CN115311441A

CN115311441A - Automatic warehousing system, goods checking method and server

Info

Publication number: CN115311441A
Application number: CN202210863761.4A
Authority: CN
Inventors: 孙伟; 杨磊
Original assignee: Beijing Kuangshi Robot Technology Co Ltd
Current assignee: Beijing Kuangshi Robot Technology Co Ltd
Priority date: 2022-07-20
Filing date: 2022-07-20
Publication date: 2022-11-08

Abstract

The application relates to the technical field of warehousing and provides an automatic warehousing system, a goods checking method and a server. Wherein, the automated storage and retrieval system includes: the system comprises a server, carrying equipment and an inventory station, wherein the inventory station comprises a camera; the server is used for issuing the checking task to the carrying equipment; the carrying equipment is used for carrying the container to be checked to the checking station according to the checking task and carrying the container to be checked to the target position after the checking is finished; the camera of the checking station is used for acquiring a cargo image of the cargo to be checked; the server is further used for obtaining current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and obtaining a goods checking result, wherein the current goods information comprises first goods information, and the first goods information is obtained from goods images. The system can be used for fully automatically checking the goods, the checking result is reliable, the checking efficiency is high, the sustainability is strong, and the labor cost is greatly reduced.

Description

Automatic warehousing system, goods checking method and server

Technical Field

The application relates to the technical field of warehousing, in particular to an automatic warehousing system, a goods checking method and a server.

Background

The goods stored in the warehouse need to be checked regularly to confirm whether the actual goods information is consistent with the pre-recorded goods information. In the prior art, goods are generally taken out of a warehouse and manually checked, however, manual checking is easy to cause errors, checking efficiency is low, and the method is difficult to be sufficient for a large warehouse storing a large number of goods and has high cost.

Disclosure of Invention

An object of the embodiments of the present application is to provide an automatic warehousing system, a cargo checking method and a server, so as to improve the above technical problems.

In order to achieve the above purpose, the present application provides the following technical solutions:

in a first aspect, an embodiment of the present application provides an automatic warehousing system, including: the system comprises a server, carrying equipment and an inventory station, wherein the inventory station comprises a camera; the server is used for issuing an inventory task to the carrying equipment 2, wherein a container to be inventoried is designated in the inventory task, and goods in the container to be inventoried are goods to be inventoried; the carrying equipment is used for carrying the container to be checked to the checking station according to the checking task and carrying the container to be checked to a target position after the checking is finished; the camera of the checking station is used for acquiring the goods image of the goods to be checked; the server is further used for obtaining current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and obtaining goods checking results for the container to be checked; the current cargo information comprises first cargo information, and the first cargo information is the cargo information of the cargo to be checked, which is obtained from the cargo image.

When the automatic warehousing system is used for checking the goods, the server automatically issues a checking task and obtains a checking result based on the comparison of the goods information, the carrying equipment automatically carries the container to be checked into and out of the checking station, and the checking station automatically finishes the collection of the goods information based on the camera and the like. This automatic warehouse system can carry out full automatization's goods inventory promptly, and its inventory result is reliable, inventory efficiency is higher, sustainability is strong (can be in succession the inventory longer time), also can easily be competent at the inventory task to large-scale warehouse to still reduced the required cost of labor of inventory by a wide margin.

In an implementation manner of the first aspect, the inventory station further includes a processing device, and the processing device is configured to receive the cargo image sent by the camera, obtain the first cargo information from the cargo image, and send the current cargo information including the first cargo information to the server.

In the implementation mode, the processing equipment of the checking station extracts the first goods information by analyzing the goods image, so that the server does not need to analyze the goods image by itself to extract the first goods information, and the working pressure of the server is reduced.

In an implementation manner of the first aspect, the first goods information includes the number of the goods to be inventoried, and the processing device is configured to identify, from the goods image, a stacking manner of the goods to be inventoried in the container to be inventoried, and determine the number of the goods to be inventoried according to the stacking manner.

For many warehouses, especially some relatively standard warehouses, the stacking modes of the goods in the containers are not random, and usually one or more fixed stacking modes exist for each kind of goods, so that the implementation mode firstly identifies the stacking mode of the goods to be inventoried in the container to be inventoried, and then further deduces the quantity of the goods to be inventoried, thereby being beneficial to obtaining a relatively accurate quantity statistical result.

In an implementation manner of the first aspect, the processing device is further configured to obtain an actual size of the goods to be inventoried from the server, and check whether the number of the goods to be inventoried determined according to the stacking manner is correct according to the actual size of the goods to be inventoried.

Although the accuracy of the quantity of the goods to be checked is higher through identifying the stacking mode of the goods, the error condition exists in the algorithm, in the implementation mode, the quantity of the goods is checked for the second time according to the actual size of the goods to be checked, errors in the quantity of the goods can be found in time, and therefore corresponding measures are taken.

In one implementation manner of the first aspect, the inventory station further includes a first scanner, the first scanner is configured to scan an identification code of a surface of the cargo to be inventoried, obtain second cargo information of the cargo to be inventoried, and the processing device is configured to receive the second cargo information sent by the first scanner and send the current cargo information including the first cargo information and the second cargo information to the server.

In the implementation mode, the checking station can acquire the second cargo information of the cargo to be checked through the first scanner besides acquiring the cargo image of the cargo to be checked through the camera, so that the content of the current cargo information is richer, and the server can also perform more comprehensive checking according to the current cargo information.

In an implementation manner of the first aspect, the inventory station further includes a second barcode scanner, the second barcode scanner is configured to scan an identification code of a surface of a container arriving at the inventory station to obtain container information, and the processing device is configured to receive the container information sent by the second barcode scanner and obtain a result of determining whether the container arriving at the inventory station is correct according to the container information.

In the above implementation mode, the checking station confirms the containers arriving at the checking station by arranging the second code scanner, so that the correctness of cargo information acquisition is ensured.

In an implementation manner of the first aspect, the server is configured to receive the cargo image sent by the camera, and obtain the first cargo information from the cargo image.

In the implementation mode, the server analyzes the goods image to extract the first goods information, so that the checking station does not need to be provided with processing equipment, namely, the camera of the checking station can be directly in communication connection with the server, the design of the checking station is simplified, and the implementation cost is saved.

In one implementation manner of the first aspect, the inventory station includes a plurality of cameras disposed at different positions, wherein each camera is used for acquiring an image of one side of the goods to be inventory.

In the implementation mode, the images of different side faces of the goods to be checked are acquired through the plurality of cameras, so that the images contain richer visual information of the goods to be checked, and the information of the first goods extracted from the images is more accurate.

In one implementation manner of the first aspect, the goods to be inventoried are stacked in the goods to be inventoried container into a cuboid goods stack, and the inventoriing station comprises four cameras arranged corresponding to four side surfaces of the goods stack and a fifth camera arranged corresponding to an upper surface of the goods stack; the cargo image collected by the fifth camera is used for determining the number of the cargos to be checked, which are positioned on the top layer of the cargo pile.

In the implementation mode, the goods image that four preceding cameras were gathered can be used to confirm the stacking mode of goods in the goods heap, and then carry out the calculation of quantity of goods, but goods heap top layer is probably not piled with the goods (the goods heap is not strict cuboid promptly, there is the vacancy in the top layer probably because of reasons such as goods warehouse-out), make the quantity of goods that directly calculates at the top layer according to the stacking mode of goods inaccurate, thereby can shoot the goods heap from the angle of overlooking through setting up fifth camera, and then obtain the correct quantity of top layer goods.

In an implementation manner of the first aspect, the fifth camera is a camera with a depth information acquisition function, and the cargo image and the depth information acquired by the fifth camera are used together to determine the number of the cargo to be inventoried on the top layer of the cargo pile; or, the inventory station further comprises a sixth camera arranged corresponding to the upper surface of the goods stack, the sixth camera is used for collecting images of the upper surface of the goods stack from angles different from those of the fifth camera, and the fifth camera and the goods images collected by the sixth camera are jointly used for determining the number of the goods to be inventoried stacked on the top layer.

If the top goods have a vacancy, the goods image acquired by the fifth camera is displayed at the vacancy position and is actually the upper surface of the next layer of goods, and if the fifth camera is a common camera, due to the fact that depth information in the goods image acquired by the fifth camera is lost, when the algorithm identifies the goods image, if the conditions of poor light and the like are met, the next layer of goods are likely to be identified by mistake as the goods on the top layer, and the number statistics of the top goods is caused to be wrong. In the above implementation, either the fifth camera itself can collect depth information (for example, it is an RGB-D camera), or the fifth and sixth cameras for downward shooting (there is a parallax between the two, which is equivalent to the available depth information) are provided, so that the problem of top cargo quantity statistics error can be improved.

In one implementation of the first aspect, the inventory station is disposed in a warehouse at least one of: the warehouse comprises a warehouse entry station, a warehouse exit station, a goods position of a stereoscopic shelf of the warehouse and the inside of a hoisting machine of the warehouse.

