Automatic palletizing and intelligent stacking method and system for logistics finished products

Abstract

The application relates to the technical field of logistics automation warehousing, and provides a logistics finished product automatic palletizing and intelligent stacking method and system. The method comprises the following steps: obtaining package basic information of sub-pallets to be palletized, calling a preset palletizing algorithm according to the package basic information, and generating a palletizing and stacking scheme; conveying the sub-pallets to be palletized to a destacking station, identifying whether the sub-pallets are basic pallets meeting the requirements, conveying the basic pallets to a palletizing station as palletizing basic pallets after removing top non-combined layer packages of the basic pallets; guiding a preset robot to complete destacking of the sub-pallets except the basic pallets, collecting actual size information of the destacked packages, and updating the palletizing and stacking scheme; if it is determined that the to-be-buffered packages meet the requirements of a combined layer, the robot is used to grab and stack the packages to a target pallet after warehouse positioning; after the palletizing and stacking are completed, the target pallet is conveyed to a downstream station, and the empty pallet is conveyed to a recycling station, so that the logistics finished product automatic palletizing and intelligent stacking are realized.

Description

Technical Field

[0001] This application relates to the field of automated warehousing technology, and in particular to a method and system for automated palletizing and intelligent stacking of finished goods. Background Technology

[0002] With the rapid development of the customized home furnishing industry, finished home furnishing products are characterized by a high degree of non-standard customization, a wide range of package sizes, a large number of packages per order, and the need for separate palletizing according to distributors and transportation routes. The palletizing and stacking operations in the finished product outbound process directly determine the logistics loading efficiency and delivery cost.

[0003] Currently, the vast majority of finished product consolidation and palletizing in the customized home furnishing industry still relies on manual operation. Workers need to manually move heavy and varied cabinet bodies and door panels, and complete consolidation and palletizing according to delivery requirements. This is not only extremely labor-intensive and inefficient, but also makes it difficult to accurately grasp the weight and volume parameters of the packages. This can easily lead to problems such as irregular pallet stacking, overweight or overheight single pallets, and splitting of packages from the same customer across pallets. These issues seriously affect subsequent logistics loading rates and delivery efficiency, and also pose significant operational safety hazards.

[0004] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0005] This application provides an automated palletizing and intelligent stacking method and system for finished goods in logistics, which aims to solve the problem that existing automated palletizing related technical solutions have significant technical defects and cannot adapt to the palletizing operation needs of the customized home furnishing industry.

[0006] In a first aspect, embodiments of this application provide an automated palletizing and intelligent stacking method for finished goods in logistics, the method comprising:

[0007] Obtain the basic information of the package to be combined into a pallet, call the preset pallet combining algorithm based on the basic information of the package, and generate a pallet combining and palletizing scheme; transport the pallet to be combined into a pallet to the depalletizing station, identify whether the pallet is a basic pallet that meets the requirements, and after removing the top non-combined layer package from the basic pallet, directly transport it to the pallet combining station as a pallet combining base pallet.

[0008] The system guides a pre-programmed robot to complete the depalletization of sub-pallets, collects the actual size information of the depalletized packages, and updates the palletization and stacking scheme. Based on the palletization and stacking scheme, it determines whether the depalletized packages can be directly stacked. For packages that can be directly stacked, the system controls the robot to transfer the stacking to the target pallet at the palletization station. For packages that cannot be directly stacked, the system transports them to a horizontal buffer for reordering and buffering. If it is determined that the packages to be buffered form a combination layer that meets the requirements, the system is taken out of the buffer and positioned at the palletization station, where the robot picks up the packages and stacks them to the target pallet.

[0009] After palletizing and stacking are completed, the target pallet is transported to the downstream workstation, and the empty pallet is transported to the recycling workstation, realizing automatic palletizing and intelligent stacking of finished products in logistics.

[0010] In some embodiments, the step of calling a preset consolidation algorithm based on package basic information to generate a consolidation and palletizing scheme includes: pre-setting consolidation grouping rules and palletizing constraint rules, wherein the consolidation grouping rules include rules for grouping consolidation ...

[0011] In some embodiments, after the palletizing and stacking are completed, the target pallet is transported to the downstream station, and the empty pallet is transported to the recycling station to realize automatic palletizing and intelligent palletizing of finished logistics products. This includes: monitoring the palletizing progress at the palletizing station; when the palletizing and stacking task of the target pallet is detected to be completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top-level special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage, and return of the empty pallets for later use.

[0012] In some embodiments, the step of conveying the sub-pallet to be combined to the destacking station and identifying whether the sub-pallet is a qualified base pallet includes: after conveying the sub-pallet to be combined to the destacking station via a conveyor line, acquiring the overall stacking image and package layout information of the sub-pallet through a preset vision system; combining the pre-stored base pallet judgment rules, extracting the flatness of the bottom stacking of the sub-pallet, the regularity of the package layout, and the consistency of the package size of a single pallet, and judging whether the corresponding sub-pallet meets the requirements as a base pallet for combining pallets by comparing the parameters.

[0013] In some embodiments, the step of directly conveying the base pallet to the pallet-combining station as a base pallet after removing the top non-combined layer packages includes: when the sub-pallet is determined to be a base pallet that meets the requirements, identifying the non-combined layer packages on the top layer of the base pallet that do not meet the requirements for flat stacking according to the pallet-combining stacking scheme; controlling the robot to remove the corresponding non-combined layer packages, while retaining the regular flat structure of the bottom layer of the base pallet; after all the top non-combined layer packages have been removed, conveying the base pallet directly from the destacking station to the pallet-combining station via a conveyor line for use as a base pallet for pallet-combining stacking.

[0014] In some embodiments, guiding a pre-defined robot to complete the depalletizing of sub-pallets includes: acquiring in real time three-dimensional point cloud data and color image data of the sub-pallets at the depalletizing station using a structured light stereo vision system mounted above the depalletizing station; identifying the package pose information of the topmost package on the sub-pallet, wherein the package pose information includes at least the corresponding contour, position, and orientation information, to identify whether the package is damaged, offset, or stacked abnormally; planning the robot's depalletizing motion path based on the identified package pose information, guiding the robot to complete the precise depalletizing and material handling of a single package using a suction cup gripper, and triggering a pre-defined abnormality handling process for any identified abnormal packages.

[0015] In some embodiments, the step of collecting the actual size information of the depalletized packages and updating the palletizing scheme includes: during the process of the robot completing the depalletizing and material handling of the packages, simultaneously collecting the actual size information corresponding to the actual length, width, and height of the depalletized packages through a vision system, and obtaining the unique identification information of the packages through a barcode scanning device; after binding the collected actual size information with the unique identification information, comparing the actual size information of the packages with the pre-stored basic information of the packages; if there is an information discrepancy, based on the real-time collected actual size information, calling the corresponding palletizing algorithm to update the palletizing scheme, and simultaneously adjusting the depalletizing, caching, and palletizing scheduling strategies for subsequent packages.

[0016] In some embodiments, determining whether depalletized packages can be directly stacked according to the palletizing scheme, and controlling the robot to transfer the palletized packages to the target pallet at the palletizing station for packages that can be directly stacked, includes: according to the updated palletizing scheme, verifying whether the depalletized packages can be matched with the stacking layers of the target pallet at the current palletizing station, and verifying whether the stacking order of the packages meets the palletizing rules; if it is determined that the packages can be directly stacked, planning the robot's palletizing motion path, controlling the robot to transfer the packages directly from the depalletizing station to the corresponding position on the target pallet at the palletizing station, and automatically performing padding filling for uneven positions after stacking.

[0017] In some embodiments, the process of conveying packages that cannot be directly stacked to a horizontal buffer for reordering and caching, and then, if it is determined that the packages to be cached form a combination layer that meets the requirements, being outsourced and positioned at the palletizing station, where a robot picks up and stacks them onto the target pallet, includes: if it is determined that the depalletized packages cannot be directly stacked, the packages are conveyed to the horizontal buffer via a conveyor line and a gantry robot, and corresponding buffer locations are allocated according to the palletizing group and stacking requirements to complete the warehousing and reordering management of the packages; monitoring the package information of the same palletizing group in the horizontal buffer, and when it is detected that the cached packages in the group can form a complete combination layer that meets the palletizing requirements, the packages corresponding to the combination layer are sequentially outsourced by the gantry robot according to the stacking order, conveyed to the robot pickup position at the palletizing station via a conveyor line, and then sequentially picked up and stacked onto the corresponding positions of the target pallet according to the palletizing plan.

[0018] Secondly, this application provides an automated palletizing and intelligent stacking system for finished goods in logistics, the system comprising:

[0019] The information acquisition unit is used to acquire the basic information of the package to be combined into a pallet, call the preset pallet combining algorithm based on the basic information of the package, generate a pallet combining and palletizing scheme; transport the pallet to be combined into a pallet to the depalletizing station, identify whether the pallet is a basic pallet that meets the requirements, and after removing the top non-combined layer package from the basic pallet, directly transport it to the pallet combining station as a pallet combining base pallet.

[0020] The control gripping unit guides the preset robot to complete the depalletizing of sub-pallets, collects the actual size information of the depalletized packages, and updates the palletizing scheme. Based on the palletizing scheme, it determines whether the depalletized packages can be directly stacked. For packages that can be directly stacked, the robot is controlled to transfer the palletizing to the target pallet at the palletizing station. For packages that cannot be directly stacked, they are transported to the horizontal buffer for reordering and buffering. If it is determined that the packages to be buffered form a combination layer that meets the requirements, they are taken out of the buffer and positioned at the palletizing station, where the robot grips and stacks them to the target pallet.

[0021] The palletizing unit is used to transport the target pallet to the downstream workstation after palletizing and to transport the empty pallet to the recycling workstation, thereby realizing automatic palletizing and intelligent palletizing of finished products in logistics.

[0022] This application completely replaces traditional manual palletizing operations by implementing a closed-loop control system for the entire process of automated unpalletizing, palletizing, and stacking. It solves the core pain points of the customized home furnishing industry, such as the large weight of packages and high work intensity. The efficiency of single pallet palletizing operations is several times higher than that of manual operations, while completely avoiding the safety hazards of manual operations.

[0023] By utilizing the availability of basic package information, offline and real-time consolidation algorithms are invoked to generate consolidation and palletizing schemes. Simultaneously, the actual dimensions of the packages are collected in real time during the depalletizing process, and the palletizing scheme is dynamically updated. This approach can perfectly adapt to the consolidation needs of standardized packages and non-standard handmade packages with a wide range of sizes, solving the industry problem of poor adaptability of existing technologies to non-standard packages and realizing automated consolidation of all categories of packages.

[0024] By identifying and directly reusing basic pallets, unnecessary depalletizing and palletizing redundancy is reduced, significantly improving pallet consolidation efficiency. At the same time, differentiated scheduling, such as immediate depalletizing and stacking of packages that can be directly stacked and buffered and reordering of packages that cannot be directly stacked, reduces the inventory and hardware investment in the buffer warehouse, ensures the flatness and stability of the consolidation pallet type, and strictly follows the consolidation rules based on customers and routes to avoid cross-palletization of packages from the same customer. Ultimately, this significantly improves the logistics loading rate and reduces the cost of last-mile delivery.

[0025] Through the dynamic update mechanism of the palletizing scheme, the palletizing strategy can be adjusted in real time according to the actual size of the package, which avoids the problems of palletizing interference and pallet instability from the root. At the same time, through the automated scheduling of the whole process, the manual intervention links are reduced, which greatly improves the continuous operation capability and long-term operation stability of the palletizing system.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic flowchart illustrating the steps of an embodiment of the present application for an automated palletizing and intelligent stacking method for finished goods in logistics.

[0029] Figure 2 This is a process flow diagram of an embodiment of the automated palletizing and intelligent stacking method for finished goods in logistics provided in this application;

[0030] Figure 3 This is a schematic diagram of the 3D finished product automatic palletizing project full-process system layout simulation model provided in one embodiment of this application;

[0031] Figure 4 This is a schematic diagram illustrating the multi-scenario simulation results of the fully offline consignment and palletizing algorithm for an automated consignment project of finished products provided in one embodiment of this application;

[0032] Figure 5 This is a schematic block diagram of the structure of an automated palletizing and intelligent stacking system for finished goods provided in an embodiment of this application;

[0033] Figure 6 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0034] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0036] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0037] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0038] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0039] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0040] With the rapid development of the customized home furnishing industry, finished home furnishing products are characterized by a high degree of non-standard customization, a wide range of package sizes, a large number of packages per order, and the need for separate palletizing according to distributors and transportation routes. The palletizing and stacking operations in the finished product outbound process directly determine the logistics loading efficiency and delivery cost.

[0041] Currently, the vast majority of finished product consolidation and palletizing in the customized home furnishing industry still relies on manual operation. Workers need to manually move heavy and varied cabinet bodies and door panels, and complete consolidation and palletizing according to delivery requirements. This is not only extremely labor-intensive and inefficient, but also makes it difficult to accurately grasp the weight and volume parameters of the packages. This can easily lead to problems such as irregular pallet stacking, overweight or overheight single pallets, and splitting of packages from the same customer across pallets. These issues seriously affect subsequent logistics loading rates and delivery efficiency, and also pose significant operational safety hazards.

