Sorting method, storage medium, and electronic device

Abstract

This application discloses a sorting method, storage medium, and electronic device, comprising: receiving a sorting request sent by a sorting machine, wherein the sorting request includes package information of a package to be sorted, and the package information includes destination information; inputting the package information into a sorting decision model to obtain the sorting slot information to be placed in, wherein the sorting decision model is a model trained based on an intelligent agent algorithm, used to determine the sorting slot information for the package to be sorted based on the destination information; generating a sorting instruction including the sorting slot information, and sending the sorting instruction to the sorting machine, so that the sorting machine can place the package to be sorted in the slot corresponding to the sorting slot information according to the sorting instruction. The technical solution of this application can automatically generate the sorting slot information corresponding to the package to be sorted, thereby improving sorting efficiency.

Description

Technical Field

[0001] This application relates to the field of logistics sorting technology, specifically to a sorting method, storage medium, and electronic equipment. Background Technology

[0002] With the rapid development of e-commerce, the logistics industry is facing unprecedented challenges in the sorting of small items. Currently, sorting requires staff to manually create sorting plans based on the characteristics of small items, packing all parcels (e.g., small items) destined for the same destination into a single package for easier transport in subsequent stages. However, if the sorting plan is flawed, packages may be sent to the wrong destination, and these misdelivered packages may not be discovered until the next transit point, thus affecting delivery times.

[0003] Therefore, how to automatically generate sorting plans for small items and improve sorting efficiency has become an urgent technical problem to be solved. Summary of the Invention

[0004] In view of this, embodiments of this application provide a sorting method, a storage medium, and an electronic device that can automatically generate sorting plans for packages to be sorted, thereby improving sorting efficiency and saving labor costs.

[0005] In a first aspect, embodiments of this application provide a sorting method applied to a server, comprising: receiving a sorting request sent by a sorting machine, wherein the sorting request includes package information of a package to be sorted, and the package information includes destination information; inputting the package information into a sorting decision model to obtain the slot information to which the package to be sorted should be placed, wherein the sorting decision model is a model trained based on an intelligent agent algorithm, used to determine the slot information for the package to be sorted based on the destination information; generating a sorting instruction including the slot information, and sending the sorting instruction to the sorting machine, so that the sorting machine can place the package to be sorted into the slot corresponding to the slot information according to the sorting instruction.

[0006] Secondly, embodiments of this application provide a sorting method applied to a sorting machine, the sorting machine including a scanner, wherein the method includes: scanning a package to be sorted using the scanner to obtain package information of the package to be sorted; sending a sorting request including package information to a server, wherein the package information includes destination information; receiving a sorting instruction sent by the server, wherein the sorting instruction is generated based on a sorting decision model running in the server, the sorting decision model is a model trained based on an intelligent agent algorithm, used to determine the corresponding slot information of the package to be sorted according to the destination information, the sorting instruction including the slot information; and placing the package to be sorted into the slot position corresponding to the slot information according to the sorting instruction.

[0007] Thirdly, embodiments of this application provide a sorting system, including a sorting machine and a server, wherein the sorting machine includes a repositioning area, the repositioning area includes a scanner for scanning packages to be sorted, and the sorting machine is used to execute the sorting method described in the second aspect above; the server is used to execute the sorting method described in the first aspect above.

[0008] Fourthly, embodiments of this application provide a computationally readable storage medium storing a computer program for performing the sorting method described in either the first or second aspect above.

[0009] Fifthly, embodiments of this application provide an electronic device, including: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute the sorting method described in either the first or second aspect above.

[0010] Sixthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the sorting method described in either the first or second aspect above.

[0011] This application provides a sorting method, storage medium, and electronic device. By receiving a sorting request sent by a sorting machine, the package information included in the sorting request is input into a sorting decision model to obtain the grid information where the package to be sorted needs to be placed. Then, a sorting instruction including the grid information is generated and sent to the sorting machine so that the sorting machine can place the package to be sorted into the grid corresponding to the grid information according to the sorting instruction. This application can automatically generate a sorting plan for the package to be sorted, improving sorting efficiency. Compared with the traditional method of specifying the sorting plan by a sorting plan administrator, this application can adapt to different batches and different express delivery distributions, reducing labor costs. Attached Figure Description

[0012] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the embodiments of the present disclosure to explain the disclosure and do not constitute a limitation thereof. The above and other features and advantages will become more apparent to those skilled in the art from the detailed description of exemplary embodiments with reference to the accompanying drawings, in which:

[0013] Figure 1 This is a schematic diagram of the small parcel sorting system provided in the embodiments of this application.