In the implementation mode, the position of the checking station can be flexibly set according to requirements. In particular, if the inventory station is provided inside the shelf or on the elevator, the inventory can be completed without taking the goods out of the warehouse, the inventory efficiency is further improved, and the influence of the inventory on other operations (for example, the entry and exit) performed in the warehouse can also be reduced.

In an implementation manner of the first aspect, the target location includes an original cargo space of the container to be checked in the warehouse, or, if the checking result is abnormal, the target location includes an abnormal cargo space or an ex-warehouse station in the warehouse.

In the implementation mode, the container to be checked can be sent back to the original goods position after the checking is finished, and the non-inductive checking is realized. Or, the container to be checked with the abnormal goods in the container can be sent to the abnormal goods position, so that the subsequent uniform processing is convenient. Or the container to be checked with the abnormal goods in the container to be checked can be sent to the ex-warehouse station to be directly subjected to ex-warehouse processing.

In a second aspect, an embodiment of the present application provides a goods checking method, which is applied to a server in an automatic warehousing system, where the automatic warehousing system further includes a carrying device and a checking station, the checking station includes a camera, and the method includes: issuing an inventory task to the carrying equipment, wherein a container to be inventoried is designated in the inventory task, and the goods in the container to be inventoried are the goods to be inventoried; acquiring current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and acquiring goods checking results of the container to be checked; the current cargo information includes first cargo information, the first cargo information is the cargo information of the cargo to be inventoried, which is obtained from the cargo image of the cargo to be inventoried, and the cargo image is the image of the cargo to be inventoried, which is acquired by the camera of the inventoriing station after the carrying device carries the container to be inventoried to the inventoriing station.

In an implementation manner of the second aspect, the obtaining current cargo information of the cargo to be inventoried includes: receiving the current cargo information sent by the processing equipment of the checking station; the current cargo information comprises the first cargo information, and the first cargo information is obtained by the processing equipment from the cargo image acquired by the camera; or receiving the cargo image sent by the camera, and obtaining the first cargo information in the current cargo information from the cargo image.

In a third aspect, an embodiment of the present application provides a cargo inventory device, including: a server configured in an automated warehousing system, the automated warehousing system further comprising a carrier device and an inventory station, the inventory station comprising a camera, the apparatus comprising: the task issuing module is used for issuing an inventory task to the carrying equipment, wherein a container to be inventoried is designated in the inventory task, and goods in the container to be inventoried are goods to be inventoried; the goods checking module is used for acquiring current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and acquiring goods checking results of the container to be checked; the current cargo information includes first cargo information, the first cargo information is the cargo information of the cargo to be inventoried, which is obtained from the cargo image of the cargo to be inventoried, and the cargo image is the image of the cargo to be inventoried, which is acquired by the camera of the inventoriing station after the carrying device carries the container to be inventoried to the inventoriing station.

In a fourth aspect, an embodiment of the present application provides a server, including: a memory in which computer program instructions are stored, and a processor, wherein the computer program instructions, when read and executed by the processor, perform the method provided by the second aspect or any one of the possible implementations of the second aspect.

In a fifth aspect, the present application provides a computer-readable storage medium, where computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are read and executed by a processor, the computer program instructions perform the method provided by the second aspect or any one of the possible implementation manners of the second aspect.

In a sixth aspect, the present application provides a computer program product, which includes computer program instructions, and when the computer program instructions are read and executed by a processor, the computer program instructions perform the method provided by the second aspect or any one of the possible implementation manners of the second aspect.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.

FIG. 1 illustrates components that may be included in an automated warehousing system provided by embodiments of the present application;

FIG. 2 illustrates components that an inventory station provided by embodiments of the present application may contain;

fig. 3 illustrates a cargo inventory process that may be employed by the automated warehousing system provided by embodiments of the present application;

fig. 4 shows goods images collected when the inventory station includes five cameras and target detection results for goods to be inventoried;

fig. 5 shows a case where the stock-taking station is set in the warehousing station;

FIG. 6 shows the inventory station disposed on the cargo space of the shelf;

fig. 7 shows a possible structure of a server provided in an embodiment of the present application.

Detailed Description

With the development of Intelligent technologies such as internet of things, artificial intelligence and big data, the requirement for transformation and upgrading of the traditional Logistics industry by using the Intelligent technologies is stronger, and Intelligent Logistics (ILS for short) becomes a research hotspot in the Logistics field. The intelligent logistics utilizes artificial intelligence, big data, various information sensors, radio frequency identification technology, global Positioning System (GPS) and other Internet of things devices and technologies, is widely applied to basic activity links of material transportation, storage, delivery, packaging, loading and unloading, information service and the like, and realizes intelligent analysis and decision, automatic operation and high-efficiency optimization management in the material management process. The internet of things technology comprises sensing equipment, radio Frequency Identification (RFID for short), laser infrared scanning, infrared induction Identification and the like, the internet of things can effectively connect materials in logistics with a network, can monitor the materials in real time, can sense environmental data such as humidity and temperature of a warehouse, and guarantees the storage environment of the materials. All data in logistics can be sensed and collected through a big data technology, the data are uploaded to an information platform data layer, operations such as filtering, mining and analyzing are carried out on the data, and finally accurate data support is provided for business processes (such as links of transportation, warehousing, storing and taking, sorting, packaging, sorting, ex-warehouse, checking, distribution and the like). The application direction of artificial intelligence in logistics can be roughly divided into two types:

(1) The Artificial Intelligence (AI) technology is used to enable intelligent devices such as unmanned trucks, automated Guided Vehicles (AGVs), autonomous Mobile Robots (AMR), forklifts, shuttles, stackers, unmanned distribution vehicles, unmanned aerial vehicles, service Robots, mechanical arms, intelligent terminals and the like to replace part of manpower.

(2) The labor efficiency is improved through software systems such as a transportation equipment management system, a storage management system, an equipment scheduling system, an order distribution system and the like driven by technologies or algorithms such as computer vision, machine learning, operation optimization and the like. With the research and progress of intelligent logistics, the technology is applied to a plurality of fields, such as retail and electric commerce, electronic products, tobacco, medicine, industrial manufacturing, shoes and clothes, textile, food and the like.

The automatic warehousing system provided by the embodiment of the application also combines the technologies in the aspects of computer vision and artificial intelligence.

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.

The terms "first," "second," and the like, are used solely to distinguish one entity or action from another entity or action without necessarily being construed as indicating or implying any actual such relationship or order between such entities or actions.

Before formally introducing the scheme of the application, the storage mode of the goods is briefly explained: in the solution of the present application, the goods in the default warehouse are all stored in containers, for example, the containers may be pallets, and the goods are stacked in the pallets in some way. When the goods are checked, the checking is carried out according to the containers, namely whether the actual goods information in each container to be checked is consistent with the pre-recorded goods information or not is confirmed.

The goods in the same container may be the same kind of goods or not, but in many warehouses, the same kind of goods may be required to be stored in the same container as much as possible. The goods may be stacked in the container directly in their original form or may be stacked in the container after being appropriately packed, for example, one or more pieces of goods may be loaded into the box and then the box may be stacked in the container, in which case, one box may also be regarded as one piece of goods, and the inventory may be made by considering whether the stacked boxes in the container are identical to those in the record, regardless of the state of the goods inside the box.

Fig. 1 illustrates components that may be included in an automated warehousing system 100 provided by embodiments of the present application. Referring to fig. 1, the system includes at least (solid line box) a carrier 110, a server 130, and an inventory station 140, and in some implementations may include (dashed line box) one or more of a hoist 120, an warehousing station 150, a stereoscopic shelf 160, and an ex-warehousing station 170, which may have data interactions with each other as indicated by the arrows in fig. 1 (the arrows are not shown, and do not represent that there is necessarily no data interaction).

The automated warehousing system 100 may be deployed in a warehouse, but not necessarily every component, for example, the server 130 may be deployed remotely with the remaining components deployed in the warehouse. Note that although the embodiment of the present application mainly describes the application of the automatic warehousing system 100 in the aspect of automatic inventory of goods, it does not mean that the automatic warehousing system 100 only has a function of inventory of goods, and obviously, the automatic warehousing system may also have functions of a general warehousing system, such as warehousing of goods, delivery of goods, and management of goods.

The server 130 manages the overall operation of the automated warehousing system 100. For example, in one aspect, the server 130 may issue control instructions to other components to cause the components to perform respective tasks; alternatively, server 130 may receive information sent to it by the components, perform analysis, storage, etc. For example, for a cargo checking scenario, the server 130 may issue a checking task to the carrier device 110, and may also receive cargo information uploaded by the checking station 140, and obtain a cargo checking result according to the received cargo information.

Optionally, a Warehouse Management System (WMS for short) may be installed on the server 130 to implement the above-mentioned Warehouse Management function. Since the rest of the components mainly operate under the control of the server 130, that is, the components can be regarded as subordinate components of the server 130, the server 130 can also be referred to as an upper computer.

Fig. 7 shows a structure that may be inside the server 130 (in fig. 7, the server 300), which may be referred to as the following description about fig. 7.

