods from the inventory container, picking failure or picking abnormality may occur.

[0050] For example, the taking and placing device 40 may not be able to identify the target goods from the inventory container, so as to fail to accurately take out the target goods, resulting in picking failure; or the taking and placing device 40 may identify the target goods, but due to the improper selection of the picking point, the taking and placing device cannot accurately and efficiently take out the target goods, resulting in picking failure. The picking failure or picking abnormality of the taking and placing device 40 will affect the overall work efficiency of the warehouse system.

[0051] In some examples, as shown in FIG. 1, the workstation 30 (such as a container taking and placing mechanism of the workstation 30) can take the hit container corresponding to the order from the carrier 10 according to the order demand, and place the hit container on a picking station of the workstation 30; then, a picking object picks the to-be-seeded goods from the hit container on the picking station and seeds the to-be-seeded goods into the order container 42 corresponding to the order. The picking object can be a picking personnel or a picking device. When the picking object is a picking personnel, the labor cost is large and the picking efficiency is low; when the picking object is a picking device, the picking device may fail to pick the to-be-seeded goods from the hit container or may have picking abnormalities. For example, the picking device may fail to identify the to-be-seeded goods from the hit container, so as to fail to accurately take out the to-be-seeded goods; or the picking device takes out the to-be-seeded goods, but the to-be-seeded goods fall off in the transfer process of the to-be-seeded goods due to insecure gripping or the like. The picking failure or picking abnormality of the picking device also affects the picking efficiency of the warehouse system.

[0052] To solve the above problems, the warehouse system provided by the embodiments of the present disclosure can realize automatic picking, reduce picking failures and picking abnormalities in the picking process, improve the accuracy and safety of the picking process, and improve the picking efficiency of the warehouse system.

[0053] FIG. 2A is a schematic view of a taking and placing device provided by an embodiment of the present disclosure.

[0054] As shown in FIG. 2A, the taking and placing device 40 includes a base 41, a first connecting arm 44 connected to the base 41 through a first driving member 43, and a picking mechanism 42.

[0055] In some examples, the picking mechanism 42 can be a component such as a gripper, a claw, or a suction cup connected to the taking and placing device 40 for picking goods in a container, and the specific structure of the picking mechanism 42 is not limited in the embodiments of the present disclosure. For example, one picking mechanism 42 can be provided on the taking and placing device 40, or a plurality of picking mechanisms 42 can be provided. It should be noted that FIG. 2A only exemplarily illustrates the case where one picking mechanism 42 is provided on the taking and placing device 40.

[0056] In some examples, the first driving member 43 can be a motor, and the first driving member 43 can directly or indirectly drive the first connecting arm 44 to move in a three-dimensional space (or a two-dimensional space).

[0057] In some examples, the picking and placing device 40 further comprises a second connecting arm 45 for connecting the first connecting arm 44 and the picking mechanism 42, and the second connecting arm 45 is rotatable relative to the first connecting arm 44. For example, the first connecting arm 44 and the second connecting arm 45 can be driven to rotate relative to each other by a second driving member 46.

[0058] As shown in FIG. 2A, one end of the picking mechanism 42 can further be provided with a third driving member 47 and a fourth driving member 48 connected to the third driving member 47. The third driving member 47 can drive the picking mechanism 42 to move in the vertical direction, and the fourth driving member 48 can drive the picking mechanism 42 to rotate by a corresponding angle. For example, the picking mechanism 42 can be driven by the third driving member 47 and the fourth driving member 48 to pick up the goods required by the order from the container, and then move the picked goods to the target position by the first connecting arm 44 and the second connecting arm 45.

[0059] It should be noted that the picking and placing device 40 shown in FIG. 2A can be an independently arranged picking and placing device. For example, the picking and placing device 40 can be independently arranged in the area of the workstation 30, or can be independently arranged in the working area 102.

[0060] FIG. 2B is a structural schematic diagram of a workstation provided by an embodiment of the present disclosure. The structure and working principle of the workstation provided by an embodiment of the present disclosure will be described below in combination with FIG. 2B. It should be noted that the workstation 30 shown in FIG. 2B can be a first workstation.

[0061] In some embodiments, as shown in FIG. 2B, the workstation 30 comprises a support frame 31, a guiding mechanism 32 and a picking and placing device 40.

[0062] It should be noted that, as shown in FIG. 2B, the picking and placing device 40 can be arranged in the workstation 30 and cooperates with other components of the workstation 30 to complete the picking operation.

[0063] For example, the guiding mechanism 32 is arranged to the support frame 31. The guiding mechanism 32 is transversely movable or vertically movable relative to the support frame 31. The picking and placing device 40 is arranged to the guiding mechanism 32, so that the picking and placing device 40 can be transversely movable or vertically movable relative to the support frame 31.

[0064] Exemplarily, the guiding mechanism 32 can include a lateral moving device 34, a moving rod 35, and a vertical moving device 36. The lateral moving device 34 is arranged to the support frame 31. The lateral moving device 34 is coupled to the control device. The lateral moving device 34 is laterally movable relative to the support frame 31. The moving rod 35 extends in a vertical direction. The moving rod 35 is arranged to the lateral moving device 34, so that the moving rod 35 is laterally movable relative to the support frame 31. The vertical moving device 36 is arranged to the moving rod 35. The vertical moving device 36 is coupled to the control device. The vertical moving device 36 is vertically movable relative to the moving rod 35. The pick-and-place device 40 is arranged to the vertical moving device 36, so that the pick-and-place device 40 is vertically movable relative to the moving rod 35. Therefore, the pick-and-place device 40 is laterally and / or vertically movable on the support frame 31.

[0065] In some examples, the work station 30 can also not be provided with the support frame 31. When the work station 30 is not provided with the support frame 31, the pick-and-place device 40 can be laterally and / or vertically movable by the guiding mechanism 32.

[0066] Exemplarily, when the work station 30 is not provided with the support frame 31, the work station 30 can include at least one guiding mechanism 32 and the pick-and-place device 40. The pick-and-place device 40 is arranged to the guiding mechanism 32, so that the pick-and-place device 40 is laterally and / or vertically movable by the guiding mechanism 32.

[0067] In some embodiments, the lateral moving device 34 is coupled to the moving rod 35 and configured to drive the moving rod 35 to be laterally movable. The moving rod 35 extends in a vertical direction. The pick-and-place device 40 is arranged to the moving rod 35, so that the moving rod 35 drives the pick-and-place device 40 to be laterally movable. The vertical moving device 36 is coupled to the pick-and-place device 40 and configured to drive the pick-and-place device 40 to be vertically movable relative to the moving rod 35.

[0068] In some examples, the lateral moving device 34 can be arranged to the moving rod 35 or the pick-and-place device 40. The vertical moving device 36 can be arranged to the moving rod 35. For example, the lateral moving device 34 and the vertical moving device 36 can be two driving motors. The embodiments of the present disclosure do not limit the specific positions of the lateral moving device 34 and the vertical moving device 36, as long as the lateral moving device 34 can drive the moving rod 35 to be laterally movable and the vertical moving device 36 can drive the pick-and-place device 40 to be vertically movable relative to the moving rod 35.

[0069] In some examples, the picking and placing device 40 is coupled to the control device. When the handling device 20 carries the carrier 10 to the picking area of the work station 30 under the control of the control device, the picking and placing device 40 reaches the position corresponding to the storage location 15 in the carrier 10 by lateral movement (as shown by the L direction in FIG. 2B) and / or vertical movement (as shown by the H direction in FIG. 2B). The picking and placing device 40 can also reach the position of any sowing position on the sowing wall 40 by lateral movement and / or vertical movement.

[0070] In some embodiments, as shown in FIG. 2B, the picking and placing device 40 includes a container picking and placing mechanism 331, which can reach the position of any container on the carrier 10 parked in the carrier parking area by lateral movement and / or vertical movement, to pick / place the container on the carrier 10; or, the container picking and placing mechanism 331 can also reach the position of any sowing position on the sowing wall 40, to pick / place the order container on the sowing position.

[0071] In some examples, the container picking and placing mechanism 331 can include a container picking and placing component, which can move along the depth direction. The container picking and placing component can be a claw, a fork or a suction cup connected to the container picking and placing mechanism 331, etc.

[0072] For example, when the container picking and placing mechanism 331 reaches the position corresponding to the storage location 15 by lateral movement and / or vertical movement, the container picking and placing component can extend to the storage location 15 along the depth direction, i.e., along the direction towards the carrier, to pick / place the container from the storage location 15. The container picking and placing mechanism 331 further includes a containing component, which can place the hit container picked by the container picking and placing component on the containing component.

[0073] In some embodiments, the picking and placing device 40 further includes a goods picking and placing mechanism 332, which can pick the goods to be sowed from the hit container.

[0074] It should be noted that the goods picking and placing mechanism 332 includes a picking mechanism, such as the picking mechanism 42 in FIG. 2A. The goods picking and placing mechanism 332 can move flexibly in three-dimensional space to adapt to goods boxes of different shapes and sizes. The goods picking and placing mechanism 332 is used to pick the goods required by an order from the hit container.

[0075] In some examples, the container picking mechanism and the goods picking mechanism can be coupled, and the goods picking mechanism can move along with the container picking mechanism in the lateral direction or the vertical direction to pick the goods in the container placed on the container picking mechanism; or the container picking mechanism and the goods picking mechanism can be independently arranged, the container picking mechanism can move in the lateral direction or the vertical direction, and the goods picking mechanism can pick the goods in the container placed on the container picking mechanism. For example, the container picking mechanism can be arranged on the guide mechanism, and the goods picking mechanism can be arranged on the support frame. The specific arrangement of the container picking mechanism and the goods picking mechanism is not limited in the embodiments of the present disclosure, as long as the container picking mechanism can move to the position of any seeding position on the seeding wall (i.e., the docking position corresponding to the order container), and the goods picking mechanism can pick the goods from the container placed on the container picking mechanism and seed the goods into the order container.

[0076] For example, as shown in FIG. 2B, the goods picking mechanism 332 can be coupled with the container picking mechanism 331 and can move along with the container picking mechanism 331 in the lateral direction or the vertical direction.

[0077] In some examples, the container picking mechanism 331 and the goods picking mechanism 332 can be directly connected or indirectly connected. For example, the container picking mechanism 331 and the goods picking mechanism 332 can be directly connected through a shaft, a pin, a gear, a connecting rod or other mechanical components, and when the container picking mechanism 331 moves, the goods picking mechanism 332 can be driven to move. The connection mode between the container picking mechanism 331 and the goods picking mechanism 332 is not limited in the embodiments of the present disclosure.

[0078] For example, the container picking mechanism 331 is arranged on the guide mechanism 32, and when the container picking mechanism 331 moves in the lateral direction or the vertical direction through the guide mechanism 32, the goods picking mechanism 332 can be driven to move in the lateral direction or the vertical direction at the same time. That is, the goods picking mechanism 332 can move along with the container picking mechanism 331 in the lateral direction or the vertical direction.

[0079] For example, after the container picking mechanism 331 picks the hit container from the carrier, the container picking mechanism 331 can move from the current position (i.e., the position corresponding to the goods location 15) to the position of the order container corresponding to the hit container on the seeding wall.

[0080] In some examples, when the container picking mechanism 331 carries the hit container to the position of the order container, the goods picking component of the goods picking mechanism 332 can pick (or suck) the goods required by the order container from the hit container placed by the containing component of the container picking mechanism 331, and deliver the picked goods to the order container, so as to complete the picking and seeding of the goods.

[0081] That is, the goods picking mechanism 332 in the embodiment of the present disclosure can pick the goods from the hit container on the picking station, or can pick the goods from the hit container corresponding to the order container carried by the container picking mechanism 331, and the embodiment of the present disclosure does not make any limitation in this regard. The following embodiments are exemplarily described by taking the goods picking mechanism picking the goods from the hit container corresponding to the order container carried by the container picking mechanism 331 as an example.

[0082] It should be noted that FIG. 2B is an example of a work station provided in the embodiment of the present disclosure, and the embodiment of the present disclosure does not make any limitation on the specific structure of the work station.

[0083] In some examples, the embodiment of the present disclosure does not make any limitation on the setting mode and specific structure of the picking device 40. The picking device 40 can be arranged on the work station, or can be independent of the work station; the picking device 40 can include the goods picking mechanism, or can include the container picking mechanism and the goods picking mechanism.

[0084] For example, as shown in FIG. 2A, taking the picking device 40 arranged independently of the work station as an example, the picking device 40 can only include the goods picking mechanism (i.e., does not include the container picking mechanism). In this case, the carrying device can carry the hit container to the work station, and then the picking device (such as at least one picking mechanism in the picking device) of the work station picks the corresponding goods from the hit container.

[0085] For another example, as shown in FIG. 2B, taking the picking device 40 arranged on the work station and including the container picking mechanism and the goods picking mechanism as an example, in the case that the carrying device carries the hit carrier to the work station, the container picking mechanism in the picking device 40 can take the hit container from the carrier, and then the goods picking mechanism (i.e., at least one picking mechanism) picks the goods required by the order from the hit container and places the picked goods in the order container on the seeding wall.

[0086] The goods picking method provided in the embodiment of the present disclosure will be described below in combination with FIG. 3.

[0087] FIG. 3 is a schematic diagram of a goods picking method provided in the embodiment of the present disclosure. As shown in FIG. 3, the goods picking method includes the following steps 310 to 340.

[0088] Step 310: Obtain the order to be processed corresponding to the first work station, and determine the target container in which at least one goods to be picked corresponding to the order to be processed is placed.

[0089] It should be noted that the first workstation can be the workstation in the above embodiments (for example, the workstation 30 in FIG. 1). The first workstation can also be referred to as a picking workstation. The structure of the first workstation can be the same as that of the workstation in FIG. 2B or different, and the embodiments of the present disclosure do not limit this.

[0090] In some examples, before performing the picking operation, the control device can obtain the order to be processed corresponding to the first workstation. The control device determines the container required by the first workstation to perform the picking operation according to the goods required by the order to be processed corresponding to the first workstation, or the container and the carrier on which the container is placed.

[0091] For example, the plurality of goods required by the order to be processed corresponding to the first workstation can be placed in at least one container, wherein the container in which the goods required by the order to be processed are placed can be referred to as a hit container of the order to be processed (hereinafter referred to as a hit container).

[0092] In some examples, the number of hit containers is one, and the target container can be the hit container; the number of hit containers is a plurality, and the target container can be any one of the plurality of hit containers.

[0093] In other examples, after determining the target container, the target carrier on which the target container is placed, i.e., the target carrier, can be further determined. For example, in the case where the target container is placed on each of a plurality of carriers, the target carrier can be any one of the plurality of carriers, or the carrier closest to the first workstation, and the embodiments of the present disclosure do not limit this.

[0094] In step 320, the control device controls the image acquisition device to acquire a first image corresponding to the target container in the case where the conveying device conveys the target container to the first workstation.

[0095] In some examples, the conveying device can convey the target container to the first workstation, or convey the target carrier on which the target container is placed to the first workstation. That is, the conveying device conveying the target container to the first workstation includes the conveying device directly conveying the target container to the first workstation, or the conveying device conveying the target carrier on which the target container is placed to the first workstation.

[0096] It should be noted that the conveying device conveying the target container and the target carrier can be different types of conveying devices, for example, the conveying device conveying the container can be a container conveying robot, and the conveying device conveying the target carrier can be a carrier conveying robot, and the embodiments of the present disclosure do not limit this.

[0097] Exemplarily, in a case that the conveying device conveys a target container, the to-be-picked objects can be goods in the target container that need to be transferred, and the taking and placing device 40 can transfer the to-be-picked objects in the target container to a target position, so as to complete the picking operation.

[0098] For example, after the target container is determined, the position of the target container can be determined, so as to determine how to convey the target container to the picking area where the taking and placing device is located (such as the area of the workstation 30 and / or the working area 102 in FIG. 1). In a case that it is determined that the target container is not located in the picking area, the control device can dispatch a conveying device (such as the conveying device 20 in the above embodiment) to convey the target container to the picking area, so as to perform the picking operation on the target container by the taking and placing device in the picking area.

[0099] In some examples, as described in the above embodiment, the picking area where the taking and placing device is located can be located in a workstation, or can be located in a position other than the workstation, such as a storage area (such as the working area 102 in the storage area 101 in the above embodiment), and the like. For example, one workstation can be one picking area, and the storage area can be provided with at least one picking area.

[0100] Exemplarily, one picking area is provided with at least one taking and placing device, and the taking and placing device can perform the picking operation on the to-be-picked objects (which can also be referred to as to-be-picked goods in the following embodiments) in the container. The to-be-picked objects can be goods themselves or goods with packaging, which is not limited in the embodiments of the present disclosure, and the embodiments of the present disclosure are exemplarily described by taking the to-be-picked objects as goods.

[0101] For example, for the taking and placing device of the workstation, after the conveying device conveys the target container to the workstation, the taking and placing device can perform the picking operation on the to-be-picked objects in the target container, so as to pick up the to-be-picked objects in the target container and transfer them to the order container in the workstation, thereby realizing the outbound operation of the goods in the target container.

[0102] For another example, for the taking and placing device of the storage area, after the conveying device conveys the target container to the picking area of the storage area, the taking and placing device in the picking area can pick up the to-be-picked objects in the target container from the target container and transfer them to another storage container, and after the picking is completed, the storage container is conveyed to the carrier, thereby realizing the inbound operation of the goods.

[0103] It should be noted that the position of the taking and placing device is not limited in the embodiments of the present disclosure, and the use scenarios of the taking and placing devices in different positions can be different, but the picking processes of the taking and placing devices are similar, and to avoid repetition, the taking and placing device located in the workstation is exemplarily described in the embodiments of the present disclosure.

[0104] Exemplarily, in a case that the handling device handles the target carrier, the control device can dispatch the handling device to move to a position of the target carrier to handle the target carrier or a target container on the target carrier. Taking the handling device handling the target carrier as an example, the handling device can handle the target carrier to the first working station, for example, the handling device can handle the target carrier to a carrier parking area (such as a picking area or a queuing area) of the first working station. Wherein, the handling device handling the target carrier can be any idle handling device in the warehouse system, such as the idle handling device closest to the target carrier.

[0105] For example, in a case that there is no carrier being picked in the picking area of the first working station, the handling device can handle the target carrier to the picking area; in a case that there is a carrier being picked in the picking area of the first working station, the handling device can handle the target carrier to the queuing area to wait, and then enter the picking area in turn to pick after the carrier in the picking area leaves.

[0106] In some examples, after the handling device handles the target carrier to the picking area of the first working station, the container taking and placing mechanism in the taking and placing device can also reach the position of the target container through lateral movement and / or vertical movement from the current position, take the target container from the target carrier, and place the target container on the picking station of the first working station. During the process that the container taking and placing mechanism handles the target container to the picking station, the image acquisition device can acquire an image (i.e., a first image) of at least one to-be-picked item in the target container, and send the acquired first image to the control device.

[0107] It should be noted that the handling device handling the target container to the first working station in the following embodiments can also include the handling device directly handling the target container to the first working station, and can also include the handling device handling the target carrier on which the target container is placed to the first working station, and the embodiments of the present disclosure are not limited thereto.

[0108] In some examples, when the handling device handles the target container to the picking area of the first working station, the first image corresponding to the target container can be acquired by the image acquisition device.

[0109] In some examples, the warehouse system can include a plurality of image acquisition devices, and the image acquisition device (hereinafter referred to as the first target acquisition device) used to acquire the first image can be one or more of the plurality of image acquisition devices. In a case that the number of the first target acquisition devices is multiple, the first target acquisition devices can include image acquisition devices arranged independently of the taking and placing device, or can include image acquisition devices arranged on the taking and placing device.

[0110] For example, the first target acquisition device for acquiring the first image can include at least one image acquisition device arranged on the support frame corresponding to the workstation. The at least one image acquisition device can include an image acquisition device arranged above the target container and an image acquisition device arranged in the direction around the target container, and the disclosure does not limit this.

[0111] In some examples, the image acquisition device can acquire a first image corresponding to the target container, and the first image is an image of at least one to-be-picked goods placed in the target container. In the case where the target container includes a plurality of to-be-picked goods, the first image can include the first goods or other goods. After the control device obtains the first image, the control device can send the acquired first image to the control device, and the control device identifies the first goods through the first image.

[0112] For example, the control device can identify the goods inside the target container and the position and layout of each good according to the first image after the image acquisition device acquires the first image, so as to determine the first goods in the target container, so that the picking mechanism in the taking and placing device can accurately grasp the first goods.

[0113] FIG. 4 is a schematic diagram of another goods picking method provided by an embodiment of the disclosure. The process of acquiring the first image by the image acquisition device will be described below in combination with FIG. 4. It should be noted that the taking and placing device includes a container taking and placing device and a goods taking and placing mechanism in FIG. 4. As shown in FIG. 4, the above step 320 includes steps 3210 to 3220.

[0114] Step 3210: The control container taking and placing mechanism moves the target container to a first docking position corresponding to the first order container on the seeding wall.

[0115] For example, the taking and placing device of the first workstation includes a container taking and placing mechanism, and the container taking and placing mechanism can take the target container from the target carrier.