[0014] Figure 2 This is a flowchart illustrating a sorting method provided in an exemplary embodiment of this application.

[0015] Figure 3This is a flowchart illustrating a sorting method provided in another exemplary embodiment of this application.

[0016] Figure 4 This is a schematic diagram of the structure of a sorting device provided in an exemplary embodiment of this application.

[0017] Figure 5 This is a schematic diagram of the structure of a sorting device provided in another exemplary embodiment of this application.

[0018] Figure 6 This is a block diagram of an electronic device for sorting provided in an exemplary embodiment of this application. Detailed Implementation

[0019] 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, and 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.

[0020] Logistics transit centers are crucial nodes in the lifecycle of express parcels. Upon arrival, parcels are sorted by automated assembly line equipment and manual labor according to their destinations. The speed of sorting directly impacts delivery time. Equipment within transit centers includes rack-type sorters, small-parcel sorters, and sorting cabinets. Most parcels sorted at the outpost are small packages. Outlets pack these small packages into smaller parcels and transport them to the transit center for sorting. The transit center unpacks these smaller parcels and sorts them through the sorting machines to their respective destination compartments for repacking before loading them onto trucks for dispatch. Therefore, the efficiency of small-parcel sorting significantly affects delivery time.

[0021] In related technologies, sorting small parcels requires a sorting planner to develop a sorting plan based on the incoming parcels (i.e., small parcels) to group all small parcels destined for the same destination into a single package for easier transport in subsequent stages. However, since the incoming parcel situation (i.e., the destination of each small parcel within a batch) varies for each shift and even each day, the sorting efficiency of the sorting machine will be affected. Although a well-developed sorting plan by the sorting planner can address these differences to some extent, this method requires the sorting planner to create multiple sorting plans for use in different waves (i.e., different time periods and different parcel distributions).

[0022] Furthermore, if the upstream and downstream logistics routes of the transit center change, the sorting plan also needs to be modified accordingly. Even if it takes a lot of time and effort, the most suitable sorting plan may not be obtained.

[0023] To address the aforementioned problems, this application provides a sorting method. Various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 This is a schematic diagram of a small parcel sorting system provided in an embodiment of this application. The sorting system includes an unloading port 110, a sorting machine 120, a server 130, and a loading port 140. The sorting machine 120 may include a repositioning area 121, a trolley 122, a scanner 123, and a sorting slot 124. The repositioning area 121 may include a scanner for scanning the barcode (or QR code) of the parcel to be sorted to obtain parcel information. The server 130 may include a sorting management system 131 and a programmable logic controller (PLC) system 132. The sorting management system 131 may be an Automated Sorting and Conveying System (ASCS), which is responsible for querying and matching the destination slots of the parcels to be sorted. The PLC system 132 is responsible for the motion control of the entire sorting machine.

[0025] After being unloaded from unloading port 110, the parcels awaiting sorting are transported to the unloading area 121 of sorting machine 120 via a swing wheel. Workers unseal the parcels and empty them. A scanner on unloading area 121 scans the unloaded parcels and sends the scanned parcel information to sorting management system 131. Workers on feeding station 125 place the unloaded parcels onto the feeding station 125, which then transfers them to the trolley 122 of sorting machine 120. The trolley 122, carrying the parcels, rotates along the sorting line. When the cart 122 passes the scanner 123, the scanner 123 will scan the packages to be sorted on the cart 122 again and send a sorting request to the sorting management system 131. The sorting management system 131 calculates the destination slot (i.e., slot information) corresponding to the package to be sorted through its internal sorting decision model. The slot can be understood as a location used to store multiple packages to be sorted with the same sorting requirements, such as storing packages with the same destination.

[0026] Furthermore, once the sorting management system 131 obtains the corresponding slot for the package to be sorted, it issues a sorting instruction to the programmable logic controller (PLC) system 132. This allows the PLC system 132 to control the trolley 122 of the sorting machine 120 to move to the corresponding slot 124 and place the package to be sorted into the slot 124. When the number of packages to be sorted in the slot 124 reaches a certain amount (e.g., the slot is full), the packages in the slot 124 are packaged and sent to the loading port 140, and the slot is released to allow it to be linked to other package flows, thus completing the package sorting process.

[0027] Figure 2 This is a flowchart illustrating a sorting method provided in an exemplary embodiment of this application. Figure 2 The method is provided by computing devices, such as servers (i.e., Figure 1 Use server 130 to execute. Figure 2 As shown, the sorting method includes the following:

[0028] S210: Receives a sorting request sent by the sorting machine.