The carrier 110 is generally referred to as a device having cargo handling capability, and there may be one or more carriers 110 in a warehouse. For example, the carrier 110 may transport containers filled with goods from one location to another in a warehouse under the control of the server 130, and the carrier may be an AGV, an AMR, a forklift, a stacker, a two-way shuttle, a four-way shuttle, or the like. For example, for the inventory scenario, the carrier device 110 may transport the container to be inventoried containing the inventory to the inventory station 140 from its storage location (e.g., in the stereoscopic shelf 160) according to the inventory task issued by the server 130, and transport the container to be inventoried away from the inventory station 140 after the inventory is completed. Of course, it is not excluded that some of the carriers 110 have more abundant functions, such as cargo picking functions.

The inventory station 140 is a facility for performing inventory, and of course, if the server 130 obtains the inventory result from the cargo information, it is regarded as a narrow cargo inventory operation, the inventory station 140 is at least used for collecting various information necessary for inventory, which includes at least a cargo image of the cargo to be inventory. The inventory station 140 includes at least a camera 141 for capturing images of the cargo to be inventoried, although in different implementations, the inventory station 140 may include other components, such as one or more of the processing device 143, the first scanner 142, and the second scanner 144 shown in fig. 2. In one warehouse, one or more inventory stations 140 may be provided, and if a plurality of inventory stations 140 are provided, the inventory stations 140 may operate simultaneously to improve inventory efficiency.

Referring to fig. 2, the camera 141 is a component that the inventory station 140 necessarily includes (a solid-line box, and an optional component is a dashed-line box), and if the inventory station 140 only includes the camera 141, the camera 141 may be connected to the server 130 (which refers to a wired or wireless communication connection, and the connection may be understood as follows), and after the camera 141 acquires the cargo image, the cargo image is sent to the server 130, and the server may perform cargo inventory based on the cargo image. The type of the camera 141 is not limited, and for example, an RGB camera, a grayscale camera, an RGB-D camera, an infrared camera, or the like may be used as needed. The number of the cameras 141 and the positions thereof are described in detail later.

In some implementations, the inventory station 140 also includes a processing device 143, and the camera 141 may be coupled to the processing device 143, and the processing device 143 may be coupled to the server 130. In an alternative, the processing device 143 only receives the cargo images collected by the camera 141 and forwards them to the server 130 (without excluding some processing before forwarding), and the server 130 may perform cargo inventory based on these cargo images; in another alternative, after receiving the image of the goods collected by the camera 141, the processing device 143 extracts the information of the goods (for example, the number of the goods to be checked) in the image, and then sends the extracted information of the goods to the server 130, and the server 130 can perform goods checking based on the information of the goods, in this alternative, the checking station 140 is not only used for information collection, but also can perform certain processing on the collected information. The processing device 143 may be a PC, an embedded device, or the like. The purpose of the processing apparatus 143 will be described in detail later.

In some implementations, the inventory station 140 further includes a first scanner 142, the first scanner 142 may be connected to the processing device 143, the first scanner 142 is configured to scan an identification code (e.g., a bar code, a two-dimensional code) of a surface of an item to be inventoried, send the scan result to the processing device 143, and forward it to the server 130 by the processing device 143 (without excluding certain processes from being performed before forwarding), or, if the inventory station 140 does not include the processing device 143, the first scanner 142 may be directly connected to the server 130 and send the scan result directly to the server 130. The scanning result either contains the goods information of the goods to be checked, or the server 130 can inquire the goods information of the goods to be checked according to the scanning result. The purpose of the first scanner 142 will be described in detail later.

In some implementations, the inventory station 140 further includes a second barcode scanner 144, the second barcode scanner 144 may be connected to the processing device 143, the second barcode scanner 144 is configured to scan an identification code (e.g., a barcode, a two-dimensional code) of a surface of a container to be inventory, send the scanning result to the processing device 143, and optionally, may be forwarded to the server 130 by the processing device 143 (without excluding some processing before forwarding), or, if the inventory station 140 does not include the processing device 143, the second barcode scanner 144 may be directly connected to the server 130, and send the scanning result directly to the server 130. The scanning result either contains the container information of the container to be inventoried, or the server 130 can query the container information of the container to be inventoried according to the scanning result. The purpose of the second scanner 144 will be described in more detail below.

The shelves 160 are multi-layer shelves, and can be used to realize the dense storage of goods, and one or more shelves 160 may be disposed in one warehouse. Each layer of the stereoscopic shelf 160 includes a plurality of cargo spaces, each of which can be used for receiving a container, and the containers can store goods. In some implementations, the travel area of the carrier 110 and the stereo rack 160 overlap, for example, tracks are laid inside the stereo rack 160, and a four-way shuttle can travel inside the rack by means of the tracks to carry goods (which can be carried by containers). Of course, in other implementations, the travel area of the carrier 110 and the sky shelves 160 are separate, for example, a stacker may travel on the ground between the sky shelves 160 and access and transport the goods on the sky shelves 160 (which may be accessed and transported in containers).

One or more hoists 120 may be provided in association with the shelf 160, the hoists 120 functioning like elevators and being capable of transferring the carrier 110 and/or the goods (which may be containers of goods) between different levels of the shelf 160. The lift 120 is typically usable with a carrier device 110 such as a four-way shuttle that may be operated within the space frame 160. A single elevator 120 may be unidirectional (e.g., only raising, only lowering), or bidirectional (e.g., both raising and lowering).

The lift 120 is an optional component, for example, if a ramp is provided between the tiers of the space frame 160 to allow the carrier 110 to move up and down, the lift 120 may be omitted. For example, the hoist 120 may not be provided for the carrier device 110 such as a stacker that has a cargo lifting function, and the hoist 120 may be provided for a device such as a four-way shuttle that does not have a cargo lifting function. The shelf 160 is also an optional component, for example, for some relatively simple warehouses, the goods (which may be containers with goods) may be directly stacked on the ground, and the shelf 160 is not needed, but the area on the ground can be divided as a goods space.

The warehousing site 150 is a facility for warehousing goods in a warehouse, and one or more warehousing sites 150 may be provided in one warehouse, and when a plurality of warehousing sites 150 are provided, these warehousing sites 150 may operate simultaneously. The warehousing station 150 may include a conveyor line extending to the vicinity of the stereoscopic shelf 160 (or, alternatively, a portion of the conveyor line may be considered as a warehousing station), and may further include one or more devices such as a robot arm, a scanner, and the like for warehousing goods. For example, when the goods are put in storage, the mechanical arm may put the goods (which may be containers filled with goods) onto the conveying line, the conveying line transports the goods to the vicinity of the stereoscopic shelf 160, and the carrying device 110 puts the goods into the cargo space.

The delivery site 170 is a facility for delivering goods in a warehouse, and one or more delivery sites 170 may be provided in one warehouse, and when a plurality of delivery sites 170 are provided, the delivery sites 170 may operate simultaneously. The outbound site 170 is similar to the inbound site 150 except that the flow of goods is reversed from the inbound site 150 and will not be described in detail.

In an alternative scheme, one warehousing site 150 and one ex-warehouse site 170 may also be integrated into one work site, that is, the work site may be used for warehousing goods or ex-warehouse goods.

Having briefly described the structure of the stocker system 100 and the structure of the inventory station 140, the basic process of the stocker system 100 for inventory will be described with reference to fig. 1 and 2:

step A: when the cargo inventory is needed, the server 130 issues an inventory task to the carrier 110.

When the inventory needs to be performed, the server 130 may decide itself (see the example in fig. 3), or the server 130 may decide to start the inventory according to the designation issued by the user.

The server 130 may issue the inventory task to all the vehicles 110 in the warehouse, or may issue the inventory task to only some of the vehicles 110, and each vehicle 110 performs similar operations after receiving the inventory task.

In the inventory task, containers to be inventoried need to be specified, the containers to be inventoried may be all or part of containers in a warehouse, and since the inventory manner of each container to be inventoried is the same, hereinafter, when referring to the containers to be inventoried, it may be understood as any one of the containers to be inventoried, and the goods actually stored in the container to be inventoried are the goods to be inventoried. By "specifying the container to be inventoried", it may be specifically specifying the position of the container to be inventoried in the warehouse, the identification of the container to be inventoried, and the like.

So-called inventory, the operations of which comprise at least: and through comparison, whether the current goods information of the goods to be checked is consistent with the inventory goods information corresponding to the container to be checked is confirmed. The goods information may be understood as the attribute of the goods, and the goods information may include one or more items of information such as the quantity and the category of the goods. The current cargo information of the cargo to be inventoried is the cargo information actually owned by the cargo to be inventoried determined by some technical means (e.g., image recognition, code scanning), and the inventory cargo information corresponding to the container to be inventoried is the cargo information owned by the cargo in the container to be inventoried recorded by the automated warehousing system 100 before the inventory is started, for example, the inventory cargo information may be recorded in a database or a file system, which may be the database or the file system of the server 130, or may be the database or the file system accessible by the server 130, and the inventory cargo information is updated by the server 130 when the cargo is entered and exited.

