Blockchain-based agv task allocation and management method

Abstract

The application discloses a kind of AGV task allocation and management method based on block chain, comprising the following steps: (1) hardware layer construction;(2) software layer configuration;(3) warehouse management server node is based on high SKU order demand, generates the AGV task containing task information, and task information is uploaded to block chain network node by encryption, and task state account book is written by block chain network node;(4) AGV node real-time acquisition state data of itself is uploaded to block chain network node;Improvement PoS consensus module calculates the equity weight value of each AGV node and submits the proposal of competition, determines optimal AGV node;(5) smart contract execution module presets smart contract rule, and whether the SKU attribute in task information involves sensitive SKU field is judged: (6) optimal AGV node receives instruction, plans path in combination with real-time warehouse map and executes task;(7) multi-agent collaborative module triggers data filtering, then starts data sharing mechanism, and synchronizes information to different nodes and distributes corresponding permissions.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of warehouse logistics and blockchain technology, and particularly relates to an AGV task allocation and management method based on a blockchain. BACKGROUND

[0002] The traditional AGV task allocation scheme on the market currently adopts a centralized control system, and an upper computer is responsible for allocating tasks, planning paths and mediating conflicts for all AGVs, and makes a lower computer execute tasks according to the path planned by the upper computer through a classical path planning algorithm (A-star algorithm, Dijkstra algorithm, etc.). Under this architecture, the upper computer bears a large amount of calculation and coordination work, and the computing performance requirement thereof is extremely high. Once the upper computer fails, the entire system will face problems such as traffic congestion and system crash. Moreover, although the classical path planning algorithm such as the A-star algorithm and the Dijkstra algorithm has a good effect when processing single-AGV path planning, when multiple AGVs run simultaneously, if only the single-AGV path planning algorithm is used to plan the picking path of each AGV without considering the mutual influence between multiple AGVs, the phenomenon of conflict and deadlock between several AGVs may occur.

[0003] In a high-SKU warehouse scenario, the number of tasks and AGVs is large, and the disadvantages of the centralized architecture are magnified, resulting in low scheduling efficiency and slow dynamic response. If the shelf position changes frequently, the path planning algorithm that does not consider the interaction between AGVs cannot meet the dynamic response demand. Moreover, the traditional scheme lacks trust between participating subjects, making it difficult to achieve efficient multi-subject collaboration in a high-SKU warehouse involving multiple parties, and there may be problems such as information opacity and low trust between subjects, affecting the overall warehouse operation efficiency.

[0004] Therefore, the traditional AGV task allocation scheme has problems such as dependence on a centralized architecture, low efficiency of traditional algorithms and insufficient application of blockchains, resulting in low scheduling efficiency, slow dynamic response, poor data consistency, low resource utilization and difficulty in multi-subject collaboration in a high-SKU warehouse, which cannot meet the intelligent warehouse demand for a large number of SKUs. SUMMARY

[0005] The technical problem to be solved by the present application is to provide an AGV task allocation and management method based on a blockchain, which can improve scheduling efficiency, dynamic response, data consistency and resource utilization, collaborate with multiple subjects and meet the intelligent warehouse demand for a large number of SKUs in a high-SKU warehouse.

[0006] To solve the above technical problems, the technical scheme adopted by the present application is as follows:

[0007] A blockchain-based AGV task allocation and management method, characterized by comprising the following steps:

[0008] (1) Hardware layer construction: the hardware layer includes a blockchain network node, an AGV node, a warehouse management server node, a cargo owner authority node, and a 5G communication module;

[0009] (2) Software layer configuration: the software layer includes an improved PoS consensus module, an AGV state real-time acquisition and transmission module, an intelligent contract execution module, a multi-agent collaboration module, an SKU dimension authority control module, and a multi-agent data interaction and authority management module;

[0010] The AGV state real-time acquisition and transmission module and the improved PoS consensus module are arranged in the AGV node; the intelligent contract execution module, the multi-agent collaboration module, the SKU dimension authority control module, and the multi-agent data interaction and authority management module are arranged in the blockchain network node;

[0011] (3) The warehouse management server node generates an AGV task containing task information based on high SKU order demand, and uploads the task information to the blockchain network node after encryption, and the blockchain network node writes the task state ledger;

[0012] (4) The AGV node acquires its own state data in real time through the AGV state real-time acquisition and transmission module, and uploads the state data of the AGV node to the blockchain network node using the 5G communication module;