[0042] To address the aforementioned issues, a few automated palletizing solutions have emerged in the industry. However, these solutions all suffer from significant technical flaws and are unsuitable for the consignment operations required by the customized furniture industry. Firstly, most existing automated palletizing solutions are designed for standardized packages with uniform specifications and small dimensional deviations. They cannot achieve stable depalletizing and accurate palletizing for non-standard packages in the customized furniture industry, which exhibit significant variations in length, width, and height, resulting in extremely poor adaptability. Secondly, existing consignment solutions employ a single operational logic, failing to differentiate scheduling based on the availability of package information. They either require sending all packages to a cache and then reassembling them, leading to excessive cache occupancy and a lengthy consignment process. First, the existing solutions suffer from several drawbacks. First, they are inefficient, only allowing for simple unpacking and stacking without ensuring the regularity and stability of the pallet arrangement, leading to instability and package drop. Second, they cannot classify and reuse incoming pallets, requiring all packages to be completely unpacked and repacked, resulting in significant redundancy. This not only drastically reduces pallet arrangement efficiency but also increases the risk of package damage due to repeated unpacking and stacking. Third, the existing pallet arrangement solutions have fixed stacking planning logic, making dynamic adaptation impossible for manual non-standard packages whose dimensions cannot be obtained in advance. This leads to stacking interference, insufficient pallet arrangement accuracy, poor automation compatibility, and an inability to achieve automated pallet arrangement for all types of packages.

[0043] To solve the above problem, please refer to Figure 1 This application provides an automated palletizing and intelligent stacking method for finished goods in logistics, applied to computer equipment. The computer equipment can be deployed on a single server or a server cluster. It can also be deployed on handheld terminals, laptops, wearable devices, or robots, etc. It should be noted that all information involved in the method provided in this application is extracted with the authorization of the relevant user and in accordance with relevant regulations, and will not infringe on user privacy.

[0044] The provided method for automated palletizing and intelligent stacking of finished goods includes steps S101 to S103. Details are as follows:

[0045] Step S101. Obtain the basic information of the package to be combined and the sub-pallet. Based on the basic information of the package, call the preset pallet combining algorithm to generate a pallet combining and palletizing scheme. Transport the pallet to be combined to the depalletizing station, identify whether the sub-pallet is a basic pallet that meets the requirements, and after removing the top non-combined layer package from the basic pallet, directly transport it to the pallet combining station as the pallet combining basic pallet.

[0046] Specifically, this step is the preparatory stage for palletizing operations, and it involves two core actions: First, based on the basic information of the packages to be palletized, a differentiated palletizing algorithm is matched to complete the pre-planning of the palletizing scheme; second, the incoming material pallets are classified and identified, and reusable basic pallets are pre-processed and directly put into the palletizing stage to reduce redundant depalletizing operations.

[0047] The pre-generation of the palletizing scheme includes:

[0048] Step 1: The system connects with the upstream production MES system, ERP system, and WMS finished product warehousing system to obtain basic information of the packages corresponding to the pallets to be combined in batches. The information includes, but is not limited to, package SKU, pre-stored size, pre-stored weight, distributor, receiving customer, delivery route, and order number, and completes the full data collection of the pallet to be combined task.

[0049] Step 2: The system has built-in consignment rule judgment logic. For each set of packages to be consigned to a consignment, it judges whether the complete size and weight information of the packages can be obtained in advance from the upstream production process system. The judgment criteria are: whether the package is a standardized cabinet body or standard door panel package, whether there is complete process size data uploaded by paper cutter, edge sealing machine and packaging machine, and whether there is any manual packaging or non-standard size modification.

[0050] Step 3: Based on the information availability determination result, match the corresponding consolidation algorithm: For the set of packages whose information can be completely obtained in advance, call the offline consolidation algorithm; for the set of manual packages and non-standard irregular-shaped packages whose information cannot be obtained in advance, call the real-time consolidation algorithm to complete consolidation grouping, stacking type planning, and operation sequence arrangement, and generate the final consolidation and palletizing scheme. The scheme clearly defines the single pallet stacking constraints, package stacking order, and cache resource reservation requirements.

[0051] Basic tray identification and preprocessing include:

[0052] Step 1: According to the generated palletizing scheme, the WMS system will take out the pallets to be palletized in the order of shipment. The pallets will be transported by manual forklifts to the pallet loading station of the palletizing system. After the photoelectric sensor at the loading station detects that the pallet is in place, it will trigger the roller conveyor to start and smoothly transport the pallet to the depalletizing station. After it is in place, the limit switch will be triggered and the conveyor will stop, completing the pallet loading and positioning.

[0053] Step 2: The 3D structured light vision system above the depalletizing station is activated to perform a full-area scan of the sub-pallets on the station, collect the overall 3D point cloud data and color image data of the sub-pallets, filter background interference data such as pallets and the ground, and extract the full information of package arrangement, layer structure and stacking flatness of the sub-pallets.

[0054] Step 3: The system calls the pre-stored basic pallet judgment rules and extracts three core parameters from the collected palletizing information: the flatness of the bottom layer of the sub-pallet, the regularity of the package layout, and the consistency of the package size of a single pallet. By comparing the parameter thresholds, it determines whether the sub-pallet meets the requirements for being a basic pallet for a combined pallet.

[0055] Step 4: For the base pallet that is determined to be qualified, the system identifies non-combined layer packages (loose narrow strip packages, hardware packages, extra-wide irregularly shaped packages, etc.) on the top layer of the base pallet that do not meet the requirements for flat stacking according to the pallet stacking scheme. The system issues a dismantling instruction to the depalletizing robot. The robot completes the dismantling of all non-combined layer packages in the order from top to bottom, leaving only the regular flat structure at the bottom layer.

[0056] Step 5: After the non-combination layer is removed, the vision system performs a second scan to verify the flatness and integrity of the base pallet. Once the verification is successful, the system sends a transfer instruction to the conveyor line. Through the lifting and transfer mechanism, the base pallet is directly transported from the destacking station to the pallet-combining station and used as the base pallet for pallet stacking. There is no need to completely destacking and re-stack the entire pallet.

[0057] Step S102. Guide the preset robot to complete the depalletizing of the sub-pallet, collect the actual size information of the depalletized packages and update the palletizing scheme; according to the palletizing scheme, determine whether the depalletized packages can be directly stacked; for packages that can be directly stacked, control the robot to transfer the palletizing to the target pallet at the palletizing station; for packages that cannot be directly stacked, transport them to the horizontal buffer for sorting and buffering; if it is determined that the packages to be buffered form a combination layer that meets the requirements, they are taken out of the buffer and located at the palletizing station, where the robot picks them up and stacks them to the target pallet.

[0058] Specifically, this step is the core execution link of the consignment operation, which completes three major actions: automated depalletizing, dynamic calibration of the palletizing scheme, and differentiated scheduling of packages. It realizes the collaborative operation of immediate depalletizing and palletizing and buffer reordering, taking into account both consignment efficiency and pallet quality.

[0059] Robot depalletizing guidance and package actual size acquisition include:

[0060] Step 1: For ordinary sub-carriers that are not on the base tray, and non-assembled layer packages that are removed from the base tray, the system uses a 3D structured light vision system to collect the three-dimensional point cloud data of the top layer package on the sub-carrier at the destacking station in real time, identify the package's outline, center coordinates, rotation angle and other pose information, and at the same time detect whether the package has any abnormal states such as damage, displacement or stacking abnormalities.

[0061] Step 2: The system transmits the identified package position information to the depalletizing robot controller in real time via industrial Ethernet. Based on preset obstacle avoidance rules, the controller plans the optimal motion path for the robot from the standby position to the package picking position, guiding the six-axis industrial robot to complete the precise depalletizing and picking of a single package through the vacuum suction cup gripper. For any abnormal packages identified, the system immediately triggers the preset abnormal handling process, suspends the depalletizing operation, and triggers an audible and visual alarm to remind manual intervention.

[0062] Step 3: After the robot completes the depalletizing and unpacking of packages, it scans the barcode label on the surface of the package using an industrial barcode scanner mounted on the robot's gripper or barcode scanners located on both sides of the corresponding conveyor line. This obtains the package's unique identifier (ID), and the actual size information is then bound to the package ID before being uploaded to the palletizing control system. Simultaneously, the length and width dimensions obtained during depalletizing and the height dimension measured by a distance sensor upon warehousing can also be bound to the unique identifier (ID).

[0063] Dynamic updates to the palletizing solution include:

[0064] Step 1: The HeTuo control system compares the actual size information of the package collected with the pre-stored basic package information and calculates the size deviation value. The preset deviation threshold is: length and width deviation ≥ 10mm, height deviation ≥ 5mm. If the deviation exceeds the threshold, it is determined that there is a deviation in the information.

[0065] Step 2: If the system determines that there is a discrepancy in the information, it will re-invoke the corresponding palletizing algorithm based on the actual dimensions of the package collected in real time, update the palletizing grouping, palletizing type planning, and package stacking order, and simultaneously adjust the depalletizing priority, cache resource allocation, and palletizing scheduling strategy of subsequent packages. The updated palletizing scheme will then be simultaneously sent to the controllers of the robot, conveyor line, and gantry robot to ensure that the operation logic of the entire system is updated synchronously.

[0066] The instant-unpacking and instant-coding features that allow for direct assembly of layered packages include:

[0067] Step 1: Based on the updated palletizing scheme, the system performs two compliance checks on the unpalletized packages: First, it checks whether the palletizing grouping of the package is consistent with the grouping of the target pallet at the palletizing station (same distributor, same delivery route, same receiving customer); second, it checks whether the size of the package can be matched with the empty position of the current layer to be stacked on the target pallet, and whether the stacking order complies with the palletizing rules.

[0068] Step 2: If both checks pass, the package is determined to be directly stackable. The system immediately sends a stacking instruction to the robot, plans a collision-free movement path for the robot from the depalletizing and material-picking position to the palletizing position, and controls the robot to transfer the package directly from the depalletizing position to the corresponding position on the target pallet at the palletizing position.

[0069] Step 3: After stacking is completed, the vision system scans and verifies the flatness of the stacking. If the height difference between adjacent packages is ≥2mm, the system controls the foam pad automatic placement mechanism on the robot fixture to place foam pads of corresponding thickness on the surface of the lower package to complete the flatness compensation process and ensure that the overall flatness of the stacking layer meets the requirements.

[0070] Cache reordering and layered outbound operations for packages that cannot be directly grouped include:

[0071] Step 1: If the package fails the stacking verification, it is determined that it cannot be directly stacked and palletized. The system sends a buffer storage instruction to the conveyor line PLC, which transports the package to the storage connection position of the horizontal buffer warehouse. Once in place, the gantry telescopic fork robot arm is triggered.

[0072] Step 2: The gantry robot uses a 12-tooth telescopic fork to stably pick up packages (single package load ≥ 60kg), and allocates corresponding buffer storage locations to the packages according to the requirements of consignment grouping and layering. The packages are then stored in the horizontal buffer warehouse, completing the inbound caching and reordering management of packages. Packages in the same consignment group are assigned adjacent storage locations to improve the efficiency of subsequent layered outbound operations.

[0073] Step 3: The system monitors the size and quantity of all cache packages in the same pallet group within the horizontal cache library in real time, calculates the overall size of the packages after splicing in real time, and immediately triggers the group layer outbound process when it detects that the cache packages in the group can be spliced ​​into a complete combination layer that meets the palletizing requirements (overall flatness ≥ 85%).

[0074] Step 4: The system sends a warehouse instruction to the gantry robot. According to the stacking priority, the corresponding packages of the combined layer are taken out from the buffer warehouse in sequence and placed on the outbound conveyor line. The packages are then transported to the robot's package picking position at the palletizing station. The robot picks up the packages in sequence and places them on the corresponding positions of the target pallet according to the palletizing and stacking scheme, thus completing the automated stacking of the combined layer.

[0075] Step S103. After palletizing is completed, the target pallet is transported to the downstream workstation, and the empty pallet is transported to the recycling workstation, realizing automatic palletizing and intelligent palletizing of finished products in logistics.

[0076] Specifically, this step is the closed-loop final stage of the consignment operation, which mainly completes the offline circulation of finished consignment pallets and the recycling and reuse of empty pallets, realizing the automated closed loop of the entire consignment process, while connecting with the downstream manual post-processing stage.

[0077] Pallet consolidation completion judgment and finished product pallet delivery include:

[0078] Step 1: The system monitors the palletizing progress of the palletizing station in real time. It uses a combination of robot stacking count and vision system layer count verification. When the palletizing task of the target pallet is completed, or the stacking height and total weight of a single pallet reach the preset upper limit of constraints, the pallet palletizing task is determined to be completed.

[0079] Step 2: The system sends a drop-off command to the conveyor line at the palletizing station. The roller conveyor line starts and smoothly transports the target pallet that has been palletized to the downstream manual post-processing station. Once in place, the Andon system is triggered to remind the staff to perform auxiliary processing such as top-level stacking and wrapping of special offline packages such as transparent packaging and long strips, and complete the final palletized finished product operation.

[0080] Empty pallet recycling and reuse includes:

[0081] Step 1: The system monitors the pallet status at the depalletizing station in real time. Through dual detection of photoelectric sensors and vision system, when it is detected that all packages on the pallet have been depalletized and no packages remain on the pallet, it is determined that the pallet is empty.