[0029] In one embodiment, the sorting request includes package information of the parcel to be sorted, including destination information. The parcel to be sorted may display an automatic identification code such as a barcode or QR code.

[0030] In one embodiment, the sorting machine may include a packing area, a scanner, a feeding station, and a cart, wherein the packing area includes the scanner.

[0031] Specifically, staff will unpack the logistics packages in the unpacking area and empty them to obtain packages to be sorted. The packages to be sorted will then be placed on the feeding platform, which will transfer the packages to a trolley. The trolley carrying the packages to be sorted will rotate along the sorting line. When the trolley passes the scanner, the scanner will scan the QR code on the package to obtain the package information and send a sorting request including the package information to the server.

[0032] In one embodiment, the package information may include the destination information of the package to be sorted (e.g., detailed address such as city, urban area or street), mode of transportation (e.g., air, land or sea transport), recipient and time of receipt.

[0033] S220: Input the package information into the sorting decision model to obtain the sorting slot information for the packages to be sorted.

[0034] In one embodiment, the sorting decision model is a model trained based on an agent algorithm, used to determine the compartment information for the packages to be sorted based on the destination information.

[0035] Specifically, a sorting decision model runs on the server. This model is trained based on an agent algorithm. The package information of the parcel to be sorted serves as the input to the sorting decision model, while the information about the appropriate sorting slot for that parcel is the output. In other words, the parcel information is input into the sorting decision model, and the output is the corresponding sorting slot information. This slot information can include slot numbering or other information used to identify the sorting slot.

[0036] For example, taking the destination information as a city (e.g., Tianjin), the destination of the package to be sorted is input into the sorting decision model to obtain information on which the package to be sorted needs to be placed in 16 slots (i.e., slot information).

[0037] The sorting decision model can be understood as an intelligent agent. This sorting decision model can perceive and adapt to the state of the sorting environment and make sorting decisions autonomously during the sorting process, that is, it can autonomously make the sorting slot information that the package to be sorted should be placed in based on the destination information of the package to be sorted.

[0038] It should be noted that the intelligent agent in this embodiment has functions such as perception, reasoning, decision-making, and learning, and its goal is to make adaptive behaviors based on feedback from the sorting environment. The core of the intelligent agent is its interaction with the sorting environment; it selects behaviors and optimizes goals based on the environmental state.

[0039] S230: Generate a sorting instruction including grid information and send the sorting instruction to the sorting machine so that the sorting machine can place the package to be sorted into the grid corresponding to the grid information according to the sorting instruction.

[0040] Specifically, the sorting management system in the server generates sorting instructions that include grid information. Then, the sorting management system sends the sorting instructions to the programmable logic controller (PLC) system, so that the PLC system can control the trolley in the sorting machine that is loaded with packages to be sorted and put the packages to be sorted into the grid indicated by the grid information.

[0041] For example, parcels destined for Tianjin are placed into compartment number 16 using a mobile cart.

[0042] Therefore, this embodiment of the application receives a sorting request sent by a sorting machine, inputs the package information included in the sorting request into a sorting decision model, obtains the grid information where the package to be sorted needs to be placed, generates a sorting instruction including the grid information, and sends the sorting instruction to the sorting machine so that the sorting machine can place the package to be sorted into the grid corresponding to the grid information according to the sorting instruction. This embodiment of the application can automatically generate a sorting plan for the package to be sorted, improving sorting efficiency. Compared with the traditional method of specifying the sorting plan by a sorting plan administrator, this embodiment of the application can adapt to different batches and different item distributions, reducing labor costs.

[0043] In one embodiment of this application, package information is input into a sorting decision model to obtain the sorting slot information for the packages to be sorted. This includes: the sorting decision model determining the receiving information of the packages to be sorted based on the package information; determining whether there is a current receiving slot for the receiving information based on preset sorting rules and receiving information; the sorting rules instructing the slot flow direction of the packages to be sorted based on the receiving information, and the slot flow direction corresponds to the receiving slot; if there is no current receiving slot, the slot information that meets preset conditions and is in an idle state is used as the slot information.

[0044] Specifically, sorting rules can also be called package creation rules. These rules can be pre-set by the sorting plan administrator and entered into the server. Sorting rules can be a mapping relationship between the package information (e.g., destination information or transportation mode information) of the packages to be sorted and the flow direction of the packages in the sorting slots. This allows the sorting decision model to learn from the sorting rules which packages to be sorted can be bundled into one package for transportation.

