A warehouse inventory method, device and medium based on an intelligent robot

Abstract

The application discloses a kind of based on intelligent robot warehousing inventory method, equipment and medium, it is related to warehousing management technical field, comprising: according to the container to be distributed in warehouse inventory account in the demand of being distributed out of warehouse, generates the inventory data and the range of inventory before being distributed out of warehouse;Control intelligent robot to collect goods location pick side image and container number, generate pick side measured data;When measured container number is consistent with the container number under the registration goods location of inventory account, generate registration goods location distributed out of warehouse processing record;When measured container number does not belong to the inventory account record under the robot collection goods location, but exist in other registration goods location, and satisfy the demand of being distributed out of warehouse, generate robot collection goods location distributed out of warehouse processing record and registration goods location suspension allocation record;Container number is collected in the distributed out of warehouse review position, when matching consistent, execute inventory deduction and remove suspension allocation.This application reduces inventory data amount, so that misposition container can be controlled to be distributed out of warehouse without being arranged first, and prevent repeated allocation.

Description

Technical Field

[0001] This invention relates to the field of warehouse management technology, and in particular to a warehouse inventory method, equipment and medium based on intelligent robots. Background Technology

[0002] Inventory counting is a crucial step in warehouse management to confirm inventory quantities, location status, and order fulfillment capabilities. With the application of intelligent robots, barcode recognition, QR code recognition, and RFID tags in warehouses, inventory counting methods are gradually shifting from manual item-by-item verification to robots collecting images of storage locations, container numbers, and storage location numbers along shelf aisles. The warehouse management system then compares the collected information with inventory ledgers, outbound tasks, and storage location records to support inventory correction, order allocation, and anomaly verification.

[0003] In scenarios involving frequent order outbound shipments, temporary misplacement of containers, or untimely write-back of inventory transfers, conventional robotic inventory checks often treat containers found in non-registered locations as abnormal objects. This typically requires generating a sorting task before resuming outbound shipments, affecting the timeliness of inventory participation in fulfillment. At the same time, there is a lack of continuous control between the robot's actual measured location, the location registered in the inventory ledger, and the location verified for outbound shipments. Misplaced containers may have insufficient constraints on repeated allocation before and after outbound shipments. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a warehouse inventory method based on intelligent robots to solve the problems of misplaced containers needing to be sorted before they can be sent out and insufficient control of duplicate allocation during the outbound process of misplaced containers.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a warehouse inventory counting method based on an intelligent robot, comprising: determining containers to be shipped from an inventory ledger according to pending outbound demand, and generating outbound verification data and a pre-outbound inventory count range by registering the location number and the location of the picking side in the inventory ledger; controlling the intelligent robot to collect images of the picking side of the location and container number data according to the pre-outbound inventory count range, determining whether there are containers on the picking side based on the images of the picking side, and generating actual measurement data of the picking side by combining the container number data; when the measured container number matches the container number registered under the location number in the inventory ledger, generating a processing record for outbound from the location registered in the inventory ledger; when the picking side is empty and there are no incomplete outbound deduction records or incomplete transfer write-back records, suspending the relevant inventory participation order. In the allocation process, when the measured container number does not belong to the inventory ledger record under the robot-collected location number but exists in other inventory ledger registered locations, and can meet the outbound allocation requirements, a processing record for outbound from the robot-collected location is generated, and a pause allocation record for the inventory ledger registered location is generated, so that the outbound task is executed from the robot-collected location. At the outbound verification location, the container number is collected, and the collected container number is matched with the container number in the outbound processing record. If the match is consistent, an outbound confirmation record is generated, inventory is deducted according to the inventory ledger registered location number and container number, and the pause allocation record for the inventory ledger registered location is lifted. If the match is inconsistent or no container number is collected, the relevant pause allocation status is maintained, and the scope of the next intelligent robot inventory count is updated according to the verification results.

[0007] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the generation of outbound verification data and the pre-outbound inventory scope includes: establishing inventory ledger records according to container numbers, and writing the inventory ledger registration location number, material code, batch number, container number, and inventory quantity into the inventory ledger records; filtering the inventory ledger records according to the material code, batch requirements, and material batch outbound rules in the outbound demand to be allocated, and deleting containers with records of unreleased suspension of allocation; extracting containers that can cover the demand quantity from the filtered inventory ledger records to form a list of containers to be outbound; finding the location of the picking side of the location according to the inventory ledger registration location number in the list of containers to be outbound, and writing the order number, inventory ledger registration location number, material code, batch number, container number, demand quantity, and location of the picking side into the outbound verification data; merging outbound verification data with the same inventory ledger registration location number and the same location of the picking side into one location picking side collection task to form the pre-outbound inventory scope.

[0008] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the generation of actual measurement data on the picking side includes: after the intelligent robot arrives at the picking side position of the storage location, it reads the storage location tag and writes the read value of the storage location tag into the storage location number collected by the robot; when the storage location tag cannot be read, the storage location number collected by the robot is determined according to the aisle number, shelf column number and shelf layer number in the warehouse electronic map, and the unread status of the storage location tag is written into the collection remarks; according to the storage location number collected by the robot, the shelf layer number, shelf column number and picking direction in the storage location file are obtained, the left and right column edges and the upper and lower shelf edges of the storage location are located in the storage location picking side image, and the storage location picking side area is enclosed by the left and right column edges and the upper and lower shelf edges of the storage location.

[0009] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the step of determining whether a container exists on the picking side based on the image of the picking side of the storage location includes: identifying a contour line continuous with the outer edge of the storage container in the picking side area of ​​the storage location; when the contour line occupies the back panel area of ​​the storage location, the empty background area of ​​the storage location, or the inner background area of ​​the shelf, the presence of a container on the picking side is recorded as the presence of a container; when the back panel area of ​​the storage location, the empty background area of ​​the storage location, or the inner background area of ​​the shelf is displayed in the picking side area of ​​the storage location, and no contour line continuous with the outer edge of the storage container is identified, the presence of a container on the picking side is recorded as empty.