[0082] Step 2: The system sends an empty pallet recycling instruction to the conveyor line at the depalletizing station. The conveyor line transports the empty pallets to the stacking station at the end, where an automatic stacking machine completes the stacking and storage of the empty pallets. The preset stacking quantity is 10 pallets. After stacking is completed, a full material reminder is triggered, and the pallets are transferred to the finished product warehouse by forklift, completing the return and reuse of empty pallets and realizing closed-loop management of pallets throughout the entire process.

[0083] In some embodiments, the step of calling a preset consolidation algorithm based on package basic information to generate a consolidation and palletizing scheme includes: pre-setting consolidation grouping rules and palletizing constraint rules, wherein the consolidation grouping rules include rules for grouping consolidation ...

[0084] This embodiment is a detailed implementation of the co-palletizing scheme generation steps. The core is to build a dual-track adaptation mechanism of "offline co-palletizing algorithm and real-time co-palletizing algorithm" to address the industry characteristics of large differences in the availability of package information in the customized home furnishing industry. Co-palletizing rules and constraints are set in advance, and corresponding algorithms are matched based on the availability of package information, taking into account both the advance planning of co-palletizing and the adaptability of non-standard packages.

[0085] The system pre-configures consignment rules and palletizing constraints. The consignment grouping rules are configured as follows: for delivery distances <150km, consignment is based on the receiving customer; for delivery distances 150-500km, consignment is based on the shopping mall; and for delivery distances >500km, consignment is based on the delivery route. The palletizing constraints are configured as follows: the total height of a single pallet is ≤1800mm, the total weight is ≤1500kg, the number of packages per pallet is ≤100, and packages from the same customer cannot be split across pallets.

[0086] The system connects to the upstream production MES system to extract basic information of the packages to be assembled. For each package, it determines whether its size and weight information can be completely obtained in advance from the process data of the paper cutter and packaging machine. Among them, the complete information of the standardized wardrobe body and standard door panel packages can be obtained in advance, accounting for about 38.5%; the complete information of the hand-packaged hardware packages and irregularly shaped door panel packages cannot be obtained in advance.

[0087] For a set of packages for which complete information can be obtained in advance, an offline consolidation algorithm is invoked to complete consolidation grouping and palletization planning 24 hours in advance. The algorithm supports two strategies: one is to concentrate the packages into the optimal solution with the fewest pallets, which is suitable for long-distance trunk transportation scenarios; the other is to distribute the packages evenly among multiple pallets, which is a suboptimal solution, suitable for short-distance same-city delivery scenarios. The algorithm generates a complete consolidation and palletizing scheme, which clarifies the unpalletizing order and stacking position of the packages.

[0088] For a set of packages for which complete information cannot be obtained in advance, the real-time consolidation algorithm is invoked to reserve inventory redundancy in the cache group layer according to the consolidation group. Each group reserves no less than 5 cache locations to generate an initial consolidation and palletizing scheme. After the actual size of the package is obtained in the subsequent depalletizing process, the pallet type planning is dynamically updated to ensure the automatic consolidation adaptability of non-standard packages.

[0089] In some embodiments, after the palletizing and stacking are completed, the target pallet is transported to the downstream station, and the empty pallet is transported to the recycling station to realize automatic palletizing and intelligent palletizing of finished logistics products. This includes: monitoring the palletizing progress at the palletizing station; when the palletizing and stacking task of the target pallet is detected to be completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top-level special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage, and return of the empty pallets for later use.

[0090] This embodiment is a detailed implementation of the pallet diversion step after pallet consolidation is completed. The core is to achieve accurate determination of the consolidation completion status through a dual verification mechanism, and at the same time to build a dual-path diversion process for finished pallets and empty pallets, connecting the downstream manual post-processing link and the empty pallet return link, so as to realize the automated closed loop of the entire consolidation process.

[0091] The system collects real-time operational data from the palletizing station and monitors the palletizing progress through a dual mechanism of robot stacking and counting, and 3D vision layer count verification: the system increments the count by 1 for each package the robot completes stacking; and the vision system scans and verifies the layer count and stacking integrity after each layer is completed to avoid counting errors.

[0092] When the system detects that all packages for the target pallet have been stacked, or that the height and weight of a single pallet have reached the preset limit, it immediately locks the pallet's task and sends a finished pallet unloading command to the conveyor line PLC. The conveyor line transports the finished pallet to the manual post-processing station at a stable speed of 0.3m / s. Once in place, a blue audible and visual alert is triggered, notifying the operator to complete auxiliary processing such as stacking and wrapping special packages on the top layer.

[0093] The system synchronously monitors the pallet status in real time through the photoelectric sensor and 3D vision system at the depalletizing station: the photoelectric sensor detects the height of the upper surface of the pallet, and if the height is ≤10mm, it is determined that there are no remaining packages; the vision system performs a second scan to verify that there are no remaining packages on the pallet, and after confirming that there are no remaining packages on the pallet, it is determined that the pallet is empty.

[0094] The system sends an empty pallet recycling instruction to the depalletizing station's conveyor line, which then transports the empty pallets to the automatic pallet stacking machine station at the end. The stacking machine's lifting mechanism lifts the empty pallets and stacks them onto the existing empty pallet stack. When 10 empty pallets are stacked in a single stack, a red full material reminder is triggered, notifying a forklift to transfer the stacked empty pallets to the finished goods warehouse for reuse, thus completing the fully automated recycling and return of empty pallets.

[0095] In some embodiments, the step of conveying the sub-pallet to be combined to the destacking station and identifying whether the sub-pallet is a qualified base pallet includes: after conveying the sub-pallet to be combined to the destacking station via a conveyor line, acquiring the overall stacking image and package layout information of the sub-pallet through a preset vision system; combining the pre-stored base pallet judgment rules, extracting the flatness of the bottom stacking of the sub-pallet, the regularity of the package layout, and the consistency of the package size of a single pallet, and judging whether the corresponding sub-pallet meets the requirements as a base pallet for combining pallets by comparing the parameters.

[0096] This embodiment is a detailed implementation of the basic tray identification step. The core is to collect the full stacking information of the sub-trays through a 3D structured light vision system, extract three core judgment parameters, and realize the automatic and accurate identification of the basic trays through threshold comparison. This provides a basis for the direct reuse of the basic trays and reduces unnecessary destacking and redundant operations.

[0097] The pallet to be stacked is conveyed to the destacking station via a conveyor line. Once in place, a limit switch is triggered, the conveyor line stops, and the pallet is accurately positioned with a positioning accuracy of ±5mm. The Hikvision MV-DB1300A structured light stereo camera, mounted directly above the station, starts up and completes a full-area scan of the pallet at a scanning frame rate of 1.5fps, collecting the pallet's 3D point cloud data and color image data. The camera's installation height completely covers the maximum size range of the pallet (2800mm long × 1200mm wide).

[0098] The system preprocesses the collected point cloud data: it filters out the background point cloud of the ground and conveyor rack through a pass-through filter, identifies the point cloud clusters of pallets and each independent package through Euclidean clustering, and then uses a hierarchical clustering algorithm to separate each layer of the sub-pallet stacking structure and extract the bottom layer of package point cloud data.

[0099] The system extracts three core judgment parameters from the preprocessed point cloud data: first, the flatness of the bottom layer palletizing, which calculates the height value of the upper surface of all packages at the bottom layer and counts the maximum height difference; second, the regularity of package arrangement, which calculates the edge offset of adjacent packages and counts the maximum offset value; and third, the consistency of package size of a single pallet, which calculates the height value of packages in the same layer and counts the maximum height difference.

[0100] The system compares the extracted parameters with the pre-stored basic pallet judgment thresholds. The thresholds are set as follows: the maximum height difference of the bottom stack flatness is ≤5mm, the maximum offset of the package layout regularity is ≤10mm, and the maximum height difference of the package size consistency of a single pallet is ≤3mm. If all three parameters meet the threshold requirements, it is judged as a qualified pallet; if any one of them does not meet the requirements, it is judged as an ordinary sub-pallet, and the entire pallet must be completely disassembled and reassembled.

[0101] In some embodiments, the step of directly conveying the base pallet to the pallet-combining station as a base pallet after removing the top non-combined layer packages includes: when the sub-pallet is determined to be a base pallet that meets the requirements, identifying the non-combined layer packages on the top layer of the base pallet that do not meet the requirements for flat stacking according to the pallet-combining stacking scheme; controlling the robot to remove the corresponding non-combined layer packages, while retaining the regular flat structure of the bottom layer of the base pallet; after all the top non-combined layer packages have been removed, conveying the base pallet directly from the destacking station to the pallet-combining station via a conveyor line for use as a base pallet for pallet-combining stacking.

[0102] This embodiment is a detailed implementation of the pre-processing steps for the basic pallet. The core of this embodiment is based on the pallet stacking scheme, which accurately identifies the non-assembled layer packages on the top layer of the basic pallet, and completes the automated removal by a robot. The bottom layer retains its regular flat structure and is then directly transferred to the pallet stacking station for reuse via a conveyor line. This significantly reduces redundant operations in destacking and stacking and improves pallet stacking efficiency.

[0103] Once a sub-carrier is determined to be a compliant base carrier, the system retrieves the corresponding palletizing scheme for that sub-carrier. Combined with the sub-carrier's layered structure information collected by the vision system, it accurately identifies the non-assembled layer packages on the top layer of the base carrier: namely, loose packages, narrow strip packages, hardware packages, and extra-wide irregularly shaped packages that do not meet the requirements for flat stacking. The system then clarifies the number of non-assembled layers and the positional information of each package.

[0104] The system issues a non-assembled layer removal command to the depalletizing robot. The robot removes the top non-assembled layer packages in sequence from top to bottom and from outside to inside. Each time a package is removed, the system simultaneously scans the package barcode and collects its dimensions. It then determines whether the package can be directly stacked according to the palletizing rules. Packages that can be directly stacked are placed directly at the palletizing station, while those that cannot be directly stacked are transported to the horizontal buffer.

[0105] After all the non-combined layer packages on the top floor are removed, the vision system performs a second scan and verification of the remaining base support layer structure. It confirms that the layer structure is undamaged, there are no packages shifted, and the flatness meets the stacking requirements. After the verification is passed, the system sends a transfer instruction to the conveyor line.

[0106] The lifting and transfer mechanism between the destacking station and the palletizing station is activated, smoothly transferring the base pallet from the destacking main line to the palletizing main line, and then conveying it to the palletizing positioning position of the palletizing station to be used as the base pallet for palletizing. Subsequent palletized packages can be directly placed on the regular flat layer of the base pallet without the need to re-stack the bottom structure. The palletizing operation time for a single pallet can be shortened by more than 30%.

[0107] In some embodiments, guiding a pre-defined robot to complete the depalletizing of sub-pallets includes: acquiring in real time three-dimensional point cloud data and color image data of the sub-pallets at the depalletizing station using a structured light stereo vision system mounted above the depalletizing station; identifying the package pose information of the topmost package on the sub-pallet, wherein the package pose information includes at least the corresponding contour, position, and orientation information, to identify whether the package is damaged, offset, or stacked abnormally; planning the robot's depalletizing motion path based on the identified package pose information, guiding the robot to complete the precise depalletizing and material handling of a single package using a suction cup gripper, and triggering a pre-defined abnormality handling process for any identified abnormal packages.

[0108] This embodiment is a detailed implementation of the robot depalletizing process. The core is to use a 3D structured light vision system to collect the three-dimensional point cloud data of the sub-pallets in real time, accurately identify the pose information and abnormal status of the top layer of packages, plan the optimal depalletizing path for the robot, guide the robot to complete the accurate depalletizing of non-standard packages, and at the same time realize the automatic identification and processing of abnormal packages, ensuring the stability and safety of the depalletizing operation.

[0109] The structured light stereo camera is mounted directly above the depalletizing station at a height of 3400mm. Its field of view completely covers the pallet area of ​​the depalletizing station. After the sub-pallet is in place, the system triggers the camera to start scanning and acquire the 3D point cloud data and color image data of the top layer of the sub-pallet. The scanning cycle is synchronized with the robot's depalletizing cycle. A new scan is automatically triggered after each package is depalletized.

[0110] The system segments the collected point cloud data, identifies each independent package at the top layer, fits the minimum bounding rectangle of each package, and obtains complete pose information such as the package's outline, center X / Y / Z coordinates, and rotation angle. The pose detection accuracy is ≤5mm@1.2m in the XY direction and ≤1mm@1.2m in the Z direction, which fully meets the accuracy requirements of robot depalletizing.

[0111] The system synchronously detects abnormal states of packages based on point cloud data and color images: if the point cloud indentation depth on the package surface is ≥10mm, the package is judged as damaged; if the offset between the edge of the package and the edge of the lower package is ≥20mm, the package is judged as stacking offset; if the point cloud of the package cannot fit a complete rectangular outline, the package is judged as stacking abnormality, and all abnormal states are marked.

[0112] The system transmits the pose information of the normal package to the six-axis industrial robot controller in real time via Profinet industrial Ethernet. The controller plans a collision-free depalletizing motion path based on the RRT algorithm, guiding the robot to move the vacuum suction cup gripper to directly above the package picking position, and slowly descend to 5mm above the package surface. The vacuum generator is activated, and the suction cup picks up the package.