[0116] In some examples, after the container taking and placing mechanism moves to the position of the target container, the container taking and placing component on the container taking and placing mechanism extends to the goods location where the target container is located along the depth direction towards the target carrier, so as to take the target container from the goods location and place the taken target container on the containing component. After the target container is placed on the containing component, the container taking and placing mechanism can move the target container to the first docking position corresponding to the first order container on the seeding wall through lateral movement and / or vertical movement from the current position.

[0117] For example, the first order container can be a container corresponding to an order to be processed, that is, the first order container is an order container for placing the first goods. In order to realize the picking of the first goods into the first order container, the target container where the first goods are located can be carried to the first docking position corresponding to the first order container by the taking and placing device, so as to facilitate the subsequent picking operation.

[0118] In some examples, the docking position can be a position at the same level as the order container position, or the docking position can be a position adjacent to the order container. For example, when the seeding wall faces the picking workstation, the adjacent position in front of the order container at the same level as the order container. The embodiments of the present disclosure do not limit the docking position, as long as the goods taking and placing mechanism in the taking and placing device can deliver the picked goods into the order container when the container taking and placing mechanism is located at the docking position. For example, the first docking position is a position adjacent to or at the same level as the first order container on the seeding wall.

[0119] Step 3220, in the process of moving the container taking and placing mechanism to the first docking position, the image acquisition device acquires a first image corresponding to the target container.

[0120] In some examples, in the process of taking out the target container by the container taking and placing mechanism and carrying it to the first docking position, the image acquisition device can pre-acquire the first image and send it to the control device, and the control device can pre-identify the first goods required by the first order container in the target container to control the goods taking and placing mechanism to grasp the first goods in the target container in advance, thereby improving the picking efficiency.

[0121] The goods picking method provided by the embodiments of the present disclosure improves the picking efficiency by directly carrying the target container to the first docking position of the first order container by the container taking and placing mechanism, compared with directly placing it in the picking station of the first workstation for picking. At the same time, the image acquisition device pre-acquires and obtains the first image in the process of moving the container taking and placing mechanism to the first docking position, so as to pre-identify the first goods, thereby further improving the subsequent picking efficiency.

[0122] Step 330, determining the first goods from the at least one goods to be picked according to the first image, and determining the first target picking information of the first goods in the case that the type of the first goods is the first goods type.

[0123] The process of determining the first goods from the at least one goods to be picked according to the first image in step 330 is described below.

[0124] Exemplarily, the control device can send a picking instruction to the taking and placing device, the picking instruction can instruct the taking and placing device to pick one of the goods in the target container, and after the taking and placing device completes picking the good, the control device sends a next to-be-picked good to the taking and placing device, thereby completing the picking operation on the at least one to-be-picked object in the target container. The first good is the good to be picked by the taking and placing device at the current time sent by the control device.

[0125] For example, the picking instruction sent by the control device to the taking and placing device can also instruct the taking and placing device to pick multiple goods in the target container, and the taking and placing device can execute picking operations on the multiple goods in turn according to the picking instruction.

[0126] Exemplarily, after determining the required first good to be picked, the control device needs to identify the first good in the target container, that is, determine the position of the first good in the target container, so that the taking and placing device picks up the first good subsequently.

[0127] In some examples, the target container can include multiple compartments, and the compartments can be arranged to be separated from each other, and each compartment can place the same good. For example, the target container can include two compartments, and the two compartments can be arranged to be separated by a partition. For example, multiple keyboards can be placed in one compartment, and multiple mice can be placed in the other compartment. In the case where the first good is determined to be a keyboard, the control device can determine the compartment where the keyboard is located, that is, determine the position of the first good in the target container.

[0128] In another example, in the case where the target container is not arranged with compartments, or different goods are placed in one compartment of the target container, the first image corresponding to the target container can be acquired by the image acquisition device, and the position of the first good in the target container can be determined based on the first image.

[0129] For example, the first good can be determined based on the first image and a preset feature information library, or can be determined based on the first image and a trained recognition model, which is not limited in the embodiments of the present disclosure. The trained recognition model can obtain richer feature information corresponding to the goods. The recognition model can identify the first good from the target container by extracting the feature information of each good in the to-be-identified image and performing feature matching according to the input to-be-identified image.

[0130] It should be noted that the embodiments of the present disclosure provide two recognition models, which are a first recognition model and a second recognition model. The first recognition model is used to identify goods (which will be described in detail in subsequent embodiments), and the second recognition model is used to identify picking information of the goods. The first recognition model and the second recognition model provided by the embodiments of the present disclosure are both trained models.

[0131] The process of determining the first goods by the first image and the preset characteristic information library is described below.

[0132] In some embodiments, the determining the first goods from the at least one to-be-picked goods according to the first image in step 330 comprises: determining characteristic information of each to-be-picked goods from the at least one to-be-picked goods according to the first image; and determining the first goods from the at least one to-be-picked goods according to the characteristic information of each to-be-picked goods, the preset characteristic information library, and the goods information of the first goods.

[0133] In some embodiments, based on the goods information of the first goods, candidate characteristic information corresponding to the goods information of the first goods is determined in the preset characteristic information library; and in a case where there is first characteristic information matching the candidate characteristic information in the characteristic information of each to-be-picked goods, the to-be-picked goods corresponding to the first characteristic information is determined as the first goods.

[0134] For example, the preset characteristic information library includes a plurality of corresponding relationships between a plurality of characteristic information and a plurality of goods information; the characteristic information includes image information of the goods and / or identification information of the goods. The image information of the goods can include at least one image corresponding to the goods; and the identification information of the goods can include at least one of a shape of the goods, a text, an image on a package of the goods, and a name (such as a product name) of the goods.

[0135] In some examples, the preset characteristic information library can store a matching relationship between the image information of the goods and the goods information; or can also store a matching relationship between the identification information of the goods and the goods information; or can also store a matching relationship between the image information of the goods and the identification information of the goods and the goods information.

[0136] In some examples, before picking the first goods, the control device can obtain the goods information of the first goods, which includes a goods ID (Identity document) or a goods number, etc. Then, the characteristic information corresponding to the first goods (such as candidate characteristic information) is determined in the preset characteristic information library according to the goods information of the first goods. Secondly, after obtaining the first image corresponding to the target container, the control device can obtain the characteristic information (such as image information) of each to-be-picked goods according to the first image, and match the characteristic information of each to-be-picked goods with each characteristic information in the preset characteristic information library.

[0137] In some examples, in a case where there is no characteristic information matching the candidate characteristic information in the characteristic information of each to-be-picked goods, it indicates that the first goods cannot be identified in the target container, and subsequent secondary identification is required.

[0138] In some examples, in a case where there is feature information (e.g., the first feature information) matching the candidate feature information among the feature information of each to-be-picked item, the item corresponding to the first feature information is the first item.

[0139] After the first item is determined (e.g., identified) from the at least one to-be-picked item, the type of the first item can be determined.

[0140] Exemplarily, the types of items include a first item type and a second item type. The picking difficulty of the items of the first item type is different from that of the items of the second item type, and the picking difficulty of the items of the first item type is lower than that of the items of the second item type. That is, the picking difficulty of the items of the first item type is lower, and the picking difficulty of the items of the second item type is higher.

[0141] It should be noted that the picking difficulty of an item is used to indicate the difficulty of successfully picking the item from a container by the item taking and placing mechanism. In a case where an item is easily grabbed (or sucked) by the item taking and placing mechanism, the item can be of the first item type, which can be referred to as an easy-to-pick item. In a case where an item is difficult to be grabbed (or sucked) by the item taking and placing mechanism, or the container is likely to fall after being grabbed (or sucked), the item can be of the second item type, which can be referred to as a difficult-to-pick item.

[0142] In some examples, in a case where the surface of an item is relatively smooth, the volume of the item is relatively small, the thickness of the item is relatively thin, or the weight of the item is too heavy, the item can be determined as the second type of item (i.e., the difficult-to-pick item) because it is difficult for the item taking and placing mechanism to pick the item from the target container, or the item is likely to fall when picked by the item taking and placing mechanism. The relatively regular items in terms of size, shape, and material are determined as the first type of item (i.e., the easy-to-pick item).

[0143] Exemplarily, the preset item information library can be generated according to the types of items. The preset item information library can include a plurality of items of the second item type.

[0144] For example, the preset item information library can include item information corresponding to a plurality of items of the second item type. The item information can be an identity document, a code, or a serial number of an item, which can be used to uniquely identify the item. Alternatively, the preset item information library can include image information of a plurality of items of the second item type, which is not limited in the embodiments of the present disclosure.

[0145] For example, after the control device identifies the first goods according to the first image, the control device can determine whether the type of the first goods is the first goods type according to the first goods and the preset goods information library.

[0146] For example, the control device can determine whether the first goods exist in the preset goods information library according to the identification information of the first goods or the image of the first goods. In a case where the first goods do not exist in the preset goods information library, it is determined that the type of the first goods is the first goods type. In a case where the first goods exist in the preset goods information library, it is determined that the type of the first goods is the second goods type.

[0147] In some examples, the preset goods information library also includes a plurality of goods of the first goods type and a plurality of goods of the second goods type. For example, the preset goods information library includes a correspondence between each goods and its type. After the first goods are determined, the control device can determine whether the type of the first goods is the first goods type or the second goods type according to the first goods and the preset goods information library.

[0148] In other examples, the preset goods information library also includes a plurality of goods of the first goods type. For example, after the first goods are determined, the control device can determine whether the first goods exist in the preset goods information library according to the first goods and the preset goods information library, so as to determine whether the type of the first goods is the first goods type.

[0149] For example, in a case where the type of the first goods is determined to be the first goods type, the control device can continue to determine the picking information of the first goods according to the first image. Since the type of the first goods is the first goods type, it indicates that the first goods can be picked up by the goods picking mechanism, and thus the control device can continue to determine the first target picking information of the first goods, so that the goods picking mechanism can accurately pick up the first goods from the target container.

[0150] For example, the first target picking information of the first goods includes a picking position and a picking direction. The picking position can be a picking point (such as a grabbing point or a suction point) and the picking direction when the picking device picks the first goods. For example, the picking point can be located at the center point of the goods, and the picking direction can be the direction of the center point. The picking device can grab or suck the goods through the center point position and the center point direction of the goods.

[0151] In some examples, after the control device acquires the first image, the control device can determine the first target picking information of the first goods based on the first image.

[0152] In some embodiments, before the above-mentioned determination of the first target picking information of the first goods, the goods picking method can further include determining the picking device and determining a target picking mechanism on the picking device for picking the first goods.

[0153] Exemplarily, in a case where the picking area includes one picking device, the picking device can be determined as a picking device for picking the first goods (which can be referred to as a target picking device below); in a case where the picking area includes multiple picking devices, the target picking device needs to be further determined from the multiple picking devices, and the picking operation on the first goods in the target container is performed through the target picking device.

[0154] For example, the distance between the target container and each picking device can be determined based on the position of the target container and the position of each picking device, and one picking device closer to the target container can be determined as the target picking device.

[0155] For another example, one picking device in an idle state from the multiple picking devices can be determined as the target picking device based on the working state of each picking device.

[0156] For another example, one picking device in an idle state and closest to the target container can be determined as the target picking device based on the distance between each picking device and the target container and the working state of each picking device. It should be noted that the target picking device can also be randomly determined from the multiple picking devices, which is not limited in the embodiments of the present disclosure.

[0157] Exemplarily, after the picking device is determined, a picking mechanism (i.e., a target picking mechanism) on the picking device for picking the first goods can be further determined. The picking device can include multiple picking mechanisms, and the target picking mechanism is at least one of the multiple picking mechanisms.

[0158] In some examples, the number of target picking mechanisms can be one or multiple, that is, the target picking mechanism can be one single picking mechanism or a picking mechanism group composed of multiple picking mechanisms.

[0159] For example, the number of target picking mechanisms can be determined based on the attribute information of the first goods; or can be determined based on experience data; or can be determined by trying to pick, that is, in a case where one suction mechanism cannot suck, the number of picking mechanisms is increased to achieve picking of the first goods.

[0160] In some embodiments, determining the target picking mechanism includes: determining the target picking mechanism based on at least one candidate picking mechanism indicated by the received picking task of the first goods; or determining the target picking mechanism from multiple picking mechanisms based on the attribute information of the first goods.

[0161] The embodiments of the present disclosure provide two methods for determining the target picking mechanism, one of which is to determine the target picking mechanism based on the picking task received by the control device, and the other of which is to determine the target picking mechanism based on the attribute information of the first goods.

[0162] The process of determining the target picking mechanism based on the attribute information of the first cargo is described below.

[0163] Exemplarily, the attribute information of the first cargo includes at least one of a type, a size, a weight, and a shape of the first cargo.

[0164] In some examples, the cargo in the warehouse system can be divided into multiple cargo categories, such as daily necessities, food, electronic products, and the like. The size of the cargo can include the length, width, height, volume, length-height ratio, or length-width ratio of the cargo, and the like. The weight of the cargo can be the actual weight of the cargo, or the weight with packaging. The shape of the cargo can be a regular cuboid, a square, a cylinder, and other irregular shapes, and the like.

[0165] In some embodiments, based on the attribute information of the first cargo and the device parameters of each picking mechanism in the plurality of picking mechanisms, a target picking mechanism is determined from the plurality of picking mechanisms.

[0166] Exemplarily, the device parameters of the picking mechanism include at least one of a size, a material, a hardness, a picking force, a shape, and a weight of the picking mechanism.

[0167] In some examples, the picking mechanism can be a gripper, a claw, or a suction cup, and the device parameters corresponding to the picking mechanism can be different in the case of different structures of the picking mechanism. The device information corresponding to the picking mechanism is exemplarily described in the embodiments of the present disclosure taking the suction cup as an example.

[0168] For example, the device parameters of the suction cup can include the size of the suction cup, the material of the suction cup, the hardness of the suction cup, the suction force of the suction cup, the shape of the suction cup, and the weight of the suction cup, and the like. The size of the suction cup can include the diameter or radius of the suction cup, the height or thickness of the suction cup, and the like. The material of the suction cup can include silica gel material or other materials, and the like. The hardness of the suction cup can be divided into soft suction cup, medium-hardness suction cup, and hard suction cup, and the like. The suction force of the suction cup can also be divided into small suction force, medium suction force, and large suction force. The shape of the suction cup includes flat type, circular, oval, or other customized shape, and the like. The weight of the suction cup can be divided into small suction cup, medium suction cup, and large suction cup, and the like.

[0169] In some examples, after determining the attribute information of the first cargo and the device information of each picking mechanism, the device information that is suitable for the attribute information of the first cargo can be determined, and the picking mechanism corresponding to the device information is determined as the target picking mechanism.

[0170] For example, the suction force required for picking up the first goods can be determined based on the goods category, size, shape and weight of the first goods, and then a target suction cup suitable for the first goods can be selected based on the suction force of each suction cup. For example, when the size or weight of the first goods is large, a suction cup with a larger suction force can be selected as the target suction cup.

[0171] In some embodiments, the target picking mechanism is determined from a plurality of picking mechanisms based on the attribute information of the first goods and the first knowledge base.

[0172] In some examples, a correspondence (e.g., the first correspondence) between the attribute information of each goods and the equipment information of each picking mechanism can be established based on historical experience data, and the first knowledge base can be obtained based on the first correspondence. That is, the first correspondence between different attribute information and the equipment parameters of different picking mechanisms can be included in the first knowledge base.

[0173] For example, after the first goods are determined, the equipment information suitable for the attribute information of the first goods can be queried in the first knowledge base based on the attribute information of the first goods, and the picking mechanism corresponding to the equipment information is determined as the target picking mechanism.

[0174] The process of determining the target picking mechanism based on the received picking task will be described below.

[0175] For example, the control device can receive a picking task sent by an upstream system (e.g., a server), and the picking task can be used to instruct the control device to perform a picking operation on the first goods. The picking task can also include at least one candidate picking mechanism for picking the first goods. The control device can determine the target picking mechanism based on the at least one candidate picking mechanism in the picking task.

[0176] In some examples, when the number of candidate picking mechanisms in the picking task is one, the candidate picking mechanism can be determined as the target picking mechanism; when the number of candidate picking mechanisms in the picking task is more than one, one or more of the candidate picking mechanisms can be selected as the target picking mechanism based on the attribute information of the first goods.

[0177] It should be noted that the upstream system can determine at least one candidate picking mechanism corresponding to the first goods based on the attribute information of the first goods, or the trained recognition model can also be used to determine at least one candidate picking mechanism corresponding to the first goods, and the present disclosure does not limit this.

[0178] In some embodiments, determining the first target picking information of the first cargo includes: determining the first target picking information of the first cargo based on at least one of the first image including the first cargo, the attribute information of the first cargo, and the picking prompt information corresponding to the first cargo collected by the image collection device.

[0179] It should be noted that the present embodiment does not limit the execution order of determining the target picking mechanism and determining the first target picking information. For example, the determination of the first target picking information can be performed before the determination of the target picking mechanism, can be performed after the determination of the target picking mechanism, or can be performed simultaneously with the determination of the target picking mechanism.

[0180] In some examples, after determining the target picking mechanism for picking the first cargo and the first target picking information, the control device can control the target picking mechanism to reach the position of the target container from the current position through lateral movement and / or vertical movement, and pick the first cargo in the target container based on the picking information corresponding to the first cargo (i.e., the first target picking information). Wherein, the picking information corresponding to each cargo is different, and inaccurate picking information can cause the target picking mechanism to fail to pick up.

[0181] In some examples, the present embodiment provides three methods for determining the first target picking information, which are determined by a second identification model, determined by a preset picking information library, and determined by a preset picking information library and a second identification model.

[0182] The implementation process of determining the first target picking information of the first cargo by the second identification model will be described below.

[0183] In some embodiments, based on the first image, the attribute information of the first cargo, and the picking prompt information, the first target picking information of the first cargo is determined by the second identification model.

[0184] In some examples, the second identification model can be a pre-trained identification model, and the second identification model can identify the picking information corresponding to each cargo.

[0185] FIG. 5 is a schematic diagram of a second identification model provided by the present embodiment. As shown in FIG. 5, the second identification model 500 can include a visual processing network 510, an encoding network 520, and a feature fusion network 530.

[0186] In some examples, the visual processing network 510 outputs, by processing the input first image, a respective segmentation result, depth information and stacking relationship information corresponding to each of the goods in the first image. The encoding network 520 is configured to perform text encoding processing on the attribute information of the first goods and the picking prompt information of the first goods, to output features corresponding to the attribute information and the picking prompt information, respectively. The feature fusion network 530 is configured to perform fusion analysis processing on the features output by the visual processing network 510 and the features output by the encoding network 520, so as to identify the first target picking information corresponding to the first goods.

[0187] In some embodiments, based on the first image, a first feature corresponding to the first goods is obtained by a visual processing network in the second identification model.

[0188] In some embodiments, the above obtaining, based on the first image, the first feature corresponding to the first goods by the visual processing network in the second identification model comprises: performing block processing on each of the visual images by an image block module in the visual processing network to obtain a first output sequence; wherein the first output sequence comprises image blocks corresponding to each of the visual images. Performing position encoding on each of the image blocks in the first output sequence by a position encoding module in the visual processing network to obtain a second output sequence. Performing feature extraction processing on the second output sequence by a feature extraction module in the visual processing network to obtain an output feature. Inputting the output feature into a task module in the visual processing network to obtain the first feature.

[0189] In some examples, the first image can include multiple visual images. For example, the first image can include multiple Red-Green-Blue (RGB) images, i.e., the first image can be RBG images taken by cameras with different perspectives.

[0190] For example, taking the first image including 4 RGB images as an example, the 4 RGB images can be respectively images (i.e., the first image) of the target container taken by cameras with front, rear, left and right perspectives, respectively.

[0191] In some examples, before the multiple images (i.e., the first image) are input into the visual processing network 510, the first image can be pre-processed. The image pre-processing process can include normalization processing, cropping or scaling processing, and the image after the image pre-processing can meet the input requirements of the visual processing network 510.

[0192] Exemplarily, the first image (or the first image after image preprocessing) is input into the visual processing network 510, and the first image is processed by the visual processing network 510 to extract first features. The first features are used to represent at least one of the segmentation result information, the depth information and the stacking relationship information corresponding to each of the goods (including the first goods) in the first image.

[0193] In some examples, the visual processing network 510 can include a vision transformer (ViT), which can model the relationship between all pixels (or image blocks) in an image through a global attention mechanism, thereby capturing long-distance dependencies.

[0194] Exemplarily, the visual processing network 510 can include an image blocking module, a position encoding module, a feature extraction module and a task module. The ViT can include the image blocking module, the position encoding module and the feature extraction module.

[0195] Exemplarily, the image blocking module performs blocking processing on each of the received visual images to obtain a first output sequence, and the first output sequence includes image blocks corresponding to each of the visual images.

[0196] In some examples, taking the resolution of each image in the multi-channel visual image as 224x224x3 for example, the image blocking module can divide each image into N image blocks, such as 256 image blocks (e.g., 16x16), and the dimension of each image block is 16x16x3. The image blocking module can perform linear projection processing on each image block to map each image block to the input dimension D of the Transformer to obtain a first output sequence, and the dimension of the first output sequence is NxD.