[0010] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the step of generating actual measurement data on the picking side by combining container number collection data includes: when it is determined that there is a container on the picking side, obtaining container number collection data through barcode collection, QR code collection, or RFID tag collection; when multiple RFID tags are read at the same picking side location, determining the actual container number by combining the visible tag area in the picking side image and the RFID tag reading order; when there is a container outline in the picking side area, but no container number is obtained from barcode, QR code, or RFID tag, the container number collection status is recorded as "no number obtained"; the actual measurement data on the picking side includes the robot-collected location number, whether there is a container on the picking side, the actual container number, the container number collection status, the collection time, and the robot number.

[0011] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the step of suspending the relevant inventory from participating in order allocation includes: when the actual measurement data of the picking side shows that the picking side is empty and the inventory ledger registers a container in the storage location, retrieving the warehouse operation status data; if there is an incomplete outbound deduction record, an outbound retrieved but not deducted record is generated, and the order number, the inventory ledger registered storage location number, the container number, and the incomplete outbound deduction record are written into the inventory deduction verification range; if there is an incomplete transfer write-back record, a transfer removed but not written-back record is generated, and the moved storage location, the moved storage location, the container number, and the transfer task number are written into the transfer write-back verification range; if there is no incomplete outbound deduction record and no incomplete transfer write-back record, a record with goods on the books but empty on the picking side is generated, and the same container number in the inventory ledger registered storage location is suspended from participating in order allocation.

[0012] As a preferred embodiment of the intelligent robot-based warehouse inventory method of the present invention, the generation of processing records for outbound shipments from the robot-collected storage locations includes: when the measured container number in the actual measurement data on the picking side is inconsistent with the inventory ledger record under the storage location number collected by the robot, retrieving the measured container number from all inventory ledgers; if the measured container number exists in other inventory ledger registered storage locations, writing misplaced container judgment data; determining whether the measured container number can meet the outbound shipment requirements according to the material code, batch number, and inventory quantity; if it can meet the outbound shipment requirements... If a demand is generated, a processing record for outbound pickup from the robot's pickup location is created, the pickup location number is written to the robot's pickup location number, and a location allocation suspension record is generated in the inventory ledger. If there is already an incomplete processing record for outbound pickup from the robot's pickup location for the same actual container number, new orders will not be allocated to the same actual container number. If the actual container number cannot meet the pending outbound pickup demand, a location organization record is generated. If the actual container number does not exist in the inventory ledger, unregistered container judgment data is generated, and the robot's pickup location is suspended from receiving inbound tasks.

[0013] As a preferred embodiment of the intelligent robot-based warehouse inventory method described in this invention, the step of updating the scope of the next intelligent robot inventory count based on the verification results includes: when the container number collected at the outbound verification location matches the container number in the outbound processing record, an outbound confirmation record is generated; inventory is deducted based on the storage location number and container number registered in the inventory ledger; and the pickup storage location number is written into the actual pickup storage location record; when the outbound processing record is a processing record of outbound from the storage location collected by the robot, the storage location allocation suspension record registered in the inventory ledger is lifted; when the outbound verification location... When the collected container number does not match the container number in the outbound processing record, an outbound anomaly review record is generated, the inventory ledger records the location allocation is suspended, and the pickup location number, the inventory ledger registered location number, the location associated with the outbound review location, and the pickup side location of adjacent locations are written into the next intelligent robot inventory count range; when no container number is collected at the outbound review location, a container number review record is generated, the pickup location number and the inventory ledger registered location number are written into the next intelligent robot inventory count range, and the collection requirement is set to prioritize the collection of container numbers.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein the computer program, when executed by the processor, implements any step of the warehouse inventory method based on intelligent robots as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the warehouse inventory method based on intelligent robots as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: by limiting the intelligent robot to only collect the location of the picking side of the storage location, the amount of inventory data before outbound is reduced; by generating a processing record of the outbound from the storage location collected by the robot when the misaligned container meets the outbound allocation requirements, the outbound of the misaligned container can be controlled without prior sorting; by generating an inventory ledger to register the storage location suspension record and performing inventory deduction after outbound verification, duplicate allocation is prevented and inventory updates are consistent. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a warehouse inventory counting method based on intelligent robots.

[0019] Figure 2 This is a schematic diagram of the outbound verification and actual measurement at the pickup side.

[0020] Figure 3 This is a schematic diagram illustrating the control and determination of misplaced containers leaving the warehouse.

[0021] Figure 4 This is a diagram illustrating the process of verifying outbound shipments and updating the scope of inventory counts.

[0022] Figure 5 This is a comparative chart of average order fulfillment time in misaligned container scenarios.

[0023] Figure 6 This is a comparative data chart showing abnormal quantities under the linkage between the suspension of allocation and the outbound verification. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 As one embodiment of the present invention, this embodiment provides a warehouse inventory counting method based on intelligent robots, including the following steps: S1. Based on the pending outbound demand, determine the containers to be outbound in the inventory ledger, and generate outbound verification data and the scope of pre-outbound inventory count by registering the location number and the location of the picking side of the location in the inventory ledger.

[0028] Records are created according to container numbers. Each record is written into the inventory ledger to register the location number, material code, batch number, container number, and inventory quantity.

[0029] The inventory quantity adopts the inventory measurement unit used by the warehouse order allocation system. When the inventory measurement unit in the inventory ledger is inconsistent with the demand measurement unit in the pending outbound demand, the inventory quantity is converted into the allocatable quantity under the demand measurement unit according to the packaging conversion relationship in the material master data.

[0030] The storage location file is established according to the storage location number. Each record includes the shelf level number, shelf column number, picking direction, and the picking side of the storage location.

[0031] Specifically, the picking direction is the direction in which a forklift, handling equipment, or manual picking can approach the storage location; the storage location picking side is the storage location opening area on the side where the picking direction is located, and is represented by the storage location number, shelf level number, shelf column number, and picking side number.

[0032] Material batch outbound rules are recorded according to material codes.

[0033] Preferably, the material batch release rule is a first-in-first-out (FIFO) rule, a near-expiry-date first-out (FIFO) rule, or a designated batch release rule.

[0034] The first-in, first-out (FIFO) rule arranges inventory records in order of container arrival time from earliest to latest; the near-expiry first-out (IF) rule arranges inventory records in order of batch expiry date from earliest to latest; and the designated batch outbound rule filters inventory records according to the batch requirements in the outbound demand to be allocated.

[0035] The pending outbound requests are generated by the order allocation system and must include at least the order number, material code, batch requirements, required quantity, and outbound time limit.