[0113] Once the vacuum pressure sensor detects that the pressure inside the suction cup reaches -80kPa or higher, and confirms that the package is firmly adsorbed, the robot lifts the package according to the planned path to complete the destacking and unpacking of a single package. For any abnormal packages detected, the system immediately suspends the destacking operation, triggers an audible and visual alarm, and displays the type and location of the abnormality on the host computer interface, reminding the operator to intervene manually to avoid destacking failures and package damage.

[0114] In some embodiments, the step of collecting the actual size information of the depalletized packages and updating the palletizing scheme includes: during the process of the robot completing the depalletizing and material handling of the packages, simultaneously collecting the actual size information corresponding to the actual length, width, and height of the depalletized packages through a vision system, and obtaining the unique identification information of the packages through a barcode scanning device; after binding the collected actual size information with the unique identification information, comparing the actual size information of the packages with the pre-stored basic information of the packages; if there is an information discrepancy, based on the real-time collected actual size information, calling the corresponding palletizing algorithm to update the palletizing scheme, and simultaneously adjusting the depalletizing, caching, and palletizing scheduling strategies for subsequent packages.

[0115] This embodiment is a detailed implementation of the dynamic update steps of the palletizing scheme. The core is to simultaneously collect the actual size and unique identification information of the package during the depalletizing process, compare it with the pre-stored basic information, and dynamically update the palletizing scheme and the overall system scheduling strategy based on the deviation value. This solves the problems of palletizing interference and pallet instability caused by the size deviation of non-standard packages, and improves the adaptability and palletizing stability of the palletizing scheme.

[0116] Using an industrial fixed barcode scanner mounted on a robot suction cup gripper, the barcode or QR code label on the surface of the package is scanned to read the package's unique identifier ID. After a successful scan, the system binds the collected actual size information of the package with the package's unique ID and uploads it to the database of the HeTuo control system.

[0117] The HeTuo control system retrieves the pre-stored basic information of the package ID from the database, compares the actual size with the pre-stored size item by item, calculates the size deviation value, and determines that the package information has significant deviation when the length and width deviation is ≥10mm and the height deviation is ≥5mm, and cannot be adapted to the original HeTuo palletizing scheme.

[0118] If a significant deviation in the information is determined, the system will re-invoke the corresponding stacking algorithm based on the actual size of the package: packages originally belonging to the offline algorithm will have their stacking plan recalculated and their placement adjusted; packages originally belonging to the real-time algorithm will have their cache grouping plan updated and the grouping logic of packages in the same group adjusted.

[0119] After the system completes the update of the palletizing scheme, the new scheme is simultaneously distributed to the depalletizing robot, palletizing robot, conveyor line PLC, and gantry robot controller. The system also adjusts the scheduling strategies for the entire process, such as the depalletizing priority, buffer location allocation, and stacking order of subsequent packages, to ensure that the operating logic of all equipment is completely consistent with the updated scheme, thus avoiding problems such as palletizing interference and stack instability.

[0120] In some embodiments, determining whether depalletized packages can be directly stacked according to the palletizing scheme, and controlling the robot to transfer the palletized packages to the target pallet at the palletizing station for packages that can be directly stacked, includes: according to the updated palletizing scheme, verifying whether the depalletized packages can be matched with the stacking layers of the target pallet at the current palletizing station, and verifying whether the stacking order of the packages meets the palletizing rules; if it is determined that the packages can be directly stacked, planning the robot's palletizing motion path, controlling the robot to transfer the packages directly from the depalletizing station to the corresponding position on the target pallet at the palletizing station, and automatically performing padding filling for uneven positions after stacking.

[0121] This embodiment is a detailed implementation of the direct stacking and unpacking process for packaged items. The core is based on the updated palletizing scheme, which performs dual verification of grouping and stacking of unpacked packages. For packages that pass the verification, the stacking path is directly planned to complete cross-workstation stacking without the need for a buffering process, which greatly shortens the package flow path and improves the efficiency of palletizing operations.

[0122] The pallet consolidation control system, based on the updated pallet consolidation and palletizing scheme, acquires the current operational status of the target pallet at the consolidation workstation in real time, including the consolidation grouping information of the target pallet, the size of the empty position of the current layer to be stacked, the required number of packages, and the stacking priority order, which serves as the benchmark for package group layer verification.

[0123] For packages that have just been depalletized and their dimensions collected, the system first performs pallet grouping verification: it checks whether the distributor, receiving customer, and delivery route of the package are completely consistent with the pallet grouping of the target pallet. If they are inconsistent, the package is directly determined to be unsuitable for direct layering, thus avoiding violations such as cross-group code mixing.

[0124] After the group verification is passed, the system performs layer matching verification: the actual size of the package is compared with the size of the empty position of the layer to be stacked to verify whether the package can be completely placed in the empty position and whether the overall size of the layer after stacking does not exceed the standard size of the pallet. At the same time, it is verified whether the stacking order of the packages meets the priority requirements of the pallet stacking scheme. If both of them are met, it is determined that the package can be directly stacked.

[0125] Once the judgment is passed, the system immediately sends an instant dismantling and coding instruction to the robot controller. Based on the spatial layout of the destacking and palletizing stations, the controller plans the optimal collision-free movement path for the robot from the current destacking and material picking position to the target stacking position in the palletizing station. The path planning avoids obstacles such as conveyor lines, equipment frames, and personnel work areas to ensure operational safety.

[0126] Following the planned path, the robot moves the package directly from the depalletizing station to the palletizing station. Once it reaches directly above the target stacking position, it slowly descends at a speed of 5mm / s, accurately placing the package into the corresponding position on the target pallet with a stacking positioning accuracy of ±5mm. After stacking, the vacuum generator is turned off, the suction cup releases the package, and the robot is raised to a safe height, completing the single unpacking and stacking operation.

[0127] After stacking is completed, the vision system scans and verifies the flatness of the stacking layer. If the height difference between adjacent packages is ≥20mm, the system controls the automatic foam pad placement mechanism on the robot fixture to place EPE foam pads of corresponding thickness on the upper surface of the lower package to complete the flatness compensation, ensure that the overall flatness of the stacking layer meets the requirements, and avoid the upper layer from becoming unstable.

[0128] In some embodiments, the process of conveying packages that cannot be directly stacked to a horizontal buffer for reordering and caching, and then, if it is determined that the packages to be cached form a combination layer that meets the requirements, being outsourced and positioned at the palletizing station, where a robot picks up and stacks them onto the target pallet, includes: if it is determined that the depalletized packages cannot be directly stacked, the packages are conveyed to the horizontal buffer via a conveyor line and a gantry robot, and corresponding buffer locations are allocated according to the palletizing group and stacking requirements to complete the warehousing and reordering management of the packages; monitoring the package information of the same palletizing group in the horizontal buffer, and when it is detected that the cached packages in the group can form a complete combination layer that meets the palletizing requirements, the packages corresponding to the combination layer are sequentially outsourced by the gantry robot according to the stacking order, conveyed to the robot pickup position at the palletizing station via a conveyor line, and then sequentially picked up and stacked onto the corresponding positions of the target pallet according to the palletizing plan.

[0129] This embodiment is a detailed implementation of the package caching and reordering steps. The core is to manage the grouping and reordering of packages that cannot be directly grouped into layers through a horizontal cache library. It monitors the feasibility of grouping packages in the same group in real time. When a complete and compliant combination layer can be formed, it completes automated outbound and stacking according to the stacking order, which reduces the inventory occupation of the cache library and ensures the regularity and stability of the pallet type.

[0130] After the control system completes the package verification, it determines that the unpalletized packages cannot be directly stacked. It immediately sends a buffer storage instruction to the conveyor line PLC. The diversion mechanism of the conveyor line is activated, diverting the packages from the main unpalletizing line to the buffer storage conveyor line. The conveyor line tracks the package position in real time through photoelectric sensors and smoothly transports the packages to the storage docking position of the horizontal buffer warehouse. Once the package is in place, the limit switch is triggered, and the conveyor line stops.

[0131] The system issues an inbound command to the gantry telescopic fork robot in the horizontal buffer warehouse. At the same time, based on the grouping of packages, it assigns corresponding buffer storage locations to the packages. Packages in the same group are assigned adjacent storage locations to improve the efficiency of subsequent layer outbound operations. After the storage location is assigned, the gantry robot moves along the X and Y axes to the inbound connection position, and the telescopic fork extends to pick up the package. The telescopic fork has a 7-tooth integral design and a single package load of ≥60kg, which is fully adapted to the weight requirements of customized home furnishing finished packages.

[0132] After the gantry robot picks up a package, it moves it to the assigned target storage location and smoothly places the package into the storage location, completing the package's inbound caching. At the same time, the package ID, storage location, size information, and grouping information are entered into the caching management system to complete the real-time update of the inventory ledger.

[0133] The cache management system monitors all cache packages in each co-location group within the horizontal cache library in real time. Based on the actual size of the packages, it iterates and calculates all possible package splicing combinations in real time. When a cache package in a certain group is detected that can be spliced ​​into a complete composite layer that meets the requirements, the layer outbound process is immediately triggered. The criteria for determining the composite layer are: total length after splicing ≤ 2800mm, total width ≤ 1200mm, total gap between packages ≤ 20mm, and layer flatness ≤ 20mm.

[0134] Based on the palletizing scheme, the system determines the order of packing within the combined layer and issues a warehouse instruction to the gantry robot. The robot then takes out the corresponding packings from the buffer location and places them on the buffer outbound conveyor line in the order of "packings placed first are out first".

[0135] The outbound conveyor line transports packages in the order of stacking to the robot pickup position at the palletizing station. Once in place, the robot is triggered to pick up the packages. The robot picks up the packages in sequence according to the palletizing plan and accurately places them into the corresponding positions on the target pallet, completing the automated stacking of the entire stacking layer. After the stacking of the stacking layer is completed, the system updates the cache inventory ledger and releases the corresponding cache storage positions for subsequent cached storage of packages.

[0136] In some embodiments, this embodiment addresses the core pain points of frequent changes in terminal orders and a high proportion of urgent orders in the customized home furnishing industry. It solves the industry problems that once the existing consignment scheme is generated, the insertion of orders will cause the original consignment grouping to be disrupted, cache resources to conflict, urgent orders to wait for the original batch of operations to be completed before they can be started, the delivery cycle to be long, and it is easy to cause cross-consignment of packages from the same customer and unstable stacking.

[0137] This embodiment constructs a linkage mechanism of offline baseline scheme locking + real-time order insertion dynamic reconstruction + cache resource priority hierarchical scheduling. The order insertion response logic is embedded in the scheme generation stage of S101, and the priority dynamic control is realized in the depalletizing and scheduling stage of S102. Under the premise of not interrupting the original normal order consolidation operation, it realizes the rapid consolidation of urgent orders and the closed-loop management of the whole process, while ensuring the stability and compliance of the original order consolidation scheme.

[0138] By pre-configuring order priority rules and order insertion response boundary conditions in the system, the priority is divided into three levels: top-level urgent orders (delivery within 2 hours), first-level urgent orders (delivery within 4 hours), and normal orders. The order insertion response boundary conditions include: the minimum number of buffer storage locations required for joint consignment of urgent orders, the minimum percentage of operation time that can be allocated to the robot, and the joint consignment grouping lock rules that cannot be split from the original order, so as to avoid the original order's compliance becoming invalid due to order insertion.

[0139] When an urgent order insertion instruction is received from the WMS system, the dynamic reconstruction process of the consolidation scheme in step S101 is triggered: First, the system locks the offline baseline scheme of the original normal order that has completed consolidation grouping and entered the depalletizing process, prohibiting modification of this part of the scheme to avoid interruption of the original order operation; at the same time, the basic information of the urgent order package is extracted, the availability of the urgent order package information is determined, the offline consolidation algorithm is called to complete the urgent order palletization pre-planning for the urgent order package that can obtain information in advance, and the real-time consolidation algorithm is called for the package that cannot obtain information in advance, reserving dedicated cache resources and generating an urgent order-specific consolidation palletizing scheme.

[0140] The system prioritizes urgent orders and performs tiered scheduling of cached resources: For top-priority urgent orders, a dedicated urgent order cache area is allocated in the horizontal cache library, reserving no less than 120% of the total number of urgent order packages, and prohibiting normal order packages from occupying these spaces; for first-priority urgent orders, no less than 100% of the total number of packages are reserved in dynamic storage locations, and normal order packages must reserve storage locations for urgent order packages. When urgent order packages are received into the warehouse, the storage locations of low-priority packages from normal orders can be dynamically adjusted and transferred to adjacent non-dedicated storage locations to ensure the caching and warehousing needs of urgent order packages are met.

[0141] In the depalletizing and palletizing scheduling of S102, the system implements dynamic queue management of robot tasks based on the priority of urgent orders: for every normal order package depalletized by the depalletizing robot, one urgent order package depalletizing task (top-level urgent order) is inserted, or for every three normal order packages completed, one urgent order package (first-level urgent order) is inserted to ensure the efficiency of urgent order depalletizing; the palletizing robot prioritizes the palletizing task of urgent order packages. After the urgent order package arrives at the pickup position, the current unlocked normal order palletizing task is immediately interrupted, and the urgent order package is completed first. At the same time, the status of the interrupted task is recorded, and the task is automatically resumed after the urgent order task is completed.