[0197] Exemplarily, after obtaining the first output sequence, the position encoding module can perform position encoding on each image block in the first output sequence to obtain a second output sequence.

[0198] In some examples, the position encoding module can add position encoding to each image block to retain the spatial position information of each image block, so that the blocking order of each image, i.e., the order of each image block, can be determined.

[0199] For example, after the 4-way visual image is processed by the avatar partitioning module, the first output sequence T = [T1.1, T1.2, …, T1.256, T2.1, T2.2, …, T2.256, …, T4.1, T4.2, …, T4.256] is output. The position encoding is added to each image block in the first output sequence. The position encoding of the first 256 image blocks (i.e., view 1) is p1, the position encoding of the next 256 image blocks (i.e., view 2) is p2, and so on. The second output sequence is obtained, which is the first output sequence T + position encoding.

[0200] For example, the feature extraction module is used to perform feature extraction on the second output sequence to obtain output features.

[0201] In some examples, the feature extraction module can be a Transformer encoder. The Transformer encoder can gradually fuse multi-view information and extract global features through multi-head self-attention and a feedforward neural network. The multi-head self-attention calculates the attention weights between all blocks to capture global relationships, and the feedforward neural network can perform nonlinear transformation to obtain the output features corresponding to the first image.

[0202] For example, the output features are input into the task module to obtain first features. The task module can include multiple task branches, which can include a first task branch, a second task branch, and a third task branch.

[0203] In some examples, after the output features are processed by the ViT, the output features are input into different task branches to obtain segmentation result information, depth information, and stacking relationship information corresponding to the first image, respectively.

[0204] For example, the output features can be upsampled by the decoder in the first task branch to obtain a pixel-level segmentation mask, thereby identifying the regions of each cargo in the first image. The segmentation results of multiple views are combined to generate the segmentation results by voting or geometric consistency algorithm. The output features are processed by the second task branch to estimate the depth of each pixel and generate the depth map or 3D coordinates (i.e., depth information) of each cargo. The output features are processed by the classification or regression module in the third task branch to predict the stacking relationship between each cargo and model the spatial relationship of each cargo by a graph neural network, thereby outputting the stacking order.

[0205] In some examples, the visual processing network 510 can generate respective implicit embeddings corresponding to the above-mentioned tasks while completing the tasks. For example, the visual processing network 510 can generate a segmentation embedding for encoding shape and contour information of the goods based on the segmentation result, generate a depth embedding for encoding 3D structure and spatial position of the goods based on the depth information, and generate a layout embedding for encoding the stacking relationship and scene layout of the goods based on the stacking relationship information. That is, the first feature can include the segmentation embedding, the depth embedding, and the layout embedding.

[0206] In some embodiments, based on the attribute information of the first goods and the picking prompt information, a second feature corresponding to the attribute information and a third feature corresponding to the picking prompt information are obtained through an encoding network in the second recognition model.

[0207] In some embodiments, the attribute information of the first goods is textually encoded through a first encoding module in the encoding network to obtain the first feature, and the picking prompt information is textually encoded through a second encoding module in the encoding network to obtain the second feature.

[0208] For example, the encoding network 520 can include a first encoding module 521 and a second encoding module 522. The first encoding module 521 is configured to encode the attribute information of the first goods, and the second encoding module 522 is configured to encode the picking prompt information of the first goods.

[0209] In some examples, the attribute information of the first goods can include information such as the type, size, and position of the first goods. For example, the attribute information can include “keyboard” (type of goods), “30cm x 15cm x 3cm” (size of goods), and “second compartment” (position of goods).

[0210] In some examples, the picking prompt information is used to indicate at least one constraint information for picking the first goods. The picking prompt information can guide the second recognition model to generate a text description of a reasonable picking strategy. The picking prompt information can be generated based on the attribute information of the first goods and historical prior data.

[0211] For example, the picking prompt information can include at least one of a picking position constraint, a picking priority constraint, a picking interaction constraint, and a picking obstacle avoidance constraint. Taking the first goods as socks as an example, the picking position constraint can include “pick the label on the sock”, and the picking obstacle avoidance constraint can include “avoid the sock body”. Taking the first goods as a glass as an example, the picking priority constraint can include “preferentially pick the upper layer”, and the picking interaction constraint can include “handle with care”, etc.

[0212] Exemplarily, the attribute information corresponding to the first cargo is input to the first encoding module 521, and the picking prompt information corresponding to the first cargo is input to the second encoding module 522, and the first encoding module 521 and the second encoding module 522 can convert the attribute information and the picking prompt information into semantic information that can be recognized and processed by the second recognition model.

[0213] For example, the structures of the first encoding module 521 and the second encoding module 522 can be the same or different, and the embodiments of the present disclosure do not limit this. For example, the first encoding module 521 and the second encoding module 522 can be a text encoder (Text Encoder), such as a CLIP (Contrastive Language-Image Pre-training) network or a BERT (Bidirectional Encoder Representations from Transformers) network based on Transformer, and the embodiments of the present disclosure do not limit this. It should be noted that the process of text encoding of the attribute information and the picking prompt information by the text encoder is similar to the encoding process in the related art, which will not be described here.

[0214] In some embodiments, in a case where the first feature, the second feature and the third feature are determined, the first target picking information corresponding to the first cargo can be determined based on the first feature, the second feature and the third feature by a feature fusion network in the second recognition model.

[0215] In some embodiments, the above determination of the first target picking information corresponding to the first cargo based on the first feature, the second feature and the third feature by the feature fusion network in the second recognition model comprises: performing splicing processing on the first feature, the second feature and the third feature by a splicing module in the feature fusion network to obtain a joint feature; and processing the joint feature by an identification analysis module in the feature fusion network to obtain the first target picking information.

[0216] Exemplarily, the feature fusion network 530 can be a pre-trained multi-modal model, and the feature fusion network 530 can fuse and process the first feature output by the visual processing network 510 and the second feature and the third feature output by the encoding network 520 to generate the first target picking information corresponding to the first cargo, and the first target picking information includes a picking position and a picking direction.

[0217] In some examples, the feature fusion network 530 can include a splicing module and an identification analysis module. After the first feature, the second feature, and the third feature are input into the feature fusion network 530, the first feature, the second feature, and the third feature are first processed by the splicing module to generate a total output sequence (i.e., a joint feature). Second, the joint feature is input into the identification analysis module for analysis processing to generate the first target picking information corresponding to the first cargo. The first target picking information can be a picking pose, which can include a 3D coordinate and a rotation angle.

[0218] In some embodiments, after the above-mentioned determination of the pick-and-place device, the method further includes: acquiring a first image including the first cargo collected by the image acquisition device; based on the first image, the attribute information of the first cargo, the picking prompt information corresponding to the first cargo, and the device parameters of each picking mechanism of the plurality of picking mechanisms of the pick-and-place device, determining, by the second identification model, a target picking mechanism and first target picking information corresponding to the first cargo among the plurality of picking mechanisms; and based on the first target picking information, controlling the target picking mechanism to pick the first cargo from the target container and transfer the first cargo to the target position.

[0219] In some examples, as described in the above embodiments, the target picking mechanism and the first target picking information can be determined simultaneously. In this case, the second identification model (such as the second identification model 500 in FIG. 5) can be used to simultaneously determine the target picking mechanism and the first target picking information. That is, after the first cargo is determined, the second identification model can be used to determine the target picking mechanism and the first target picking information corresponding to the first cargo.

[0220] For example, in the case where the second identification model 500 determines the target picking mechanism, the input information of the second identification model 500 further includes the device parameters of each picking mechanism of the pick-and-place device. The device parameters of each picking mechanism include at least one of the size, material, hardness, picking force, shape, and weight of the picking mechanism.

[0221] In some embodiments, based on the first image, a first feature corresponding to the first cargo is obtained by a visual processing network in the second identification model; the first feature is used to represent at least one of a segmentation result, depth information, and stacking relationship information corresponding to the first cargo; based on the attribute information of the first cargo, the picking prompt information, and the device parameters of each picking mechanism, a second feature corresponding to the attribute information, a third feature corresponding to the picking prompt information, and a fourth feature corresponding to the device parameters of each picking mechanism are obtained by an encoding network in the second identification model; and based on the first feature, the second feature, the third feature, and the fourth feature, a target picking mechanism and first target picking information corresponding to the first cargo are determined by a feature fusion network in the second identification model.

[0222] It should be noted that the process of determining the first feature, the second feature and the third feature has been described in detail in the above embodiment, and will not be repeated here.

[0223] In some examples, when the target picking mechanism is determined through the second identification model 500, the device parameters of each picking mechanism can be input into the encoding network 520 for text encoding processing of the device parameters of each picking mechanism through the encoding network 520, so as to obtain the fourth feature.

[0224] For example, the encoding network 500 can further include a third encoding module, and the device parameters of the picking mechanism can be input into the third encoding module for encoding of the device parameters of the picking mechanism through the third encoding module. Alternatively, the device parameters of the picking mechanism can also be input into the first encoding module 510 or the second encoding module 520 for encoding of the device parameters of the picking mechanism through the first encoding module 510 or the second encoding module 520 to obtain the fourth feature, which is not limited in the embodiment of the present disclosure.

[0225] In some examples, after obtaining the fourth feature, the first feature, the second feature, the third feature and the fourth feature are input into the feature fusion network, the first feature, the second feature, the third feature and the fourth feature are spliced through the splicing module in the feature fusion network, and a new joint feature is obtained. Then, the target picking mechanism and the first target picking information are determined through the identification analysis module. Since the device information of each picking mechanism is included in the input information, the identification analysis module can comprehensively consider the attribute information of the first cargo and the device parameters of each picking mechanism, so as to determine the target picking mechanism more suitable for the first cargo.

[0226] It should be noted that the embodiment of the present disclosure can also first determine the picking mechanism corresponding to the first cargo through the first knowledge base, and in the case that the device parameters corresponding to the attribute information of the first cargo do not exist in the first knowledge base, i.e., the first knowledge base cannot determine the picking mechanism corresponding to the first cargo, the target picking mechanism corresponding to the first cargo is further determined based on the second identification model. After the target picking mechanism corresponding to the first cargo is determined through the second identification model and the first cargo is successfully picked based on the target picking mechanism, the first knowledge base can be updated, so that the first cargo can be directly determined through the first knowledge base in subsequent picking.

[0227] The implementation process of determining the first target picking information of the first cargo through the preset picking information base will be described below.

[0228] In some examples, to further improve the picking efficiency, the control device can determine the first target picking information of the first goods according to the first characteristic information and the preset picking information library after determining the first characteristic information of the first goods through the above embodiments.

[0229] FIG. 6 is a schematic diagram of another goods picking method provided by the embodiments of the present disclosure. The process of determining the first target picking information through the preset picking information library will be described below in combination with FIG. 6. As shown in FIG. 6, the above step 330 includes the following steps:

[0230] Step 3310, obtaining a preset picking information library.

[0231] Exemplarily, the preset picking information library can include the correspondence between the characteristic information of each goods and each picking information.

[0232] In some examples, the preset picking information library includes the picking information corresponding to each characteristic information of each goods in the plurality of goods. For example, when the characteristic information is image information of the goods, each goods can correspond to a plurality of images, each image can correspond to a picking information, and the picking information corresponding to different images of the same goods can be the same or different, which is not limited by the embodiments of the present disclosure.

[0233] In some examples, the preset picking information library can be generated according to historical picking information. For example, in the historical picking process, when the picking information of the goods is determined through the characteristic information of the goods, the correspondence between the characteristic information of the goods and the picking information can be stored in the preset picking information library, and the preset picking information library can be stored and updated after each picking is completed.

[0234] Exemplarily, the first characteristic information of the first goods can be determined in the process of determining the first goods, and the first target picking information of the first goods can be further determined according to the first characteristic information and the preset picking information library.

[0235] In some examples, the above preset characteristic information library and the preset picking information library can be one information library or different information libraries, which is not limited by the embodiments of the present disclosure.

[0236] For example, when the preset characteristic information library and the preset picking information library can be one information library, the information library can include the correspondence relationship among the goods information of the goods, the characteristic information of the goods, and the picking information of the goods. Whether the first characteristic information corresponding to the characteristic information of the first goods matches the first characteristic information can be found in the information library according to the characteristic information of the at least one to-be-picked goods. In the case that the characteristic information of the at least one to-be-picked goods exists the first characteristic information, the first target picking information corresponding to the first characteristic information can be further determined in the information library. That is, in the case that the preset characteristic information library and the preset picking information library can be one information library, the first goods and the first target picking information of the first goods can be determined according to the information library.

[0237] It should be noted that the preset characteristic information library and the preset picking information library are taken as different information libraries for example in the embodiments of the present disclosure.

[0238] In some embodiments, after the preset picking information library is acquired, the first target picking information of the first goods can be determined according to the first characteristic information and the preset picking information library.

[0239] Step 3320, determining whether the picking information corresponding to the first characteristic information exists in the preset information library.

[0240] In the case that the picking information corresponding to the first characteristic information exists in the preset picking information library, step 3330 is performed; in the case that the picking information corresponding to the first characteristic information does not exist in the preset picking information library, step 3341 or step 3351 is performed.

[0241] Step 3330, determining the picking information corresponding to the first characteristic information as the first target picking information.

[0242] In some examples, in the case that the picking information corresponding to the first characteristic information exists in the preset picking information library, the picking information corresponding to the first characteristic information in the preset picking information library is determined as the first target picking information.

[0243] Step 3341, controlling the taking and placing device to adjust the position of the at least one to-be-picked goods in the target container.

[0244] For example, in the case that the picking information corresponding to the first characteristic information does not exist in the preset picking information library, the position of the at least one to-be-picked goods in the target container can be adjusted to change the placement position, placement posture, and placement direction of the first goods in the target container, and the like.

[0245] The taking and placing device can adjust the position of one, part, or all of the goods in the target container, and the embodiments of the present disclosure do not limit this.

[0246] At step 3342, after adjusting the position of the at least one to-be-picked goods in the target container, the image acquisition device is controlled to acquire a second image corresponding to the target container.

[0247] In some examples, after adjusting the position of the at least one to-be-picked goods in the target container, the position of the first goods can change, and therefore, the image acquisition device can be controlled to reacquire and acquire a second image corresponding to the target container. In the second image, the position of the first goods is different from that in the first image.

[0248] At step 3343, the first target picking information of the first goods is determined according to the second image.

[0249] In some examples, the image acquisition device can send the acquired second image to the control device, the control device acquires the second image, determines the first goods among the at least one to-be-picked goods according to the second image, determines the second feature information of the first goods, and re-determines the first target picking information of the first goods according to the second feature information and the preset picking information library.

[0250] It should be noted that, in the case where the second feature information of the first goods does not exist in the preset picking information library, step 3341 can be returned and re-executed to adjust the position of the first goods again until the first target picking information of the first goods is determined or until the number of adjustments exceeds a preset threshold number of times.

[0251] For example, the preset threshold number of times can be set according to requirements, such as the preset threshold number of times can be set to 5. When the position of the goods in the target container is adjusted by the taking and placing device for more than 5 times, the first target picking information of the first goods still cannot be determined, and therefore, it is not necessary to adjust the position again, and step 3351 can be continued to be executed.

[0252] That is, in the case where the first target picking information of the first goods cannot be determined by executing steps 3341 to 3343, or the number of repeated executions of steps 3341 to 3343 exceeds the preset threshold number of times, the first target picking information of the first goods still cannot be determined, and therefore, step 3351 can be continued to be executed to further determine the first target picking information of the first goods.

[0253] In some examples, the first target picking information of the first goods can also be identified by a shadow of the illumination of the first goods in the target container. In a case where the first target picking information of the first goods cannot be identified by the current illumination shadow, the position of the first goods in the target container is adjusted to change the illumination shadow of the first goods, so as to further determine the first target picking information of the first goods.

[0254] In step 3351, the first image is sent to the interactive device.

[0255] In some examples, the warehouse system includes an interactive device, such as a display. In a case where the first target picking information of the first goods does not exist in the preset picking information library, the first image can be sent to the interactive device. The interactive device can be a device with input and output functions, and the specific form of the interactive device is not limited in the embodiments of the present disclosure.

[0256] In step 3352, the target contour information corresponding to the first goods input by the user on the interactive device is obtained.

[0257] In some examples, after the interactive device receives the first image, the first image is displayed. The user can draw the contour information (i.e., the target contour information) of the first goods based on the first image displayed by the interactive device, such as drawing the contour line of the first goods, and send the drawn target contour information to the control device.

[0258] For example, the interactive device has an output function, such as a display function, and the interactive device can output and display the first image. The interactive device also has an input function, and the user can input the target contour information corresponding to the first goods based on the first image displayed by the interactive device.

[0259] In step 3353, the first target picking information of the first goods is determined according to the target contour information.

[0260] In some examples, after the control device obtains the target contour information of the first goods, the control device can determine the first target picking information of the first goods, i.e., the picking position and picking direction of the first goods, based on the target contour information.

[0261] For example, in a case where the control device cannot automatically identify the first target picking information of the first goods, the position of the first goods in the target container can be adjusted by performing steps 3341 to 3343, so as to reacquire the second image corresponding to the target container, and re-determine the first target picking information of the first goods according to the second image. Alternatively, the target contour information of the first goods can also be obtained in an artificial assisted manner by performing steps 3351 to 3353, so as to further determine the first target picking information of the first goods according to the target contour information.

[0262] In some examples, if the control device cannot automatically identify the first target pickup information of the first cargo, the first target pickup information of the first cargo can be determined by first performing position adjustment of steps 3341 to 3343. If the first target pickup information of the first cargo cannot be determined after performing steps 3341 to 3343 more times than a preset threshold, steps 3351 to 3353 can be performed to further determine the first target pickup information of the first cargo through manual assistance.

[0263] For example, the control device can determine the pickup information of the goods by recognizing the outline of the goods in the image. If the control device cannot determine the first target pickup information of the first goods, it means that the outline of the first goods cannot be recognized in the first image. Therefore, the outline of the first goods can be manually drawn by human assistance to further determine the first target pickup information of the first goods.

[0264] In some embodiments, first feature information and first target pickup information are stored in a preset pickup information database to update the preset pickup information database.

[0265] In some examples, after obtaining the pickup information of the goods through manual assistance or by adjusting the position, the characteristic information of the goods and the corresponding pickup information can be stored in a preset pickup information database so that when the goods appear again in the future, their corresponding pickup information can be determined more quickly.

[0266] The goods picking method provided in this disclosure can determine the first target picking information of the first goods through the first feature information of the first goods. When the first target picking information of the first goods cannot be determined, the first target picking information of the first goods can be further determined by position adjustment and / or manual contour drawing, thereby improving the efficiency of identifying the first goods and improving the accuracy of subsequent picking of the first goods.

[0267] The following describes the process of determining the first target pickup information of the first cargo by jointly using a preset pickup information database and a second recognition model.

[0268] In some embodiments, if the first target pickup information of the first cargo does not exist in the preset pickup information database, the first target pickup information of the first cargo is determined by a second recognition model based on the first image, the attribute information of the first cargo, and the pickup prompt information corresponding to the first cargo.

[0269] For example, the preset pickup information database may or may not include pickup information corresponding to the first item. After acquiring the first image, it can first be determined whether pickup information corresponding to the first item exists in the preset pickup information database based on the first image. If pickup information corresponding to the first item does not exist in the preset pickup information database, the first target pickup information corresponding to the first item can be determined by a second recognition model based on the first image, the attribute information of the first item, and the pickup prompt information corresponding to the first item.

[0270] In some embodiments, if the pickup information for the first item is not found in the preset pickup information database, the preset pickup information database is updated based on the first image and the first target pickup information.

[0271] In some examples, since the pickup information corresponding to the first item does not exist in the preset pickup information database, after the first pickup information corresponding to the first item is identified by the second recognition model, the preset pickup information database can be updated based on the first image and the first target pickup information.

[0272] For example, the correspondence between the first image and the first target picking information can be stored in a preset picking information database. This makes it easier to directly obtain the picking information corresponding to the first goods when picking the first goods in the future, thereby further improving the picking efficiency of the picking and placing device and the working efficiency of the warehousing system.

[0273] It should be noted that the update of the preset pickup information database involved in this embodiment may be based on the first image and the first target pickup information after the pickup device (i.e., the target pickup mechanism) picks up the first goods based on the first target pickup information.

[0274] Step 340: Based on the first target picking information, control the picking and placing device to pick out the first goods from the target container and transfer the first goods to the target location.

[0275] In some embodiments, step 340 may include: based on first target picking information, controlling the target picking mechanism to pick up the first goods from the target container and transfer the first goods to the target location.

[0276] For example, after determining the first target pickup information of the first cargo, the target pickup mechanism in the pick-and-place device can be controlled to pick up the first cargo from the target container based on the first target pickup information. After picking up the first cargo, the target pickup mechanism transfers the first cargo to the target location. Specifically, the target pickup mechanism can pick up the first cargo from the target container based on the pickup position and pickup direction, and transfer the first cargo to the target location according to the target path.