[0036] The warehouse operation status data records include incomplete outbound deduction records, incomplete transfer write-back records, incomplete inbound registration records, and incomplete suspension allocation records.

[0037] Among them, the unresolved suspension allocation record is used to restrict containers that have been suspended from allocation by the previous inventory process from entering the outbound verification data again.

[0038] Furthermore, first, retrieve inventory records in the inventory ledger according to the material code, then filter the retrieved inventory records according to batch requirements and material batch outbound rules, and delete containers with unreleased suspension allocation records.

[0039] After screening, the remaining inventory records are arranged according to the material batch outbound rules, and the container number and inventory quantity are extracted from the inventory records at the top of the list until the extracted container inventory quantity can cover the demand quantity in the outbound demand, thus forming the outbound container record.

[0040] If the inventory of a single container can cover the demand, then the single container is written to the pending outbound container record; if the inventory of a single container cannot cover the demand, then multiple containers are retrieved consecutively in order of arrangement, and each of the multiple containers is written to the pending outbound container record.

[0041] After a record of containers to be shipped is created, the location number is registered in the inventory ledger in the record of containers to be shipped. The location of the picking side of the location is found in the location file. The order number, the location number registered in the inventory ledger, the material code, the batch number, the container number, the required quantity, and the location of the picking side of the location are written into the outbound verification data.

[0042] As an example, if an order requires 10 boxes of material M01 in batch B02, and the inventory ledger records 6 boxes of container C01 in location H01 and 4 boxes of container C02 in location H02, and neither container has any pending allocation records, then two outbound verification data entries containing containers C01 and C02 are generated respectively and written to the pickup side of locations H01 and H02 respectively.

[0043] Furthermore, the scope of the pre-shipment inventory check is generated based on the outbound verification data.

[0044] When generating the pre-outbound inventory count range, outbound verification data with the same inventory ledger registered location number and the same location pickup side position are merged into a single location pickup side collection task, and all order numbers and container numbers are retained in the location pickup side collection task.

[0045] If the same order generates multiple pending outbound container records, then each pending outbound container record will be collected by the pickup side of its respective storage location.

[0046] If multiple orders involve the same pickup location, only one pickup task will be generated, and multiple order numbers and multiple container numbers will be written into the same pickup task.

[0047] It should be noted that when the same pickup task contains multiple order numbers and multiple container numbers, separate matching records are created according to the container numbers. For the actual container numbers in the actual pickup data, they are matched one by one with the unmatched matching records in the order of the outbound verification data generation. After a successful match, the actual container number is not assigned to other orders. The unmatched matching records are retained in the pre-outbound inventory count and are collected again after the pickup container leaves the warehouse or is moved from the pickup location.

[0048] S2. Control the intelligent robot to collect images of the picking location and container number data according to the pre-outbound inventory range. Determine whether there is a container on the picking side based on the images of the picking location, and generate actual measurement data of the picking side by combining the container number data.

[0049] Specifically, the pre-outbound inventory count consists of a collection task on the location pickup side, which records the location number, pickup location, order number, and container number in the inventory ledger.

[0050] After receiving the data collection task from the pickup side of the storage location, the intelligent robot arrives at the target shelf aisle according to the storage location number and stops collecting data in front of the pickup side of the storage location.

[0051] When the intelligent robot stops collecting data, it reads the location tag and writes the read value of the location tag into the location number collected by the robot. When the location tag cannot be read, the intelligent robot determines the location number collected by the robot based on the aisle number, shelf column number, and shelf layer number in the warehouse electronic map, and writes the location tag unread status into the collection remarks.

[0052] Furthermore, when the intelligent robot collects images of the picking side of the storage location, the camera's field of view covers the picking side of the storage location, the edges of the left and right uprights of the storage location, the edges of the storage location shelves, and the area of ​​the storage location back panel.

[0053] The image of the pickup location does not need to cover all containers deep within the location, nor does it need to cover all locations on the same shelf.

[0054] Among them, the image of the picking side of the storage location is used to determine whether there is an outer outline of the container at the picking side of the storage location, and the container number collection data is used to obtain the container number currently located at the picking side of the storage location.

[0055] Furthermore, the image area of ​​the picking side of the storage location is determined by using the storage location file and the storage location number collected by the robot.

[0056] The intelligent robot first obtains the shelf layer number, shelf column number and picking direction from the storage location file based on the storage location number collected by the robot. Then, it locates the left and right column edges and the upper and lower shelf edges in the storage location picking side image and forms the storage location picking side area by the left and right column edges and the upper and lower shelf edges.

[0057] Preferably, the edges of the left and right uprights of the storage location are determined by the shelf edge lines that extend continuously in the vertical direction in the image, and the edges of the upper and lower shelves are determined by the shelf edge lines that extend continuously in the horizontal direction in the image; if the edges of the left and right uprights or the upper and lower shelves cannot be fully displayed due to container obstruction, the intelligent robot uses the extension lines of the upright edges and shelf edges of adjacent storage locations to determine the storage location retrieval side area.

[0058] After determining the pickup area of ​​the storage location, the intelligent robot checks whether there is a container on the pickup side within the pickup area.

[0059] If there is a continuous outline line on the outer edge of the storage container in the picking area of ​​the storage location, and the corresponding outline line obscures the back panel area of ​​the storage location, the empty background area of ​​the storage location, or the background area inside the shelf, then the presence of a container on the picking side is recorded as the presence of a container.

[0060] If the pickup side area of ​​the storage location can display the storage location back panel area, the storage location vacancy background area, or the shelf inner background area, and no outline line continuous with the outer edge of the storage container is identified, then the presence of a container on the pickup side will be recorded as vacancy.

[0061] As a preferred approach, the intelligent robot does not use the proportion of the container's outer contour area as the basis for judging the occupancy on the picking side, but rather uses whether the outer edge of the container forms a continuous occupancy relationship within the picking area of ​​the storage location.

[0062] Furthermore, after determining that a container exists on the pickup side, the intelligent robot collects the container number and other data.

[0063] Container number data is collected through barcode collection, QR code collection, or RFID tag collection.

[0064] Preferably, when a barcode or QR code label is affixed to the outside of the container, the intelligent robot extracts the label area from the image of the goods retrieval location, decodes the label area, and writes the decoded number into the measured container number. When an RFID tag is affixed to the container, the intelligent robot reads the RFID tag in front of the goods retrieval location and writes the read container number into the measured container number.