[0045] For example, sorting rules could be to package parcels destined for a certain city into one parcel for transport; or, sorting rules could be to package parcels awaiting sorting, transported by air, into one parcel for transport.

[0046] It should be noted that the sorting plan administrator who formulates the sorting rules can also add their desired constraints to the dynamic sorting system of this application embodiment. For example, packages to be sorted with different destinations can be placed in one compartment (i.e., mixed packages). Another example is specifying that compartments numbered 1 to 99 are for sorting packages from outside the province, and compartments numbered 100 to 200 are for sorting packages from within the province, etc.

[0047] Therefore, the embodiments of this application omit the process of manually creating sorting plans and directly input the sorting rules into the server, allowing the sorting decision model running on the server to determine which compartment to bind the package to maximize sorting efficiency, thus improving sorting efficiency.

[0048] Specifically, the package information may include destination information and mode of transport. Destination information may be a detailed address including province, city, district, and street. Recipient information may be a specific address from the destination information; for example, recipient information may be city information (e.g., Beijing) or district information (e.g., Tongzhou District, Beijing). Recipient information may also be the mode of transport, such as air or land transport. It should be noted that the specific content of the recipient information can be flexibly set according to actual circumstances, and this embodiment does not impose specific limitations on it.

[0049] In one embodiment, the sorting decision model can extract the required receiving information from the package information, and then determine whether there is a current receiving slot for the receiving information. If there is no current receiving slot, the sorting decision model can further obtain the information of slots that are in an idle state, and set the slot information that is closest to the package to be sorted and is in an idle state as the slot information to be placed for the package to be sorted, based on the position of the trolley loading the package to be sorted.

[0050] In one embodiment, the information of the compartment on the sorting line that is closest to the package to be sorted and is in an idle state is used as the compartment information to which the package to be sorted should be placed.

[0051] Specifically, the cart loaded with packages to be sorted moves along the sorting line. The sorting decision model first obtains information on multiple idle slots, and then further determines the position of the cart loaded with packages to be sorted on the sorting line. It then selects the slot closest to the package from the multiple slot information and uses the slot closest to the package as the slot information to which the package needs to be placed, thus saving sorting time.

[0052] Therefore, the embodiments of this application automatically generate the sorting slot information of the packages to be sorted based on the receiving information of the packages to be sorted, avoiding the need for manual sorting planning and reducing labor costs. At the same time, the sorting decision model automatically generates the sorting slot information, improving the efficiency and accuracy of sorting decisions.

[0053] In one embodiment of this application, before receiving a sorting request sent by a sorting machine, the method further includes: training an initial sorting decision model based on a digital twin model to simulate an actual sorting scenario and a reinforcement learning method to obtain a sorting decision model, wherein the reinforcement learning method includes learning a sorting strategy through the interaction between the initial sorting decision model and the virtual sorting scenario constructed by the digital twin model.

[0054] In one embodiment, multiple packages to be sorted and operational data of the sorting machine are acquired; based on the operational data, a virtual sorting scenario corresponding to the actual sorting scenario is constructed using a digital twin model; in the virtual sorting scenario, an initial sorting decision model uses different sorting strategies to simulate sorting multiple packages to be sorted, obtaining multiple sorting results, including a first sorting result that meets the reward conditions; with the goal of obtaining multiple first sorting results, the initial sorting decision model is trained until a sorting decision model is obtained.

[0055] It should be noted that obtaining multiple first sorting results can be understood as repeatedly executing the steps of "in a virtual sorting scenario, the initial sorting decision model uses different sorting strategies to simulate sorting multiple packages to be sorted and obtain multiple sorting results".

[0056] Specifically, the multiple packages to be sorted can be historical data or a set of multiple packages to be sorted in the current time period (wherein, if it is a set of multiple packages to be sorted in the current time period, then the multiple packages to be sorted can include the packages to be sorted in the current time period). This application embodiment does not specifically limit this. The operating data of the sorting machine can be the operating data generated by the sorting machine when sorting packages. The digital twin model can simulate a virtual sorting scenario with a high degree of simulation based on the operating data.

[0057] Different sorting strategies can be understood as different sorting methods. For example, a strategy that applies the shortest sorting time, or a strategy that applies the fewest sorting personnel. This application's embodiments specifically define the sorting strategy. The sorting results correspond to the sorting strategy. For example, based on the strategy of shortest sorting time, multiple sorting results are obtained. These multiple results can be numerical values ​​of sorting time, such as sorting results of 12 minutes, 10 minutes, or 8 minutes. In this case, the first sorting result can be the result of completing sorting in 8 minutes.