[0277] In some examples, the target picking mechanism can determine the target picking method and target path based on the first target picking information. For example, the target picking mechanism can move to the location of the first item (such as the picking location) according to the target path based on the first target picking information, retrieve the first item using the target picking method, and transfer the first item to the target location.

[0278] For example, the target path can be the optimal path for the target picking mechanism to move from its current position to the picking position. This optimal path can be the shortest path or the path with the fewest obstacles. The control device can adjust the position and orientation of the target picking mechanism to ensure that it can stably and safely pick up the first item and avoid damage to adjacent items.

[0279] In some examples, the target location can be the location of the delivery mechanism on the distribution device; or, the target location can be the location corresponding to the first order container of the required first goods, such as the location within the first order container.

[0280] In some embodiments, step 340 further includes: controlling the cargo picking mechanism to pick out the first cargo from the target container and transfer the first cargo to the first order container based on the first target picking information.

[0281] In some examples, the target location includes the location corresponding to the first order container, i.e., the location within the first order container. When the container pick-and-place mechanism and the goods pick-and-place mechanism are coupled, the container pick-and-place mechanism moves the target container to the first docking position of the first order container, and the goods pick-and-place mechanism (i.e., the target picking mechanism) can pick the first goods in the target container at the first docking position and deliver the picked first goods into the first order container.

[0282] In some examples, after determining the images of each item and the corresponding picking information, the images and corresponding picking information can be used as a training dataset to train the picking process of the goods picking mechanism (such as intelligent training). This allows the goods picking mechanism to determine the picking information through training, thereby facilitating subsequent picking and improving picking efficiency.

[0283] The goods picking method provided in this disclosure identifies a first item from at least one item to be picked in the target container by acquiring a first image corresponding to the target container, and determines the type of the first item based on the first item and a preset goods information database. When the type of the first item is a first type of goods, the method further determines the first target picking information corresponding to the first item. Therefore, by pre-determining the type of goods before picking, and further determining the picking information of the first item when it is an easy-to-pick item (i.e., the first type of goods), this disclosure can avoid picking anomalies or picking failures during the picking process of the first item by the goods picking and placing mechanism, thereby improving the safety of the picking process.

[0284] Meanwhile, since the first target pickup information determined in this embodiment is based on the first image corresponding to the first goods, the attribute information of the first goods, and the pickup prompt information corresponding to the first goods, the first target pickup information is more in line with the pickup requirements of the first goods. This makes it easier for the picking and placing device (such as the target pickup mechanism) to pick up the first goods more accurately and efficiently, thus improving the accuracy and efficiency of the picking and placing device. Therefore, this embodiment can improve the picking efficiency and accuracy of the picking and placing device, as well as the overall working efficiency of the warehousing system.

[0285] Figure 7 is a schematic diagram of another goods picking method provided in an embodiment of this disclosure. As shown in Figure 7, after the target picking mechanism picks up the first goods in the target container based on the first target picking information (i.e., step 340), the method further includes steps 710 to 750 as shown below.

[0286] Step 710: If the target picking mechanism fails to pick up the first goods based on the first target picking information, control the target picking mechanism to pick up the first goods again based on the first target picking information.

[0287] In some examples, the target picking mechanism may succeed or fail in picking up the first item based on the first target picking information. In the event of a first failed pick-up, the control device can control the target picking mechanism to attempt to pick up the item again based on the first target picking information.

[0288] Step 720: When the number of times the target picking mechanism fails to pick up the first cargo based on the first target picking information reaches a first preset number, the target picking mechanism is controlled to adjust the position of the first cargo in the target container.

[0289] In some examples, the first preset number of times can be set according to requirements, such as 2 times, 3 times or 4 times. The specific value of the first preset number of times is not limited in the embodiments of this disclosure.

[0290] For example, the control device controls the target picking mechanism to pick up the first item based on the first target picking information no more than a first preset number of times. That is, after the target picking mechanism fails to pick up the item the first time, it can pick it up again multiple times, as long as the number of times the target picking mechanism picks up the item does not exceed the first preset number of times.

[0291] In some examples, if the target picking mechanism successfully picks up the first item within the first preset number of attempts, it indicates that the first item has been successfully picked up; if the target picking mechanism fails to pick up the first item within the first preset number of attempts, it indicates that the first item has failed to be picked up. In this case, the control device can control the target picking mechanism to adjust the position of the first item.

[0292] For example, the target picking mechanism can adjust the placement position and / or orientation of the first cargo in the target container to obtain the adjusted first cargo.

[0293] Step 730: Acquire a third image of the adjusted target container captured by the image acquisition device.

[0294] In some examples, after adjusting the pose of the first cargo, a third image corresponding to the adjusted first cargo can be re-acquired by an image acquisition device. This third image includes the first cargo. The image acquisition device can then send the acquired third image to the control device. The third image differs from the first image.

[0295] It should be noted that the process of the image acquisition device acquiring the third image is similar to the process of acquiring the first image in the above embodiment. To avoid repetition, it will not be described again here.

[0296] Step 740: Based on at least one of the third image, the attribute information of the first cargo, and the pickup prompt information corresponding to the first cargo, determine the second target pickup information corresponding to the first cargo.

[0297] In some examples, the control device can determine the second target pickup information corresponding to the first cargo based on at least one of the third image, the attribute information of the first cargo, and the pickup prompt information corresponding to the first cargo. For example, the control device can also determine the second target pickup information corresponding to the first cargo through a second recognition model, a preset feature information library, or a preset feature information library and a second recognition model.

[0298] It should be noted that the process of determining the second target picking information is similar to the process of determining the first target picking information in the above embodiment (such as step 330). To avoid repetition, it will not be described again here.

[0299] Step 750: Based on the second target picking information, control the target picking mechanism to pick up the first item from the target container.

[0300] In some examples, the target picking mechanism can pick up the first item from the target container based on the redefined second target picking information. If the target picking mechanism fails to pick up the item on the first attempt based on the second target picking information, it can also attempt to pick it up again based on the second target picking information. If it succeeds within a first preset number of attempts, it indicates that the first item has been successfully picked.

[0301] In some examples, if the target picking mechanism fails to pick up the first item within a first preset number of times based on the second target picking information, the control device can control the target picking mechanism to adjust the position of the first item in the target container again, that is, to re-execute the above steps 730 to 750.

[0302] For example, to further improve the picking efficiency of the pick-and-place device and avoid the device continuously adjusting the pose of a single item in the event of a pickup failure, the number of pose adjustments for each item can be limited. For instance, the number of pose adjustments for each item can be set as a threshold for the first time.

[0303] In some examples, if the number of pose adjustments for the first item reaches the first threshold and picking still fails, it indicates that picking the first item has failed. The pick-and-place device does not need to adjust the pose of the first item again, nor does it need to pick the first item again. For example, the control device can mark the first item as a picking anomaly and control the pick-and-place device to perform the picking operation on the next item to be picked in the target container.

[0304] In some embodiments, if the target picking mechanism successfully picks up the first goods based on the second target picking information, the preset feature information database is updated based on the third image and the second target picking information.

[0305] In some examples, since the picking and placing device fails to pick up the first item based on the first target picking information but successfully picks up the first item based on the second target picking information, the second picking information is the picking information that is more suitable for the first item. The control device can update the preset feature information database based on the third image and the second target picking information, so that when picking the first item in the future, the second target picking information corresponding to the first item can be directly obtained through the preset feature information database, thereby improving the picking efficiency of the picking and placing device and the working efficiency of the warehousing system.

[0306] In some embodiments, after step 750 above, the method further includes: if the target picking mechanism fails to pick up the first goods based on the second target picking information, controlling the output device to output an abnormal prompt message; instructing the execution object to operate the target picking mechanism based on the abnormal prompt message so that the target picking mechanism performs a picking operation on the first goods; if the target picking mechanism successfully picks up the first goods, updating the preset feature information database based on the picking information of the first goods when the picking is successful.

[0307] In some examples, the warehousing system may also include output devices, which may include at least one of a display, indicator lights, and a voice device. If the target picking mechanism fails to pick up the first item based on the second target picking information, an error message can be generated, and the output device can be controlled to output the error message. For example, the control device can control the display, indicator lights, or voice device to output the error message; alternatively, at least two of the display, indicator lights, and voice devices may simultaneously output the error message. This disclosure does not limit the scope of the embodiments.

[0308] In some examples, the error message may include previously attempted pickup information for the first item (such as first target pickup information and second target pickup information). After the output device outputs the error message, the executing object can remotely control the target pickup mechanism of the pick-and-place device to re-perform the pickup operation for the first item based on the error message. The executing object can be a worker, who can remotely control the target pickup mechanism to pick up the first item based on the error message; alternatively, the executing object can also be an operating device, which is not limited in this embodiment.

[0309] For example, the number of times the execution object controls the target picking mechanism to pick up the first goods does not exceed the second number threshold, or the duration of the execution object controlling the target picking mechanism to pick up the first goods does not exceed the preset duration threshold.

[0310] For example, to improve the efficiency of a warehousing system, the number of times or the duration for which the executing object controls the target picking mechanism to pick up the first item can be limited. If the executing object successfully picks up the item within a second number threshold or a preset duration threshold, it indicates that the first item has been successfully picked. Based on the successful picking of the first item by the executing object, the preset feature information database can also be updated based on the picking information corresponding to this success.

[0311] In some embodiments, if the target picking mechanism fails to pick up the first item, the first item is marked as an abnormal picking object, and the next item to be picked is determined in the target container.

[0312] For example, if the target picking mechanism fails to pick up the first item within the second number of attempts or a preset time threshold, the first item can be marked as an abnormal picking object, and picking of the first item will no longer be performed. The control device can determine the next item to be picked in the target container, thereby controlling the picking and placing device to perform a picking operation on the next item to be picked.

[0313] In some examples, after all the items to be picked in the target container have been picked, if there is at least one abnormal item in the target container, the handling equipment can be controlled to move the target container to the manual operation area in the warehousing system so that the abnormal item in the target container can be manually picked by the operators in the manual operation area.

[0314] The goods picking method provided in this disclosure can handle picking anomalies when the target picking mechanism fails to pick up the first goods based on the first target picking information. First, the control device can control the target picking mechanism to attempt to pick up the first goods multiple times based on the first target picking information. If multiple attempts fail, the target picking mechanism adjusts the pose of the first goods, re-determines the second target picking information, and attempts to pick up the goods again. If picking still fails after pose adjustment, an anomaly message can be output to instruct the execution object to remotely operate the target picking mechanism to pick up the first goods. After the above anomaly handling, if the target picking mechanism successfully picks up the first goods, the preset feature information database can be updated based on the picking information corresponding to the successful pick-up; if the target picking mechanism still fails to pick up the first goods, the first goods picking anomaly is marked, and manual processing is performed subsequently.

[0315] Figure 8 is a schematic diagram of another goods picking method provided in an embodiment of this disclosure.

[0316] In some warehousing scenarios, after the picking and placing device (such as the target picking mechanism of the picking and placing device) picks up the first item from the target container, the first item can be identified and verified to ensure the accuracy of the picked first item. The verification of the first item and the cases of verification failure are explained below with reference to Figure 8.

[0317] As shown in Figure 8, the goods picking method provided in this embodiment further includes steps 810 to 860 as shown below. It should be noted that step 810 can be performed after step 340 in the above embodiment.

[0318] Step 810: If the target picking mechanism picks up the first cargo, at least one fourth image including the first cargo is acquired by the image acquisition device.

[0319] In some examples, each item may be labeled with an identification code, which can be a QR code, barcode, or similar format. Each item has a unique identification code that can uniquely identify each item.

[0320] For example, a target identification code may be affixed to the first item. When the target picking mechanism picks up the first item, it needs to remove it from the target container and transfer it to the target location. Before removing the first item and transferring it to the target location, an image acquisition device can acquire at least one fourth image including the first item. The image acquisition device can then send the acquired fourth image to a control device. The fourth image includes the first item.

[0321] In some examples, there may be multiple image acquisition devices (hereinafter referred to as second target acquisition devices) used to acquire at least one fourth image. For example, the second target acquisition device may be an image acquisition device positioned around the target container. The second target acquisition device may be a camera positioned at the front, rear, left, right, and top of the target container. The second target acquisition device can acquire images of the first cargo from multiple perspectives.

[0322] Step 820: Determine whether at least one fourth image includes a target identifier code.

[0323] In some examples, the control device can determine whether at least one fourth image includes a target identification code on the first cargo. The control device can search for the target identification code in each fourth image separately. For example, at least one fourth image may contain an image that includes the target identification code; or, none of the fourth images may contain the target identification code.

[0324] For example, if at least one fourth image includes the target identifier code, proceed to step 830; if at least one fourth image does not include the target identifier code, proceed to step 850.

[0325] Step 830: Identify the target identification code, and update the inventory information of the first item if the target identification code is used to indicate the first item.

[0326] In some examples, the target identification code uniquely identifies the first item. When the fourth image includes the target identification code, identification is performed based on the target identification code. The target identification code indicates that the item is the first item, suggesting that the pickup by the target picking mechanism was accurate, and therefore, the inventory information can be updated. For example, the status of the first item in the inventory information can be set to "picked successfully."

[0327] Step 840: If the target identification code cannot be recognized, or if the first item cannot be determined based on the target identification code, mark the first item as an abnormal picking object, put the first item back into the target container, and determine the next item to be picked in the target container.

[0328] In some examples, the fourth image includes a target identification code, but this code cannot identify the corresponding goods. For instance, if the target identification code is contaminated or damaged, it may be impossible to identify the indicated goods; or, the target identification code may identify the corresponding goods, but those goods are not the first goods. In these abnormal situations, it can be assumed that the first goods picked up by the target picking mechanism are incorrect, and therefore, the first goods can be marked as abnormal picking objects. Simultaneously, the target picking mechanism can return the picked first goods to the target container and re-identify the picking goods within the target container.

[0329] Step 850: Control the pick-and-place device to adjust the angle of the target picking mechanism.

[0330] In some examples, when the target identification code is not included in any of the fourth images, it indicates that the direction in which the target picking mechanism picks up the first item obscures the target identification code. Therefore, the angle of the picking mechanism can be adjusted by the picking and placing device. The present disclosure does not limit the adjustment angle or the adjustment direction of the target picking mechanism.

[0331] Step 860: Based on the adjusted target picking mechanism, at least one fifth image including the first cargo is re-acquired by the image acquisition device.

[0332] In some examples, after adjusting the angle of the target pickup mechanism, the position of the first cargo may change accordingly, and the target identification code may be exposed. Therefore, at least one fifth image including the first cargo can be re-acquired using an image acquisition device (such as the second target acquisition device described above). After acquiring the fifth image, the target identification code is re-determined and identified. This determination and identification process is similar to that of the fourth image described above, and will not be repeated here to avoid repetition.

[0333] For example, the number of times the angle of the target picking mechanism is adjusted does not exceed a second preset number. The second preset number can be set according to needs, and this embodiment does not limit it. That is to say, if the target identification code cannot be found after the number of angle adjustments of the target picking mechanism reaches the second preset number, it indicates that the identification code may not exist on the first item. Therefore, the first item can be marked as an abnormal picking object. At the same time, the target picking mechanism can put the picked first item back into the target container and re-determine the picking item in the target container.

[0334] In some examples, after all the items to be picked in the target container have been picked, the handling equipment can be controlled to move the target container to the manual operation area in the warehousing system, so that the operators in the manual operation area can manually pick the abnormal items in the target container.

[0335] The goods picking method provided in this disclosure, after picking up the first goods through the target picking mechanism of the pick-and-place device, can further determine the target identification code corresponding to the first goods by acquiring a fourth image including the first goods through an image acquisition device. If the target identification code indicates the first goods, the method determines that the first goods have been picked correctly. If the first goods cannot be identified through the target identification code, the method determines that the picking of the first goods has failed. Therefore, this disclosure embodiment can verify the goods picked up by the pick-and-place device (such as the picking mechanism of the pick-and-place device) to ensure the accuracy of the picked goods.

[0336] In some warehousing scenarios, double-picking anomalies may occur during the process of a pick-and-place device (such as the target picking mechanism of the pick-and-place device) picking up the first item from a target container. For example, when the target picking mechanism picks up the first item, it may also pick up other items in the target container besides the first item, thus causing a picking anomaly.

[0337] The following explanation, in conjunction with Figures 9 and 10, describes the handling process for a double-picking anomaly in the pick-and-place device.

[0338] Figure 9 is a schematic diagram of another goods picking method provided by an embodiment of the present disclosure. As shown in Figure 9, the goods picking method provided by this embodiment of the present disclosure further includes steps 910 to 930 as shown below. It should be noted that step 910 can be performed after step 340 in the above embodiment.

[0339] Step 910: After the target picking mechanism picks up the first cargo from the target container, at least one sixth image including the first cargo is acquired by the image acquisition device.

[0340] In some examples, the number of image acquisition devices (hereinafter referred to as third target acquisition devices) acquiring at least one sixth image can be one or more. For example, at least one sixth image can be an image acquired by one image acquisition device of the picking process before the target picking mechanism picks up the first goods from the target container and transfers them to the target location; or, at least one sixth image can also be an image of the picking process acquired by multiple image acquisition devices from different perspectives, which is not limited in this embodiment. The image acquisition device can send the acquired at least one sixth image to the control device.

[0341] Step 920: Based on at least one sixth image, if it is determined that the target picking mechanism has picked up multiple candidate objects, it is determined that a double picking anomaly has occurred in the target picking mechanism.

[0342] In some examples, the control device determines the number of candidate objects picked up by the target picking mechanism in each of the six images based on at least one sixth image. If the number of candidate objects picked up by the target picking mechanism is multiple (i.e., multiple candidate objects), it can be determined that a double-picking anomaly has occurred. Here, the multiple candidate objects include a first item and at least one other item besides the first item.

[0343] Step 930: If a double-picking anomaly is determined to have occurred at the target picking mechanism, the target picking mechanism is controlled to return the picked-up candidate objects to the target container, and the target picking mechanism is controlled to pick up the first goods again based on the first target picking information.

[0344] In some examples, if a double-picking anomaly is determined to have occurred at the target picking mechanism, the mechanism is controlled to return the multiple candidate objects picked up to the target container and then pick up the first item again based on the first target picking information. This first target picking information refers to the picking information corresponding to the first item when the previous picking was successful.

[0345] It should be noted that after the target picking mechanism picks up the first item again, images of its picking process can still be captured by the image acquisition device to ensure that no double picking anomalies occur during the picking process.

[0346] Figure 10 is a schematic diagram of another goods picking method provided by an embodiment of the present disclosure. As shown in Figure 10, the goods picking method provided by this embodiment of the present disclosure further includes steps 1010 to 1030 as shown below. It should be noted that step 1010 can be performed after step 340 in the above embodiment.

[0347] Step 1010: After the target picking mechanism transfers the picked-up first goods to the first order container, the weight information of the first order container is determined by the weight collection device corresponding to the first order container.

[0348] In some examples, the target location includes the first order container corresponding to the first item. After the target picking mechanism transfers the picked first item from the target container to the first order container, the weight of the first order container can also be used to determine whether a double-picking anomaly occurred during the picking process of the first item.

[0349] For example, the first order container can be placed on the picking station in the picking area, and a weight acquisition device, such as a weight sensor, can be installed below the first order container. When the target picking mechanism places the first item into the first order container, the weight of the first order container will change, and the weight information of the first order container collected by the weight acquisition device will also change. The weight acquisition device will then send the collected weight information to the control device.

[0350] For example, the weight acquisition device can dynamically send the weight information of the first order container it detects to the control device. For instance, the weight acquisition device can send the weight information of the first order container to the control device at preset time intervals.

[0351] Step 1020: Based on the weight information of the first order container and the weight information of the first goods, determine whether there is a double-picking anomaly in the target picking mechanism.

[0352] In some examples, the control device can store attribute information of the first item, such as weight information. The control device first determines a predicted weight difference based on a first weight and a second weight of the first order container sent by the weight acquisition device, where the first weight is the weight of the first order container before the first item is placed, and the second weight is the weight of the first order container after the first item is placed. Secondly, the control device determines whether there is a double-pickup anomaly in the target picking mechanism based on the weight error between the predicted weight difference and the weight of the first item.

[0353] For example, if the weight error between the predicted weight difference and the weight of the first item exceeds a preset error threshold, it indicates that other items besides the first item were placed in the first order container, thus determining that the target picking mechanism has a double-picking anomaly. If the weight error between the predicted weight difference and the weight of the first item does not exceed the preset error threshold, it indicates that no other items besides the first item were placed in the first order container, thus determining that the target picking mechanism does not have a double-picking anomaly.

[0354] Step 1030: If a double-picking anomaly is determined to have occurred at the target picking mechanism, the first order container is marked as an abnormal order container, and the output device is controlled to output an anomaly prompt message to instruct the execution object to handle the abnormal order container.

[0355] In some examples, if a double-picking exception is determined to have occurred, the control device can mark the first order container as an abnormal order container. At the same time, the control output device outputs an exception message, which is used to instruct the execution object to perform exception handling for the abnormal order container.

[0356] For example, the execution object can be a staff member, who can retrieve the abnormal object from the first order container based on the abnormal message.