Ideally, the current goods information of the goods to be checked and the inventory goods information corresponding to the container to be checked should be identical, but actually, the inventory goods information and the actual change of the goods in the warehouse may not be completely synchronous, so that the two pieces of goods information are different from each other, and the normal operation of the warehouse is affected, which is the meaning of checking the goods. For example, the server 130 issues an instruction to the delivery site 170 to deliver one container of goods in the container a, but the delivery site 170 actually delivers two containers of goods in the container a due to improper operation, so that the actual number (number of containers) of goods in the container a is 1 less than the number recorded by the server 130, and the difference in the number can be found through goods inventory.

It should be noted that the above operation compared with the current cargo information of the cargo to be checked and the inventory cargo information corresponding to the container to be checked is only a basic operation of the cargo checking, and some other operations for confirming the cargo state, such as detecting whether the cargo is damaged or not, whether the surface is stained or not, etc., may also be performed in the process of the cargo checking, which will be further described later.

It should be understood that in addition to specifying the containers to be inventoried, other information may be included in the inventory task, such as one or more of the location of the inventory station 140 in the warehouse, the movement trajectory of the carrier device 110, and the like.

And B: the carrier 110 carries the container to be inventoried to the inventorying station 140 according to the inventorying task.

Depending on the inventory task, the carrier 110 may find the container to be inventoried in the warehouse, pick up the container to be inventoried, and transport it to the inventory station 140. This step belongs to the basic function of the carrier device 110 and will not be explained in detail.

And C: the camera 141 of the inventory station 140 collects images of the goods to be inventoried.

After the containers to be inventoried arrive at the inventory station 140, the camera 141 may capture an image of the goods to be inventoried, referred to as a goods image. If the inventory station includes processing device 143, camera 141 may be controlled by processing device 143 for image capture, and if the inventory station does not include processing device 143, camera 141 may be controlled by server 130 for image capture. The number and arrangement of the cameras 141 will be described in detail later. For one camera 141, the image of the cargo collected by it may be one frame or multiple frames.

If the inventory station 140 also includes other components, such as a first scanner 142 and a second scanner 144, the inventory station 140 may also collect other information (but not necessarily cargo information), although this is not discussed here for the sake of brevity.

Step D: the server 130 obtains current cargo information of the cargo to be checked, compares the current cargo information with inventory cargo information corresponding to the container to be checked, and obtains a cargo checking result for the container to be checked.

The current cargo information at least comprises first cargo information, and the first cargo information is cargo information of cargos to be checked, which is obtained from the cargo image. For example, the number of items to be inventoried is identified from the image of the items using an image recognition algorithm, and so on. It should be understood that the current cargo information may also include cargo information for the cargo to be inventoried obtained from other channels (e.g., code scanning), which will not be set forth herein for the time being.

There are different ways for the server 130 to obtain the current cargo information of the cargo to be checked, three of which are listed below:

mode 1: if the inventory station 140 does not include the processing device 143, the server 130 may directly receive the original information collected by other components of the inventory station 140, and obtain the current cargo information of the cargo to be inventoried according to the original information. For example, the server 130 may directly receive the cargo image of the cargo to be inventoried sent by the camera 141 and obtain the first cargo information from the cargo image, for example, the server 130 may deploy an image recognition algorithm, and recognize the quantity of the cargo to be inventoried from the cargo image by executing the image recognition algorithm.

Mode 2: if the inventory station 140 includes the processing device 143, the processing device 143 may receive the original information collected by other components of the inventory station 140, obtain the current goods information of the goods to be inventoried according to the original information, and send the current goods information of the goods to be inventoried to the server 130. For example, the processing device 143 may receive a cargo image of the cargo to be checked sent by the camera 141, and obtain the first cargo information from the cargo image, for example, an image recognition algorithm may be deployed on the processing device 143, and the number of the cargo to be checked is recognized from the cargo image by executing the image recognition algorithm. The processing device 143 may then send the number of items to be inventoried to the server 130 along with information on the items, if any, obtained from other channels.

It should be noted that the inventory station 140 includes the processing device 143, and it is not necessarily the case that the manner 2 is adopted, for example, the processing device 143 may be only used for collecting the original information collected by other components of the inventory station 140 and forwarding the original information to the server 130, and the server 130 further obtains the current cargo information of the cargo to be inventoried based on the original information, and this case may also be considered as the manner 1.

Mode 3: mode 3 is a combination of mode 1 and mode 2, where the inventory station 140 includes a processing device 143, the processing device 143 receives original information collected by other components of the inventory station 140, processes part of the original information, obtains goods information of corresponding goods to be inventoried, and sends the goods information to the server 130, and the remaining original information is directly forwarded to the server 130, and the server 130 obtains the goods information of corresponding goods to be inventoried based on the part of the original information, where the two parts of the goods information together are the current goods information of the goods to be inventoried.

Simple analysis modes 1 to 3, wherein in the mode 1, the server 130 automatically analyzes the original information collected by the checking station 140 to obtain the current cargo information of the cargo to be checked, so that the working pressure of the checking station 140 is reduced, and at this time, the checking station 140 does not need to be provided with the processing device 143, thereby not only simplifying the structure of the checking station 140, but also saving the implementation cost. Mode 2 is that the processing device 143 of the inventory station 140 obtains the current cargo information of the cargo to be inventoried by analyzing the raw information collected by other components of the inventory station 140 and sends it to the server 130, so that the work pressure of the server 130 can be reduced, and in some implementations, the amount of data interaction between the inventory station 140 and the server 130 can also be reduced (e.g., without sending the cargo image to the server 130). Mode 3 is more flexible and can choose whether to analyze the original information locally at the inventory station 140 (in the processing device 143) or at the server 130 according to the characteristics of different information.

The cargo information alignment is explained in step a and will not be described in detail. For the goods inventory result, one possible implementation is to include two results: one is normal, namely the current goods information of the goods to be checked is consistent with the inventory goods information corresponding to the container to be checked; one is abnormal, that is, the current goods information of the goods to be checked is inconsistent with the inventory goods information corresponding to the container to be checked. For a normal condition, the server 130 may not perform any processing, but directly execute the subsequent steps, and of course, may output a prompt message, and for an abnormal condition, the server 130 may output an alarm message according to a set rule, so that the user may know the abnormality found in the inventory in time and take corresponding measures.

And E, step E: after the inventory is completed, the carrier 110 carries the container to be inventoried to the target location.

After completing the inventory, the server 130 may issue a control command to the carrier 110 instructing it to transport the container to be inventoried to the target location. The target position may be set freely according to requirements, for example, one of the following positions may be selected:

(1) Original cargo space of container to be checked in warehouse

And the original goods space is the position for the carrying equipment to take out the container to be checked in the step B. If the position (1) is selected, after all the containers to be checked in the warehouse are checked, the containers to be checked are returned, so that the storage positions of the containers in the warehouse before and after the checking are not changed, and the condition that the checking is not carried out is realized, namely, the so-called 'non-inductive checking' is realized. Of course, at this time, the goods inventory result is already recorded by the server 130, and if there is an abnormality therein, the user may subsequently instruct the container with the abnormal inventory result to be taken out of the warehouse.

(2) Abnormal goods position in warehouse

The abnormal cargo space may be a cargo space dedicated to a container in the warehouse for storing the abnormal checking result, for example, the abnormal cargo space may be located in a specially opened area of the shelf 160 or a specially reserved floor, or the abnormal cargo space may be located in a specially set shelf 160, and so on. The containers with abnormal states are uniformly transported to the abnormal goods positions for storage, so that the containers can be uniformly processed in the follow-up process.

(3) Ex-warehouse site 150

If the position (3) is selected, the containers with abnormal states are indicated to be directly subjected to warehouse-out processing, so that the goods information of the remaining goods in the warehouse is consistent with the inventory goods information. Certainly, for goods delivered from a warehouse, the goods can be warehoused again at the later stage so as to re-input the corresponding goods information.

Note that for a particular one of the above three positions may be selected by the target location, but the selected target location is not necessarily the same for different ones of the inventory containers. For example, if the goods counting result corresponding to the to-be-counted container M is normal, it may be transported back to the original goods location, and if the goods counting result corresponding to the to-be-counted container N is abnormal, it may be transported to the abnormal goods location, and so on.

In one implementation, "after the inventory is completed" in step E does not necessarily mean that the server 130 obtains the inventory result, but may also mean that the inventory station 140 has collected necessary information (for example, an image of the goods) required for inventory, and at this time, although the inventory result is not obtained, the condition that the inventory result can be obtained is satisfied, and thus the server 130 may instruct the carrier device 110 to leave the inventory station 140.

According to the descriptions of the steps a to E, when the automatic warehousing system 100 performs inventory, the server 130 automatically issues an inventory task and obtains an inventory result based on cargo information comparison, the carrying device 110 automatically carries a container to be inventoried into and out of the inventory station 140, and the inventory station 140 automatically completes cargo information acquisition based on the camera 141 and the like, that is, the automatic warehousing system 100 can perform full-automatic inventory of the cargo, and has reliable inventory result, high inventory efficiency, strong sustainability (can perform continuous inventory for a long time), and the inventory task of a large warehouse can be easily performed, and the labor cost required by inventory is greatly reduced.