[0065] When multiple RFID tags are read at the same pickup location, the intelligent robot generates RFID tag reading data according to the reading order of the RFID tags.

[0066] Specifically, if there is a decodable tag area in the image of the cargo location pickup side, the container number obtained by decoding is verified one by one in the reading order to see if it exists in the RFID tag reading data; if it exists, the decoded container number is written into the measured container number; if it does not exist, or there is no decodable tag area in the image of the cargo location pickup side, the container number in any RFID tag is not directly written into the measured container number, the container number acquisition status is recorded as number not obtained, and the reason for not obtaining is recorded as multi-tag conflict.

[0067] In addition, when there is a container outline in the pickup area of ​​the storage location, but the barcode, QR code and RFID tag collection has not obtained a unique container number, the container number collection status will be recorded as "no number obtained".

[0068] When the container number collection status is "No number obtained", the intelligent robot still retains information on whether there is a container on the picking side, the robot's collection location number, the collection time, and the robot number, so that it can distinguish between two situations: the picking side is empty and the picking side has a container but the number has not been obtained.

[0069] The actual data measured on the picking side includes the robot collecting the location number, whether a container exists on the picking side, the measured container number, the container number collection status, the collection time, and the robot number.

[0070] It should be noted that the robot collects the location number to determine whether the location is the one registered in the inventory ledger; the presence of a container on the picking side is used to determine whether the location registered in the inventory ledger is empty on the picking side; the measured container number is used to determine whether the container on the picking side is the container in the outbound verification data, or whether it is a container registered in another location in the inventory ledger; the container number collection status is used to trigger pre-outbound re-collection or label maintenance; the collection time and robot number are used to write outbound processing records, anomaly review scope, and subsequent inventory records.

[0071] In some implementations, if the intelligent robot continuously collects multiple images from the same pickup location, the intelligent robot first verifies whether the pickup location numbers in the multiple images are consistent. When the pickup location numbers are consistent, the robot decodes the visible barcode label area or visible QR code label area in each of the multiple images. When only one image can be decoded to obtain the container number, the decoded container number is written into the measured container number. When multiple images can be decoded and the decoded container numbers are consistent, the consistent container number is written into the measured container number. When multiple images can be decoded but the decoded container numbers are inconsistent, the generation of pickup-side measured data is stopped, and a container number conflict record is generated. When the robot collects inconsistent pickup location numbers in multiple images, the generation of pickup-side measured data is stopped, and a pickup location verification record is generated.

[0072] It should be noted that the location verification record is used to prompt the intelligent robot to reconfirm the location number and avoid writing images of adjacent locations into the current location.

[0073] Optionally, when the warehouse uses image-text acquisition to obtain container numbers, image-text acquisition is only a supplementary method when barcode acquisition, QR code acquisition, and RFID tag acquisition cannot obtain container numbers. The input of image-text acquisition is the label text area in the image of the goods picking location, and the output is the container number character sequence. The label text area is determined by a rectangular area with characters arranged on the container surface. After obtaining the container number character sequence, it is verified according to the container number format rules pre-configured by the warehouse. When the container number character sequence conforms to the container number format rules, the container number character sequence is written into the measured container number. When the container number character sequence does not conform to the container number format rules, the container number acquisition status is recorded as "no number obtained". Barcode acquisition, QR code acquisition, or RFID tag acquisition are the main implementation methods for container number acquisition.

[0074] S3. When the measured container number matches the container number under the inventory ledger registered location number, a processing record for outbound from the inventory ledger registered location is generated. When the picking side is empty and there are no incomplete outbound deduction records or incomplete transfer write-back records, the relevant inventory is suspended from participating in order allocation. When the measured container number does not belong to the inventory ledger record under the robot-collected location number but exists in other inventory ledger registered locations and can meet the outbound allocation requirements, a processing record for outbound from the robot-collected location is generated, and an inventory ledger registered location allocation suspension record is generated, so that the outbound task is executed from the robot-collected location.

[0075] Specifically, the outbound verification data is first read according to the order number and container number, and then the actual measurement data on the picking side is read according to the storage location number registered in the inventory ledger.

[0076] If the location number collected by the robot in the actual data of picking matches the location number registered in the inventory ledger in the outbound verification data, and the measured container number in the actual data of picking matches the container number in the outbound verification data, it means that the container to be outbound is located at the picking position of the location registered in the inventory ledger; generate a processing record for outbound from the location registered in the inventory ledger, and allow the outbound demand to be allocated to be outbound from the location registered in the inventory ledger.

[0077] The process record for outbound goods from the inventory ledger is written with the order number, pickup location number, inventory ledger location number, container number, material code, batch number, required quantity, collection time, robot number, and processing status.

[0078] The pickup location number is entered into the inventory ledger as the location number, and the processing status is entered as "pending review". The collection time and robot number are derived from actual measurement data on the pickup side and are used for subsequent outbound review and anomaly checking.

[0079] Furthermore, if the actual measurement data on the picking side shows that the picking side is empty, but the inventory ledger records that there are containers in the inventory ledger locations, retrieve the warehousing operation status data.

[0080] If there are incomplete outbound deduction records in the warehouse operation status data, it means that the container may have been taken but the inventory ledger has not yet been deducted. Generate an outbound taken but not deducted record, and write the order number, inventory ledger registered location number, container number, and incomplete outbound deduction record into the inventory deduction verification scope.

[0081] If there are incomplete transfer write-back records in the warehouse operation status data, it means that the container may have left the storage location registered in the inventory ledger, but the transfer information has not been written back. Generate a transfer-out-but-not-written-back record, and write the out storage location, the in storage location, the container number, and the transfer task number into the transfer write-back verification scope.

[0082] If there are no incomplete outbound deduction records and no incomplete transfer write-back records in the warehouse operation status data, a record is generated showing goods on the books but empty on the picking side, and the registration of the same container number in the inventory ledger for order allocation is suspended.

[0083] Preferably, if the inventory is recorded as in stock but the pickup side is empty, the inventory deduction will not be triggered directly. Instead, the same container number will be removed from the order allocation range, and then the system will wait for subsequent review or feedback on the storage location. This will prevent the system from continuing to allocate orders when the inventory ledger still shows that the goods are in stock but the pickup side is already empty.