[0357] The goods picking method provided in this disclosure verifies whether a double picking anomaly occurs at the target picking mechanism during the process of the target picking mechanism transferring the first goods from the target container to the first order container, thereby ensuring the accuracy of the picking process of the picking device and improving the picking efficiency of the picking device and the working efficiency of the warehousing system.

[0358] In some embodiments, the warehousing system further includes a second workstation, and the method further includes: in the event that the first target pickup information of the first goods cannot be determined, controlling a handling device to transport the target container to the second workstation so that the first goods can be picked from the target container by a picker at the second workstation.

[0359] In some examples, if the first target pickup information for the first item cannot be determined after performing steps 3341 to 3343 (position adjustment) and steps 3351 to 3353 (manual assistance), the first item can be picked manually. It should be noted that the target container may include at least one abnormal object remaining in the target container after the picking and placing device of the first workstation performs the picking operation.

[0360] For example, the warehousing system also includes a second workstation, which can also be called a manual workstation, where the picking objects are picked by picking personnel. That is, the second workstation is where picking personnel perform picking operations on the goods in the containers.

[0361] For example, when the first target picking information of the first item cannot be determined, the pick-and-place device (i.e., the target picking mechanism) cannot accurately and safely pick out the first item from the target container. In other words, the pick-and-place device may find abnormalities such as dropping during the transfer of the first item. In order to avoid the occurrence of such abnormalities, the handling equipment can be controlled to move the target container (or target carrier) where the first item is located to the manual workstation, so that the picking personnel at the manual workstation can perform the picking operation on the first item.

[0362] The goods picking method provided in this disclosure acquires a first image corresponding to a target container using an image acquisition device, and determines the first goods and their type based on the first image. If the first goods are of a first goods type (e.g., goods that can be picked by a pick-and-place device), the picking position and picking direction of the first goods are further determined, so that the pick-and-place device can accurately pick the first goods from the target container based on the picking position and picking direction. The picking method provided in this disclosure enables automatic picking through the pick-and-place device, improving picking efficiency compared to manual picking. Secondly, since this disclosure only picks goods of the first goods type using the pick-and-place device, it reduces the occurrence of abnormal situations such as goods falling due to picking failures, improving the safety of the picking process. Simultaneously, this disclosure can determine the picking position and picking direction of the goods, making it easier for the pick-and-place device to pick up the goods more accurately, improving picking accuracy. Therefore, the goods picking method provided in this disclosure can improve the picking efficiency of the warehousing system and improve the accuracy and safety of the picking process.

[0363] The above embodiments disclose the picking process when the first goods are of a first type of goods. The picking process when the first goods are of a second type of goods will be described below with reference to FIG11.

[0364] Figure 11 is a schematic diagram of another goods picking method provided by an embodiment of this disclosure. As shown in Figure 11, after step 320, the method further includes steps 350 to 360 as shown below.

[0365] Step 350: If the type of the first goods is the second goods type, control the pick-and-place device to pick at least one second goods from the goods to be picked in the target container.

[0366] In some examples, if the type of the first item is determined to be the second item type based on the first item and a preset item information database, the picking operation may not be performed on the first item.

[0367] For example, if a second item exists in addition to the first item in the target container, the picking operation can continue to be performed on the second item in the target container. Here, the second item is any item of the type of the first item in the target container that is to be picked; for example, the second item can be the next item to be picked after the first item.

[0368] For example, the type of the second item can be determined based on the second item and a preset item information database. If the type of the second item is determined to be the second item type, a picking operation can be performed on the second item. In other words, the picking information corresponding to the second item can be determined based on the image information of the second item and the second item.

[0369] In some examples, when the type of the first item is the type of the second item, it indicates that the first item cannot (or is not easy) be picked up by the pick-and-place device (such as at least one picking mechanism in the item pick-and-place mechanism). Therefore, the picking operation can be performed on the first item instead of the first item, and the picking operation can be performed on the first item type (such as the second item) in the target container first, thereby improving picking efficiency.

[0370] For example, the picking mechanism on the pick-and-place device for picking the second item may be the same as or different from the picking mechanism for picking the first item, and this embodiment does not limit this. It should be noted that the process of the pick-and-place device picking the second item in the target container is similar to the process of the pick-and-place device picking the first item in the target container in the above implementation (such as steps 320 to 340), and will not be described again here to avoid repetition.

[0371] Step 360: After the pick-and-place device has picked the first type of goods to be picked from the target container, the control handling equipment is used to move the target container to the second workstation so that the picking personnel at the second workstation can pick out the first goods from the target container.

[0372] In some examples, after the pick-and-place device completes picking the second type of goods, it determines whether there are still goods of the second type in the target container. If there are still goods of the second type in the target container, the pick-and-place device continues to perform picking operations on other goods of the second type in the target container. After picking all goods of the second type in the target container, the control handling equipment moves the target container to the second workstation.

[0373] For example, if the first type of goods in the target container has been picked at the first workstation, and there are no goods (such as the second type of goods) that can be picked by the pick-and-place device in the target container, then the target container (or the carrier containing the target container) can be moved to the second workstation, so that the picking personnel at the second workstation can perform the picking operation on the second type of goods in the target container. It should be noted that the above picking process can also be called relay picking.

[0374] In some embodiments, prior to step 310, the method further includes: determining whether there are second-type goods among the at least one goods to be picked based on at least one goods to be picked corresponding to the pending order and a preset goods information database; if there are second-type goods among the at least one goods to be picked, controlling the handling equipment to transport the target container containing the second-type goods to the second workstation, so that the picking personnel at the second workstation can pick out the second-type goods from the target container.

[0375] In some examples, before dispatching orders to each workstation, the goods required for the pending orders can be predetermined. If a second type of goods is among the goods required for an pending order, the container containing the second type of goods can be moved to the second workstation for manual picking. This avoids having to move the goods back to the first workstation if they cannot be picked, thus improving picking efficiency. It should be noted that the above picking process can also be called a pre-dispatch picking method.

[0376] In other words, after the pick-and-place device at the first workstation completes the picking of the target container in this embodiment of the present disclosure, there may still be goods in the target container that the pick-and-place device cannot pick. These unpicked goods may include goods that the pick-and-place device failed to pick, or they may include goods of a second type. Therefore, when the pick-and-place device completes the picking of the target container and determines that there are goods to be picked in the target container, the control device schedules the handling equipment to move the target container to the second workstation, so that the picking operation can continue to be performed on the remaining goods to be picked in the target container through the second workstation, thereby fulfilling the requirements of the order to be picked.

[0377] The goods picking method provided in this disclosure can transport goods of a second type (such as second goods) that cannot be picked directly by the goods picking and placing mechanism to a manual picking workstation through relay picking and pre-ordering. This allows for manual picking to prevent abnormal situations such as picking failure or goods falling during the picking process, thereby improving the safety of picking and further increasing picking efficiency.

[0378] In some examples, the control device may fail to identify the first item. For instance, the inability of the control device to identify the first item from at least one item to be picked may include the absence of a feature information in the feature information of each item to be picked that matches the feature information of the first item in the preset feature information database. In this case, it is necessary to re-identify the first item in the target container.

[0379] In some embodiments, if there is no feature information in the feature information of each item to be picked that matches the feature information corresponding to the first item in the preset feature information database, the first item is determined among at least one item to be picked based on the first image and the first recognition model; or, the picking and placing device is controlled to adjust the position of at least one item to be picked in the target container, and the image acquisition device is controlled to acquire the adjusted image corresponding to the target container, and the first item is determined among at least one item to be picked based on the adjusted image corresponding to the target container; or, the first image is sent to the interactive device, the item marking information corresponding to the first item input by the user on the interactive device is obtained, and the first item is determined among at least one item to be picked based on the item marking information.

[0380] For example, if the first cargo cannot be directly identified through a preset feature information database, the first cargo can be re-identified in the target container through at least one of the following methods: a first identification model, position adjustment, and manual assistance.

[0381] In some examples, the first recognition model can be a pre-trained model that can acquire richer feature information corresponding to the goods (e.g., the first recognition model can recognize more feature information than the feature information corresponding to each goods in a preset feature information library). For example, the first recognition model can determine the first goods from at least one goods to be picked in the target container by extracting feature information of each goods in the first image and performing feature matching based on the input first image.

[0382] In other examples, the position of the goods in the target container can be adjusted, and the image acquisition device can be controlled to re-acquire the adjusted image (such as the second image in the above embodiment). Based on the adjusted image corresponding to the target container, the feature information of the first goods can be re-determined, thereby determining the first goods based on the re-determined feature information and the preset feature information database.

[0383] In some other examples, manual assistance can also be used to re-identify the first item. The control device can send the first image to the interactive device, which displays the first image, and the user can determine the first item based on the first image. For example, the user can mark the first item in the first image and send the item marking information to the control device, which then determines the first item based on the item marking information in the first image.

[0384] It should be noted that the three methods for re-determining the first cargo described above can be executed separately or sequentially. For example, when the first cargo cannot be directly determined using the preset feature information database, it can first be identified using the first recognition model. If the first recognition model fails to identify it, it can then be identified using a position adjustment method. If the position adjustment method also fails to identify it, it can then be identified using manual assistance. Alternatively, other execution orders can be used for identification, and this embodiment of the disclosure does not limit this.

[0385] In some embodiments, the above-described goods picking method further includes: if the first goods cannot be identified among at least one goods to be picked, controlling the handling equipment to transport the target container or target vehicle to a second workstation, so that the picking personnel at the second workstation can pick the first goods from the target container.

[0386] In some examples, if the first item in the target container cannot be identified by the first identification model, position adjustment and manual assistance described above, the handling equipment can be controlled to move the target container (or target vehicle) to a second workstation (i.e., a manual workstation) so that the picking personnel at the second workstation can pick out the first item from the target container.

[0387] For example, the preset feature information database can be dynamically updated to facilitate the identification of subsequent goods to be picked. In other words, the feature information corresponding to the goods in the preset feature information database can be optimized or supplemented to more quickly identify the first goods.

[0388] In some examples, after the first cargo in the target container is identified through the first identification model, position adjustment and manual assistance, the feature information of the first cargo can be stored in a preset feature information database and matched with the cargo information of the first cargo.

[0389] It should be noted that the above embodiments illustrate the identification of the first cargo by taking the example that the target container contains multiple different cargoes. If the target container contains only the same type of cargo, there is no need to identify the cargo. Any cargo in the target container can be taken as the first cargo, and the first target pickup information of the first cargo can be determined.

[0390] In some examples, a pre-defined feature information database can be generated or updated during the goods picking process. For instance, when the target container is a box-and-one-item container, meaning the target container contains the same type of goods, the feature information of the goods can be determined based on the image corresponding to the target container, and the feature information can be stored in the pre-defined feature information database for updating the database.

[0391] For example, after the first cargo in the target container is identified through the first identification model, position adjustment and manual assistance, the first cargo can be confirmed a second time to improve the accuracy of cargo identification. If the second confirmation is correct, the feature information of the first cargo is stored in a preset feature information database to update the preset feature information database, thereby ensuring the accuracy of the preset feature information database.

[0392] In some examples, the first workstation is equipped with a scanning device, and each item has a scanning tag. After the pick-and-place device selects the first item, it can be moved to the scanning device for scanning. The scanning device determines the item's information by scanning the tag on the item. For example, after scanning the item selected by the pick-and-place device, if the scanning device determines that the item is the first item, it delivers the item to the first order container and stores the item's feature information in a preset feature information database for future updates. If the item is determined not to be the first item, it is returned to the target container, and the first item is re-identified within the target container.

[0393] For example, the feature information of the first cargo can be stored as training data for the first recognition model, thereby training and optimizing the first recognition model and improving its recognition efficiency and accuracy.

[0394] In some embodiments, where the target location includes a first order container, the above-described goods picking method further includes: after controlling the pick-and-place device to pick up the first goods from the target container, acquiring a picking image; determining the drop position of the first goods based on the picking image; if the drop position of the first goods is located in the first order container, determining that the first goods have been picked; if the drop position of the first goods is located in the target container, controlling the pick-and-place device to pick up the first goods again from the target container and transferring the first goods to the first order container; if the drop position of the first goods is located in a candidate container, determining that the picking of the first goods has failed, controlling the pick-and-place device to pick up the first goods from the candidate container and transferring the first goods to the first order container; wherein, the candidate container is any container on the picking path of the first goods.

[0395] For example, during the process of controlling the pick-and-place device to pick up a first item from a target container and transfer the first item to a target location, the picking status of the first item by the pick-and-place device is determined. The picking status of the item includes successful picking and failed picking. If the item pick-and-place device removes the item from the target container and transfers it to the target location, the picking status of the first item is indicated as successful picking. If the item pick-and-place device fails to remove the item from the target container or fails to successfully transfer the item to the target location, the picking status of the first item is indicated as failed picking.

[0396] It should be noted that the goods picking and placing mechanism may include at least one picking mechanism (i.e., target picking mechanism) in the picking and placing device for performing picking operations on the first goods. The following embodiment is illustrated by taking the goods picking and placing mechanism performing picking operations on the first goods as an example.

[0397] For example, if the goods handling mechanism fails to retrieve the first item from the target container, the picking of the first item is considered a picking failure; or, if the goods handling mechanism retrieves the first item from the target container, but the first item falls during the transfer of the first item to the first order container, resulting in the first item not being delivered to the first order container, the picking of the first item is considered a picking failure.

[0398] For example, the control device can control the image acquisition device to acquire images of the picking process corresponding to the picking of the first item by the goods picking and placing mechanism. For instance, the image acquisition device can acquire multiple images of the goods picking and placing mechanism picking the first item in real time, or it can acquire multiple images (i.e., multiple picking process images) of the goods picking and placing mechanism picking the first item at preset time intervals and send the acquired multiple picking process images to the control device. The control device can determine the picking status of the first item by the goods picking and placing mechanism based on the multiple picking process images corresponding to the first item.

[0399] In some examples, the control device can identify the transfer direction and position of the first item in multiple picking process images based on multiple picking process images, thereby determining whether the first item has been successfully transferred to the target location (such as the first order container), and thus determining the picking status of the first item.

[0400] For example, a weight sensor may be installed on the goods picking and placing mechanism, or the picking status of the first goods may be determined based on the weight information of the goods picking and placing mechanism during the picking process, such as the weight change of the goods picking and placing mechanism during the picking process. It should be noted that the method for determining the picking status of the first goods is not limited in the embodiments of this disclosure.

[0401] The following explains the situation where the picking device fails to pick the first item.

[0402] For example, after the first item is removed from the picking and placing device (such as an item picking and placing mechanism), a picking image can be acquired, and the drop position (i.e., current position) of the first item can be determined based on the picking image. The picking image includes an image corresponding to the target container, an image corresponding to the target position (i.e., the first order container), and images along the picking path of the first item. The picking path of the first item may include multiple containers.

[0403] In some examples, after the first item leaves the pick-and-place device, an image corresponding to the first order container can be acquired first. Based on this image, it can be determined whether the first item has fallen into the first order container. If it is determined that the first item has fallen into the first order container, the picking of the first item is considered complete, and the picking operation for other items continues. It should be noted that if the first item falls into the first order container not under the control of the pick-and-place device, but because the device failed to catch it, the picking operation can also be considered a failure. In this case, it is not necessary to re-pick the first item.

[0404] In other examples, if it is determined that the first item has not fallen into the first order container, an image of the target container is acquired, and based on the image of the target container, it is further determined whether the first item has fallen into the target container. If it is determined that the first item has fallen into the target container, the pick-and-place device is controlled to pick the first item out of the target container again and transfer the first item to the first order container.

[0405] In other examples, if the first item does not fall into the target container, images of multiple containers along the picking path of the first item can be acquired, and based on the images of each container, it can be determined whether the first item has fallen into one of these containers (such as a candidate container). If it is determined that the first item has fallen into the candidate container, the pick-and-place device is controlled to pick the first item from the candidate container and transfer the first item to the first order container.

[0406] In some other examples, if the first item is not found in any of the containers on the picking path of the first order container, the target container, and the first item, the images corresponding to the first order container, the target container, and each container on the picking path of the first item can be sent to the interactive device for secondary identification through manual marking.

[0407] For example, after receiving each image, the interactive device displays them, allowing the user to determine the drop location of the first item. If the drop location is determined to be within the first order container, the target container, or any container along the picking path of the first item, the location of the first item is marked, and the marking information is sent to the control device. The control device then controls the pick-and-place device to retrieve the first item from its drop location and transfer it to the first order container based on the marking information. It should be noted that when the drop location of the first item is determined to be within the first order container, the pick-and-place device does not need to re-pick the item.

[0408] In some examples, if the location of the first item's fall is still not found after the aforementioned secondary identification, the control device acquires images within a preset range. This preset range can be the acquisition range of the image acquisition device itself, or a range set according to requirements. The control device can further determine the location of the first item's fall based on the images within the preset range.

[0409] For example, if the first item falls within the operating range of the pick-and-place device and the fall location does not affect the normal operation of the first workstation, the pick-and-place device can be controlled to pick up the first item and transfer it to the first order container based on the image captured by the image acquisition device within a preset range. If the first item falls outside the operating range of the pick-and-place device, or if the fall location may affect the normal operation of the first workstation, the interactive device can be controlled to output a prompt message to indicate the fall location of the first item and instruct the operator to handle the abnormality.

[0410] For example, if the picking status of the first item by the picking device is "picking failure", the first item is added to the preset item information database to update the preset item information database.

[0411] For example, if the picking status of the first item is "picking failed," it indicates that the first item may not be suitable for picking via a pick-and-place device. Therefore, the item can be added to a preset item information database, that is, the type of the first item can be marked as the second item type. Thus, in the next picking, the picking operation can be performed on it according to the picking method of the second item type (i.e., manual picking).

[0412] It should be noted that if the picking status of the first item is "picking failure", the picking and placing mechanism in the picking and placing device can be controlled to pick the first item multiple times. If the picking status is "picking failure" multiple times, the type of the first item can be determined as the second item type and added to the preset item information database.

[0413] The goods picking method provided in this disclosure acquires picking images of the picking process of a first item by a goods picking and placing mechanism, and determines the picking status of the first item based on the picking images. If the picking status of the first item is "picking failure," anomaly processing can be performed based on the picking images, and the first item can be picked again. Even if the picking status of the first item is "picking failure," the type of the first item can be determined to be a second item type. Therefore, in subsequent picking processes, the first item can be picked according to the picking method for the second item type (i.e., manual picking), thereby avoiding picking failures and improving picking safety and efficiency.

[0414] In some examples, during the process of the cargo handling mechanism transferring the first cargo from the target container to the first order container at the first docking position, picking failures occur, such as the first cargo falling into the gap between the first workstation and the seeding wall due to a loose grip (or suction). This requires operators to handle the abnormality, increases labor costs, and reduces picking efficiency.

[0415] To address the aforementioned issues, the container handling mechanism provided in this embodiment is equipped with a guide component that can automatically extend and retract. When the guide component is extended, it can bridge the gap between the container handling mechanism and the order container to receive goods dropped by the goods handling mechanism.

[0416] In some embodiments, step 340 may include: when the container pick-and-place mechanism moves to the first docking position, controlling the guide component to bridge between the container pick-and-place mechanism and the first order container; controlling the goods pick-and-place mechanism to pick up the first goods from the target container and transfer the first goods to the first order container; or controlling the goods pick-and-place mechanism to place the first goods on the guide component so as to transfer the first goods to the first order container through the guide component.

[0417] Figure 12A is a schematic diagram of another warehousing system provided in an embodiment of the present disclosure, and Figure 12B is a schematic diagram of yet another warehousing system provided in an embodiment of the present disclosure.

[0418] For example, the picking scenario for the first item shown in Figures 12A and 12B can be a scenario where a picking and placing device is set at the first workstation, and the picking and placing device includes a container picking and placing mechanism and a goods picking and placing mechanism. In the picking scenario of Figure 12A (or Figure 12B), the handling equipment can transport the target carrier to the first workstation. The container picking and placing mechanism 331 in the picking and placing device takes the target container from the target carrier and transfers the target container to the first docking position corresponding to the first order container. The goods picking and placing mechanism 332 in the picking and placing device picks the first item from the target container transported by the container picking and placing mechanism and transfers the first item to the first order container on the seeding wall. It should be noted that the following embodiments can all be described using the picking scenario of Figure 12A (or Figure 12B) as an example.

[0419] As shown in Figure 12A, when the container pick-and-place mechanism 331 moves the target container to the first docking position, the guide member 71 can be connected across the container pick-and-place mechanism 331 and the first order container. For example, the guide member 71 can extend from the container pick-and-place mechanism 331 in the direction toward the first order container and connect with the docking position of the first order container (the edge of the first order container).

[0420] In some examples, the guide component can be connected to the container loading and unloading mechanism via a guide rail (such as a linear guide), ball screw, rack and pinion, or other moving assembly. During extension, the guide component can move smoothly and accurately to the bridging position of the first order container.

[0421] In some examples, the container loading and unloading mechanism is also equipped with a guide drive component that can drive the extension and retraction of the guide plate.