Some detailed implementations of the cargo inventory process are described below with reference to fig. 3, and steps S200 to S211 in fig. 3 can also be regarded as an implementation manner of steps a to E.

Step S200: the server 130 confirms that the current operating state of the automated warehousing system 100 is an idle state.

The current working state of the automated warehousing system 100 may refer to the working states of the components in the automated warehousing system 100, for example, if the current warehousing station 150 does not perform warehousing operation, the ex-warehouse station 170 does not perform ex-warehouse operation, and the carrying device 110 does not carry any goods, the current working state of the automated warehousing system 100 may be considered as an idle state, otherwise, the current working state of the automated warehousing system 100 may be considered as a busy state. Since the remaining components of the automated warehousing system 100 are operating under the control of the server 130, the operating states of these components are known to the server 130 sufficient to further determine the current operating state of the automated warehousing system 100.

In some implementations, the current operating status of the automated warehousing system 100 may also be determined according to the current time, for example, if a certain automated warehousing system 100 is set to stop operating from 20 pm to 8 pm every day.

If the current operating status of the automated warehousing system 100 is idle, the inventory may be performed (i.e., the subsequent steps are performed), otherwise the inventory is not performed. That is, the priority of inventory is relatively low, and thus may be re-performed when the automated warehousing system 100 is not performing other high priority tasks.

Step S200 is optional, for example, although the current working status of the automated warehouse system 100 is busy, the user may issue an order for checking the goods, terminate other tasks immediately, prioritize the checking of the goods, or perform other tasks and the checking of the goods simultaneously.

Step S201: the server 130 confirms the containers to be inventoried in the warehouse.

There are various implementation ways for step S201:

for example, the server 130 may determine all containers in the warehouse as containers to be inventoried, i.e., all goods in the inventory warehouse.

For another example, if the quantity of the goods in the warehouse is large and the checking needs to be completed in multiple times, the server 130 may determine the containers in the warehouse that have not been checked and the containers that have been checked but have moved past the position (the movement of the containers past the position indicates that the goods in the containers may change) as the containers to be checked. Note that even if it is determined that a container is to be checked, it does not mean that the container will participate in the checking at this time, for example, the automated warehouse system 100 only takes 1 hour per day to perform the checking, and if the container is not checked after the time, the checking will not be continued.

For another example, the server 130 may determine the containers to be inventoried according to the user's plan, for example, the first day of the user's plan is 1-10 shelves, the second day is 11-20 shelves, the server 130 may determine the containers in the 1-10 shelves as the containers to be inventoried on the first day, the second day is 11-20 shelves, and so on.

Step S202: the server 130 issues an inventory task to the carrier 110.

The content of step S202 is already described in the description of step a, and is not repeated.

Step S203: the carrier 110 carries the container to be inventoried to the inventorying station 140 according to the inventorying task.

The content of step S203 is already introduced when describing step B, and the description is not repeated.

Before describing steps S204 to S208, it should be noted that these steps are all based on the premise that the inventory station 140 includes the processing device 143, and in the case that the inventory station 140 does not include the processing device 143, they will be described appropriately during description.

Step S204: the inventory station 140 confirms that the arriving container is correct by scanning the code.

Step S204 is an optional step, and if the container arriving at the inventory station 140 is the intended inventory container to be processed (i.e., the arriving container is correct), the subsequent steps may be continued, if the container arriving at the inventory station 140 is not the intended inventory container to be processed (e.g., the intended inventory container a is currently processed and the arriving container is the intended inventory container B), or even not an inventory container at all (i.e., the arriving container is incorrect), the subsequent steps may not be performed, and the server 130 may also control the carrier device 110 currently located at the inventory station 140 to carry the incorrect container away from the inventory station 140.

To implement step S204, the second barcode scanner 144 is required to be disposed in the inventory station 140, and meanwhile, an identification code is also required to be disposed on the container, the second barcode scanner 144 may scan the identification code on the surface of the container arriving at the inventory station, obtain container information (e.g., ID of the container), and send the container information to the processing device 143, the processing device 143 may further send the container information to the server 130, the server 130 determines whether the container currently arriving at the inventory station 140 is correct according to the container information, after obtaining the determination result, the server 130 may return the determination result to the processing device 143, so that the processing device 143 may determine whether to continue to collect information (e.g., image of the goods) of the goods in the container according to the determination result, and the server 130 may also determine whether to control the carrying device 110 currently located at the inventory station 140 to leave the inventory station 140 according to the determination result. Alternatively, the server 130 may issue container information of a container to be checked, which is expected to be processed, to the processing device 143 first, after the processing device 143 obtains the container information of the container arriving at the checking station, the processing device 143 may locally determine whether the container arriving at the checking station is correct, and then determine whether to continue to collect information of goods in the container, and the processing device 143 may further send the determination result to the server 130, so that the server 130 may determine whether to control the carrying device 110 currently located at the checking station 140 to leave the checking station 140 according to the determination result.

If the processing device 143 is not disposed in the inventory station 140, the second scanner 144 may directly transmit the container information to the server 130, and the server 130 may determine whether the container currently arriving at the inventory station 140 is correct according to the container information, and determine whether to control the inventory station 140 to continue collecting information of the goods in the container according to the determination result, and whether to control the carrier device 110 currently located at the inventory station 140 to leave the inventory station 140.

In step S204, the inventory station 140 confirms the containers arriving at the inventory station by setting the second barcode scanner 144, so as to ensure the correctness of the cargo information collection.

Step S205: the inventory station 140 collects images of the items to be inventoried.

Step S206: the inventory station 140 obtains first cargo information of the cargo to be inventoried from the cargo image.

Steps S205 and S206 are explained together. The camera 141 of the inventory station 140 collects an image of the goods to be inventoried and sends the image to the processing device 143, and the processing device 143 may deploy an image recognition algorithm, where the algorithm may analyze the image of the goods to obtain first information of the goods to be inventoried, such as one or more information of the number and the type of the goods to be inventoried. The image recognition algorithm may be a conventional algorithm or an artificial intelligence algorithm (e.g., a deep learning algorithm).

For example, if the first item information includes the number of items to be inventoried, one possible image recognition algorithm includes:

step a: and identifying the stacking mode of the goods to be checked in the container to be checked from the goods image.

So-called stacking mode, namely how goods are placed in the container, a reasonable stacking mode can enable the container to contain more goods, and therefore goods storage space is saved. For example, for a 3 × 2 × 5 box stack, the stacking may be: 3 boxes are placed in the x direction of each layer, 2 boxes are placed in the y direction, and 5 layers of boxes are stacked.

The step a has a plurality of implementation modes: for example, the goods image can be input into a trained neural network model, and the model can directly predict the stacking mode of goods to be inventoried; for another example, the image of the goods may be processed by using a target detection algorithm to obtain the position of the goods to be checked in the image (for example, the white box in fig. 4 is the detected position of the box), and then the stacking manner of the goods to be checked in the image of the goods may be determined according to the position of the goods to be checked and by combining a preset rule, and the like.

Step b: and determining the quantity of the goods to be checked according to the stacking mode of the goods to be checked.

A conversion relation exists between the stacking mode of the goods to be checked and the quantity of the goods to be checked, for example, the step a determines that the stacking mode of the box is as follows: if 3 boxes are placed in the x direction of each layer, 4 boxes are placed in the y direction, and 5 layers of boxes are stacked, the total number of boxes (i.e., the number of goods to be inventoried) can be calculated to be 3 × 4 × 5= 60. Of course, there may be instances where the topmost box is not full, as will be described later.

For many warehouses, especially some relatively standard warehouses, the stacking modes of the goods in the containers are not random, and usually one or more fixed stacking modes exist for each kind of goods (meaning a specific shape and size), so that the difficulty in identifying the stacking mode of the goods to be inventorized in the step a is not high, if an artificial intelligence algorithm is combined, a relatively accurate result is easily obtained, the number of the goods to be inventorized is further estimated in the step b, and a relatively accurate number statistical result is easily obtained.

Further, although the accuracy of determining the number of the goods to be inventoried by identifying the stacking manner of the goods is high, since the algorithm is adopted, the error condition is inevitable. Therefore, in an alternative, the processing device 143 may further obtain the actual size of the goods to be inventoried from the server 130, and check whether the number of the goods to be inventoried determined according to the stacking manner is correct according to the actual size of the goods to be inventoried, that is, check the number of the goods and then check the number of the goods secondarily. This is beneficial to find errors in the statistics of the quantity of the goods in time, and then take corresponding measures, such as re-collecting images of the goods, re-calculating the quantity of the goods and the like.

The actual size of the goods to be checked may be that the server 130 actively issues the goods to be checked to the processing device 143, for example, the server 130 may actively notify the processing device 143 that the size of the box in the container to be checked is 100 × 50 × 20cm. The actual size of the goods to be inventoried may also be requested by the processing device 143 from the server 130, for example, after the container information of the container to be inventoried is obtained (which may be obtained in step S204), the processing device 143 sends the container information to the server 130 (if the container information is sent in step S204, there is no need to send repeatedly), the server 130 may return the actual size of the goods in the container corresponding to the container information, and so on.