[0142] During the consolidation of urgent orders, the system synchronously updates the consolidation and palletizing scheme of the original normal orders in real time: it dynamically adjusts the unpacking and palletizing order and cache group layer planning of normal orders based on the cache resources occupied by urgent orders and the robot operation time, so as to avoid problems such as insufficient cache resources and operation stagnation in the original orders; at the same time, it strictly verifies the consolidation rules of packages of the same customer, prohibits the cross-consolidation of packages of the same customer due to order insertion, and ensures the compliance of consolidation.

[0143] After the urgent order consolidation task is completed, the urgent order-specific offline process in step S103 is triggered: After the urgent order finished pallet is completed and palletized, the system triggers the priority connection instruction of the conveyor line, bypassing the queue of normal orders and directly transporting it to the manual post-processing station, triggering a dedicated urgent order reminder to ensure the rapid offline delivery of urgent orders; at the same time, the system releases the dedicated cache resources for urgent orders, restores the full operation permissions of normal orders, and completes the closed loop of the entire process of inserting orders.

[0144] In some embodiments, this embodiment breaks through the limitations of independent reuse of existing single basic pallets, and solves the industry pain points of multiple basic pallets in the same group being unable to be optimized collaboratively, pallet center of gravity shifting after merging, uneven weight distribution between upper and lower layers of the stack causing transportation safety hazards, and stack instability and damage.

[0145] In the basic pallet identification stage of S101, this embodiment constructs a collaborative screening, hierarchical complementarity, and merging optimization mechanism for multiple basic pallets under the same pallet grouping. At the same time, it embeds a real-time calculation and dynamic optimization model of the center of gravity of a single pallet into the entire pallet merging process (S101 scheme generation, S102 palletizing scheduling, S103 finished product verification). While significantly reducing redundant depalletizing operations, it strictly controls the center of gravity deviation of the pallet after merging within a safe threshold, achieving a dual improvement in merging efficiency and transportation safety.

[0146] The system is pre-configured with multi-base pallet collaborative screening rules, hierarchical complementary matching algorithms, and single pallet center of gravity safety thresholds. The center of gravity safety thresholds are set as follows: the horizontal offset of the pallet's center of gravity is ≤ 1 / 6 of the pallet's side length, the vertical center of gravity height is ≤ 2 / 3 of the pallet's total height, and the weight of the bottom layer of a single pallet accounts for ≥ 60% of the total weight. At the same time, the flatness threshold for merging base pallets is configured to be ≤ 5mm.

[0147] In the basic trustee identification step of S101, the system first performs a full-domain scan and basic trustee determination on all sub-trustees to be combined in the same group (same dealer, same route, same customer), and filters out all basic trustees that meet the requirements to form a basic trustee candidate pool for the group, instead of making independent determinations on individual trustees, thus achieving full-domain coordination of multiple basic trustees in the same group.

[0148] The system performs collaborative screening and hierarchical complementary matching on multiple pallets in the candidate pool of basic pallets: extracting the flat structure parameters of each basic pallet, including flat dimensions, number of packages, weight of a single package, total weight, layer height, and flatness. Through the hierarchical complementary matching algorithm, the system calculates the flat merging compatibility between different basic pallets: if the flat dimensions of two basic pallets are completely matched, the sum of their layer heights is less than or equal to the upper limit of a single layer height, and the weight distribution conforms to the rule of heavier bottom layers and lighter top layers, then they are determined to be a basic pallet combination that can be merged and optimized.

[0149] For merging and optimizing basic pallet combinations, the system generates a basic pallet collaborative reuse scheme: retain the basic pallet with the larger total weight and better flatness as the main basic pallet, and directly transport it to the pallet merging station as the pallet merging base; the complete flat package of the other set of basic pallets does not need to be completely destacking, only the top non-combined layer is removed, and the robot directly transfers and stacks the entire flat package on the upper layer of the main basic pallet, realizing the flat collaborative reuse of multiple basic pallets, and the single pallet merging operation time can be shortened by more than 40%.

[0150] In the palletizing scheme generation stage of S101, the system embeds a real-time center of gravity calculation model: for each target pallet to be palletized, based on the weight, size, and stacking position of each package, the system calculates in real time the horizontal center of gravity offset, vertical center of gravity height, and weight distribution of the upper and lower layers of a single pallet. If the calculation results exceed the safety threshold, the stacking position of the packages is immediately adjusted, with heavier packages moved to the center area of ​​the bottom layer of the pallet and lighter packages moved to the upper edge area to ensure that the center of gravity of the palletizing scheme meets the safety requirements.

[0151] In the palletizing execution phase of S102, the system updates the center of gravity calculation results in real time based on the actual weight and size of the packages collected during the depalletizing process. If there is a deviation between the actual parameters and the preset calculation results, the system immediately and dynamically adjusts the placement position of subsequent packages to compensate for the center of gravity shift. After each layer of stacking is completed, the vision system scans and verifies the actual placement position of the packages, and the system calculates the center of gravity status after stacking that layer for the second time. If it exceeds the threshold, the system immediately adjusts the stacking plan of the next layer of packages to ensure that the center of gravity is controllable throughout the entire process.

[0152] In the finished pallet off-line stage of S103, the system uses weighing sensors and vision systems installed at the end of the conveyor line to perform final center of gravity verification and weight check on the finished pallets that have been palletized. It calculates the actual center of gravity offset. If it meets the safety threshold, it is normally transported to the downstream station. If it exceeds the threshold, an alarm is triggered, and the pallet is transported to the adjustment station, where manual or robotic adjustments are made to ensure that the pallets that are finally palletized fully meet the transportation safety requirements.

[0153] In some embodiments, this embodiment addresses the pain points of uneven workstation load, cache resource contention, high idle stroke rate of gantry robot, and low overall line utilization rate in the original scheme's core layout architecture of two independent robot workstations on the left and right sides and a shared horizontal cache library in the middle.

[0154] In this embodiment, a dual-station task dynamic allocation model is constructed in the sub-carrier loading stage of S101, and a cache library dynamic partitioning and cross-station collaborative scheduling mechanism is constructed in the cache management stage of S102. Real-time monitoring and balanced control logic of equipment load is embedded in the entire process to maximize the production capacity synergy of dual stations, while controlling the idle stroke rate of the gantry robot to within 30%, which greatly improves the overall equipment efficiency (OEE) of the entire line. At the same time, it has the redundancy takeover capability for fault conditions to ensure the continuous operation of the entire line.

[0155] By pre-configuring the equipment parameters of the dual workstations, the partitioning rules of the buffer library, the load balancing threshold, and the fault takeover conditions in the system, the dual workstations are divided into a left workstation (EMS36) and a right workstation (EMS38). Each workstation has independent depalletizing and palletizing functions and shares the middle horizontal buffer library. The load balancing threshold is set as follows: the difference in the operating load rate of the dual workstations ≤ 20%, the idle stroke rate of the gantry robot ≤ 30%, and the difference in the buffer library location utilization ≤ 15%.

[0156] In the sub-cart loading and solution generation stage of step S101, the system constructs a dual-workstation task dynamic allocation model: real-time collection of the current working load of the dual workstations, the number of tasks to be completed, the robot idle rate, and the congestion status of the conveyor line. At the same time, based on the sub-carts to be combined, the grouping of the sub-carts, the package type, and the caching requirements, the system allocates the optimal workstation for each sub-cart to be combined. For sub-carts that do not require caching and can be immediately unpacked and coded, they are preferentially allocated to workstations with lower current load rates. For sub-carts that require a large amount of caching and reordering, they are preferentially allocated to workstations that are closer to the corresponding partition of the cache library to reduce the length of the package conveying path.

[0157] The system dynamically partitions the shared horizontal cache library: based on the task allocation results of the two workstations, the cache library is divided into three dynamic areas: a dedicated area for the left workstation, a dedicated area for the right workstation, and a public shared area. The size of each area is adjusted in real time based on the cache requirements of the workstations. Packages from the left workstation are preferentially stored in the left dedicated area, and packages from the right workstation are preferentially stored in the right dedicated area. Urgent packages and packages scheduled across workstations are stored in the public shared area to avoid competition for cache resources between the two workstations and to shorten the material handling travel of the gantry robot.

[0158] In the depalletizing, buffering, and palletizing scheduling stage of step S102, the system monitors the equipment load status of the two workstations in real time, collects the robot utilization rate, conveyor line congestion rate, and number of packages to be processed for each workstation, and immediately triggers load balancing scheduling when the load rate difference between the two workstations exceeds the 20% threshold: the tasks of unpalletizing pallets and package stacking in the high-load workstation are transferred to the low-load workstation through the intermediate connecting conveyor line. At the same time, the corresponding package palletizing scheme and buffering resource permissions are transferred to ensure the continuity of operation after the task transfer.

[0159] For the scheduling of gantry robots in the cache library, the system constructs a storage location allocation algorithm optimized for empty travel: when allocating storage locations for packages, it prioritizes selecting the nearest available storage location to the current position of the gantry robot and adjacent to the same pallet group, while planning the picking and placing path of the gantry robot to achieve continuous "picking-placing" operation and avoid empty travel; after the gantry robot completes the warehousing of packages at the left workstation, it prioritizes the warehousing of packages at the adjacent storage location at the right workstation to achieve full-load operation of the round trip and keep the empty travel rate stably controlled within 30%.

[0160] The system constructs a cross-workstation package collaborative scheduling mechanism: when the package required by the consignment station of one workstation is stored in the dedicated buffer area of ​​another workstation, there is no need to depalletize and redistribute the package. The package is directly transported to the consignment and pickup position of the target workstation through the cross-workstation connection conveyor line of the buffer library, realizing package resource sharing between the two workstations and avoiding repeated depalletizing and buffer turnover.

[0161] The system is configured with a redundant takeover mechanism for fault conditions: when a robot or conveyor line in one of the workstations malfunctions, the system immediately transfers all tasks to be completed at that workstation to another normal workstation. At the same time, the system adjusts the partitioning of the cache library, moving the dedicated area of ​​the faulty workstation into the public shared area, so that the normal workstation can complete all the combined operations, avoiding a complete line downtime and ensuring production continuity.

[0162] In the finished product pallet unloading and empty pallet recycling stage of step S103, the system uniformly schedules the finished product pallets of the two workstations. Based on the working status of the downstream manual post-processing stations, the finished product pallets of the two workstations are sorted and transported according to priority to avoid congestion at the unloading point. At the same time, the recycled empty pallets are preferentially allocated to the workstations with insufficient empty pallet inventory, realizing the sharing of empty pallet resources between the two workstations.

[0163] In some embodiments, this embodiment addresses the industry pain points of non-standard packages such as narrow strip packages, extra-wide packages, and irregularly shaped packages in customized home furnishing products, which cannot be effectively layered with standard packages, resulting in high occupancy rates in the cache warehouse, low utilization of pallet space, and poor stack flatness. It constructs a non-standard package nested matching algorithm + cache pre-assembly mechanism. In the pallet solution generation stage (S101), it realizes the pre-planning of nested stacking of non-standard and standard packages. In the cache management stage (S102), it realizes the pre-assembly scheduling of non-standard packages, achieving collaborative layering and complementary stacking of non-standard and standard packages. This increases pallet space utilization by more than 25%, while significantly reducing invalid inventory occupancy in the cache warehouse and ensuring the flatness and stability of the stack.

[0164] By pre-configuring classification rules, nested matching algorithms, pre-assembly conditions, and stacking constraints for non-standard packages in the system, non-standard packages are divided into three categories: narrow strip packages (width ≤ 300mm), extra-wide packages (width ≥ 1000mm), and irregularly shaped packages (irregular rectangles). The nested matching constraints are: after nested stacking, the total gap between packages is ≤ 15mm, the layer flatness is ≤ 3mm, and the suspended proportion of a single package is ≤ 10%, thus preventing stacking instability from the source.

[0165] In the palletizing scheme generation stage of step S101, the system first classifies and identifies all packages to be palletized, marking the type, size, and outline features of non-standard packages. At the same time, it extracts the size information of standard packages grouped with the same pallet and calls a nested matching algorithm to calculate the feasibility of nested matching between non-standard packages and standard packages: for narrow strip packages, it matches the remaining long strip gap space after standard packages are stacked; for extra-wide packages, it matches the full-length stacking area of ​​the pallet and fills it with narrow strip packages on both sides; for irregularly shaped packages, it extracts the concave area of ​​its outline and matches it with small standard packages that can be embedded, realizing nested stacking with complementary concave and convex shapes.

[0166] Based on the nested matching results, the system generates a dedicated palletizing scheme for non-standard packages, specifying the stacking position, matching standard packages, and layering order for each non-standard package. At the same time, for non-standard packages that cannot be immediately nested and assembled, a dedicated pre-assembly area for non-standard packages is planned in the horizontal cache library, reserving corresponding storage locations for pre-assembly caching of non-standard packages and standard packages in the same group.

[0167] In the depalletizing stage of step S102, the system accurately collects the actual outline, size, and irregular features of the depalletized non-standard packages using a 3D vision system, updates the nesting matching model, and recalculates the optimal nesting assembly scheme. For non-standard packages that can be immediately nested and matched with the current layer to be stacked, they are determined to be directly stacked and palletized. The robot is then controlled to directly place the non-standard packages into the corresponding nesting positions on the target pallet, while simultaneously placing the matching standard packages to complete the nested stacking.