[0422] In some examples, after the guide member is connected between the container pick-and-place mechanism and the first order container, the goods pick-and-place mechanism can place the picked first goods at the first end of the guide member to transfer the first goods into the first order container via the guide member. For example, the guide member can be configured with a smooth surface to facilitate the transport of the first goods.

[0423] For example, as shown in Figure 12A, the guide component 71 can be tilted across the container handling mechanism 331 and the first order container. One end connecting to the container handling mechanism 331 is the first end A of the guide component, and the other end connecting to the first order container is the second end B of the guide component 71. The height of the first end A is lower than the height of the second end B. After the goods handling mechanism 332 removes the first goods from the target container, the first goods can be placed at the first end A of the guide component 71, and then transported to the first order container via the second end B of the guide component 71.

[0424] In other examples, after a guide component is installed between the container pick-and-place mechanism and the first order container, the goods pick-and-place machine can transfer the picked first goods into the first order container. In this case, the guide component can act as a fall prevention mechanism, catching goods dropped by the goods pick-and-place mechanism.

[0425] In some embodiments, the cargo handling mechanism is controlled to pick up a first cargo from a target container. During the process of transferring the first cargo to a first order container, if the first cargo falls onto a guide component, the cargo handling mechanism is controlled to pick up the first cargo from the guide component and deliver the first cargo into the first order container.

[0426] In some examples, goods may fall during the transfer of the first goods by the goods handling mechanism, and the first goods may fall onto the guide component.

[0427] For example, sensors on the goods handling mechanism can acquire information about the picking process. If the first item is detected to have fallen, the control device can send a drop information message, and the control device can control the goods handling mechanism to move to the guide component to retrieve (grab or suck up) the first item from the guide component and transfer the first item back to the first order container.

[0428] In some embodiments, the control cargo handling mechanism picks up a first cargo from a target container, and in the process of transferring the first cargo to a first order container, if the first cargo falls onto a guide member, the guide member transfers the first cargo into the first order container.

[0429] In some examples, where the first item can fall onto the guide component, the guide component can also directly convey the first item to the first order container. As shown in Figure 7, when the first item falls onto the guide component 71, the guide component 71 can convey the first item to the first order container through the second end B.

[0430] In some embodiments, the control guide component is retracted after the first goods are transferred to the first order container.

[0431] In some examples, the control device can retract the guide component based on sensor detection information on the goods handling mechanism, once it is determined that the first item has been delivered into the first order container. For example, after the guide component retracts, the container handling mechanism can be controlled to continue moving without affecting the movement of the container handling mechanism.

[0432] The goods picking method provided in this embodiment can prevent goods from falling during the picking operation of the first goods by setting a guide component on the container picking mechanism, thereby improving the safety of the picking process. At the same time, it can also deliver the goods to the location when they fall, avoiding manual operation and improving picking efficiency.

[0433] In some examples, the spacing (e.g., vertical spacing) between the seed positions (also called seed grids) on the seed wall used to place order containers is typically set relatively large to facilitate the delivery of picked goods to the corresponding order containers by the goods handling mechanism. With larger spacing between seed positions, the number of seed positions on the seed wall decreases, thus reducing the number of order containers that can be placed. This results in fewer orders that the first workstation can handle, leading to lower picking efficiency in the warehouse system when the number of orders is high. To address this, the number of seed positions on the seed wall can be increased. Increasing the number of seed positions allows the first workstation to handle a larger number of orders. However, this reduces the spacing between seed positions, resulting in smaller spacing (e.g., vertical spacing) between order containers. In this case, the goods handling mechanism may not be able to directly deliver picked goods to the order containers.

[0434] To address the aforementioned issues, in the goods picking method provided in this embodiment, the container picking mechanism can pull out the order container by pushing and pulling, thereby facilitating the goods picking mechanism to deliver the picked goods into the order container; after the goods delivery is completed, the order container is pushed back to its original position.

[0435] In some embodiments, when the container pick-and-place mechanism moves to the first docking position, the container pick-and-place mechanism is controlled to pull the first order container out from the first seeding position so that the goods pick-and-place mechanism can deliver the first goods into the first order container; the goods pick-and-place mechanism is controlled to pick the first goods from the target container and deliver the first goods into the first order container; the container pick-and-place mechanism is controlled to push the first order container back to the first seeding position.

[0436] In some examples, the preset quantity threshold is the upper limit of the number of planting positions on the planting wall. If the number of planting positions exceeds this upper limit, the spacing between the placed order containers will be too small due to the excessive number of planting positions, preventing the goods handling mechanism from directly delivering goods into the order containers. If the number of planting positions does not exceed the upper limit, the spacing between the order containers is sufficient for the goods handling mechanism to directly deliver goods into the order containers. Therefore, the corresponding planting method can be determined based on the relationship between the number of planting positions on the planting wall and the preset quantity threshold.

[0437] For example, when the container pick-and-place mechanism moves to the first docking position, it can pull the first order container out from the first seeding position where the first order container is located. After pulling out the first order container, the control of the goods pick-and-place mechanism picks out the first goods from the target container and delivers the first goods into the first order container.

[0438] For example, as shown in Figure 12B, the container picking and placing mechanism 331 can pull the first order container on the seeding wall a preset distance along the direction toward the first workstation. For example, the container picking and placing mechanism 331 can pull the first order container out along direction a. After the first order container has been pulled out a preset distance, and there are no obstacles on the first order container, the goods picking and placing mechanism 332 can deliver the first goods into the first order container.

[0439] In some examples, when the container handling mechanism pulls out the first order container, the pulling distance can be determined based on the size and shape of the delivered goods (i.e., the first goods). The preset distance for each order container can be the same or different. It should be noted that when pulling out the first order container, the pulling distance must ensure that the first order container can be placed stably and will not tip over.

[0440] For example, after the goods handling mechanism delivers the first goods to the first order container, the container handling mechanism can be controlled to push the first order container back to the first seeding position. That is, the first order container can be pushed back from a preset position to its original position. As shown in Figure 12B, the container handling mechanism 331 can push the first order container back to the first seeding position in the opposite direction to direction a.

[0441] The goods picking method provided in this embodiment allows for the removal of a first order container by pushing and pulling when there are many picking positions on the picking wall. This enables the goods picking and placing mechanism of the picking and placing device to deliver the picked first goods into the first order container. After delivery, the first order container is pushed back to its original position. Therefore, this embodiment can increase the number of orders that can be processed at the first workstation by increasing the number of picking positions, thereby further improving the picking efficiency of the warehousing system.

[0442] Figure 13 is a schematic diagram of another goods picking method provided by an embodiment of the present disclosure. As shown in Figure 13, after step 340 above, the goods picking method further includes steps 1310 to 1340 as shown below.

[0443] Step 1310: Determine if there are other goods required for pending orders in the target container.

[0444] If there are other goods required for pending orders in the target container, proceed to step 1320; if there are no other goods required for pending orders in the target container, proceed to step 1340.

[0445] In some examples, a target container can hold multiple goods, which can be matched by different pending orders. In other words, goods required by multiple order containers can be placed in the same container.

[0446] Step 1320: Control the container picking and placing mechanism to move the target container from the first docking position to the second docking position corresponding to the second order container on the seeding wall.

[0447] In some examples, if the target container also contains third goods required for other pending orders, the control container handling mechanism moves the target container from the first docking position corresponding to the first order container to the second docking position corresponding to the second order container. The second order container is different from the first order container.

[0448] For example, the second docking position can be adjacent to or level with the second order container on the seeding wall. When the container picking and placing mechanism moves the target container to the second docking position, the cargo picking and placing mechanism can deliver the picked second cargo into the second order container.

[0449] In some examples, in related technologies, picking operations are performed by picking robots. The target container also contains a second item required for a second order container. The picking robot needs to return the first order container to its original position and move to the location of the second order container to retrieve it. In this process, the picking robot needs to first return the first order container, then move to the location of the second order container, and use its forks to move up and down to retrieve the second order container. This process is time-consuming and has low picking efficiency. However, in the embodiments of this disclosure, the container handling mechanism only needs to move horizontally and / or vertically to transport the target container from the location of the first order container to the location of the second order container, simplifying the picking operation and improving picking efficiency.

[0450] Step 1330: During the process of the container picking and placing mechanism moving to the second docking position, the image acquisition device is controlled to acquire the seventh image corresponding to the target container.

[0451] For example, when the type of the third goods is the same as the type of the first goods, the first target pickup information of the third goods is determined based on the seventh image. When the container pick-up and drop mechanism moves to the second docking position, the goods pick-up and drop mechanism is controlled to pick out the third goods from the target container and transfer the third goods to the second order container based on the first target pickup information.

[0452] It should be noted that step 1330 is similar to step 320 in the above embodiment, that is, the picking process of the goods picking and placing mechanism for the first stored goods is similar to the picking process of the first goods. To avoid repetition, it will not be described again here.

[0453] Step 1340: Control the container handling mechanism to move the target container from the first docking position to the idle cargo space of the target vehicle or the idle buffer space of the first workstation.

[0454] For example, if the target container does not contain the goods required by other pending orders among multiple pending orders, the container handling mechanism can move the target container back to the target vehicle.

[0455] In some examples, after the target container has been picked, the target carrier is located in the picking area of ​​the first workstation. The container handling mechanism can then move the target container from the first docking position to an available storage location on the target carrier. For example, this available storage location can be the original storage location of the target container or any other available storage location.

[0456] In other examples, after the target container has been picked, the target carrier moves the picking area of ​​the first workstation, and the container handling mechanism can move the target container from the first docking position to an idle buffer on the first workstation. For example, the first workstation may have multiple buffer slots, each of which can hold one container. When no container is placed on a buffer slot, that buffer slot is an idle buffer slot.

[0457] In some examples, after the next vehicle arrives at the picking area of ​​the first workstation, the container pick-and-place mechanism can place the target container on an empty cache location on the vehicle.

[0458] The goods picking method provided in this disclosure can, after the first goods have been picked, continue to determine whether there are other goods required by other order containers in the target container. If there are other goods required by other order containers in the target container, the target container is moved to the docking position corresponding to the next order container for picking its required goods; if there are no other goods required by other order containers in the target container, the target container is moved back to the target vehicle or the idle buffer space of the first workstation, thereby further improving the picking efficiency of the warehousing system.

[0459] In some examples, when a full order container is present on the seeding wall, the operator needs to perform an outbound operation, that is, remove the full order container from the seeding wall and replace it with an empty order container. This outbound process of full order containers and the replenishment of empty order containers can be called the box swapping process. When there are many order containers, manually performing the box swapping operation is costly and inefficient. Furthermore, when using a picking robot to perform full-box outbound operations or replenish empty boxes, the picking robot needs to carry each order container individually, resulting in low handling efficiency.

[0460] In some embodiments, the goods picking method further includes: obtaining the sowing status of each order container among multiple order containers on the sowing wall; if there is a third order container on the sowing wall with a sowing status of "sowing completed", controlling the container picking and placing mechanism to move to the third docking position corresponding to the third order container, and taking out the third order container at the third docking position, and transporting the third order container to an idle storage location of the first vehicle parked at the first workstation or an idle buffer location of the first workstation. The third docking position may be adjacent to or level with the third order container on the sowing wall.

[0461] For example, the third order container can be any order container on the seeding wall. The third order container can be the same as or different from the first or second order container.

[0462] For example, the seeding status of an order container includes "seeding completed" and "seeding incomplete." "Seeding completed" indicates that the order container has contained all the required goods; in this case, an outbound operation needs to be performed on the order container, i.e., the order container is moved away from the seeding wall. If the order container is not seeded, it can remain on the seeding position, awaiting the next seeding cycle.

[0463] In some examples, when goods are placed in an order container, the control device can update the information on the goods required for that order container until all the required goods are placed in the order container, at which point the control device can determine that the order container has completed planting.

[0464] In other examples, a weight sensor can be installed at the sowing position. The control device can determine whether all the required goods have been placed in the order container, i.e., whether the order container has completed sowing, based on the weight information of each item and the weight information of the order container. It should be noted that the embodiments of this disclosure do not limit the method for determining the sowing status of the order container.

[0465] For example, if it is determined that the third order container has completed sowing, it is necessary to perform an outbound operation on the third order container. In this case, the container pick-and-place mechanism can be controlled to move from the previous position to the third docking position corresponding to the third order container, retrieve the third order container at the third docking position, and transport the third order container to the idle storage space of the first vehicle parked at the first workstation or the idle buffer space of the first workstation.

[0466] In some examples, if the container pick-and-place mechanism retrieves the third order container and the first carrier is docked in the picking area of ​​the first workstation, the container pick-and-place mechanism can place the third order container in an available location on the first carrier. The first carrier can be the target carrier or another carrier awaiting picking.

[0467] For example, the first carrier can also be a designated container swapping carrier. This container swapping carrier can be a carrier specifically designed for full-container outbound shipments within the warehousing system. For instance, the first carrier may include a carrier with multiple available storage locations.

[0468] For example, when the first vehicle is a container-changing vehicle, the control device can transport the first vehicle to the first workstation at preset time intervals (such as the first time interval) to receive the completed seeding order container (such as the third order container) on the seeding wall at the first workstation.

[0469] In other examples, if the container pick-and-place mechanism retrieves the third order container and the first carrier is not present in the picking area of ​​the first workstation, the container pick-and-place mechanism may place the seeded third order container on an empty buffer slot of the first workstation.

[0470] In some embodiments, after the handling equipment moves the first carrier to the first workstation, the control container handling mechanism moves the third order container on the idle buffer position to the idle storage position of the first carrier.

[0471] For example, when the handling equipment moves the first carrier to the picking area of ​​the first workstation, the container pick-and-place mechanism can move the third order container from the buffer position of the first workstation to an empty storage position on the first carrier, thereby realizing the outbound operation of the third order container.

[0472] In some embodiments, after step 340 above, the goods picking method further includes: obtaining the full status of the first order container detected by the full box detection component; when the full status of the first order container is full, controlling the container picking mechanism to remove the first order container from the seeding wall and transporting the first order container to an idle storage location of the first vehicle parked at the first workstation or an idle buffer location of the first workstation.

[0473] In some examples, a full container status is used to indicate whether the goods placed in the order container have reached the container's capacity limit. This capacity limit includes a weight limit and / or a space limit. For example, if the goods are heavy or large, the order container may have reached its capacity limit before planting is complete, in which case an outbound operation needs to be performed on the order container.

[0474] In some embodiments, the container handling mechanism is equipped with a full-container detection component, which can detect the fullness of the order container. For example, the full-container detection device can detect the placement of goods in the order container to determine the fullness of the order container.

[0475] In some examples, the full-container detection device can be an ultrasonic sensor, a laser rangefinder, or an infrared sensor, which can detect the stacking height of goods in an order container or the space occupancy. For example, when the goods reach the maximum height of the order container or reach the maximum space utilization rate, the full state of the order container can be determined as a full state.

[0476] In other examples, the fullness detection device can also be a camera, which captures images of the goods inside the order container to determine the degree of fullness. For example, if the empty space in the order container is less than a preset percentage, the container can be determined to be full.

[0477] It should be noted that the embodiments disclosed herein do not limit the method for detecting a full container state. For example, the detection result of a weight sensor installed at the seeding position can also be used to determine whether the order container has reached a full state.

[0478] For example, after the goods handling component places the first goods into the first order container, the control device can obtain the detection result of the full-box detection device and determine the full status of the first order container based on the detection result. If the first order container is in a full-storage state, an outbound operation needs to be performed on the first order container. Since a target container is currently placed on the container handling mechanism, after the container handling mechanism completes the picking of the target container and places it on the target carrier (or buffer position), the control device can move the container handling mechanism to the first docking position corresponding to the first order container and retrieve the first order container. Finally, the container handling mechanism can transport the retrieved first order container to an empty storage position on the first carrier or an empty buffer position on the first workstation.

[0479] In some embodiments, when there are vacant seeding positions among the multiple seeding positions on the seeding wall, the control container pick-up and drop mechanism removes an empty container from a second carrier docked at the first workstation and transports the empty container to an vacant seeding position on the seeding wall.

[0480] For example, if there are empty seeding slots on the seeding wall, it is necessary to perform an operation to fill empty order containers. Here, an empty seeding slot is a seeding slot that has not been filled with an order container.

[0481] In some examples, where a second vehicle is docked in the picking area of ​​the first workstation, the container pick-and-place mechanism can move empty order containers from the second vehicle to an available seeding position on the seeding wall.

[0482] In some examples, the second carrier can be a carrier currently docked in the picking area of ​​the first workstation, such as a target carrier. For example, at least one empty order container can be placed on each carrier to be picked. When the carrier determines that there is an available seeding position on the seeding wall during the picking process, the empty order container on the carrier can be moved to the available seeding position to complete the replenishment operation of the empty order container.

[0483] In other examples, the second vehicle can also be a container swapping vehicle for instructions. This swapping vehicle can be a vehicle specifically designed within the warehousing system for replenishing empty order containers. For example, the second vehicle can hold multiple empty order containers.

[0484] For example, when the second vehicle is a container-changing vehicle, the control device can move the second vehicle to the first workstation at preset time intervals (such as a second time interval) to perform the replenishment operation of empty order containers at the first workstation.

[0485] In some examples, the second vehicle and the first vehicle can be the same vehicle or different vehicles. For example, when the second vehicle and the first vehicle can be the same vehicle, some of the storage locations on this vehicle are vacant, while others contain empty order containers. The handling equipment can move this vehicle to the first workstation at preset intervals to perform outbound operations for full order containers and inbound operations for empty order containers.

[0486] The goods picking method provided in this embodiment can realize automatic box changing operation (including the outbound operation of full order containers and the replenishment operation of empty order containers) through the container picking and placing mechanism of the first workstation. Compared with the manual box changing in related technologies, it reduces labor costs and improves box changing efficiency.

[0487] For example, before performing picking operations on the containers placed on each carrier, a sorting operation can be performed on the containers placed on each carrier to improve subsequent picking efficiency. The sorting operation can include sorting between containers on different carriers, or sorting between different containers on the same carrier. For example, containers with pending orders can be transferred from multiple carriers to at least one carrier; or containers with pending orders can be transferred from one storage location on a carrier to another storage location.

[0488] In some embodiments, the warehousing system further includes a third workstation. The third workstation is used to perform sorting operations on the containers placed on each vehicle; the third workstation may also be referred to as a sorting workstation. The third workstation includes sorting objects, which can be sorting personnel or sorting equipment.

[0489] In some examples, when the sorting object is a sorting device, the sorting device can be a container robot that can pick up / place containers on a carrier; or, the third workstation can also be equipped with a robotic arm that can pick up / place containers on a carrier.

[0490] Figure 14 is a schematic diagram of another goods picking method provided by an embodiment of this disclosure. As shown in Figure 14, before step 320, the method further includes steps 1410 to 1440 as shown below. It should be noted that step 1410 can be performed after step 310 and before step 320.

[0491] Step 1410: Determine the container positions of the multiple containers corresponding to the order to be processed on the target vehicle.

[0492] In some examples, after determining the multiple containers corresponding to the orders to be processed and the target vehicle for placing the multiple containers through step 310, the positions of the multiple containers on the target vehicle can be further determined. The multiple containers include the target container. It should be noted that the container corresponding to the orders to be processed (i.e., the container hit by multiple orders to be processed) is referred to as the hit container.

[0493] For example, the container location of each hit container on the target vehicle may include the cargo location of each hit container on the target vehicle, and the location area of ​​that cargo location on the target vehicle.

[0494] Step 1420: Based on the container location, determine the cargo containers to be processed on the target vehicle.

[0495] In some examples, a container awaiting processing can be determined from multiple inventory containers on the target vehicle based on the container positions of each hit container on the target vehicle. The inventory containers on the target vehicle include hit containers and non-hit containers (containers not hit by pending orders), and the container awaiting processing is at least one inventory container on the target vehicle, which may include hit containers and / or non-hit containers.

[0496] For example, the containers to be processed include at least one first container that has been hit by multiple pending orders and whose container position is off-center from a preset area, and / or at least one second container that has not been hit by multiple pending orders and whose container position is within the preset area. The preset area is located in the middle region of the target vehicle.

[0497] In some examples, the target container can be divided into two areas, such as a central area and a surrounding area. When the target container is placed in the central area, the container pick-and-place mechanism in the pick-and-place unit removes the container from the target container and transports it to the docking position corresponding to the order container. The distance traveled by the container pick-and-place mechanism is relatively short, thus further improving picking efficiency. Therefore, by performing a sorting operation, the container corresponding to the order to be processed (i.e., the target container) can be transferred to the central area of ​​the target container.

[0498] For example, among multiple hit containers on a target vehicle, some hit containers may be located in the central area, and some hit containers may be located outside the central area (i.e., the surrounding area); or, all hit containers may be located within the central area; or, all hit containers may be located outside the central area. Here, a hit container being located within the central area means that the cargo location of the hit container is a cargo location within the central area.

[0499] In some examples, if all hit containers are located within the intermediate area, no tallying operation is required for the target vehicle. If at least one hit container is located outside the intermediate area, a tallying operation can be performed on the target vehicle. It should be noted that a tallying operation can also be performed on the target vehicle when the number of hit containers located outside the intermediate area exceeds a specified threshold; this disclosure does not limit this practice.