The rules for checking the number of the goods to be inventoried are not limited, for example, the total size of the goods to be inventoried cannot exceed the size of the container, which is determined by combining the stacking mode of the goods to be inventoried, the actual size of the goods to be inventoried and the actual size of the container. Wherein the actual size of the container may be stored in the processing device 143 in advance or may be obtained from the server 130 together with the actual size of the goods to be inventoried. For example, the length of the container in the x direction is 280cm, the length of the box in the x direction is 100cm, the identified stacking manner is that 3 boxes are placed in the x direction, but the total length of the 3 boxes in the x direction is 300cm, and exceeds 280cm, which indicates that the identified stacking manner of the boxes is wrong, and the boxes may need to be identified again.

In the inventory station 140, there may be one or more cameras 141, and if the inventory station 140 includes a plurality of cameras 141, the cameras 141 may be disposed at different positions, wherein each camera 141 is used to capture an image of one side of the goods to be inventory. The cameras 141 with more quantity are beneficial to obtaining more goods images, the images contain more visual information of goods to be checked, and then the first goods information extracted from the images is more accurate.

For example, if the goods to be inventoried are stacked in a rectangular parallelepiped goods pile (generally, rectangular, such as a top layer may not be full) in the container to be inventoried, the inventoriing station may be provided with four cameras corresponding to four sides of the goods pile, each camera collects an image of one side of the goods pile, and in combination with the images, the processing device 143 may identify the stacking manner of the goods in the goods pile, and further determine the quantity of the goods in the goods pile.

However, it should be noted that the number of the cameras 141 and the number of the sides of the goods stack are not necessarily the same, for example, although the goods stack has four sides, only one camera 141 may be provided, if the goods stacking mode in the warehouse is less, it is likely that the goods image of one side can also distinguish different goods stacking modes, and further obtain the number of the goods to be inventoried.

Further, it is highly probable that the top layer of the goods stack is not full of goods (non-top layer is full of goods generally because more goods are stacked in a container as much as possible), for example, theoretically, a goods stack should be stacked with 3 × 4 × 5 boxes, but actually, only 11 boxes are stacked on the top layer, and an empty space exists. There are various reasons for this, such as the total amount of goods is not enough to be fully piled, or the goods pile is fully piled, but the top layer is not fully piled due to the warehouse-out later, etc.

For the situation that the goods are not fully stacked on the top layer of the goods stack, if only four cameras are arranged to collect the goods images and calculate the number of the goods to be checked, an error is likely to occur, for example, if the vacant position of the top layer of the goods stack is exactly located in the center of the top layer, that is, the vacant position cannot be directly observed from the goods images on four sides, the number of the goods to be checked calculated by the processing device 143 will exceed the number of the actual goods.

To solve this problem, a fifth camera 141 may be further disposed in the inventory station 140 corresponding to the top surface of the cargo pile, and the camera 141 is configured to capture an image of the top surface of the cargo pile from a top view angle, and obviously, in the overhead cargo image, the existence of a vacancy in the top layer of the cargo pile may be recognized by an algorithm, so that the number of the cargo to be inventory located on the top layer of the cargo pile may be more accurately determined according to the cargo image (and possibly also in combination with the lateral cargo image). As for the other layers of the stack, since the stack is full, the number of the items to be inventoried contained in the stack can be calculated based on only the side images of the stack.

Fig. 4 shows the goods image collected when the inventory station 140 includes five cameras 141 and the target detection result for the goods to be inventoried. The center is the top view of the stack, the four sides are the images of the stack, and the white frame is the detected box position.

Further, if there is a gap in the cargo on the top of the cargo stack, the cargo image collected by the fifth camera 141 is actually the upper surface of the cargo on the next floor displayed at the gap position, and if the fifth camera 141 is a common camera (for example, an RGB camera or a grayscale camera), because depth information in the collected cargo image is lost, when the algorithm identifies the cargo image, if there are poor light and other conditions, the cargo on the next floor is likely to be mistakenly identified as the cargo on the top floor, that is, the position is considered to have no gap, and further the number statistics of the cargo on the top floor is wrong.

To solve this problem, the fifth camera 141 may adopt a camera (e.g., RGB-D camera) with a depth information collecting function, and such a camera may collect depth information (e.g., depth image) of a shooting scene in addition to a visual cargo image (e.g., RGB image, gray image), obviously, it is easy to identify whether a certain position of the top layer of the cargo pile is empty according to the depth information, and then, in combination with the visual cargo image, an algorithm may obtain a more accurate statistical result of the number of the top cargo. Optionally, the fifth camera 141 may still adopt a common camera, and the inventory station 140 may further separately set a depth camera corresponding to the upper surface of the cargo pile for collecting depth information, which is substantially similar to the scheme in which the fifth camera 141 collects depth information by itself and will not be described again.

As an alternative, the fifth camera 141 may still adopt a common camera, but the inventory station 140 may further set a camera 141 corresponding to the upper surface of the cargo pile, that is, the sixth camera 141, where the sixth camera 141 and the fifth camera 141 are located at different positions, and may collect images of the upper surface of the cargo pile from different angles, and because there is a parallax between the sixth camera 141 and the fifth camera 141, the depth information is included in cargo images collected by the sixth camera 141 and the fifth camera 141, so that the number of cargo to be inventoried on at the top layer of the cargo pile may be determined more accurately by combining the cargo images collected by the sixth camera 141 and the fifth camera 141 (or by combining the cargo images of the sides). Optionally, more cameras 141 may be disposed corresponding to the upper surface of the cargo stack, and the effect is similar to that of two cameras 141, and will not be repeated.

Optionally, the processing device 143 may also perform some sort of inventory operation based on the cargo image that does not involve a comparison (i.e., a comparison of the current cargo information with the inventory cargo information). For example, an algorithm deployed on the processing device 143 may detect whether a stain or damage is present on the surface of the cargo to be inventoried, and obtain a corresponding detection result, which may also be considered as part of the inventory result.

Step S207: the checking station 140 obtains second cargo information of the cargo to be checked through the code scanning.

Step S207 is an optional step, in order to implement step S207, the first scanner 142 needs to be arranged in the checking station 140, and meanwhile, an identification code needs to be arranged on the goods to be checked, and the first scanner 142 may scan the identification code on the surface of the goods to be checked, obtain the second goods information of the goods to be checked, and send the second goods information to the processing device 143. The second goods information may have different contents from the first goods information, for example, the first goods information may be the number of goods to be inventoried, the second goods information may be the kind of goods to be inventoried (for example, SKU information), and the like.

Alternatively, the scanning result may not include the second item information, for example, the scanning result is only the ID of the item to be checked, the processing device 143 needs to send the ID to the server 130, and the server 130 queries the second item information of the item to be checked from the database or the file system according to the ID. However, for simplicity, the case where the scanning result includes the second item information is hereinafter exemplified.

To the condition that the goods that wait to check are piled up the goods and are piled up in waiting to check the container, probably have all surfaces of some goods to submerge in the goods is piled up, obviously, can't sweep the sign indicating number to this kind of goods, but if the warehouse only deposits a goods in requiring every container, it is also not necessary to sweep the sign indicating number to every goods that wait to check in fact, only need scan arbitrary goods that wait to check the sign indicating number on surface can.

In step S207, the checking station 140 collects the second goods information of the goods to be checked by setting the first code scanner 142, so that the content of the current goods information of the goods to be checked is richer, and then the server 130 can perform more comprehensive checking according to the current goods information.

Alternatively, the identification code of the surface of the goods to be checked may be directly recognized by using an image recognition algorithm to obtain the second goods information, or the second goods information at this time may be considered as a part of the first goods information (because all the second goods information are recognized from the goods image).

Step S208: the inventory station 140 sends current cargo information for the cargo to be inventoried to the server 130.

The inventory station 140 may send current cargo information of the cargo to be inventoried to the server 130 through the processing device 143, where the current cargo information includes at least first cargo information and optionally second cargo information.

Alternatively, if the checking station 140 performs a part of the checking operation (for example, detecting stains or damages of the goods), the part of the goods checking result may also be sent to the server 130.

Optionally, the inventory station 140 may also send the cargo images to the server 130, although according to the implementation of fig. 3, the server 130 is not required to identify the contents of the cargo images, but these cargo images may be used for other purposes, which are explained in step S210.

Step S209: the server 130 performs inventory according to the current cargo information of the cargo to be inventory.

The server 130 may compare the current cargo information of the cargo to be checked with the inventory cargo information corresponding to the container to be checked, so as to obtain a cargo checking result. For example, if the current cargo information includes the number of the cargo to be checked (first cargo information) and the type of the cargo to be checked (second cargo information), the number of the cargo to be checked in the current cargo information may be compared with the number of the cargo in the inventory cargo information, the type of the cargo to be checked in the current cargo information may be compared with the type of the cargo in the inventory cargo information, and a cargo checking result may be obtained by combining a specific rule. For example, one possible rule is that if any one of the two comparison results is inconsistent, the goods inventory result is abnormal, otherwise it is normal. Of course, the two comparison results can be directly used as the goods checking result.