[0168] For non-standard packages that cannot be immediately nested and matched, the system transports them to the non-standard pre-assembly area of ​​the horizontal cache, assigns corresponding storage locations according to consignment grouping and type, and stores non-standard packages of the same type and group in adjacent storage locations; at the same time, the system monitors the package information in the pre-assembly area in real time, and when it detects that non-standard packages in a certain group can form a complete nested combination layer with standard packages, it immediately triggers the pre-assembly outbound process.

[0169] The system issues warehouse instructions to the gantry robot in a nested stacking order, sequentially retrieving non-standard and standard packages corresponding to the combined layers and transporting them via conveyor line to the robot's package retrieval position at the palletizing station. The robot, following a nested palletizing scheme, first stacks large standard packages, then places non-standard packages into the corresponding nested gaps, achieving precise stacking with complementary concave and convex shapes and gap filling. After each nested package is stacked, the vision system verifies the stacking position and flatness to ensure compliance with constraints.

[0170] For nested stacking of irregularly shaped packages, the system uses a 3D vision system to identify the stacking posture of the irregularly shaped packages in real time and adjusts the stacking angle of the robot to ensure that the concave area of ​​the irregularly shaped package is completely facing upwards, and the matching small packages are completely embedded in the concave area without any suspension or offset. For irregularly shaped packages that cannot be completely embedded, the system automatically places foam padding of the corresponding thickness at the bottom to compensate for flatness and prevent the upper layer from becoming unstable.

[0171] In the finished product pallet off-line stage of step S103, the system performs a full-area visual scan and verification of the finished product pallets containing non-standard nested stacking. After confirming that the stacking position, nesting matching degree, and flatness of all non-standard packages meet the requirements and there is no risk of suspension, offset, or instability, the pallets are normally transported to the downstream workstation. If there are any non-compliance issues, an alarm is triggered, and the pallets are transported to the adjustment workstation for correction to ensure the quality of the finished products.

[0172] In some embodiments, such as Figure 2 As shown, Figure 2 This document presents a process flow diagram for automated palletizing and intelligent stacking of finished goods in logistics. It addresses industry scenarios where customized home furnishing packages are categorized into standardized online packages adaptable to automated operations and special offline packages not. The document constructs a comprehensive human-machine collaborative process encompassing pre-processing and diversion, automated palletizing in a closed-loop main line, and post-processing supplementation. This fully covers the entire execution logic from finished goods outbound planning, package classification and diversion, automated depalletizing, package information collection and stack type calculation, buffer reordering, automated stacking, empty pallet return and reuse, finished goods post-processing, to final wrapping and packaging. It resolves pain points such as the inability of special irregularly shaped packages to adapt to automated palletizing, insufficient coverage of fully automated operations, and process disconnects. This enables fully unmanned palletizing of standardized packages and human-machine collaborative adaptation of special packages, ensuring the continuity, compliance, and full-scenario coverage of palletizing operations.

[0173] The entire process in this embodiment is divided into 7 core execution stages. The core automated palletizing operation area is delineated by a red dashed box, realizing fully automated closed-loop operation of standardized packages. At the same time, it achieves full-scenario adaptation for special packages through pre- and post-processing manual steps. The specific execution steps strictly match the arrow logic and functional nodes of the flowchart, as detailed below:

[0174] Step 1: Pre-planning of finished goods warehouse outbound and shipment sequence (corresponding to...) Figure 2(Number 1) This step is the pre-planning stage of the entire consignment process. The consignment control system connects to the finished goods warehouse WMS system. Based on the pre-generated consignment palletizing scheme, it plans the outbound sequence of the consignment pallets to be consigned in advance. According to the consignment rules of the same distributor, the same delivery route, and the same receiving customer, the consignment pallets to be consigned are sorted and outbound according to the operation priority to ensure that the outbound sequence of the consignment pallets is completely matched with the consignment operation sequence, avoid consignment operation waiting and process congestion, and provide pre-scheduling guarantee for subsequent automated consignment operations.

[0175] Step 2: Pre-processing and package sorting (corresponding to...) Figure 2 (Number 2) This step is the package classification and diversion link in the entire pallet consolidation process. After the consolidation sub-pallets leave the finished product warehouse, the operators complete the pre-processing: manually identify transparent packaging, long and irregularly shaped items, oversized non-standard items, and other offline packages in the sub-pallets that cannot be adapted to automated depalletizing and palletizing operations, remove them from the sub-pallets in advance, place them in a special tooling trolley, and directly transport them to the subsequent manual post-processing link through the offline flow path shown by the dotted line in the flowchart, without entering the automated consolidation main line; the remaining standardized online packages enter the subsequent automated consolidation operation link with the original pallets, realizing the accurate diversion of special packages and standardized packages, and avoiding the shutdown of the automated main line due to non-standard packages.

[0176] Step 3: Automated depalletizing and empty pallet return and reuse (corresponding to...) Figure 2 (3) This step is the starting point of the automated pallet assembly line. The pre-processed sub-pallets are transported to the depalletizing station via a conveyor line and enter the automated operation area marked by the red dotted box. Through the 3D vision system installed above the depalletizing station, the position and posture information of the packages on the sub-pallets are collected in real time. The system guides a six-axis industrial robot with a vacuum suction cup clamp to automatically depalletize and remove the standardized online packages on the sub-pallets one by one in a top-to-bottom order. Simultaneously, the empty pallet return loop is executed: After all the online packages on the sub-pallets have been depalletized and removed, the empty pallets are transported to the empty pallet stacking station via a conveyor line. The automatic stacking machine completes the stacking and storage of the empty pallets, realizing the empty pallet stacking return. The stacked empty pallets are transferred back to the finished product warehouse for reuse, completing the closed-loop management of the entire pallet process.

[0177] Step 4: Automatic collection of package information and dynamic calculation of stack type (corresponding to...) Figure 2(Number 4) This step is the information processing stage of the automated palletizing main line. After the robot completes the depalletizing and material handling of a single package, it drives the package through the automatic barcode scanning and size measurement station: the package barcode is scanned by an industrial barcode scanner to obtain the package's unique identifier and ownership information; the actual length, width, and height dimensions of the package are collected by a 3D vision measurement system; all the collected package information is uploaded to the palletizing algorithm software in real time. Based on the actual package information, combined with preset palletizing rules and palletizing constraints, the algorithm software calculates and updates the palletizing scheme in real time, providing accurate execution basis for subsequent cache reordering and automatic palletizing.

[0178] Step 5: Packet reordering and outbound scheduling in the cache library (corresponding to...) Figure 2 (Number 5) This step is the sorting and caching stage of the automated palletizing main line. Based on the calculation results of the palletizing algorithm, packages that cannot be directly stacked are transported to the cache area via a chain conveyor. The cache area is equipped with a horizontal three-dimensional cache, a chain cache line, and insertable suction cup clamps. According to the palletizing grouping and stacking requirements, corresponding cache locations are assigned to packages to complete the warehousing, caching, and sorting management of packages. The system monitors the package information of the same palletizing group in the cache in real time. When it detects that the cached packages can form a complete combination layer that meets the palletizing requirements, based on the optimal stacking order calculated by the algorithm, the packages are sorted out of the warehouse in order by insertable suction cup clamps and chain cache lines and transported to the subsequent automated palletizing station to achieve accurate sorting and on-demand outbound of packages.

[0179] Step 6: Automatic palletizing and manual post-processing completion (corresponding to...) Figure 2 (Number 6): This step is the forming stage of the palletized finished product, which is divided into two execution units: automated palletizing and manual post-processing. The first unit is automated palletizing within the main automated line: packages that have been cached and sorted out are transported to the palletizing station. The system guides an industrial robot with a vacuum suction cup clamp to automatically palletize the packages layer by layer according to the palletizing algorithm, forming the main structure of the palletized finished product stack. The second unit is manual post-processing palletizing: the finished product stacks that have completed automated palletizing are transported to the manual post-processing station via a conveyor line. The operators manually palletize the offline packages (transparent packaging, long strips, etc.) sorted out in the pre-processing stage on the top layer of the finished product stack, completing the final palletization of the finished product and realizing the palletization operation of all types of packages.

[0180] Step 7: Finished product wrapped, packaged, and removed from production line (corresponding to...) Figure 2 (Number 7): This step is the final stage of the entire palletizing process. The palletized finished product stacks that have completed manual post-processing are transported to the wrapping and packaging station, where operators complete the wrapping and packaging of the finished product stacks. This stage is a key area for non-automation improvement, and an interface is reserved for future upgrades to automated wrapping equipment. After packaging, the finished product stacks are transferred to the shipping area, completing the entire automated palletizing process.

[0181] In some embodiments, such as Figure 3 As shown, this 3D finished product automated palletizing project's full-process system layout simulation model completely restores the hardware architecture of the automated palletizing process in the finished product outbound stage of customized home furnishings, which consists of two independent robot workstations and a shared horizontal cache sorting library. It is a visual representation of the corresponding core solution, fully covering the entire closed-loop operation from sub-pallet feeding, automatic depalletizing, package information collection, cache sorting, robot palletizing and palletizing, empty pallet recycling to finished product off-line, and is adapted to the industry characteristics of customized home furnishings, which have many non-standard packages and complex palletizing rules.

[0182] The model adopts a linear layout with inlet and outlet at both ends, buffer scheduling in the middle, and symmetrical distribution of two workstations. It can be divided into four core functional units from left to right along the material conveying direction. Each unit is seamlessly connected by roller conveyor lines, and the material flow lines are clear and do not intersect. The details are as follows:

[0183] The front pallet loading and empty pallet supply and recycling unit, located on the far left of the model, is the system's inlet. Its core components include an automatic empty pallet stacking and supply system, a pallet loading roller conveyor line, and a manual loading station. After a forklift transports the pallets to be combined from the finished goods warehouse to this station, the conveyor line automatically and precisely positions the pallets and transports them to the destacking station. Simultaneously, this unit automatically recycles, stacks, and reuses empty pallets, providing an empty pallet supply for pallet combining operations and achieving closed-loop management of the entire pallet process.

[0184] The dual independent depalletizing and palletizing robot workstations are symmetrically arranged on the left and right sides of the model, forming two identical robot workstations that serve as the core execution units of the system. Each workstation is equipped with a six-axis industrial robot (with a vacuum suction cup gripper), a 3D structured light stereo vision system, and a dual-station depalletizing / palletizing conveyor line. A single workstation can independently complete the entire process of "automatic depalletizing + palletizing," while the two workstations can simultaneously and independently handle palletizing tasks for different orders, or they can collaborate to achieve a "left workstation depalletizing, right workstation palletizing" separation mode, significantly improving operational flexibility and efficiency. The vision system can accurately identify package poses, automatically measure dimensions, and detect abnormal packages, guiding the robot to stably depalletize all categories of finished home furnishing packages with lengths of 300-2800mm and widths of 200-1200mm.

[0185] The shared horizontal buffer and reordering storage area, located in the core area between the two workstations, is the system's intelligent reordering unit and a core hardware module shared by both workstations. The storage area adopts a multi-layered, dense horizontal storage design, equipped with a gantry telescopic forklift robot and a chain buffer conveyor line. It is mainly used to store packages that cannot be immediately stacked after depalletizing, enabling package warehousing, intelligent reordering, and stacking for outbound delivery. Storage locations in the area are dynamically allocated according to consignment groups (distributors / delivery routes / receiving customers). The system monitors the stacking feasibility of packages in the same group in real time. When buffered packages can form a complete stacking layer that meets the requirements, the gantry robot automatically retrieves the packages according to stacking priority and transports them to the consignment workstation for stacking. This ensures the regularity of the consignment stacking and significantly reduces the redundant occupation of buffer resources.

[0186] The finished product unloading and human-machine collaborative post-processing unit is located on the far right of the model, serving as the system's output end. It is equipped with a finished product pallet conveyor line, a manual post-processing station, and a safety guardrail. After the finished product stacks are automated and palletized, they are conveyed to this station via the conveyor line. There, operators perform auxiliary tasks such as top-level stacking and wrapping of transparent packaging, long and irregularly shaped packages, and other packages that cannot be adapted to automated operations. This achieves a collaborative connection between the automated main line and manual operations, completing the final unloading and delivery of the palletized finished products.

[0187] The overall layout of the model takes into account operational efficiency, site utilization, and redundancy backup capabilities. The dual-workstation architecture can take over all tasks in the event of a single workstation failure, avoiding downtime of the entire line. Simulation verification shows that the system's single-shift co-consignment operation efficiency is more than 3 times higher than the traditional manual mode, which can significantly reduce the labor intensity of operators and perfectly adapt to the co-consignment operation needs of customized home furnishing finished product logistics.

[0188] like Figure 4 As shown in the figure, this diagram illustrates the simulation results of the fully offline consolidation and palletizing algorithm for the finished product automated consolidation project. It is a schematic diagram of the simulation results of the algorithm in multiple scenarios. It is based on the automated consolidation and palletizing requirements of customized furniture finished products. The algorithm is generated through fully offline pre-calculation and is used to verify the adaptability of the offline consolidation and palletizing algorithm to different package specifications, different consolidation and palletizing rules, and different delivery scenarios. It provides pre-planning and feasibility verification for on-site automated consolidation and palletizing operations.