[0500] For example, if at least one hit container is located outside the intermediate area, the hit container located outside the intermediate area is referred to as the first container. That is, when the first container is present on the target vehicle, a tallying operation can be performed on the target vehicle.

[0501] Step 1430: Control the handling equipment to move the target vehicle to the third workstation.

[0502] In some examples, when it is determined that a tallying operation needs to be performed on the target vehicle, the handling equipment can be controlled to move the target vehicle to a third workstation. The handling equipment moving the target vehicle to the third workstation can be the same as the handling equipment moving the target vehicle to the first workstation in step 320, or they can be different handling equipment.

[0503] Step 1440: When the handling equipment moves the target vehicle to the third workstation, control the handling object of the third workstation to perform handling operations on the first container and / or the second container to handle the first container to a preset area.

[0504] In some examples, after the target vehicle is moved to the first workstation, a tallying operation can be performed on the target vehicle. Tallying objects (such as tallying personnel or tallying equipment) can transfer the first container on the target vehicle to a predetermined area.

[0505] In some embodiments, the number of first containers and the number of available storage spaces in a preset area are determined; based on the number of first containers and the number of available storage spaces in the preset area, the sorting object is controlled to sort the first containers to available storage spaces in the preset area, and / or to sort the second containers to available storage spaces outside the preset area.

[0506] In some embodiments, if the number of available storage locations in a preset area is greater than or equal to the number of first containers, the sorting object is controlled to sort the first containers to available storage locations in the preset area; if the number of available storage locations in the preset area is less than the number of first containers, the sorting object is controlled to sort at least one second container to an available storage location outside the preset area, and sort the first container to an available storage location within the preset area.

[0507] In some examples, if there are many vacant storage locations in a preset area (such as the middle area) that are greater than or equal to the number of the first containers, the first containers can be directly moved from the current storage location (i.e., the storage location outside the preset area) to an vacant storage location within the preset area, so that the matching containers corresponding to the pending orders are all placed in the preset area of ​​the target vehicle.

[0508] In other examples, if the number of available storage spaces within the preset area is small and less than the number of first containers, the available storage spaces within the preset area cannot accommodate all the first containers. Therefore, it can be determined whether there are any non-target containers (i.e., second containers) within the preset area. If there are second containers on the target vehicle, the handling object can be controlled to move the corresponding number of second containers from the current storage space (i.e., the storage space within the preset area) to an available storage space outside the preset area, based on the number of storage spaces required by the first containers. After moving the second containers outside the preset area, the first containers can be moved to an available storage space within the preset area.

[0509] The goods picking method provided in this embodiment can pre-sort multiple containers corresponding to the orders to be processed to the middle area of ​​the target vehicle by performing a sorting operation on the target vehicle. This allows the container picking and placing mechanism of the first workstation to retrieve the target container and other hit containers by moving less distance to perform the picking operation, thereby further improving the picking efficiency of the warehousing system.

[0510] For example, multiple first workstations in a warehousing system can be divided into at least one workstation group, and each workstation group includes multiple pick-and-place devices and multiple seeding walls.

[0511] Figure 15 is a schematic diagram of another warehousing system provided in an embodiment of the present disclosure. As shown in Figure 15, the workstation group 1510 may consist of two workstations, and the workstation group 1510 includes two pick-and-place devices, a first pick-and-place device 3310 and a second pick-and-place device 3320.

[0512] In some examples, workstation group 1510 corresponds to vehicle docking area 1511, which may include multiple picking areas. That is, multiple (e.g., two) pick-and-place devices in workstation group 1510 can simultaneously perform picking operations on multiple vehicles docked in the picking area.

[0513] As shown in Figure 15, the vehicle docking area 1511 has three picking areas, where vehicles A, B, and C are respectively docked. A first pick-and-place device 3310 and a second pick-and-place device 3320 can retrieve the target container from vehicle A, vehicle B, or vehicle C, respectively. The first pick-and-place device 3310 can transport the retrieved target container to the corresponding order container on the first seeding wall 410 to perform the picking operation; the second pick-and-place device 3320 can transport the retrieved target container to the corresponding order container on the second seeding wall 420 to perform the picking operation. For example, each vehicle can enter or leave the picking area of ​​the vehicle docking area 1511 via the direction of the arrow.

[0514] It should be noted that the first pick-and-place device 3310 and the second pick-and-place device 3320 have the same structure as the pick-and-place device 33 in the above embodiment, and the picking process is also similar. To avoid repetition, they will not be described again here.

[0515] The goods picking method provided in this disclosure improves the picking efficiency of the warehousing system by dividing the workstations into groups, increasing the number of picking vehicles and the order hit rate.

[0516] Figure 16 is a schematic diagram of a goods picking device provided in an embodiment of this disclosure. As shown in Figure 16, the goods picking device 1600 includes an acquisition module 1610, a control module 1620, and a determination module 1630. Wherein:

[0517] The acquisition module 1610 is configured to: acquire the pending orders corresponding to the first workstation, and determine the target container for placing at least one item to be picked corresponding to the pending orders.

[0518] The control module 1620 is configured to control the image acquisition device to acquire a first image corresponding to the target container when the handling equipment moves the target container to the first workstation.

[0519] The determination module 1630 is configured to: determine the first cargo based on the first image, and if the type of the first cargo is a first cargo type, determine the first target pickup information of the first cargo; wherein the first target pickup information includes pickup position and pickup direction.

[0520] The control module 1620 is configured to: based on the target picking information, control the picking and placing device of the first workstation to pick out the first goods from the target container and transfer the first goods to the target location.

[0521] In some embodiments, the determining module 1630 is configured to: determine feature information of each item to be picked among at least one item to be picked based on the first image; and determine a first item among the at least one item to be picked based on the feature information of each item to be picked, a preset feature information database, and the item information of the first item. The preset feature information database includes correspondences between multiple feature information items and multiple item information items, and the feature information includes at least one of the following: image information of the item, identification information of the item, and outline information of the item.

[0522] In some embodiments, the determining module 1630 is configured to: determine candidate feature information corresponding to the cargo information of the first cargo in a preset feature information database based on the cargo information of the first cargo; and determine the cargo to be picked corresponding to the first feature information as the first cargo if there is first feature information that matches the candidate feature information among the feature information of each cargo to be picked.

[0523] In some embodiments, the determining module 1630 is configured to: determine the picking information corresponding to the first feature information as the first target picking information in a preset picking information library.

[0524] In some embodiments, the determining module 1630 is configured to: control the picking and placing device to adjust the position of at least one item to be picked in the target container when there is no picking information corresponding to the first feature information in the preset picking information database; after adjusting the position of at least one item to be picked in the target container, control the image acquisition device to acquire a second image corresponding to the target container; wherein the second image is different from the first image; and determine the first target picking information of the first item based on the second image.

[0525] In some embodiments, the goods picking device 1100 further includes a sending module, configured to send a first image to the interactive device if no picking information corresponding to the first feature information exists in a preset picking information database. The acquisition module 1610 is configured to acquire target contour information corresponding to the first goods input by the user on the interactive device. The determination module 1630 is configured to determine first target picking information for the first goods based on the target contour information.

[0526] In some embodiments, the goods picking device 1100 further includes an update module, which is configured to store first feature information and first target picking information in a preset picking information database to update the preset picking information database.

[0527] In some embodiments, the control module 1620 is further configured to: in the event that the first target pickup information of the first cargo cannot be determined, control the handling equipment to transport the target container or target carrier to the second workstation so that the first cargo can be picked from the target container by the picking personnel at the second workstation.

[0528] In some embodiments, the control module 1620 is further configured to: when the type of the first goods is a second goods type, control the pick-and-place device to pick at least one second goods from the goods to be picked in the target container; wherein the type of the second goods is the first goods type, and the goods to be picked of the second goods type are more difficult to pick than the goods to be picked of the first goods type; after the pick-and-place device has finished picking the goods to be picked of the first goods type in the target container, control the handling equipment to transport the target container or target vehicle to the second workstation so that the picking personnel at the second workstation can pick out the first goods from the target container.

[0529] In some embodiments, the determining module 1630 is further configured to: determine whether there are any second type of goods among the at least one type of goods to be picked, based on at least one type of goods to be picked corresponding to the pending order and a preset goods information database; wherein the preset goods information database includes multiple goods of the second type. The control module 1620 is further configured to: if there are second type of goods among the at least one type of goods to be picked, control the handling equipment to transport the target container containing the second type of goods to be picked to the second workstation, so that the picking personnel at the second workstation can pick out the second type of goods from the target container.

[0530] In some embodiments, if no first feature information matching the candidate feature information exists in the feature information of each item to be picked, the determining module 1630 is further configured to: determine a first item among at least one item to be picked based on the first image and the recognition model; the acquiring module 1610 is further configured to: control the picking and placing device to adjust the position of at least one item to be picked in the target container, and control the image acquisition device to acquire the adjusted image corresponding to the target container; the determining module 1630 is further configured to: determine the first item among at least one item to be picked based on the adjusted image corresponding to the target container; and determine the first item among at least one item to be picked based on the item tag information corresponding to the first item input by the user on the interactive device.

[0531] In some embodiments, the control module 1620 is further configured to: if the first item cannot be identified among at least one item to be picked, control the handling equipment to transport the target container or target carrier to a second workstation so that the first item can be picked from the target container by the picking personnel at the second workstation.

[0532] In some embodiments, the determining module 1630 is further configured to: determine a pick-up and place device, and determine a target picking mechanism on the pick-up and place device for picking up the first goods; wherein the pick-up and place device includes a plurality of picking mechanisms, and the target picking mechanism is at least one of the plurality of picking mechanisms. The control module 1620 is configured to: based on the first target picking information, control the target picking mechanism to pick up the first goods from the target container and transfer the first goods to the target location.

[0533] In some embodiments, the determining module 1630 is configured to determine first target picking information based on at least one of a first image, attribute information of the first cargo, and picking prompt information corresponding to the first cargo.

[0534] In some embodiments, the determining module 1630 is configured to: determine a target picking agency based on at least one candidate picking agency corresponding to the picking task instruction received for the first goods; or, determine a target picking agency among a plurality of picking agencies based on the attribute information of the first goods.

[0535] In some embodiments, the determining module 1630 is configured to: determine a target picking mechanism among a plurality of picking mechanisms based on attribute information of a first cargo and equipment parameters of each picking mechanism among a plurality of picking mechanisms; wherein the attribute information includes at least one of the type, size, weight and shape of the first cargo; and the equipment parameters include at least one of the size, material, hardness, picking force, shape and weight of the picking mechanism.

[0536] In some embodiments, the determining module 1630 is configured to: determine a target picking mechanism among multiple picking mechanisms based on the attribute information of the first goods and a first knowledge base; wherein the first knowledge base includes a first correspondence between the attribute information of multiple goods and the equipment parameters of multiple picking mechanisms.

[0537] In some embodiments, the determining module 1630 is configured to: determine first target pickup information of the first cargo based on the first image, the attribute information of the first cargo, and pickup prompt information, using a second recognition model; wherein the pickup prompt information is used to indicate at least one constraint information for picking up the first cargo.

[0538] In some embodiments, the determining module 1630 is configured to: obtain first feature information corresponding to the first cargo based on the first image through a visual processing network in a second recognition model; wherein the first feature information is used to characterize at least one of the segmentation result, depth information, and stacking relationship information corresponding to the first cargo. Based on the attribute information and pickup prompt information of the first cargo, obtain second feature information corresponding to the attribute information and third feature information corresponding to the pickup prompt information through an encoding network in the second recognition model. Based on the first feature information, the second feature information, and the third feature information, determine the first target pickup information of the first cargo through a feature fusion network in the second recognition model.

[0539] In some embodiments, the first image includes multiple visual images; the determining module 1630 is configured to: perform block processing on each visual image through an image block segmentation module in the visual processing network to obtain a first output sequence; wherein the first output sequence includes image blocks corresponding to each visual image. The position encoding module in the visual processing network performs position encoding on each image block in the first output sequence to obtain a second output sequence. The feature extraction module in the visual processing network performs feature extraction processing on the second output sequence to obtain output features. The output features are input to a task module in the visual processing network to obtain first feature information.

[0540] In some embodiments, the determining module 1630 is configured to: perform text encoding on the attribute information of the first goods by a first encoding module in the encoding network to obtain first feature information; and perform text encoding on the pickup prompt information by a second encoding module in the encoding network to obtain second feature information.

[0541] In some embodiments, the determining module 1630 is configured to: perform splicing processing on the first feature information, the second feature information, and the third feature information through the splicing module in the feature fusion network to obtain joint features; and process the joint features through the recognition and analysis module in the feature fusion network to obtain first target picking information.

[0542] In some embodiments, the determining module 1630 is further configured to: determine the first target pickup information of the first cargo based on the first image, the attribute information of the first cargo, and the pickup prompt information corresponding to the first cargo, by means of a second recognition model when the first target pickup information does not exist in the preset pickup information database.

[0543] In some embodiments, the control module 1620 is further configured to: control the target picking mechanism to pick up the first goods again based on the first target picking information if the target picking mechanism fails to pick up the first goods based on the first target picking information; and control the target picking mechanism to adjust the pose of the first goods in the target container if the number of times the target picking mechanism fails to pick up the first goods based on the first target picking information reaches a first preset number. The acquisition module 1610 is further configured to: acquire a fourth image of the first goods including the adjusted first goods acquired by the image acquisition device. The determination module 1630 is further configured to: determine the second target picking information of the first goods based on at least one of the fourth image, the attribute information of the first goods, and the picking prompt information corresponding to the first goods. The control module 1620 is further configured to: control the target picking mechanism to pick up the first goods from the target container based on the second target picking information.

[0544] In some embodiments, the update control module is further configured to: update a preset pickup information library based on the first image and the first target pickup information; or, update the preset pickup information library based on the third image and the second target pickup information.

[0545] In some embodiments, the control module 1620 is further configured to: control the output device to output an error message when the target picking mechanism fails to pick up the first item based on the second target picking information; and instruct the execution object to operate the target picking mechanism based on the error message so that the target picking mechanism performs a picking operation on the first item. The update control module is further configured to: update the preset picking information database based on the picking information of the first item when the target picking mechanism successfully picks up the first item.

[0546] In some embodiments, the determining module 1630 is further configured to: mark the first item as an abnormal picking item if the target picking mechanism fails to pick up the first item, and determine the next item to be picked in the target container.

[0547] In some embodiments, the acquisition module 1610 is further configured to: acquire a first image including the first cargo acquired by the image acquisition device. The determination module 1630 is further configured to: determine the target picking mechanism and the first target picking information of the first cargo among the multiple picking mechanisms of the picking and placing device, based on the first image, the attribute information of the first cargo, the picking prompt information corresponding to the first cargo, and the device parameters of each picking mechanism among the multiple picking mechanisms of the picking and placing device, through a second recognition model.

[0548] In some embodiments, the determining module 1630 is configured to: obtain first feature information corresponding to the first cargo based on the first image through a visual processing network in a second recognition model; wherein the first feature information is used to characterize at least one of the segmentation result, depth information, and stacking relationship information corresponding to the first cargo. Based on the attribute information of the first cargo, the pickup prompt information, and the device parameters of each pickup mechanism, obtain second feature information corresponding to the attribute information, third feature information corresponding to the pickup prompt information, and fourth feature information corresponding to the device parameters of each pickup mechanism through an encoding network in the second recognition model. Based on the first feature information, the second feature information, the third feature information, and the fourth feature information, determine the target pickup mechanism and the first target pickup information of the first cargo through a feature fusion network in the second recognition model.

[0549] In some embodiments, a target identification code is provided on the first item; the determining module 1630 is further configured to: when the target picking mechanism picks up the first item, acquire at least one fourth image including the first item through an image acquisition device; when the at least one fourth image includes the target identification code, identify the target identification code; when the target identification code is used to indicate the first item, update the inventory information of the first item; when the target identification code cannot be identified, or when the first item cannot be determined based on the target identification code, mark the first item as an abnormal picking object, return the first item to the target container, and determine the next item to be picked in the target container.

[0550] In some embodiments, the control module 1620 is further configured to: control the pick-and-place device to adjust the angle of the target picking mechanism when the fourth image does not include the target identification code; wherein the number of angle adjustments of the target picking mechanism does not exceed a second preset number; and based on the adjusted target picking mechanism, re-acquire at least one fifth image including the first goods through the image acquisition device.

[0551] In some embodiments, the determining module 1630 is configured to: after the target picking mechanism picks up the first goods from the target container, acquire at least one sixth image including the first goods through an image acquisition device; based on the at least one sixth image, if it is determined that the target picking mechanism has picked up multiple candidate objects, determine that the target picking mechanism has encountered a double picking anomaly; wherein the multiple candidate objects include the first goods and at least one other object besides the first goods; if it is determined that the target picking mechanism has encountered a double picking anomaly, control the target picking mechanism to put the picked-up multiple candidate objects back into the target container, and control the target picking mechanism to pick up the first goods again based on the first target picking information.

[0552] In some embodiments, the determining module 1630 is configured to: after the target picking mechanism transfers the picked-up first goods to the first order container, determine the weight information of the first order container through the weight acquisition device corresponding to the first order container; wherein, the target location includes the first order container corresponding to the first goods; based on the weight information of the first order container and the weight information of the first goods, determine whether there is a double picking anomaly in the target picking mechanism; if it is determined that a double picking anomaly has occurred in the target picking mechanism, mark the first order container as an abnormal order container, and control the output device to output an anomaly prompt message to instruct the execution object to perform anomaly handling on the abnormal order container.

[0553] In some embodiments, the control module 1620 is further configured to: acquire a picking image after controlling the pick-and-place device to pick up the first item from the target container; wherein the picking image includes an image corresponding to the target container, an image corresponding to the target location, and an image on the picking path of the first item. The determination module 1630 is further configured to: determine the drop position of the first item based on the picking image; determine that the first item has been picked if the drop position of the first item is located in the first order container; determine that the picking of the first item has failed if the drop position of the first item is located in the target container; the control module 1620 is further configured to: control the pick-and-place device to pick up the first item again from the target container and transfer the first item to the first order container. The determination module 1630 is further configured to: determine that the picking of the first item has failed if the drop position of the first item is located in a candidate container; the control module 1620 is further configured to: control the pick-and-place device to pick up the first item from the candidate container and transfer the first item to the first order container; wherein the candidate container is any container on the picking path of the first item.

[0554] In some embodiments, the pick-and-place device includes a container pick-and-place mechanism; the control module 1620 is configured to: control the container pick-and-place mechanism to move the target container to the first docking position corresponding to the first order container on the seeding wall; wherein, the first goods are the goods required for the first order container; during the process of the container pick-and-place mechanism moving to the first docking position, control the image acquisition device to acquire the first image corresponding to the target container.

[0555] In some embodiments, the pick-and-place device further includes a goods pick-and-place mechanism, the container pick-and-place mechanism and the goods pick-and-place mechanism are coupled together, the goods pick-and-place mechanism can move laterally or vertically with the container pick-and-place mechanism; the goods pick-and-place mechanism includes a target picking mechanism; the control module 1620 is configured to: control the goods pick-and-place mechanism to pick out the first goods from the target container according to the target picking information, and transfer the first goods to the first order container; wherein, the target location includes the location corresponding to the first order container.

[0556] In some embodiments, the control module 1620 is configured to: control the container pick-and-place mechanism to pull the first order container out from the first seeding position so that the goods pick-and-place mechanism can deliver the first goods into the first order container; control the goods pick-and-place mechanism to pick the first goods from the target container and deliver the first goods into the first order container; and control the container pick-and-place mechanism to push the first order container back to the first seeding position.

[0557] In some embodiments, the control module 1620 is further configured to: when a third item required by the second order container on the seeding wall is placed in the target container, control the container picking and placing mechanism to move the target container from the first docking position to the second docking position corresponding to the second order container; and during the process of the container picking and placing mechanism moving to the second docking position, control the image acquisition device to acquire the seventh image corresponding to the target container.

[0558] In some embodiments, the acquisition module 1610 is further configured to acquire the sowing status of each order container among the multiple order containers on the sowing wall. The control module 1620 is further configured to, when there is a third order container on the sowing wall with a sowing status of "sowing completed", control the container pick-up and place mechanism to move to the third docking position corresponding to the third order container, pick up the third order container at the third docking position, and transport the third order container to the idle storage space of the first vehicle docked at the first workstation or the idle buffer space of the first workstation.

[0559] In some embodiments, the control module 1620 is further configured to: when there is an empty seeding position among the plurality of seeding positions on the seeding wall, control the container pick-up and drop mechanism to take an empty container from the second carrier docked at the first workstation and transport the empty container to the empty seeding position on the seeding wall.

[0560] In some embodiments, the control module 1620 is configured to: control the guide component to bridge between the container pick-and-place mechanism and the first order container when the container pick-and-place mechanism moves to the first docking position; control the goods pick-and-place mechanism to pick up the first goods from the target container and transfer the first goods to the first order container; or control the goods pick-and-place mechanism to place the first goods on the guide component to transfer the first goods to the first order container via the guide component; and control the guide component to retract after the first goods have been transferred to the first order container.