[0084] Furthermore, if containers exist in the actual data of the picking side, but the container number collection status is "number not obtained", and the outbound demand to be allocated needs to be registered from the inventory ledger for outbound, a pre-outbound re-collection record will be generated.

[0085] Before shipment, the re-collection record is written into the inventory ledger to register the location number, the location of the pickup side, the order number, the container number, the collection time, and the robot number; the intelligent robot returns to the location of the pickup side according to the pre-shipment re-collection record and re-collects the barcode, QR code, RFID tag, or label text.

[0086] If the outbound demand to be allocated does not require the outbound from the inventory ledger to be registered with the storage location, a label maintenance record is generated. The label maintenance record only restricts the corresponding storage location to participate in order allocation based on the container number, and does not affect other storage locations from entering the order allocation range.

[0087] In addition, if the measured container number in the actual data of the picking side is inconsistent with the inventory ledger record under the robot-collected storage location number, continue to search for the measured container number in all inventory ledgers; if the measured container number exists in other inventory ledger registered storage locations, write the measured container number, the robot-collected storage location number, the inventory ledger registered storage location number of the measured container number in the inventory ledger, the material code, the batch number and the inventory quantity into the misplaced container judgment data.

[0088] After the misaligned container judgment data is generated, the storage location sorting record is not generated directly, but the pending outbound demand is retrieved.

[0089] Determine whether the actual container number can meet the needs of the pending outbound allocation based on the material code, batch number, and inventory quantity.

[0090] If the material code registered in the inventory ledger for the measured container number matches the material code in the pending outbound demand, the batch number registered in the inventory ledger for the measured container number meets the batch requirements in the pending outbound demand, and the inventory quantity registered in the inventory ledger for the measured container number can cover the demand quantity in the pending outbound demand, a processing record for outbound from the robot's collection location will be generated, and a record for paused allocation of the storage location will be generated in the inventory ledger.

[0091] The processing records for robot-collected outbound goods are written into the order number, picking location number, robot-collected location number, inventory ledger registered location number, measured container number, material code, batch number, required quantity, collection time, robot number, and processing status.

[0092] Specifically, the pickup location number in the outbound processing record is set to the robot-collected location number, and the processing status is written as pending review. The inventory ledger records the location allocation suspension record, which includes the inventory ledger registered location number, the measured container number, the reason for suspension, and the suspension status. The reason for suspension is written as "the container has been processed outbound from the robot-collected location."

[0093] As an example, the inventory ledger records container C01 located in storage location H01. The intelligent robot collects container C01 from the picking side of storage location H02. When the outbound demand requires the material and batch to which container C01 belongs, and the inventory quantity of container C01 can cover the demand quantity, instead of generating a sorting task to move container C01 from storage location H02 back to storage location H01, a processing record for outbound from storage location H02 is generated, and a paused allocation record for storage location H01 is generated. The picking location field of the outbound task is rewritten to storage location H02, so that the outbound task is executed from storage location H02. At the same time, container C01 in storage location H01 must not re-enter the order allocation range.

[0094] When the misaligned container can meet the outbound allocation demand, a processing record for outbound from the robot-collected location is generated, and the picking location field of the outbound task is rewritten to the robot-collected location number, so that the outbound task is executed according to the robot-collected location number. At the same time, a record of suspended allocation of the inventory ledger location is generated, so that the same measured container number in the inventory ledger location is stopped from entering the order allocation range. After the container number collected at the outbound verification location matches the measured container number, the inventory is deducted according to the inventory ledger registered location number and the measured container number, and the robot-collected location number is written into the actual picking location record.

[0095] It should be noted that when a container number different from the actual measured container number is registered under the robot-collected storage location number, a robot-collected storage location inventory verification record is generated, and the original registered container under the robot-collected storage location number is suspended from participating in order allocation. The robot-collected storage location inventory verification record is written with the robot-collected storage location number, the original registered container number, the actual measured container number, and the misalignment status, until the intelligent robot collects the robot-collected storage location again and confirms the storage location where the original registered container is located, at which point the suspension status of the original registered container is lifted.

[0096] If the measured container number cannot cover the material code, batch number, or inventory quantity registered in the inventory ledger, a processing record for picking up goods from the robot location will not be generated; instead, a location sorting record will be generated.

[0097] The storage location sorting record is written into the storage location number collected by the robot, the storage location number registered in the inventory ledger, the measured container number, the material code, the batch number, and the sorting status.

[0098] Among them, the location sorting record is used to instruct warehouse personnel or handling equipment to move the measured container to the location registered in the inventory ledger, or to re-register the location of the container.

[0099] If there is already an incomplete processing record for outbound shipment from the robot's pickup location for the same measured container number, new orders will not be assigned to the same measured container number; only after the previous processing record for outbound shipment from the robot's pickup location is completed or canceled can the same measured container number re-enter the order allocation range.

[0100] When the measured container number in the actual measurement data of the picking side does not exist in the inventory ledger, unregistered container judgment data is generated, and the robot's collection of storage locations is paused to continue receiving inbound tasks.

[0101] The data for determining unregistered containers is written into the robot's data collection location number, measured container number, collection time, robot number, and unregistered status.

[0102] After the robot stops receiving inbound tasks at the collection location, it can prevent new inbound tasks from covering the location of unregistered containers, which facilitates subsequent accounting or removal processing.

[0103] The outbound control data includes at least the following: processing records of outbound shipments from inventory ledger registered locations; processing records of outbound shipments from robot-collected locations; records of suspended allocation of inventory ledger registered locations; records of outbound shipments that have been taken but not deducted; records of shipments that have been moved but not written back; records of goods on the books but empty on the picking side; records of re-collection before outbound shipments; label maintenance records; location organization records; and data on the determination of unregistered containers.

[0104] It should be noted that the outbound processing records in the outbound control data are used to verify the container number at the outbound verification location; the inventory ledger records of suspended allocation of storage locations are used to restrict the allocation of duplicate orders; and various abnormal records are used to determine the scope of the next intelligent robot inventory count and the scope of inventory processing.