[0189] The core premises and constraints of this simulation are fully aligned with the actual operational requirements of the project, specifically including:

[0190] Compatible package range: Standardized finished home furnishing packages (such as wardrobe bodies, standard door panels, etc.) whose complete dimensions, weight, and order attribution information can be obtained in advance from the production process system (paper cutter, edge banding machine, packaging machine), covering the full size range of packages required by the project: length 300-2800mm, width 200-1200mm, height 42-150mm.

[0191] Consolidation rules: Strictly follow the consolidation grouping rules of the project by distributor, receiving customer, and delivery route, while also meeting the mandatory constraints of 80-100 pieces per pallet, stacking height, total weight, and stack stability, and setting the core rule that packages of the same customer cannot be split across pallets.

[0192] The core logic of the algorithm is as follows: The fully offline algorithm completes the collection, layer structure arrangement, and pallet planning of packages in the same group 24 hours in advance. It supports two switchable strategies: one is the optimal solution strategy that concentrates packages to the minimum number of pallets (suitable for trunk transportation), and the other is the suboptimal solution strategy that evenly distributes packages to multiple pallets (suitable for same-city delivery), giving priority to ensuring the stability of the pallet structure and the utilization rate of pallet space.

[0193] Simulated variable settings: Covering two cache capacity gradients of 50 packets and 100 packets, to verify the impact of cache resources on the co-location effect.

[0194] Figure 4Divided into two rows, there are a total of 8 sets of simulation results, which fully cover the core co-delivery scenarios of the project. The specific correspondence is as follows: The top row of 4 sets of simulation results corresponds to the core co-delivery scenarios of trunk line / same-city delivery of large-sized multi-item mixed packages, which is the verification of the main operation scenarios of the project. This simulation uses large-sized cabinets and standard door panels as the core packages, with a wide size range, to adapt to different consolidation rules for different delivery distances: Upstream groups 1-2 correspond to same-city delivery within <150km, consolidating packages according to the receiving customer, verifying the consolidation effect with buffer capacities of 50 and 100 packages respectively. The stacking type mainly uses large-sized packages stacked flat, with a neat layer structure and balanced package distribution, adapting to the needs of multiple batches of same-city delivery. Upstream groups 3-4 correspond to trunk line transportation >500km, consolidating packages according to the delivery route, verifying the consolidation effect with buffer capacities of 50 and 100 packages respectively. A nested stacking method is used, with large-sized packages as the main body and smaller packages filling the gaps, maximizing the space utilization of a single pallet, reducing the number of pallets transported, and lowering trunk logistics costs. Downstream group 4 simulation results: corresponding to a mixed consolidation scenario of standard single-item packages + small irregularly shaped packages, covering verification of subdivided scenarios such as special package adaptation and human-machine collaborative operation. This simulation covers various package types, including standard single-item packages, narrow hardware packages, and small-sized irregularly shaped packages, to verify the algorithm's adaptability to complex package combinations. Downlink groups 1-2 correspond to intercity delivery scenarios of 150-500km and consignment by commercial location, verifying the consignment effects of two algorithm strategies: suboptimal solution (average distribution) and optimal solution (centralized stacking). The suboptimal solution provides a more balanced distribution of stacked packages, while the optimal solution results in a more compact stack and higher space utilization. Downlink group 3 corresponds to a mixed consignment scenario involving small-sized narrow packages and hardware packages, verifying the algorithm's ability to arrange small packages within layers and fill gaps, achieving standardized consignment of scattered packages and avoiding space waste. Downlink group 4 corresponds to a consignment scenario with a reserved top-level manual post-processing area. The top of the stack has a reserved flat stacking area, adapting to the project's requirement for "manual top-level processing of transparent packaging and long, irregularly shaped packages," achieving seamless integration between automated mainline consignment and manual post-processing.

[0195] Figure 4 The feasibility and adaptability of the offline consolidation algorithm for the project were fully verified by eight sets of fully offline simulation results: First, the algorithm can achieve compliant consolidation of finished home furnishing packages of different specifications and types, strictly meeting the consolidation rules and palletizing constraints of the project; Second, through simulation comparison of different cache capacities, it was verified that a cache capacity of 100 packages can increase the average space utilization of a single pallet by 12%, providing data support for the design of the storage location of the project's cache warehouse; Third, the dual-strategy algorithm can flexibly adapt to the consolidation needs of different delivery scenarios, providing a palletization scheme that can be directly executed for on-site automated consolidation operations, enabling the immediate unpacking and palletizing of packages with known information, greatly improving the efficiency of consolidation operations, and solving the industry pain points of low efficiency, irregular palletization, and low space utilization of traditional manual consolidation.

[0196] Please see Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of the automated palletizing and intelligent stacking system 200 for finished goods provided in this application embodiment. The automated palletizing and intelligent stacking system 200 for finished goods is used to execute the steps of the automated palletizing and intelligent stacking methods for finished goods shown in the above embodiments. The automated palletizing and intelligent stacking system 200 for finished goods can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0197] like Figure 5 As shown, the automated palletizing and intelligent stacking system 200 for finished goods logistics includes:

[0198] Information acquisition unit 201 is used to acquire basic information of packages to be combined into pallets, call a preset pallet combining algorithm based on the basic information of packages, generate a pallet combining and palletizing scheme; transport the pallet to be combined to the depalletizing station, identify whether the pallet is a basic pallet that meets the requirements, and after removing the top non-combined layer packages from the basic pallet, directly transport it to the pallet combining station as a pallet combining base pallet.

[0199] The control gripping unit 202 is used to guide the preset robot to complete the depalletizing of the sub-pallet, collect the actual size information of the depalletized packages and update the palletizing scheme; according to the palletizing scheme, it determines whether the depalletized packages can be directly stacked; for packages that can be directly stacked, it controls the robot to transfer the palletizing to the target pallet at the palletizing station; for packages that cannot be directly stacked, it is transported to the horizontal buffer for reordering and buffering; if it is determined that the packages to be buffered form a combination layer that meets the requirements, it is taken out of the buffer and positioned to the palletizing station, where the robot grips and stacks them to the target pallet.

[0200] The palletizing unit 203 is used to transport the target pallet to the downstream workstation after palletizing and to transport the empty pallet to the recycling workstation, thereby realizing automatic palletizing and intelligent palletizing of finished products in logistics.

[0201] In some embodiments, the step of calling a preset consolidation algorithm based on package basic information to generate a consolidation and palletizing scheme includes: pre-setting consolidation grouping rules and palletizing constraint rules, wherein the consolidation grouping rules include rules for grouping consolidation ...

[0202] In some embodiments, after the palletizing and stacking are completed, the target pallet is transported to the downstream station, and the empty pallet is transported to the recycling station to realize automatic palletizing and intelligent palletizing of finished logistics products. This includes: monitoring the palletizing progress at the palletizing station; when the palletizing and stacking task of the target pallet is detected to be completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top-level special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage, and return of the empty pallets for later use.

[0203] In some embodiments, the step of conveying the sub-pallet to be combined to the destacking station and identifying whether the sub-pallet is a qualified base pallet includes: after conveying the sub-pallet to be combined to the destacking station via a conveyor line, acquiring the overall stacking image and package layout information of the sub-pallet through a preset vision system; combining the pre-stored base pallet judgment rules, extracting the flatness of the bottom stacking of the sub-pallet, the regularity of the package layout, and the consistency of the package size of a single pallet, and judging whether the corresponding sub-pallet meets the requirements as a base pallet for combining pallets by comparing the parameters.

[0204] In some embodiments, the step of directly conveying the base pallet to the pallet-combining station as a base pallet after removing the top non-combined layer packages includes: when the sub-pallet is determined to be a base pallet that meets the requirements, identifying the non-combined layer packages on the top layer of the base pallet that do not meet the requirements for flat stacking according to the pallet-combining stacking scheme; controlling the robot to remove the corresponding non-combined layer packages, while retaining the regular flat structure of the bottom layer of the base pallet; after all the top non-combined layer packages have been removed, conveying the base pallet directly from the destacking station to the pallet-combining station via a conveyor line for use as a base pallet for pallet-combining stacking.

[0205] In some embodiments, guiding a pre-defined robot to complete the depalletizing of sub-pallets includes: acquiring in real time three-dimensional point cloud data and color image data of the sub-pallets at the depalletizing station using a structured light stereo vision system mounted above the depalletizing station; identifying the package pose information of the topmost package on the sub-pallet, wherein the package pose information includes at least the corresponding contour, position, and orientation information, to identify whether the package is damaged, offset, or stacked abnormally; planning the robot's depalletizing motion path based on the identified package pose information, guiding the robot to complete the precise depalletizing and material handling of a single package using a suction cup gripper, and triggering a pre-defined abnormality handling process for any identified abnormal packages.

[0206] In some embodiments, the step of collecting the actual size information of the depalletized packages and updating the palletizing scheme includes: during the process of the robot completing the depalletizing and material handling of the packages, simultaneously collecting the actual size information corresponding to the actual length, width, and height of the depalletized packages through a vision system, and obtaining the unique identification information of the packages through a barcode scanning device; after binding the collected actual size information with the unique identification information, comparing the actual size information of the packages with the pre-stored basic information of the packages; if there is an information discrepancy, based on the real-time collected actual size information, calling the corresponding palletizing algorithm to update the palletizing scheme, and simultaneously adjusting the depalletizing, caching, and palletizing scheduling strategies for subsequent packages.

[0207] In some embodiments, determining whether depalletized packages can be directly stacked according to the palletizing scheme, and controlling the robot to transfer the palletized packages to the target pallet at the palletizing station for packages that can be directly stacked, includes: according to the updated palletizing scheme, verifying whether the depalletized packages can be matched with the stacking layers of the target pallet at the current palletizing station, and verifying whether the stacking order of the packages meets the palletizing rules; if it is determined that the packages can be directly stacked, planning the robot's palletizing motion path, controlling the robot to transfer the packages directly from the depalletizing station to the corresponding position on the target pallet at the palletizing station, and automatically performing padding filling for uneven positions after stacking.

[0208] In some embodiments, the process of conveying packages that cannot be directly stacked to a horizontal buffer for reordering and caching, and then, if it is determined that the packages to be cached form a combination layer that meets the requirements, being outsourced and positioned at the palletizing station, where a robot picks up and stacks them onto the target pallet, includes: if it is determined that the depalletized packages cannot be directly stacked, the packages are conveyed to the horizontal buffer via a conveyor line and a gantry robot, and corresponding buffer locations are allocated according to the palletizing group and stacking requirements to complete the warehousing and reordering management of the packages; monitoring the package information of the same palletizing group in the horizontal buffer, and when it is detected that the cached packages in the group can form a complete combination layer that meets the palletizing requirements, the packages corresponding to the combination layer are sequentially outsourced by the gantry robot according to the stacking order, conveyed to the robot pickup position at the palletizing station via a conveyor line, and then sequentially picked up and stacked onto the corresponding positions of the target pallet according to the palletizing plan.

[0209] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the above-described automated palletizing and intelligent stacking system for finished goods and its modules can be found in the corresponding contents of the above-described embodiments of the automated palletizing and intelligent stacking method for finished goods, and will not be repeated here.

[0210] The aforementioned automated palletizing and intelligent stacking method for finished goods in logistics can be implemented as a computer program, which can be used in, for example... Figure 5It runs on the device shown.

[0211] Please see Figure 6 , Figure 6 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0212] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any method of automated palletizing and intelligent stacking of finished goods in logistics.

[0213] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0214] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When these computer programs are executed by a processor, the processor can execute any method of automated palletizing and intelligent stacking of finished goods in logistics.

[0215] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0216] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0217] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0218] Obtain the basic information of the package to be combined into a pallet, call the preset pallet combining algorithm based on the basic information of the package, and generate a pallet combining and palletizing scheme; transport the pallet to be combined into a pallet to the depalletizing station, identify whether the pallet is a basic pallet that meets the requirements, and after removing the top non-combined layer package from the basic pallet, directly transport it to the pallet combining station as a pallet combining base pallet.

[0219] The system guides a pre-programmed robot to complete the depalletization of sub-pallets, collects the actual size information of the depalletized packages, and updates the palletization and stacking scheme. Based on the palletization and stacking scheme, it determines whether the depalletized packages can be directly stacked. For packages that can be directly stacked, the system controls the robot to transfer the stacking to the target pallet at the palletization station. For packages that cannot be directly stacked, the system transports them to a horizontal buffer for reordering and buffering. If it is determined that the packages to be buffered form a combination layer that meets the requirements, the system is taken out of the buffer and positioned at the palletization station, where the robot picks up the packages and stacks them to the target pallet.

[0220] After palletizing and stacking are completed, the target pallet is transported to the downstream workstation, and the empty pallet is transported to the recycling workstation, realizing automatic palletizing and intelligent stacking of finished products in logistics.

[0221] In some embodiments, the step of calling a preset consolidation algorithm based on package basic information to generate a consolidation and palletizing scheme includes: pre-setting consolidation grouping rules and palletizing constraint rules, wherein the consolidation grouping rules include rules for grouping consolidation ...

[0222] In some embodiments, after the palletizing and stacking are completed, the target pallet is transported to the downstream station, and the empty pallet is transported to the recycling station to realize automatic palletizing and intelligent palletizing of finished logistics products. This includes: monitoring the palletizing progress at the palletizing station; when the palletizing and stacking task of the target pallet is detected to be completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top-level special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage, and return of the empty pallets for later use.