[0058] For example, based on the strategy of using the fewest sorting personnel during sorting, multiple sorting results can be obtained, where the multiple sorting results can be the number of sorting personnel, such as 15 or 14 people. In this case, the first sorting result can be the result of the strategy of using 14 people to complete the sorting.

[0059] Specifically, the training of the intelligent agent (i.e., the initial sorting decision model) uses a digital twin model to replace the real-world scenario, making data acquisition and status return more convenient and iteration efficiency higher. Digital twin technology makes the reinforcement learning environment more realistic. Furthermore, the digital twin model has a high degree of simulation fidelity, and after the sorting process is verified in a simulation environment, it can be directly applied to the field without causing major accidents.

[0060] Reinforcement learning learns optimal strategies through interactions between an agent and its environment (i.e., a virtual sorting scenario constructed using a digital twin model). It trains an initial sorting decision model using rewards. For example, in reinforcement learning, the initial sorting decision model makes different sorting plans based on the environment and different sorting strategies, with correct actions earning rewards. That is, the initial sorting decision model receives a reward when it achieves a first sorting result that meets preset reward conditions (e.g., the shortest sorting time or the fewest sorting personnel used in the sorting process). This application's embodiments aim to obtain more rewards by achieving optimal interaction between the agent and the environment, thereby training the initial sorting decision model until the desired sorting decision model is obtained.

[0061] The agent algorithm in this application is a reinforcement algorithm based on policy optimization, such as Proximal Policy Optimization (PPO) to optimize the grid allocation policy. Digital twin models are mainly used for data augmentation and policy evaluation. The real-world scenarios simulated by digital twin models can help agents better understand the dynamic characteristics of the environment and assist them in making better decisions.

[0062] It should be noted that by building digital twin models of sorting machines with different parameters and training them in a simulation environment, the sorting decision model can learn its performance under various environmental conditions, thereby improving its generalization ability and enabling more efficient application to the control of small-item sorting machines in transit hubs nationwide. Users can also fine-tune strategies by building their own digital twin models of sorting machines to better suit their specific needs and environments.

[0063] In one embodiment, reinforcement learning is typically described using a Markov Decision Process (MDP). The main elements of an MDP can be state, action, environment, and reward.

[0064] Specifically, the status can include the number of currently idle slots, the number of bound slots (i.e., the number of slots bound to a flow direction), the number of slots currently being packaged, the distance from the current item (i.e., the package to be sorted) to the slot, the flow direction bound to each slot (e.g., city or mode of transport), the current number of items in the slot, the current item distribution on the sorting machine, and the expected distribution of incoming items. Decisions can be made by assigning a slot to the package to be sorted and binding the mapping relationship between the flow direction and the slot. Rewards can include negative feedback calculated based on the number of times the package to be sorted has been scanned, or positive feedback calculated based on factors such as the proportion of trolleys performing transport tasks and the return flow rate. Rewards can be accumulated over a period of time. The number of times the package to be sorted has been scanned can be the number of times the package to be sorted has been scanned by scanners on the sorting line (if the package to be sorted is not placed in a slot, it will continue to move on the sorting line and be scanned multiple times by scanners on the sorting line). The environment is a digital twin model of the sorting machine. The update step size can be adjusted by moving the sorting machine cart one step, accumulating a certain reward, or at intervals (e.g., every 1 or 2 seconds) until all packages to be sorted are sorted.

[0065] It should be noted that the embodiments of this application use reinforcement learning to train the agent to replace the original grid allocation logic of ASCS, making the sorting process of the small item sorting machine more dynamic and able to adapt to more sorting sites.

[0066] Therefore, it can be seen that the embodiments of this application train the model through the intelligent agent model, making the sorting process more intelligent and dynamic, thereby improving sorting efficiency and ensuring that the model can adapt to different waves and different item distributions.

[0067] In one embodiment of this application, the sorting machine includes multiple component modules, wherein the method for training a digital twin model includes: constructing a model for each component module among the multiple component modules, wherein the model for each component module includes a mechanistic model, a data model, and a display model for each component module; and obtaining a digital twin model based on the model for each component module.

[0068] Specifically, a sorting machine can be understood as being constructed from multiple component modules, which may include a feeding station, sorting line, sorting carts, and sorting slots. Building a digital twin model of the sorting machine requires building models for each of these component modules, such as models of the feeding station, sorting line, sorting carts, sorting slots, and packing workers. Furthermore, it requires building a mechanistic model, a data model, and a visual model (also known as a 3D model) for each component module. After building the models for each component module, these models are combined to obtain the digital twin model.