[0105] S4. Collect container numbers at the outbound verification location, match the collected container numbers with the container numbers in the outbound processing record. If they match, generate an outbound confirmation record. Perform inventory deduction based on the storage location number and container number registered in the inventory ledger, and remove the storage location suspension record registered in the inventory ledger. If they do not match or no container number is collected, maintain the relevant suspension status, and update the scope of the next intelligent robot inventory count based on the verification results.

[0106] Outbound processing records include processing records of goods leaving the warehouse from the inventory ledger and processing records of goods leaving the warehouse from the robot's collection of goods from the warehouse.

[0107] In the processing records of goods leaving the warehouse from the inventory ledger, the warehouse location number is the same as the warehouse location number registered in the inventory ledger; in the processing records of goods leaving the warehouse from the robot, the warehouse location number is the same as the warehouse location number collected by the robot.

[0108] Both types of outbound processing records include the order number, pickup location number, inventory ledger registration location number, container number, material code, batch number, required quantity, collection time, robot number, and processing status.

[0109] Specifically, after the outbound task reaches the outbound verification position, the outbound verification device reads the outbound task number and obtains the outbound processing record based on the outbound task number.

[0110] The outbound verification device collects the container numbers that actually arrive at the outbound verification location. The container numbers can be obtained by barcode scanning, QR code scanning, RFID tag reading, or manual verification and entry.

[0111] Preferably, barcode scanning or QR code scanning is used as the primary data collection method for outbound verification, while RFID tag reading is used as a supplementary data collection method when the label on the container surface is obscured.

[0112] When the container number collected at the outbound verification location matches the container number in the outbound processing record, an outbound confirmation record is generated.

[0113] The outbound confirmation record includes the order number, outbound verification location, inventory ledger registration location number, pickup location number, container number, material code, batch number, required quantity, verification time, and verification pass status.

[0114] Inventory deductions are performed based on the location number and container number recorded in the inventory ledger, and the pickup location number is written into the actual pickup location record.

[0115] When the outbound processing record is a processing record of outbound from the location collected by the robot, after the outbound confirmation record is generated, the record of the location being temporarily suspended from being registered in the inventory ledger is lifted.

[0116] It should be noted that when there is still a remaining inventory quantity of containers after inventory deduction, the inventory ledger record of the storage location suspension is lifted, and a record of the remaining inventory suspension is generated, so that the remaining inventory does not participate in order allocation for the time being. When the remaining inventory is placed back into the storage location, the intelligent robot collects the storage location number and container number of the repositioned storage location. If the collected container number is consistent with the container number in the outbound processing record, the storage location number of the repositioned storage location is written into the inventory ledger record where the container number is located, and the record of the remaining inventory suspension is lifted.

[0117] When canceling the suspended allocation record in the inventory ledger, the suspended allocation status will be changed to canceled status, and the order number and review time in the outbound confirmation record will be written into the release reason field.

[0118] Inventory deductions are still performed according to the inventory ledger's registered location number and container number. The actual pickup location record is written into the robot's collected location number, so that changes in inventory quantity and on-site pickup location are both preserved.

[0119] When the container number collected at the outbound verification location is inconsistent with the container number in the outbound processing record, the outbound processing record is canceled, an outbound anomaly verification record is generated, and the pickup location number, the inventory ledger registered location number, and the outbound verification location are written into the anomaly verification scope.

[0120] The outbound anomaly review record is written with the order number, the container number in the outbound processing record, the container number collected from the outbound review location, the picking location number, the location number registered in the inventory ledger, the review time, and the anomaly type; if the anomaly type is inconsistent with the container number; after the outbound anomaly review record is generated, no inventory deduction is performed.

[0121] When the outbound processing record is a processing record of outbound from the robot, and the container number collected at the outbound verification location is inconsistent with the container number in the outbound processing record, the inventory ledger should maintain the record of suspended allocation of the storage location.

[0122] When maintaining a record of suspended allocation of inventory locations in the inventory ledger, the same container number in the inventory ledger location is not allowed to enter the new order allocation range until the pickup location, inventory ledger location, and outbound verification location in the abnormal review range are re-verified.

[0123] When no container number is collected at the outbound verification location, the outbound processing record is maintained, and a container number verification record is generated. Inventory deduction is not performed. The container number verification record includes the order number, pickup location number, inventory ledger registered location number, outbound verification location, verification time, and reason for non-collection. The reason for non-collection is written by the outbound verification device, including label obstruction, label damage, RFID reading failure, or incomplete manual verification. After the container number verification record is generated, the same container number is suspended from participating in new order allocations.

[0124] Furthermore, based on outbound confirmation records, outbound anomaly verification records, container number verification records, outbound retrieved but not deducted records, relocated but not rewritten records, records of goods on the books but empty on the picking side, label maintenance records, storage location organization records, and unregistered container judgment data, the scope of the next intelligent robot inventory count is generated.

[0125] The next inventory check by the intelligent robot will include the location number, the location of the goods being picked up, the trigger record type, the associated container number, and the collection requirements.

[0126] When an outbound confirmation record is generated, the next inventory count range of the intelligent robot is written with the pickup location number and the inventory ledger registered location number, enabling the intelligent robot to collect the pickup location and the inventory ledger registered location on the pickup side.

[0127] For outbound confirmation records collected by the robot from the warehouse location, if the pickup location number is different from the location number registered in the inventory ledger, the next time the intelligent robot collects the pickup side positions of the two locations respectively, to confirm whether the location collected by the robot is already vacant, and whether the location registered in the inventory ledger still has the same container number.

[0128] When an outbound anomaly verification record is generated, the next inventory check scope of the intelligent robot includes the pickup location number, the location number registered in the inventory ledger, the location associated with the outbound verification location, and the pickup-side location of adjacent locations. In the next data collection, the intelligent robot reads the location number, pickup-side image, and container number of the aforementioned locations to verify whether the container has been misplaced among the pickup location, the location registered in the inventory ledger, the location associated with the outbound verification location, or adjacent locations.

[0129] When a container number verification record is generated, the next inventory check scope of the intelligent robot is written with the pickup location number and the inventory ledger registration location number, and the collection requirement is set to prioritize the collection of container numbers; the intelligent robot prioritizes the collection of barcodes, QR codes and RFID tags; when no container number is obtained from barcodes, QR codes and RFID tags, a tag maintenance record is generated.