[0223] In some embodiments, the step of conveying the sub-pallet to be combined to the destacking station and identifying whether the sub-pallet is a qualified base pallet includes: after conveying the sub-pallet to be combined to the destacking station via a conveyor line, acquiring the overall stacking image and package layout information of the sub-pallet through a preset vision system; combining the pre-stored base pallet judgment rules, extracting the flatness of the bottom stacking of the sub-pallet, the regularity of the package layout, and the consistency of the package size of a single pallet, and judging whether the corresponding sub-pallet meets the requirements as a base pallet for combining pallets by comparing the parameters.

[0224] In some embodiments, the step of directly conveying the base pallet to the pallet-combining station as a base pallet after removing the top non-combined layer packages includes: when the sub-pallet is determined to be a base pallet that meets the requirements, identifying the non-combined layer packages on the top layer of the base pallet that do not meet the requirements for flat stacking according to the pallet-combining stacking scheme; controlling the robot to remove the corresponding non-combined layer packages, while retaining the regular flat structure of the bottom layer of the base pallet; after all the top non-combined layer packages have been removed, conveying the base pallet directly from the destacking station to the pallet-combining station via a conveyor line for use as a base pallet for pallet-combining stacking.

[0225] In some embodiments, guiding a pre-defined robot to complete the depalletizing of sub-pallets includes: acquiring in real time three-dimensional point cloud data and color image data of the sub-pallets at the depalletizing station using a structured light stereo vision system mounted above the depalletizing station; identifying the package pose information of the topmost package on the sub-pallet, wherein the package pose information includes at least the corresponding contour, position, and orientation information, to identify whether the package is damaged, offset, or stacked abnormally; planning the robot's depalletizing motion path based on the identified package pose information, guiding the robot to complete the precise depalletizing and material handling of a single package using a suction cup gripper, and triggering a pre-defined abnormality handling process for any identified abnormal packages.

[0226] In some embodiments, the step of collecting the actual size information of the depalletized packages and updating the palletizing scheme includes: during the process of the robot completing the depalletizing and material handling of the packages, simultaneously collecting the actual size information corresponding to the actual length, width, and height of the depalletized packages through a vision system, and obtaining the unique identification information of the packages through a barcode scanning device; after binding the collected actual size information with the unique identification information, comparing the actual size information of the packages with the pre-stored basic information of the packages; if there is an information discrepancy, based on the real-time collected actual size information, calling the corresponding palletizing algorithm to update the palletizing scheme, and simultaneously adjusting the depalletizing, caching, and palletizing scheduling strategies for subsequent packages.

[0227] In some embodiments, determining whether depalletized packages can be directly stacked according to the palletizing scheme, and controlling the robot to transfer the palletized packages to the target pallet at the palletizing station for packages that can be directly stacked, includes: according to the updated palletizing scheme, verifying whether the depalletized packages can be matched with the stacking layers of the target pallet at the current palletizing station, and verifying whether the stacking order of the packages meets the palletizing rules; if it is determined that the packages can be directly stacked, planning the robot's palletizing motion path, controlling the robot to transfer the packages directly from the depalletizing station to the corresponding position on the target pallet at the palletizing station, and automatically performing padding filling for uneven positions after stacking.

[0228] In some embodiments, the process of conveying packages that cannot be directly stacked to a horizontal buffer for reordering and caching, and then, if it is determined that the packages to be cached form a combination layer that meets the requirements, being outsourced and positioned at the palletizing station, where a robot picks up and stacks them onto the target pallet, includes: if it is determined that the depalletized packages cannot be directly stacked, the packages are conveyed to the horizontal buffer via a conveyor line and a gantry robot, and corresponding buffer locations are allocated according to the palletizing group and stacking requirements to complete the warehousing and reordering management of the packages; monitoring the package information of the same palletizing group in the horizontal buffer, and when it is detected that the cached packages in the group can form a complete combination layer that meets the palletizing requirements, the packages corresponding to the combination layer are sequentially outsourced by the gantry robot according to the stacking order, conveyed to the robot pickup position at the palletizing station via a conveyor line, and then sequentially picked up and stacked onto the corresponding positions of the target pallet according to the palletizing plan.

[0229] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the automated consignment and intelligent palletizing method for finished goods logistics provided in any embodiment of this application.

[0230] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0231] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for automatic palletizing and intelligent stacking of finished goods in logistics, characterized in that, include: The process involves: acquiring basic package information for the sub-pallets to be combined; invoking a pre-defined pallet-combining algorithm based on the package information to generate a pallet-combining and palletizing scheme; pre-setting pallet-combining grouping rules and palletizing constraint rules; the pallet-combining grouping rules including rules for grouping pallets according to distributor information, delivery route information, and receiving customer information; and the palletizing constraint rules including restrictions on single pallet stacking height, total weight, and number of packages; extracting basic package information for the sub-pallets to be combined and determining whether the package size and weight information can be obtained in advance from the production process system; if the package information can be obtained in advance, invoking an offline pallet-combining algorithm to complete pallet grouping and pallet type planning in advance and generating a pallet-combining and palletizing scheme; if the package information cannot be obtained in advance, invoking a real-time pallet-combining algorithm, reserving inventory redundancy for the cache group layer, and generating an initial pallet-combining and palletizing scheme; and transporting the sub-pallets to be combined to the depalletizing station, identifying whether the sub-pallets are compliant base pallets, and directly transporting the base pallets after removing the top non-combination layer packages as pallet-combining base pallets. The process of guiding a pre-programmed robot to destacking sub-pallets (excluding the base pallet) includes: real-time acquisition of 3D point cloud data and color image data of the sub-pallets at the destacking station using a structured light stereo vision system mounted above the destacking station; identification of the pose information of the topmost package on the sub-pallet, including at least its contour, position, and orientation information, to identify whether the package is damaged, misaligned, or has stacking anomalies; and planning the robot's destacking motion path based on the identified package pose information, guiding the robot to precisely destacking single packages using suction cup grippers. The system picks up materials and triggers a preset exception handling process for any identified abnormal packages. It also collects the actual size information of the depalletized packages and updates the palletizing scheme. Based on the palletizing scheme, it determines whether the depalletized packages can be directly stacked. For packages that can be directly stacked, the robot is controlled to transfer the pallet to the target pallet at the palletizing station. For packages that cannot be directly stacked, they are transported to the horizontal buffer for sorting and caching. If it is determined that the packages to be cached form a combination layer that meets the requirements, the packages to be cached are taken out of the buffer and moved to the palletizing station, where the robot picks them up and stacks them onto the target pallet. After palletizing is completed, the target pallet is transported to the downstream station, and the empty pallet is transported to the recycling station, realizing automatic palletizing and intelligent palletizing of finished products in logistics. This includes: monitoring the palletizing progress at the palletizing station; when the palletizing task of the target pallet is completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage, and return of the empty pallets for later use.

2. The method according to claim 1, characterized in that, The process of conveying the pallet to be assembled to the destacking station and identifying whether the pallet is a qualified base pallet includes: After the pallets to be stacked are transported to the destacking station via the conveyor line, the overall stacking image of the pallets and the package layout information are collected by the preset vision system. Based on the pre-stored basic pallet judgment rules, the flatness of the bottom stacking of the sub-pallet, the regularity of the package arrangement, and the consistency of the package size of a single pallet are extracted. By comparing the parameters, it is determined whether the corresponding sub-pallet meets the requirements for being a basic pallet for a combined pallet.

3. The method according to claim 2, characterized in that, After removing the top non-assembly layer packaging from the base pallet, it is directly transported to the pallet assembly station as the pallet assembly base pallet, including: When a sub-pastor is determined to be a compliant base pallet, the non-combined layer packages that do not meet the flat stacking requirements on the top layer of the base pallet are identified according to the combined pallet stacking scheme. Control the robot to remove the corresponding non-combined layer packages, while preserving the regular flat structure of the bottom layer of the base tray; After all the non-assembled packages on the top layer are removed, the base pallet is transported directly from the destacking station to the pallet assembly station via a conveyor line, and used as the base pallet for pallet assembly and stacking.

4. The method according to claim 1, characterized in that, The process of collecting the actual size information of the unpacked packages and updating the palletizing scheme includes: During the process of the robot unpacking and picking up packages, the actual size information corresponding to the actual length, width and height of the unpacked packages is collected synchronously through the vision system, and the unique identification information of the packages is obtained through the barcode scanning device. After binding the collected actual size information with the unique identifier information, the actual size information of the package is compared with the pre-stored basic package information. If there is a discrepancy, the corresponding palletizing algorithm is called to update the palletizing scheme based on the real-time collected actual size information, and the depalletizing, caching and palletizing scheduling strategies of subsequent packages are adjusted synchronously.

5. The method according to claim 4, characterized in that, The step of determining whether depalletized packages can be directly stacked according to the palletizing scheme, and controlling the robot to transfer the pallet to the target pallet at the palletizing station for packages that can be directly stacked, includes: According to the updated palletizing scheme, for packages that have been depalletized, it is verified whether the packages can be matched with the stacking layer of the target pallet at the current palletizing station, and at the same time, it is verified whether the stacking order of the packages meets the requirements of the palletizing rules. If it is determined that the package can be directly stacked, the robot's stacking motion path is planned, and the robot is controlled to transfer the package directly from the depalletizing station to the corresponding position on the target pallet at the palletizing station. At the same time, padding and filling are automatically performed on any uneven areas after stacking.

6. The method according to claim 5, characterized in that, For packages that cannot be directly stacked, they are transported to a horizontal buffer for sorting and caching. If it is determined that the packages to be cached form a suitable stacking layer, the packages to be cached are taken out of the buffer and moved to the palletizing station, where a robot picks them up and stacks them onto the target pallet. This includes: If it is determined that the depalletized packages cannot be directly stacked, the packages are transported to the horizontal buffer warehouse via a conveyor line and gantry robot. The corresponding buffer locations are allocated according to the consignment grouping and stacking requirements to complete the warehousing and ordering management of the packages. The system monitors package information within the same pallet group in the horizontal buffer warehouse. When it detects that the buffer packages within the group can form a complete combination layer that meets the palletizing requirements, the gantry robot will sequentially remove the packages corresponding to the combination layer from the warehouse according to the stacking order. The packages will then be transported via a conveyor line to the robot pickup position at the palletizing station, where the robot will sequentially grab and stack the packages to the corresponding positions on the target pallet according to the palletizing plan.

7. An automated palletizing and intelligent stacking system for finished goods in logistics, characterized in that, The method applied to any one of claims 1-6 includes: The information acquisition unit is used to acquire the basic information of packages in the sub-pallets to be combined, and to call a preset consolidation algorithm based on the basic information of the packages to generate a consolidation and palletizing scheme. This includes: pre-setting consolidation grouping rules and palletizing constraint rules, wherein the consolidation grouping rules include rules for grouping consolidation pallets according to distributor information, delivery route information, and receiving customer information, and the palletizing constraint rules include restrictions on single pallet stacking height, total weight, and number of packages; extracting the basic information of packages in the sub-pallets to be combined, and determining whether the package size and weight information can be obtained in advance from the production process system; if the package information can be obtained in advance, calling the offline consolidation algorithm to complete the consolidation grouping and pallet type planning in advance, and generating a consolidation and palletizing scheme; if the package information cannot be obtained in advance, calling the real-time consolidation algorithm, reserving inventory redundancy for the cache group layer, and generating an initial consolidation and palletizing scheme; transporting the sub-pallets to be combined to the depalletizing station, identifying whether the sub-pallets are compliant base pallets, and after removing the top non-combination layer packages from the base pallets, directly transporting them to the consolidation station as consolidation base pallets; The control gripping unit guides a pre-programmed robot to destacking sub-pallets (excluding the base pallet), including: real-time acquisition of 3D point cloud data and color image data of the sub-pallets at the destacking station using a structured light stereo vision system mounted above the destacking station; identification of the package pose information of the topmost package on the sub-pallet, the package pose information including at least the corresponding contour, position, and orientation information, to identify whether the package is damaged, misaligned, or has stacking abnormalities; and planning the robot's destacking motion path based on the identified package pose information, guiding the robot to destacking a single package using a suction cup gripper. The system accurately depalletizes and retrieves materials, while triggering a preset exception handling process for identified abnormal packages. It also collects the actual size information of the depalletized packages and updates the palletizing scheme. Based on the palletizing scheme, it determines whether the depalletized packages can be directly stacked. For packages that can be directly stacked, the robot is controlled to transfer the pallet to the target pallet at the palletizing station. For packages that cannot be directly stacked, they are transported to the horizontal buffer for reordering and caching. If it is determined that the packages to be cached form a combination layer that meets the requirements, the packages to be cached are taken out of the buffer and moved to the palletizing station, where the robot grabs and stacks them onto the target pallet. The palletizing unit is used to transport the target pallet to the downstream station after palletizing and to transport the empty pallet to the recycling station, realizing automatic palletizing and intelligent palletizing of finished products. It includes: monitoring the palletizing progress at the palletizing station; when the palletizing task of the target pallet is completed, the target pallet is transported to the manual post-processing station via a conveyor line to complete the stacking and auxiliary processing of the top special packages; and monitoring the pallet status at the depalletizing station; when all packages on the pallet have been depalletized and unpacked, the empty pallet is transported to the stacking station via a conveyor line to complete the stacking, storage and return of the empty pallets for later use.

Images