[0069] The mechanism model can refer to the actual operating logic of the component modules. For example, the mechanism model of the feeding station is to determine whether there are packages to be sorted in the cart in front of the feeding station. If there are no packages to be sorted in the cart, the packages to be sorted placed on the feeding station by the worker are transferred to the cart.

[0070] A data model can refer to some attributes of the constituent modules that cannot be directly set, and which need to be statistically obtained through mathematical methods such as machine learning. For example, the no-read rate of a scanner.

[0071] The demonstration model can refer to a model showcasing the appearance and dimensions of the constituent modules, which must be consistent with the dimensions and operational actions of real-world objects. It should be noted that after all the constituent modules of the sorting machine are built into a digital twin model library using digital twin technology, simulation work can be carried out by creating sorting machine models of different transfer stations through the front end.

[0072] In one embodiment, the interface for building the digital twin simulation platform interacts with the user. When building a small-item sorting machine through this interface, the user can select the corresponding simulation model by dragging and dropping to create sorting equipment identical to the on-site equipment, including the shape of the sorting line, the location and number of the feeding stations, and the location and number of the slots. The user can also select historical shift data to perform a more realistic fit of the twin sorting machine, making it closer to the state of the on-site sorting machine.

[0073] It should be noted that the basic parameters set by the digital twin simulation platform are consistent with those of the actual sorting machine on site, which improves the user experience.

[0074] Therefore, the embodiments of this application enable users to flexibly build the required digital twin model according to actual needs, and the digital twin model can be displayed in real time and with exquisite 3D interface, thereby improving the user's immersion and experience.

[0075] Figure 3 This is a flowchart illustrating a sorting method provided in another exemplary embodiment of this application. Figure 3 The method is based on computing devices, such as sorting machines (i.e., Figure 1 The sorting machine (120) is used to perform this task. Figure 3 The example is Figure 1 The similarities in the examples of the embodiments will not be repeated here; the focus here is on describing the differences, such as... Figure 3 As shown, the sorting method includes the following:

[0076] In one embodiment, the sorting machine includes a repacking area, which includes a scanner.

[0077] S310: Scans packages to be sorted using a scanner to obtain package information.

[0078] S320: Sends a sorting request, including package information, to the server.

[0079] In one embodiment, the package information includes destination information.

[0080] S330: Receives sorting instructions sent by the server.

[0081] In one embodiment, the sorting instruction is generated based on a sorting decision model running on a server. The sorting decision model is a model trained based on an intelligent agent algorithm, used to determine the corresponding compartment information of the package to be sorted based on the destination information. The sorting instruction includes the compartment information.

[0082] S340: According to the sorting instructions, the packages to be sorted are placed into the corresponding slot locations.

[0083] Therefore, this embodiment of the application sends a sorting request including package information to the server and then receives sorting instructions sent by the server. The sorting instructions are generated based on the sorting decision model running on the server, and the packages to be sorted are placed into the corresponding slots according to the sorting instructions. This allows this embodiment of the application to automatically generate a sorting plan for the packages to be sorted, thereby improving sorting efficiency.

[0084] In one embodiment of this application, the sorting machine further includes a repackaging area, which includes a scanner. Before scanning the packages to be sorted by the scanner to obtain the package information, the machine further includes: scanning the packages to be sorted by the scanner to obtain the package information; and uploading the package information to a server.

[0085] Specifically, the unpacking area may include a scanner for scanning the automatic identification codes on the packages to be sorted in order to obtain package information.

[0086] In one embodiment, a scanner in the repackaging area scans the packages to be sorted in the repackaging area (e.g., the QR code on the package to be sorted) to obtain the package information of the package to be sorted. The scanned package information is then uploaded to the server so that the server can compare the package information obtained from the repackaging area with the package information obtained from the scanner in the sorting machine (i.e., the package information included in the sorting request), ensuring the accuracy of the package information.

[0087] Therefore, it can be seen that the embodiments of this application, by adding a scanning step for the package transfer area, have improved the accuracy of package information input.

[0088] Figure 4 This is a schematic diagram of the structure of a sorting device 400 provided in an exemplary embodiment of this application. Figure 4As shown, the sorting device 400 includes: a receiving module 410, an input module 420, a generating and sending module 430, and a training module 440.