[0130] When a record of outbound items being retrieved but not deducted is generated, the next intelligent robot inventory count will record the location number and the location of the retrieval side in the inventory ledger, and the corresponding outbound task will be written into the inventory deduction verification scope. In the inventory deduction verification scope, it will check whether the outbound task has completed the physical handover. If the outbound task has completed the physical handover, the inventory deduction will be performed. If the outbound task has not completed the physical handover, the inventory allocation will remain in a suspended state.

[0131] When a record of a warehouse transfer that has been moved but not yet written back is generated, the next inventory count by the intelligent robot will include the location of the warehouse that was moved out, the location of the warehouse that was moved in, and the location of the warehouse pickup side. The corresponding warehouse transfer task will also be written into the warehouse transfer write-back verification range. The intelligent robot will collect the pickup side locations of the warehouse that was moved out and the warehouse that was moved in to verify whether the container has left the warehouse that was moved out and arrived at the warehouse that was moved in.

[0132] When a record is generated that shows goods in the inventory ledger but is empty on the pickup side, the next intelligent robot inventory count will be written to the pickup side location of the inventory ledger's registered location and adjacent locations, and the relevant inventory will be suspended from participating in order allocation. The intelligent robot collects the pickup side location of the inventory ledger's registered location and adjacent locations to determine whether the container has been misplaced from the inventory ledger's registered location to an adjacent location, or whether it has left the location but has not been recorded in the business record.

[0133] When a label maintenance record is generated, before the label maintenance is completed, the corresponding storage location is only judged based on whether there is a container on the picking side to participate in the inventory freeze, and is not judged based on the container number; after the label maintenance is completed, the warehouse management system removes the label maintenance restriction of the corresponding storage location and allows the corresponding storage location to participate in the order allocation judgment again based on the container number.

[0134] When a storage location organization record is generated, the next inventory count range of the intelligent robot is written into the picking-side location of the robot-collected storage location and the storage location registered in the inventory ledger. The intelligent robot collects the picking-side location of the robot-collected storage location and the storage location registered in the inventory ledger respectively to verify whether the measured container has been moved to the storage location registered in the inventory ledger. If the robot-collected storage location no longer collects the measured container number, and the storage location registered in the inventory ledger collects the same measured container number, the organization status of the storage location organization record is changed to the completed status.

[0135] When unregistered container detection data is generated, the robot's collection of storage locations is paused from receiving inbound goods, and the robot's collection location is added to the next intelligent robot inventory count range; the intelligent robot continues to collect the picking-side location of the collection location. If the unregistered container is successfully recorded, the warehouse management system removes the inbound restriction on the robot's collection location; if the unregistered container is removed, the system confirms whether the robot's collection location has returned to an empty or ready-to-receive-inbound state based on the next picking-side measurement data.

[0136] like Figure 5 As shown, before the introduction of the misaligned container scenario, the average order fulfillment time of the three methods was basically the same, indicating that different outbound control methods had little impact on fulfillment efficiency under normal storage conditions. After the introduction of the misaligned container scenario, the fulfillment time of the traditional sorting-then-outbound method increased rapidly and remained highly volatile for a long time. This is because misaligned containers require the generation of sorting tasks before resuming outbound operations. Although the method of verifying only the registered storage location can avoid some sorting waits, it is still affected by the inconsistency between the registered storage location and the actual measured storage location. The method of this invention quickly dropped back and stabilized at a low level after a brief fluctuation, indicating that by identifying misaligned containers through actual measurement data on the picking side and directly generating processing records for outbound operations from the robot-collected storage location when the outbound demand is met, it can reduce invalid sorting waits and improve outbound continuity and order fulfillment efficiency.

[0137] like Figure 6 As shown, during the process of increasing outbound rounds after verification, the method without pausing allocation consistently maintains a high level of duplicate allocation and inventory inconsistency anomalies. This indicates that relying solely on the regular outbound process, misplaced containers may still re-enter order allocation due to the uncontrolled location registration in the inventory ledger. The outbound verification method alone can reduce some anomalies, but the anomaly level still fluctuates at a moderate level, indicating that relying solely on verification is insufficient to constrain front-end allocation in a timely manner. After the method of this invention performs the linkage between pausing allocation and outbound verification, the anomaly level drops rapidly and remains stable at a low level. This indicates that the pausing allocation record of the location registration in the inventory ledger can limit duplicate allocation during the outbound of misplaced containers. After the outbound verification is consistent, inventory deduction is performed and the pausing allocation is lifted, thereby improving the consistency of inventory updates and the security of warehouse outbound operations.

[0138] This embodiment also provides a computer device applicable to the warehouse inventory method based on intelligent robots, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the warehouse inventory method based on intelligent robots as proposed in the above embodiment.