If the checking station 140 further sends the partial goods checking result (e.g., the stain and damage detection result) to the server 130 in step S208, the server 130 may obtain a final goods checking result (normal or abnormal) by combining the partial goods checking result, the information comparison result, and the specific rule.

For the case that the processing device 143 is not provided in the inventory station 140, the arrangement manner of the middle camera 141 in steps S205 to S206 is similar, except that the camera 141 may be directly connected to the server 130, and sends the acquired cargo image to the server 130, and the server 130 may also obtain the first cargo information of the cargo to be inventoried by using the image recognition algorithm mentioned in steps S205 to S206, and perform the subsequent information comparison step. Further, the server 130 may detect stains, damages, etc. on the cargo surface from the cargo image as part of the cargo inventory result. If the inventory station 140 further includes a first scanner 142, the first scanner 142 may also be directly connected to the server 130 to send the second cargo information directly to the server 130 for inventory.

Step S210: the server 130 generates alarm information according to the goods checking result.

The server 130 may output alarm information according to the goods checking result and the set rule, so that the user can timely learn the abnormality found in the checking and take corresponding measures. For example, one possible rule is to output an alarm message whenever the goods inventory result is abnormal; for another example, another possible rule is to further determine the severity of the abnormality when the cargo inventory result is abnormal, and output an alarm message only when the abnormality is serious (for example, a small amount of dirt on the surface of the cargo can be ignored).

Optionally, if the inventory station 140 sends the goods image to the server 130 in step S208, the server 130 may output the corresponding goods image while outputting the alarm information, so that the user may confirm the inventory result. Because the current goods checking result is automatically obtained by a machine, the result is difficult to ensure 100% accuracy due to certain limitations of the algorithm, so that a user can confirm the result manually to ensure the reliability of the checking result. On the other hand, since the abnormal counting result is not always present, the burden of the user is not excessively increased.

Step S211: the server 130 instructs the carrier device 110 to transport the container to be inventoried to the target location.

Step S211 is similar to step E above and will not be repeated. However, it should be noted that if the target location is the original goods location of the container to be checked in the warehouse, and then the target location is matched with the appropriate checking time, then "non-sensitive checking" may be achieved, for example, checking is performed when no person is in the warehouse at night, each container is returned after the checking is completed, and after the next day worker gets on work, the goods cannot be sensed and checked unless checking the checking record in the server 130, which may improve the user experience and avoid the user from generating doubt to the behavior of the automatic warehousing system 100 (for example, the storage location of the goods is changed without knowing). Of course, even if the "non-sensory checking" is adopted, the abnormality generated during the checking process is recorded, and the user can subsequently perform targeted measures such as shipment of goods on the server 130 according to the abnormality.

Continuing with the description of possible locations for the inventory station 140, based on the above embodiment, the inventory station 140 may be located in at least one of the following locations in the warehouse:

(1) In the warehouse site 150

(2) In the delivery station 170

(3) On the goods position of the three-dimensional goods shelf 160

(4) On the hoisting machine 120

Fig. 5 shows a case where the inventory station 140 is provided in the warehousing station 150. Referring to fig. 5, the inventory station 140 includes 5 cameras 141, which are numbered 1 to 5 respectively and are shown as gray triangles/rectangles, wherein the cameras 141 No. 1 to 4 are used for collecting images of the side surfaces of the goods piles, the camera 141 No. 5 is used for collecting images of the upper surfaces of the goods piles, and each camera 141 is disposed above the conveying line (the portion between the two black horizontal lines) of the warehousing station 150.

With reference to fig. 5, one possible cargo inventory process is: the carrying device 110 takes out the to-be-checked container from the original goods position of the three-dimensional shelf 160, carries the to-be-checked container to the conveying line, operates the conveying line until the to-be-checked container is conveyed to the position below the checking station 140, stops conveying as shown in fig. 5, collects the goods image of the to-be-checked goods by the camera 141 of the checking station 140 and checks the goods image, operates the conveying line again after the checking is finished, conveys the to-be-checked container to the vicinity of the three-dimensional shelf 160, takes the to-be-checked container from the conveying line by the carrying device 110, and carries the to-be-checked container back to the original goods position on the three-dimensional shelf 160 for storage.

If the checking station 140 is disposed at the warehousing station 150, the checking can be completed by using the goods transportation capability of the warehousing station 150, thereby reducing the checking cost and improving the checking efficiency.

In the case where the inventory station 140 is provided in the outbound station 170, similar to the inbound station 150, the description will not be repeated.

Fig. 6 shows a case where the inventory station 140 is disposed on the cargo space of the shelf 160. Referring to fig. 6, the inventory station 140 includes 5 cameras 141, numbered 1-5 respectively, shown as gray triangles/rectangles, wherein cameras 1-4 141 are used to capture images of the sides of the stack of goods, and camera 5 is used to capture images of the top surface of the stack of goods, and each camera 141 may be positioned above the cargo space (black squares). Alternatively, one inventory station 140 may occupy a plurality of adjacent cargo spaces, such as in fig. 6, the inventory station 140 occupies five adjacent cargo spaces arranged in a cross, wherein each cargo space is provided with a camera 141, thereby facilitating the image of the cargo to be inventoried from a better perspective.

If the inventory station 140 is located at a cargo space of the shelf 160, the carrier device 110 may employ a two-way or four-way shuttle, such as a four-way shuttle, which can move directly inside the shelf 160 to conveniently move containers from one cargo space to another cargo space, and thus is well suited for performing inventory within the shelf 160. With continued reference to fig. 6, the shelf 160 may include a trunk for a four-way shuttle, where no cargo space is provided, and the four-way shuttle may move along the trunk to a location near the container before moving to the cargo space to pick up the container.

With reference to fig. 6, one possible cargo inventory process is: the carrying device 110 takes the container to be checked out from the original position of the stereoscopic shelf 160 and carries the container to be checked out to the position under the checking station 140, as shown in fig. 6, the camera 141 of the checking station 140 can then collect the image of the goods to be checked out and perform checking, and after the checking is completed, the carrying device 110 carries the container to be checked out back to the original position.

Optionally, an inventory station 140 may be disposed on each floor of the shelf 160, so that each floor can independently and concurrently complete inventory (in the case of sufficient vehicles 110), and the vehicles 110 do not need to move across floors, which is more efficient.

For the case where the inventory station 140 is provided on the hoist 120, reference may be made to the arrangement of the camera 141 in fig. 5 and 6. One or more inventory stations 140 may be provided within a single lift 120, and if multiple inventory stations 140 are provided, the inventory stations may be located at different levels of the lift 120. After the carrying device 110 transports the container to be inventoried to the elevator 120, the carrying device 110 (on which the container to be inventoried is carried) or the container to be inventoried may be transported by the elevator 120 from the floor where the original cargo space is located to the floor where the inventory station 140 is disposed according to the inventory requirement for image or other information acquisition, and transport it back to the floor where the original cargo space is located after the inventory is completed, so that the inventory process becomes flexible.

The location of the inventory station 140 may be flexibly set as desired in practice. In particular, if the inventory station 140 is disposed inside the stereoscopic shelf 160 or on the lift 120, the inventory can be completed without taking the goods out of the shelf (here, the taking out can be understood as moving away from the shelf rather than leaving the warehouse), the inventory efficiency is high, and the influence of the inventory on other operations (e.g., entering and exiting the warehouse) performed in the warehouse can also be reduced.

The embodiment of the application also provides a goods checking method, which is applied to a server in an automatic warehousing system, wherein a possible structure of the server can be shown in fig. 7. The goods inventory method comprises the following steps:

issuing an inventory task to the carrying equipment, wherein a container to be inventoried is designated in the inventory task, and the goods in the container to be inventoried are the goods to be inventoried;

acquiring current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and acquiring goods checking results of the container to be checked; the current cargo information includes first cargo information, the first cargo information is the cargo information of the cargo to be inventoryed, which is obtained from the cargo image of the cargo to be inventoryed, and the cargo image is the image of the cargo to be inventoryed, which is acquired by the camera of the inventorying station after the container to be inventoryed is transported to the inventorying station by the carrying device.

In an implementation manner of the cargo checking method, the obtaining current cargo information of the cargo to be checked includes: receiving the current cargo information sent by the processing equipment of the checking station; the current cargo information comprises the first cargo information, and the first cargo information is obtained by the processing equipment from the cargo image acquired by the camera; or receiving the cargo image sent by the camera, and obtaining the first cargo information in the current cargo information from the cargo image.

The implementation principle and the generated technical effects of the cargo inventory method provided by the embodiment of the application are introduced in the foregoing system embodiment, and for brief description, parts of the method embodiment that are not mentioned may refer to corresponding contents in the system embodiment.