[0089] The receiving module 410 is used to receive a sorting request sent by the sorting machine, wherein the sorting request includes package information of the package to be sorted, and the package information includes destination information; the input module 420 is used to input the package information into the sorting decision model to obtain the slot information to be placed in for the package to be sorted, wherein the sorting decision model is a model trained based on the intelligent agent algorithm, used to determine the slot information for the package to be sorted based on the destination information; the generating and sending module 430 is used to generate a sorting instruction including the slot information and send the sorting instruction to the sorting machine so that the sorting machine can place the package to be sorted in the slot corresponding to the slot information according to the sorting instruction.

[0090] This application provides a sorting device that receives sorting requests from a sorting machine, inputs the package information included in the sorting request into a sorting decision model to obtain the grid information where the packages to be sorted need to be placed, and then generates sorting instructions including the grid information. The sorting instructions are then sent to the sorting machine so that the sorting machine can place the packages to be sorted into the grids corresponding to the grid information according to the sorting instructions. This application can automatically generate sorting plans for the packages to be sorted, improving sorting efficiency. Compared with the traditional method of specifying sorting plans by a sorting plan administrator, this application can adapt to different batches and different item distributions, reducing labor costs.

[0091] It should be understood that the specific working process and functions of the receiving module 410, input module 420, generation and transmission module 430, and training module 440 in the above embodiments can be referred to the above. Figure 2 The description of the sorting method provided in the embodiments will not be repeated here to avoid repetition.

[0092] Figure 5 This is a schematic diagram of the structure of a sorting device 500 provided in another exemplary embodiment of this application. Figure 5 As shown, the sorting device 500 includes: a scanning module 510, a sending module 520, a receiving instruction module 530, and a delivery module 540.

[0093] The scanning module 510 is used to scan the packages to be sorted using a scanner to obtain the package information of the packages to be sorted; the sending module 520 is used to send a sorting request including package information, such as destination information, to the server; the receiving instruction module 530 is used to receive sorting instructions sent by the server, wherein the sorting instructions are generated based on the sorting decision model running in the server, and the sorting decision model is a model trained based on an intelligent agent algorithm, used to determine the corresponding grid information of the package to be sorted according to the destination information, and the sorting instructions include grid information; the delivery module 540 is used to deliver the package to be sorted to the grid position corresponding to the grid information according to the sorting instructions.

[0094] This application provides a sorting device that sends a sorting request including package information to a server and then receives sorting instructions from the server. The sorting instructions are generated based on a sorting decision model running on the server. The device then places the packages to be sorted into the corresponding slots according to the sorting instructions. This allows the application to automatically generate a sorting plan for the packages to be sorted, thereby improving sorting efficiency.

[0095] According to one embodiment of this application, the sorting machine further includes a repackaging area, which includes a scanner. The scanning module 510 is further used to scan the packages to be sorted by the scanner to obtain package information and upload the package information to the server.

[0096] It should be understood that the specific working process and functions of the scanning module 510, sending module 520, receiving instruction module 530, and delivery module 540 in the above embodiments can be referred to the above description. Figure 3 The description of the sorting method provided in the embodiments will not be repeated here to avoid repetition.

[0097] Figure 6 This is a block diagram of an electronic device 600 for sorting provided in an exemplary embodiment of this application.

[0098] Reference Figure 6 The electronic device 600 includes a processing component 610, which further includes one or more processors, and memory resources represented by memory 620 for storing instructions, such as application programs, that can be executed by the processing component 610. The application programs stored in memory 620 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 610 is configured to execute instructions to perform the sorting method described above.

[0099] Electronic device 600 may also include a power supply component configured to perform power management of electronic device 600, a wired or wireless network interface configured to connect electronic device 600 to a network, and an input / output (I / O) interface. Electronic device 600 can be operated based on an operating system stored in memory 620, such as Windows Server. TM Mac OSX TM Unix TM Linux TM FreeBSD TM Or similar.

[0100] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of the aforementioned electronic device 600, enables the electronic device 600 to perform a sorting method, comprising: receiving a sorting request sent by a sorting machine, wherein the sorting request includes package information of a package to be sorted, the package information including destination information; inputting the package information into a sorting decision model to obtain the slot information to which the package to be sorted should be placed, wherein the sorting decision model is used to determine the slot information for the package to be sorted based on the destination information; generating a sorting instruction including the slot information, and sending the sorting instruction to the sorting machine so that the sorting machine can place the package to be sorted into the slot corresponding to the slot information according to the sorting instruction;

[0101] Alternatively, the system scans the packages to be sorted using a scanner to obtain their information; sends a sorting request to the server, including the destination information; receives sorting instructions from the server, which are generated based on a sorting decision model running on the server. This model determines the corresponding compartment information for the packages to be sorted based on the destination information, and the sorting instructions include the compartment information; and then places the packages to be sorted into the compartments corresponding to the specified compartment information according to the sorting instructions.