[0139] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0140] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the warehouse inventory method based on an intelligent robot as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0141] In summary, this invention reduces the amount of pre-outbound inventory data by limiting the intelligent robot to collecting data only from the pickup side of the storage location; it enables controlled outbound processing of misaligned containers without prior sorting by generating a processing record of the outbound pickup from the robot when the misaligned container meets the outbound allocation requirements; and it prevents duplicate allocation and ensures consistent inventory updates by generating an inventory ledger to record the storage location allocation suspension and performing inventory deduction after outbound verification.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A warehouse inventory counting method based on intelligent robots, characterized in that: include: Based on the pending outbound demand, determine the containers to be outbound in the inventory ledger, and generate outbound verification data and the scope of pre-outbound inventory count by registering the location number and the location of the picking side of the location in the inventory ledger. The intelligent robot is controlled to collect images of the picking location and container numbers according to the pre-outbound inventory scope. Based on the images of the picking location, it is determined whether there are containers on the picking side, and the actual measurement data of the picking side is generated by combining the container number data. When the measured container number matches the container number under the inventory ledger registered location number, a processing record for outbound from the inventory ledger registered location is generated. When the picking side is empty and there are no incomplete outbound deduction records or incomplete transfer write-back records, the relevant inventory is suspended from participating in order allocation. When the measured container number does not belong to the inventory ledger record under the robot-collected location number but exists in other inventory ledger registered locations and can meet the outbound allocation requirements, a processing record for outbound from the robot-collected location is generated, and an inventory ledger registered location allocation suspension record is generated, so that the outbound task is executed from the robot-collected location. The container number is collected at the outbound verification location. The collected container number is matched with the container number in the outbound processing record. If they match, an outbound confirmation record is generated. The inventory is deducted according to the storage location number and container number registered in the inventory ledger, and the storage location suspension record registered in the inventory ledger is lifted. If they do not match or the container number is not collected, the relevant suspension status is maintained, and the scope of the next intelligent robot inventory is updated according to the verification results. The scope of generating outbound verification data and pre-outbound inventory count includes: establishing an inventory ledger record according to the container number, and writing the inventory ledger record with the inventory ledger registration location number, material code, batch number, container number and inventory quantity; Filter inventory ledger records according to the material code, batch requirements and material batch outbound rules in the pending outbound demand, and delete containers with unreleased suspension records; Extract containers that can cover the required quantity from the filtered inventory ledger records to form a list of containers to be shipped out. Based on the inventory ledger location number in the container to be shipped record, locate the pickup side of the storage location, and write the order number, inventory ledger location number, material code, batch number, container number, required quantity, and pickup side location of the storage location into the outbound verification data; Outbound verification data with the same inventory ledger registration location number and the same location pickup side position are merged into one location pickup side collection task to form the scope of pre-outbound inventory count. The process of generating the outbound processing record from the robot-collected storage location includes: when the measured container number in the actual measurement data on the picking side is inconsistent with the inventory ledger record under the robot-collected storage location number, the measured container number is retrieved from all inventory ledgers. If the measured container number exists in other inventory ledger registered locations, then write the misplaced container judgment data; Determine whether the actual container number can meet the pending outbound allocation requirements based on the material code, batch number, and inventory quantity. If the pending outbound allocation needs can be met, a processing record for outbound from the robot-collected location will be generated, the pickup location number will be written into the robot-collected location number, and an inventory ledger record for the location's suspended allocation will be generated. If there is already an incomplete processing record for outbound from the robot's collection location for the same measured container number, new orders will no longer be assigned to the same measured container number. If the measured container numbers cannot meet the outbound allocation requirements, a storage location organization record will be generated. If the measured container number does not exist in the inventory ledger, unregistered container judgment data will be generated, and the robot will be paused from collecting data from the storage location to continue receiving inbound tasks. The step of updating the scope of the next intelligent robot inventory based on the review results includes: when the container number collected at the outbound review location matches the container number in the outbound processing record, an outbound confirmation record is generated, inventory is deducted based on the storage location number and container number registered in the inventory ledger, and the pickup storage location number is written into the actual pickup storage location record; When the outbound processing record is a processing record of outbound from the location collected by the robot, the record of the location being temporarily suspended from the inventory ledger is removed; When the container number collected at the outbound verification location is inconsistent with the container number in the outbound processing record, an outbound abnormal verification record is generated, the inventory ledger records the location allocation is suspended, and the pickup location number, the inventory ledger registered location number, the outbound verification location associated location, and the pickup side location of adjacent locations are written into the scope of the next intelligent robot inventory count. When no container number is collected at the outbound verification location, a container number verification record is generated. The pickup location number and the inventory ledger registration location number are written into the scope of the next intelligent robot inventory count, and the collection requirement is set to prioritize the collection of container numbers.

2. The warehouse inventory method based on intelligent robots as described in claim 1, characterized in that, The generated pickup-side measured data includes: After the intelligent robot arrives at the pickup side of the storage location, it reads the storage location tag and writes the read value of the storage location tag into the storage location number collected by the robot. When a location tag cannot be read, the robot determines the location number to be collected based on the aisle number, shelf column number, and shelf level number in the warehouse electronic map, and writes the unread location tag status into the collection notes. Based on the location number collected by the robot, the shelf layer number, shelf column number, and picking direction in the location file are obtained. The edges of the left and right uprights and the upper and lower shelves of the location are located in the location picking side image, and the location picking side area is formed by the edges of the left and right uprights and the upper and lower shelves.

3. The warehouse inventory method based on intelligent robots as described in claim 2, characterized in that, The step of determining whether a container exists at the pickup side based on the image of the pickup side includes: Identify the outline that is continuous with the outer edge of the storage container in the pickup area of ​​the storage location; When the outline of the storage location obscures the back panel area, the empty background area of ​​the storage location, or the background area inside the shelf, the presence of a container on the picking side will be recorded as the presence of a container. When the pickup side area displays the back panel area of ​​the pickup location, the empty background area of ​​the pickup location, or the inner background area of ​​the shelf, and no outline continuous with the outer edge of the storage container is identified, the presence of a container on the pickup side is recorded as empty.

4. The warehouse inventory method based on intelligent robots as described in claim 2 or 3, characterized in that, The process of generating actual pickup-side measurement data by combining container number data includes: Once it is determined that a container exists at the pickup side, the container number data is obtained through barcode collection, QR code collection, or RFID tag collection. When multiple RFID tags are read from the same location on the pickup side, the actual container number is determined by combining the visible tag area in the pickup side image and the RFID tag reading order. When a container outline exists in the pickup area of ​​the storage location, but no container number is obtained from the barcode, QR code, or RFID tag, the container number collection status is recorded as "no number obtained". The measured data on the picking side includes the robot-collected location number, whether a container exists on the picking side, the measured container number, the container number collection status, the collection time, and the robot number.

5. The warehouse inventory method based on intelligent robots as described in claim 1, characterized in that, The suspension of related inventory from participating in order allocation includes: When the actual measurement data on the picking side shows that the picking side is empty and the inventory ledger shows that there are containers in the storage locations, retrieve the warehouse operation status data; If there are any incomplete outbound deduction records, an outbound retrieved but not deducted record will be generated, and the order number, inventory ledger registration location number, container number, and incomplete outbound deduction record will be written into the inventory deduction verification scope; If there are incomplete transfer write-back records, a transfer has been removed but not written back record will be generated, and the outgoing storage location, incoming storage location, container number, and transfer task number will be written into the transfer write-back verification scope; If there are no incomplete outbound deduction records and no incomplete transfer back records, a record will be generated showing goods on the books but empty on the picking side, and the registration of the same container number in the inventory ledger location for order allocation will be suspended.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the warehouse inventory method based on intelligent robots as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the warehouse inventory method based on intelligent robots as described in any one of claims 1 to 5.

Images