[0561] In some embodiments, the control module 1620 is further configured to: obtain the full status of the first order container detected by the full container detection component; and, if the full status of the first order container is full, control the container pick-and-place mechanism to remove the first order container from the seeding wall and move the first order container to an idle storage location of the first vehicle parked at the first workstation or an idle buffer location of the first workstation.

[0562] In some embodiments, the determining module 1630 is further configured to: determine the container positions of a plurality of containers corresponding to an order to be processed on a target vehicle; wherein the plurality of containers includes a target container; determine, based on the container positions, a container to be sorted on the target vehicle; wherein the container to be sorted includes at least one first container that is hit by an order to be processed and whose container position is off-center from a preset area, and / or at least one second container that is not hit by an order to be processed and whose container position is within the preset area; the preset area is located in the middle area of ​​the target vehicle. The control module 1620 is further configured to: control a handling device to transport the target vehicle to a third workstation; and, when the handling device transports the target vehicle to the third workstation, control the sorting object at the third workstation to perform sorting operations on the first container and / or the second container to sort the first container into the preset area.

[0563] Figure 17 is a schematic diagram of an electronic device provided in an embodiment of this disclosure. In some embodiments, the electronic device includes one or more processors and a memory. The memory is configured to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the goods picking method in the above embodiments.

[0564] As shown in Figure 17, the electronic device 1000 includes a processor 1001 and a memory 1002. Exemplarily, the electronic device 1000 may also include a communications interface 1003 and a communications bus 1004.

[0565] The processor 1001, memory 1002, and communication interface 1003 communicate with each other via communication bus 1004. Communication interface 1003 is used to communicate with other network elements such as clients or other servers.

[0566] In some embodiments, the processor 1001 is used to execute program 1005, specifically performing the relevant steps in the above-described embodiments of the goods picking method. Specifically, program 1005 may include program code, which includes computer-executable instructions.

[0567] For example, processor 1001 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present disclosure. Electronic device 1000 may include one or more processors, which may be processors of the same type, such as one or more CPUs; or they may be processors of different types, such as one or more CPUs and one or more ASICs.

[0568] In some embodiments, memory 1002 is used to store program 1005. Memory 1002 may include high-speed RAM memory, and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0569] Specifically, program 1005 can be called by processor 1001 to enable electronic device 1000 to perform the goods picking method.

[0570] This disclosure provides a computer-readable storage medium storing at least one executable instruction that, when executed on an electronic device 1000, causes the electronic device 1000 to perform the goods picking method described in the above embodiments.

[0571] The executable instructions can be used to cause the electronic device 1000 to perform the goods picking method.

[0572] For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0573] In some embodiments, this disclosure provides a computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the goods picking method described in any of the above embodiments.

[0574] In some embodiments, this disclosure also provides a computer program that, when executed by a processor, can implement the goods picking method described in any of the above embodiments.

[0575] The beneficial effects that the goods picking device, electronic device, computer-readable storage medium, computer program product and computer program provided in this disclosure can achieve are the same as the beneficial effects of the goods picking method provided above, and will not be repeated here.

[0576] The embodiments described above do not constitute a limitation on the scope of protection of this disclosure.

Claims

1. A cargo picking method, comprising: obtaining a to-be-processed order corresponding to a first workstation, and determining a target container in which at least one to-be-picked cargo corresponding to the to-be-processed order is placed; controlling an image acquisition device to acquire a first image corresponding to the target container in a case where a carrying device carries the target container to the first workstation; determining a first cargo from the at least one to-be-picked cargo according to the first image, and determining first target picking information of the first cargo in a case where a type of the first cargo is a first cargo type; wherein the first target picking information comprises a picking position and a picking direction; controlling a taking and placing device to pick the first cargo from the target container and transfer the first cargo to a target position according to the first target picking information.

2. The method of claim 1, wherein, The determining of the first cargo from the at least one to-be-picked cargo comprises: determining characteristic information of each to-be-picked cargo from the at least one to-be-picked cargo according to the first image; determining the first cargo from the at least one to-be-picked cargo according to the characteristic information of each to-be-picked cargo, a preset characteristic information library, and cargo information of the first cargo; wherein the preset characteristic information library comprises a correspondence relationship between a plurality of characteristic information and a plurality of cargo information, and the characteristic information comprises image information of a cargo and / or identification information of the cargo.

3. The method of claim 2, wherein, The determining of the first cargo from the at least one to-be-picked cargo according to the characteristic information of each to-be-picked cargo, the preset characteristic information library, and the cargo information of the first cargo comprises: determining candidate characteristic information corresponding to the cargo information of the first cargo in the preset characteristic information library based on the cargo information of the first cargo; determining a to-be-picked cargo corresponding to first characteristic information that matches the candidate characteristic information as the first cargo in a case where the first characteristic information exists in the characteristic information of each to-be-picked cargo.

4. The method of claim 3, further comprising: determining the first cargo from the at least one to-be-picked cargo based on the first image and a first identification model in a case where the first characteristic information that matches the candidate characteristic information does not exist in the characteristic information of each to-be-picked cargo; or controlling the taking and placing device to adjust a position of at least one to-be-picked cargo in the target container, controlling the image acquisition device to acquire an adjusted image corresponding to the target container, and determining the first cargo from the at least one to-be-picked cargo according to the adjusted image corresponding to the target container; or sending the first image to an interactive device, obtaining cargo marker information corresponding to the first cargo input by a user on the interactive device, and determining the first cargo from the at least one to-be-picked cargo according to the cargo marker information.

5. The method of claim 3 or 4, wherein, The determining of the first target picking information of the first cargo comprises: obtaining a preset picking information library; wherein the preset picking information library comprises a correspondence relationship between a plurality of characteristic information and a plurality of picking information. determine the first target picking information of the first cargo according to the first feature information and the preset picking information library.

6. The method of claim 5, wherein, The determining of the first target picking information of the first cargo according to the first feature information and the preset picking information library comprises: in a case where the preset picking information library contains picking information corresponding to the first feature information, determining the picking information corresponding to the first feature information as the first target picking information; or in a case where the preset picking information library does not contain picking information corresponding to the first feature information, controlling the taking and placing device to adjust the position of at least one to-be-picked cargo in the target container; after the position of at least one to-be-picked cargo in the target container is adjusted, controlling the image acquisition device to acquire a second image corresponding to the target container; wherein the second image is different from the first image; determining the first target picking information of the first cargo according to the second image.

7. The method of claim 5, wherein, The determining of the first target picking information of the first cargo according to the first feature information and the preset picking information library comprises: in a case where the preset picking information library does not contain picking information corresponding to the first feature information, sending the first image to an interactive device; acquiring target contour information of the first cargo input by a user on the interactive device; determining the first target picking information of the first cargo according to the target contour information.

8. The method of claim 7, further comprising: storing the first feature information and the first target picking information in the preset picking information library to update the preset picking information library.

9. The method of any one of claims 1-8, further comprising: in a case where the first target picking information of the first cargo cannot be determined, controlling the carrying device to carry the target container to a second working station, so that a picker at the second working station picks the first cargo from the target container.

10. The method of any one of claims 1-8, further comprising: in a case where the type of the first cargo is a second cargo type, controlling the taking and placing device to pick a second cargo in the at least one to-be-picked cargo in the target container; wherein the type of the second cargo is the first cargo type, and the to-be-picked cargo of the second cargo type is more difficult to pick than the to-be-picked cargo of the first cargo type; after the taking and placing device finishes picking the to-be-picked cargo of the first cargo type in the target container, controlling the carrying device to carry the target container to a second working station, so that a picker at the second working station picks the first cargo from the target container.

11. The method of any one of claims 1-8, further comprising: determine, according to the at least one to-be-picked goods corresponding to the to-be-processed order and a preset goods information library, whether the at least one to-be-picked goods includes a to-be-picked goods of a second goods type; the preset goods information library includes a plurality of goods of the second goods type; in a case where the at least one to-be-picked goods includes the to-be-picked goods of the second goods type, control the carrying device to carry the target container in which the to-be-picked goods of the second goods type is placed to a second work station, so that a picker at the second work station picks the to-be-picked goods of the second goods type from the target container.

12. The method of claim 3 or 4, further comprising: in a case where the first goods cannot be determined from the at least one to-be-picked goods, control the carrying device to carry the target container to a second work station, so that a picker at the second work station picks the first goods from the target container.

13. The method of claim 1, further comprising: determine the picking and placing device, and determine a target picking mechanism on the picking and placing device for picking the first goods; the picking and placing device includes a plurality of picking mechanisms, and the target picking mechanism is at least one of the plurality of picking mechanisms; the control, according to the first target picking information, of the picking and placing device to pick the first goods from the target container and transfer the first goods to a target position includes: based on the first target picking information, control the target picking mechanism to pick the first goods from the target container and transfer the first goods to the target position.

14. The method of claim 13, wherein, the determination of the first target picking information of the first goods further includes: based on at least one of the first image, attribute information of the first goods, and picking prompt information corresponding to the first goods, determine the first target picking information.

15. The method of claim 14, wherein, the determination of the target picking mechanism on the picking and placing device for picking the first goods includes: based on at least one candidate picking mechanism indicated by the received picking task of the first goods, determine the target picking mechanism; or based on the attribute information of the first goods, determine the target picking mechanism from the plurality of picking mechanisms.

16. The method of claim 15, wherein, the determination of the target picking mechanism from the plurality of picking mechanisms based on the attribute information of the first goods includes: based on the attribute information of the first goods and device parameters of each picking mechanism in the plurality of picking mechanisms, determine the target picking mechanism from the plurality of picking mechanisms; wherein the attribute information includes at least one of a type, a size, a weight, and a shape of the first goods; and the device parameters include at least one of a size, a material, a hardness, a picking force, a shape, and a weight of the picking mechanism.

17. The method of claim 15, wherein, the determination of the target picking mechanism from the plurality of picking mechanisms based on the attribute information of the first goods includes: based on the attribute information of the first goods and a first knowledge base, determine the target picking mechanism from the plurality of picking mechanisms. The first knowledge base comprises a plurality of first corresponding relationships between attribute information of a plurality of goods and device parameters of a plurality of picking mechanisms.

18. The method of any one of claims 14-17, wherein, The first target picking information is determined based on at least one of the first image, the attribute information of the first goods, and the picking prompt information corresponding to the first goods. The first target picking information of the first goods is determined based on the first image, the attribute information of the first goods, and the picking prompt information by a second recognition model, wherein the picking prompt information is used to indicate at least one constraint information for picking the first goods.

19. The method of claim 18, wherein, The first target picking information of the first goods is determined based on the first image, the attribute information of the first goods, and the picking prompt information by a second recognition model, wherein the picking prompt information is used to indicate at least one constraint information for picking the first goods. The first feature information corresponding to the first goods is obtained based on the first image by a visual processing network in the second recognition model, wherein the first feature information is used to represent at least one of a segmentation result, depth information, and stacking relationship information corresponding to the first goods. The second feature information corresponding to the attribute information and the third feature information corresponding to the picking prompt information are obtained based on the attribute information of the first goods and the picking prompt information by an encoding network in the second recognition model. The first target picking information of the first goods is determined based on the first feature information, the second feature information, and the third feature information by a feature fusion network in the second recognition model.

20. The method of claim 19, wherein, The first image comprises a plurality of visual images, and the first feature information corresponding to the first goods is obtained based on the first image by a visual processing network in the second recognition model, comprising: The image blocks corresponding to the visual images are obtained by an image block module in the visual processing network, wherein the first output sequence comprises the image blocks corresponding to the visual images. The position of each image block in the first output sequence is encoded by a position encoding module in the visual processing network to obtain a second output sequence. The output feature is obtained by a feature extraction module in the visual processing network for feature extraction processing on the second output sequence. The output feature is input into a task module in the visual processing network to obtain the first feature information.

21. The method of claim 14, further comprising: in the case that the first target picking information does not exist in the preset picking information library, determining the first target picking information of the first goods based on the first image, the attribute information of the first goods, and the picking prompt information corresponding to the first goods by a second recognition model.

22. The method of any one of claims 14-21, further comprising: in the case that the target picking mechanism fails to pick the first goods based on the first target picking information, controlling the target picking mechanism to pick the first goods again based on the first target picking information. In a case where the number of times of failure of the target picking mechanism picking up the first goods based on the first target picking information reaches a first preset number of times, the target picking mechanism is controlled to adjust the pose of the first goods in the target container; a third image including the first goods after the adjustment is acquired by the image acquisition device; second target picking information of the first goods is determined based on at least one of the third image, attribute information of the first goods, and picking prompt information corresponding to the first goods; the target picking mechanism is controlled to pick up the first goods from the target container based on the second target picking information.

23. The method of claim 22, further comprising: updating a preset picking information library based on the first image and the first target picking information; or, updating the preset picking information library based on the third image and the second target picking information.

24. The method of claim 23, further comprising: in a case where the target picking mechanism fails to pick up the first goods based on the second target picking information, controlling an output device to output abnormal prompt information; instructing a performing object to operate the target picking mechanism based on the abnormal prompt information, so that the target picking mechanism performs a picking operation on the first goods; in a case where the target picking mechanism successfully picks up the first goods, updating a preset picking information library based on picking information of the first goods when the picking is successful.

25. The method of claim 24, further comprising: in a case where the target picking mechanism fails to pick up the first goods, marking the first goods as an abnormal sorting goods, and determining a next to-be-picked goods in the target container.

26. The method of claim 13, wherein, After the determining the picking and placing device, the method further comprises: acquiring a first image including the first goods acquired by the image acquisition device; determining, by a second recognition model, the target picking mechanism and first target picking information of the first goods from a plurality of picking mechanisms of the picking and placing device based on the first image, attribute information of the first goods, picking prompt information corresponding to the first goods, and device parameters of each picking mechanism of the plurality of picking mechanisms.

27. The method of any one of claims 13-26, wherein, The first goods are provided with a target identification code, and the method further comprises: in a case where the target picking mechanism picks up the first goods, acquiring at least one fourth image including the first goods by the image acquisition device; in a case where the at least one fourth image includes the target identification code, identifying the target identification code; in a case where the target identification code is used to indicate the first goods, updating inventory information of the first goods; in a case where the target identification code cannot be identified, or the first goods cannot be determined based on the target identification code, marking the first goods as an abnormal sorting object, and returning the first goods to the target container, and determining a next to-be-picked goods in the target container.

28. The method of claim 27, further comprising: controlling the pick-and-place device to adjust an angle of the target picking mechanism if the fourth image does not include the target identification code; wherein the target picking mechanism is adjusted at a number of times not more than a second preset number of times; reacquiring, by the image acquisition device, at least one fifth image including the first item based on the adjusted target picking mechanism.

29. The method of any one of claims 13-28, further comprising: acquiring, by the image acquisition device, at least one sixth image including the first item after the target picking mechanism picks up the first item from the target container; determining that the target picking mechanism has a double-pick anomaly if it is determined, based on the at least one sixth image, that the target picking mechanism picks up a plurality of candidate objects; wherein the plurality of candidate objects includes the first item and at least one other object other than the first item; controlling the target picking mechanism to return the plurality of candidate objects picked up by the target picking mechanism to the target container and controlling the target picking mechanism to pick up the first item again based on the first target picking information if it is determined that the target picking mechanism has a double-pick anomaly.

30. The method of any one of claims 13-29, further comprising: acquiring, by a weight acquisition device corresponding to the first order container, weight information of the first order container after the target picking mechanism transfers the first item picked up to the first order container; wherein the target location includes a first order container corresponding to the first item; determining whether the target picking mechanism has a double-pick anomaly based on the weight information of the first order container and the weight information of the first item; labeling the first order container as an abnormal order container and controlling an output device to output an abnormal prompt information to indicate that a performing object performs an abnormal processing on the abnormal order container if it is determined that the target picking mechanism has a double-pick anomaly.

31. The method of any one of claims 1-30, wherein, The target location includes a first order container, and the method further comprises: acquiring a picking image after controlling the pick-and-place device to pick the first item from the target container; wherein the picking image includes an image corresponding to the target container, an image corresponding to the target location, and an image on a picking path of the first item; determining a drop-off location of the first item according to the picking image; determining that the first item has completed picking if the drop-off location of the first item is located in the first order container; controlling the pick-and-place device to pick the first item from the target container again and transfer the first item to the first order container if the drop-off location of the first item is located in the target container. In a case where the drop position of the first goods is located in a candidate container, it is determined that the first goods picking fails, the picking device is controlled to pick the first goods from the candidate container, and the first goods are transferred to the first order container; wherein the candidate container is any container on the picking path of the first goods.

32. The method of any one of claims 1-31, wherein, The picking device comprises a container picking mechanism; the image acquisition device is controlled to acquire a first image corresponding to the target container, which comprises: The container picking mechanism is controlled to carry the target container to a first docking position corresponding to a first order container on the seeding wall; wherein the first goods are goods required by the first order container; In the process of moving the container picking mechanism to the first docking position, the image acquisition device is controlled to acquire a first image corresponding to the target container.

33. The method of claim 32, wherein, The picking device further comprises a goods picking mechanism, the container picking mechanism and the goods picking mechanism are coupled, and the goods picking mechanism can move along the transverse direction or the vertical direction with the container picking mechanism; wherein the goods picking mechanism comprises a target picking mechanism; The target picking information is used to control the picking device to pick the first goods from the target container and transfer the first goods to a target position, which comprises: According to the target picking information, the goods picking mechanism is controlled to pick the first goods from the target container and transfer the first goods to the first order container; wherein the target position comprises a position corresponding to the first order container.

34. The method of claim 33, wherein, The goods picking mechanism is controlled to pick the first goods from the target container and transfer the first goods to the first order container, which comprises: The container picking mechanism is controlled to pull the first order container out of a first seeding position, so that the goods picking mechanism can deliver the first goods into the first order container; The goods picking mechanism is controlled to pick the first goods from the target container and deliver the first goods into the first order container; After the first goods are delivered into the first order container, the method further comprises: The container picking mechanism is controlled to push the first order container back to the first seeding position.

35. The method of claim 33 or 34, wherein, After the goods picking mechanism is controlled to pick the first goods from the target container and transfer the first goods to the first order container, the method further comprises: In a case where the target container contains third goods required by a second order container on the seeding wall, the container picking mechanism is controlled to carry the target container from the first docking position to a second docking position corresponding to the second order container; In the process of moving the container picking mechanism to the second docking position, the image acquisition device is controlled to acquire a seventh image corresponding to the target container.

36. The method of any one of claims 33-35, further comprising: acquiring a seeding state of each order container in a plurality of order containers on the seeding wall; In the case that a third order container with a sowing state of completed sowing exists on the sowing wall, the container taking and placing mechanism is controlled to move to a third docking position corresponding to the third order container, the third order container is taken out at the third docking position, and the third order container is transported to an idle bin of a first carrier parked at the first workstation or an idle buffer position of the first workstation.

37. The method of any one of claims 33-36, wherein, The container taking and placing mechanism is provided with a full-box detection component; after the first goods are picked from the target container and transferred into the first order container by the goods taking and placing mechanism, the method further comprises: obtaining the full-box state of the first order container detected by the full-box detection component; in the case that the full-box state of the first order container is a full storage state, the container taking and placing mechanism is controlled to take out the first order container from the sowing wall and transport the first order container to an idle bin of a first carrier parked at the first workstation or an idle buffer position of the first workstation.

38. A warehousing system characterized by, comprising: a first workstation comprising a taking and placing device; a control device configured to: obtain a to-be-processed order corresponding to the first workstation, determine a target container in which at least one to-be-picked good corresponding to the to-be-processed order is placed, and generate collection instructions according to the target container; an image collection device configured to: in the case that the target container is transported to the first workstation by a transportation device, collect a first image corresponding to the target container according to the collection instructions; the control device is configured to: determine a first good in the at least one to-be-picked good according to the first image, and in the case that the type of the first good is a first good type, determine first target picking information of the first good, and generate picking instructions according to the target picking information; wherein the first target picking information comprises a picking position and a picking direction; the taking and placing device is configured to: pick the first good from the target container and transfer the first good to a target position according to the picking instructions.

39. A goods picking apparatus, characterised in that, comprising: an obtaining module configured to: obtain a to-be-processed order corresponding to a first workstation, and determine a target container in which at least one to-be-picked good corresponding to the to-be-processed order is placed; a control module configured to: in the case that the target container is transported to the first workstation by a transportation device, control an image collection device to collect a first image corresponding to the target container; a determination module configured to: determine a first good in the at least one to-be-picked good according to the first image, and in the case that the type of the first good is a first good type, determine first target picking information of the first good; wherein the first target picking information comprises a picking position and a picking direction; the control module is configured to: control a taking and placing device of the first workstation to pick the first good from the target container and transfer the first good to a target position according to the target picking information.

40. An electronic device, comprising: comprising: one or more processors and a memory; the memory is configured to store one or more programs; wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-37.

41. A computer-readable storage medium, comprising: a computer program stored thereon, which when executed by a processor, performs the method of any of claims 1-37.

42. A computer program product, characterised in that, a computer program, which when executed by a processor, performs the method of any of claims 1-37.

Images