The embodiment of the present application further provides a cargo checking device, where the cargo checking device may be configured in a server of an automatic warehousing system, and a possible structure of the server may refer to fig. 7, the automatic warehousing system further includes a carrying device and a checking station, and the checking station includes a camera. This goods device of checing includes:

the task issuing module is used for issuing an inventory task to the carrying equipment, wherein a container to be inventoried is appointed in the inventory task, and goods in the container to be inventoried are goods to be inventoried;

the goods counting module is used for acquiring current goods information of the goods to be counted, comparing the current goods information with inventory goods information corresponding to the container to be counted and acquiring goods counting results aiming at the container to be counted; the current cargo information includes first cargo information, the first cargo information is the cargo information of the cargo to be inventoried, which is obtained from the cargo image of the cargo to be inventoried, and the cargo image is the image of the cargo to be inventoried, which is acquired by the camera of the inventoriing station after the carrying device carries the container to be inventoried to the inventoriing station.

In an implementation manner of the cargo checking device, the acquiring, by the cargo checking module, current cargo information of the cargo to be checked includes: receiving the current cargo information sent by the processing equipment of the checking station; the current cargo information comprises the first cargo information, and the first cargo information is obtained by the processing equipment from the cargo image acquired by the camera; or receiving the cargo image sent by the camera, and obtaining the first cargo information in the current cargo information from the cargo image.

The cargo inventory device provided by the embodiment of the present application can be used for executing the cargo inventory method provided by the embodiment of the present application, and the implementation principle and the resulting technical effects thereof have been introduced in the foregoing method embodiments, and for the sake of brief description, reference may be made to corresponding contents in the method embodiments where no part is mentioned in the embodiment of the device.

Fig. 7 illustrates a possible structure of the server 300 provided in the embodiment of the present application. Referring to fig. 7, the server 300 includes: a processor 310, a memory 320, and a communication interface 330, which are interconnected and in communication with each other via a communication bus 340 and/or other form of connection mechanism (not shown).

The processor 310 includes one or more (only one is shown), which may be an integrated circuit chip having signal processing capability. The Processor 410 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Micro Control Unit (MCU), a Network Processor (NP), or other conventional processors; the Processor may also be a dedicated Processor, including a Graphics Processing Unit (GPU), a Neural-Network Processing Unit (NPU), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, and a discrete hardware component. Also, when there are a plurality of processors 310, some of them may be general-purpose processors, and the other may be special-purpose processors.

The Memory 320 includes one or more (Only one is shown in the figure), which may be, but not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an electrically Erasable Programmable Read-Only Memory (EEPROM), and the like.

The processor 310, as well as possibly other components, may access, read from, and/or write to the memory 320. In particular, one or more computer program instructions may be stored in the memory 320, and may be read and executed by the processor 310 to implement the cargo inventory method provided by the embodiments of the present application.

The communication interface 330 includes one or more (only one shown) that may be used to communicate directly or indirectly with other devices (e.g., carrier devices, inventory stations, etc.) for interaction with data. Communication interface 330 may include an interface to communicate wired and/or wireless.

It will be appreciated that the configuration shown in fig. 7 is merely illustrative and that the server 300 may include more or fewer components than shown in fig. 7 or have a different configuration than shown in fig. 7. The components shown in fig. 7 may be implemented in hardware, software, or a combination thereof. The server 300 may be a physical device or a virtual device, and the server 300 is not limited to a single server, and may be a combination of a plurality of servers or a cluster formed by a large number of servers.

In addition, it should be noted that, in the solution of the present application, the server 300 should be understood as a device capable of providing a service to the outside in a functional sense, and should not be understood as only a device specifically used as a server, for example, a common PC may also be used as the server 300 when a warehouse management system is installed.

Embodiments of the present application further provide a computer-readable storage medium, where computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are read and executed by a processor of a computer, the computer-readable storage medium performs inventory of goods provided in embodiments of the present application. For example, the computer-readable storage medium may be implemented as memory 320 in server 300 in FIG. 7.

Embodiments of the present application further provide a computer program product, which includes computer program instructions, and when the computer program instructions are read and executed by a processor, the depth estimation method and/or the target detection method provided by the embodiments of the present application are performed.

The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An automated warehousing system, comprising: the system comprises a server, carrying equipment and an inventory station, wherein the inventory station comprises a camera;

the server is used for issuing an inventory task to the carrying equipment, wherein a container to be inventoried is designated in the inventory task, and goods in the container to be inventoried are goods to be inventoried;

the carrying equipment is used for carrying the container to be checked to the checking station according to the checking task and carrying the container to be checked to a target position after the checking is finished;

the camera of the checking station is used for acquiring a cargo image of the cargo to be checked;

the server is further used for obtaining current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and obtaining goods checking results for the container to be checked; the current cargo information comprises first cargo information, and the first cargo information is the cargo information of the cargo to be checked, which is obtained from the cargo image.

2. The automated warehousing system of claim 1, wherein the inventory station further comprises a processing device, the processing device is configured to receive the cargo image sent by the camera, obtain the first cargo information from the cargo image, and send the current cargo information including the first cargo information to the server.

3. The automated warehousing system of claim 2, wherein the first goods information comprises the quantity of the goods to be inventoried, and the processing device is configured to identify a stacking manner of the goods to be inventoried in the container to be inventoried from the goods image and determine the quantity of the goods to be inventoried according to the stacking manner.

4. The automated warehousing system of claim 3, wherein the processing device is further configured to obtain the actual size of the goods to be inventoried from the server and verify whether the quantity of the goods to be inventoried determined from the stacking method is correct according to the actual size of the goods to be inventoried.

5. The automated warehousing system of any of claims 2-4, wherein the inventory station further comprises a first scanner for scanning an identification code of a surface of the goods to be inventoried to obtain second goods information of the goods to be inventoried, and the processing device is configured to receive the second goods information sent by the first scanner and send the current goods information including the first goods information and the second goods information to the server.

6. The automated warehousing system according to any of claims 2-5, wherein the inventory station further comprises a second scanner for scanning an identification code of a surface of a container arriving at the inventory station to obtain container information, and the processing device is configured to receive the container information sent by the second scanner and obtain a determination result whether the container arriving at the inventory station is correct according to the container information.

7. The automated warehousing system of claim 1, wherein the server is configured to receive the cargo image sent by the camera and obtain the first cargo information from the cargo image.

8. The automated warehousing system of any of claims 1-7 wherein the inventory station comprises a plurality of cameras positioned at different locations, wherein each camera is configured to capture an image of a side of the goods to be inventoried.

9. The automated storage and retrieval system of claim 8, wherein the goods to be inventoried are stacked in the container to be inventoried into a rectangular parallelepiped pile, the inventoriing station includes four cameras disposed corresponding to four sides of the pile and a fifth camera disposed corresponding to an upper surface of the pile; the cargo image collected by the fifth camera is used for determining the number of the cargos to be checked, which are positioned on the top layer of the cargo pile.

10. The automated warehousing system of claim 9, wherein the fifth camera is a camera with a depth information acquisition function, and the cargo image and the depth information acquired by the fifth camera are used together to determine the number of the cargo to be inventoried on the top of the cargo pile; alternatively, the first and second liquid crystal display panels may be,

the inventory station further comprises a sixth camera arranged on the upper surface of the cargo pile, the sixth camera is used for collecting images of the upper surface of the cargo pile from angles different from those of the fifth camera, and the fifth camera and the images of the cargos collected by the sixth camera are jointly used for determining the number of the cargos to be inventoried stacked on the top layer.

11. The automated warehousing system of any of claims 1-10 wherein the inventory station is located in at least one of the following locations in the warehouse:

the warehouse comprises a warehouse entry station, a warehouse exit station, a goods position of a stereoscopic shelf of the warehouse and the inside of a lifting machine of the warehouse;

the target position comprises an original goods position of the container to be checked in the warehouse, or if the checking result is abnormal, the target position comprises an abnormal goods position in the warehouse or the ex-warehouse station.

12. The goods checking method is applied to a server in an automatic warehousing system, the automatic warehousing system further comprises a carrying device and a checking station, the checking station comprises a camera, and the method comprises the following steps:

acquiring current goods information of the goods to be checked, comparing the current goods information with inventory goods information corresponding to the container to be checked, and acquiring goods checking results of the container to be checked;

the current cargo information includes first cargo information, the first cargo information is the cargo information of the cargo to be inventoried, which is obtained from the cargo image of the cargo to be inventoried, and the cargo image is the image of the cargo to be inventoried, which is acquired by the camera of the inventoriing station after the carrying device carries the container to be inventoried to the inventoriing station.

13. The method for checking the goods according to claim 12, wherein the obtaining the current goods information of the goods to be checked comprises:

receiving the current cargo information sent by the processing equipment of the checking station; the current cargo information comprises the first cargo information, and the first cargo information is obtained by the processing equipment from the cargo image acquired by the camera; alternatively, the first and second electrodes may be,

and receiving the cargo image sent by the camera, and obtaining the first cargo information in the current cargo information from the cargo image.

14. A server, comprising: a memory having stored therein computer program instructions which, when read and executed by the processor, perform the method of claim 12 or 13.