[0102] This application also provides a computer program product, including a computer program used to execute the steps of the sorting method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here. This computer program product can be implemented through hardware, software, or a combination thereof. In one optional embodiment, the computer program product is specifically embodied as a computer storage medium; in another optional embodiment, the computer program product is specifically embodied as a software product, such as a software development kit (SDK), etc.

[0103] All of the above-mentioned optional technical solutions can be combined in any way to form optional embodiments of this application, and will not be described in detail here.

[0104] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0109] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program verification codes, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] It should be noted that in the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0111] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A sorting method, characterized in that, Applied to servers, including: Receive a sorting request sent by a sorting machine, wherein the sorting request includes package information of the package to be sorted, and the package information includes destination information; The package information is input into the sorting decision model to obtain the sorting slot information of the package to be sorted. The sorting decision model is a model trained based on an intelligent agent algorithm and is used to determine the sorting slot information for the package to be sorted based on the destination information. A sorting instruction including the grid information is generated and sent to the sorting machine so that the sorting machine can place the package to be sorted into the grid corresponding to the grid information according to the sorting instruction.

2. The sorting method according to claim 1, characterized in that, The step of inputting the package information into the sorting decision model to obtain the sorting slot information for the package to be sorted includes: The sorting decision model determines the receiving information of the package to be sorted based on the package information; it determines whether there is a current receiving slot for the receiving information based on the preset sorting rules and the receiving information; the sorting rules indicate the slot flow direction of the package to be sorted based on the receiving information, and the slot flow direction has a corresponding relationship with the receiving slot; if there is no current receiving slot, the slot information that meets the preset conditions and is in an idle state is used as the slot information.

3. The sorting method according to claim 1, characterized in that, Before receiving the sorting request from the sorting machine, it also includes: Based on a digital twin model simulating a real-world sorting scenario and using reinforcement learning methods, an initial sorting decision model is trained to obtain the sorting decision model. The reinforcement learning method includes learning sorting strategies through the interaction between the initial sorting decision model and the virtual sorting scenario constructed by the digital twin model.

4. The sorting method according to claim 3, characterized in that, The method based on digital twin model to simulate actual sorting scenarios and reinforcement learning trains an initial sorting decision model to obtain the sorting decision model, including: Acquire multiple packages to be sorted and operational data of the sorting machine during operation; Based on the operational data, a virtual sorting scenario corresponding to the actual sorting scenario is constructed using the digital twin model; In the virtual sorting scenario, the initial sorting decision model uses different sorting strategies to simulate sorting the multiple packages to be sorted, and obtains multiple sorting results, wherein the multiple sorting results include the first sorting result that meets the reward conditions; With the goal of obtaining multiple first sorting results, the initial sorting decision model is trained until the sorting decision model is obtained.

5. The sorting method according to claim 3, characterized in that, The sorting machine includes multiple component modules, wherein the method for training the digital twin model includes: Construct a model for each of the plurality of component modules, wherein the model for each component module includes a mechanism model, a data model, and a presentation model for each component module; The digital twin model is obtained based on the model of each component module.

6. A sorting method, characterized in that, Applied to a sorting machine, the sorting machine including a scanner, wherein the method includes: The scanner scans the packages to be sorted to obtain the package information of the packages to be sorted. Send a sorting request to the server, including the package information, wherein the package information includes destination information; The server receives a sorting instruction, wherein the sorting instruction is generated based on a sorting decision model running on the server, the sorting decision model is a model trained based on an intelligent agent algorithm, and is used to determine the compartment information corresponding to the package to be sorted based on the destination information, and the sorting instruction includes the compartment information. According to the sorting instruction, the package to be sorted is placed into the slot location corresponding to the slot information.

7. The sorting method according to claim 6, characterized in that, The sorting machine also includes a repackaging area, which includes a scanner. The process includes, prior to scanning the packages to be sorted using the scanner to obtain their information, the following steps: The package to be sorted is scanned by a scanner to obtain the package information; The package information is uploaded to the server.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program for executing the sorting method according to any one of claims 1 to 7.

9. An electronic device, characterized in that, include: processor; A memory for storing processor-executable instructions, wherein the processor is used to execute the sorting method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the sorting method according to any one of claims 1 to 7.

Images