[0013] The improved PoS consensus module calculates the equity weight value of each AGV node according to the task information and the state data and submits a competition proposal, and determines the optimal AGV node through a weighted voting mechanism;

[0014] (5) The intelligent contract execution module presets intelligent contract rules, and judges whether the SKU attribute in the task information involves a sensitive SKU field:

[0015] (5-1) If the SKU attribute involves a sensitive SKU field, the intelligent contract execution module issues a task instruction to the optimal AGV node and proceeds to step (6), and sends an authority control request to the multi-agent collaboration module and proceeds to step (7);

[0016] (5-2) If the SKU attribute does not involve a sensitive SKU field, the intelligent contract execution module issues a task instruction to the optimal AGV node and proceeds to step (6);

[0017] (6) The intelligent contract execution module automatically allocates tasks and updates the task state ledger; the optimal AGV node receives instructions, plans a path in combination with the real-time warehouse map and executes the task, and reports the progress to the blockchain network node; after the task is completed, the intelligent contract is verified and stored on the chain, and step (5) is returned;

[0018] (7) The multi-agent collaboration module triggers the SKU dimension permission control module to filter data, and then starts the data sharing mechanism through the multi-agent data interaction and permission management module, synchronizes information to the warehouse management server node and the consignee permission node, and allocates corresponding permissions.

[0019] In the above step (1), the blockchain network node is the basis for data interaction and storage; the AGV node is used to collect its own state data and participate in task competition; the warehouse management server node is used to generate task information and store it on the chain; and the consignee permission node is used to realize safe access to data.

[0020] In the above step (2), the improved PoS consensus module calculates the equity weight of each AGV node based on the state data and task information uploaded by the AGV node and submits a competition proposal; the intelligent contract execution module predefines intelligent contract rules to automatically execute task allocation operations, permission control operations and data storage operations; the AGV state real-time acquisition and transmission module continuously provides dynamic data support for the improved PoS consensus module and task execution; the multi-agent collaboration module receives the permission control request of the intelligent contract execution module, and realizes field-level data filtering through the SKU dimension permission control module; the SKU dimension permission control module dynamically authorizes according to the pre-set intelligent contract rules, and sends safe data to the multi-agent data interaction and permission management module; the multi-agent data interaction and permission management module ensures the credible sharing of data between nodes, and according to the pre-set intelligent contract rules, allocates corresponding permissions for different subjects, realizes the collaborative operation of each part, and completes the efficient management of AGV tasks in high-SKU scenarios.

[0021] In the above steps (1)-(4), the blockchain distributed architecture is applied to AGV task allocation in high-SKU warehouses, and the improved PoS consensus mechanism combining AGV state and SKU attributes is used to realize decentralized task allocation. The system has strong fault tolerance and can fairly and efficiently select the optimal AGV node among multiple AGVs, opening up a new way for AGV scheduling in high-SKU warehouses.

[0022] The task information in the step (3) is chained in real time, and the task progress in the step (6) is chained at a high frequency, which constitutes an innovative double-layer data real-time chaining method. The data is encrypted and protected during transmission and storage, ensuring the real-time and integrity of the data, the timeliness of task publishing, and the real-time monitoring of the AGV execution process. This design ensures a trusted closed loop of dynamic data and task execution, significantly improving the dynamic task processing efficiency and response speed.

[0023] In the step (5), when sensitive SKU fields are involved, the task instruction is issued to the optimal AGV node and the permission control request is sent to the multi-agent collaboration module in a double-path parallel execution manner. The sensitive SKU fields are generally "temperature control required", "medicine information", and "sensitive data".

[0024] In the step (6), the smart contract execution module automatically allocates tasks and generates task instructions to the optimal AGV node, and synchronously updates the task state ledger of the blockchain network node. In the step (6), a dynamic task optimization algorithm and path planning strategy suitable for high SKU characteristics are proposed, realizing adaptive management of warehouse layout adjustment and SKU attribute changes, and highly adapting to dynamic changes in high SKU scenarios with high resource utilization.

[0025] In the step (7), a fine-grained permission control system based on smart contract is constructed in the SKU dimension, and different data access permissions are allocated to different subjects through smart contract rules, effectively solving the privacy protection problem in multi-agent collaboration, and ensuring the trusted sharing and privacy security of multi-agent data. Generally, the authorized specific fields can be "open medicine expiration date to consignee" or "hide path details from warehouse management server node".

[0026] In the preferred scheme, in the step (3), the task information includes the number of tasks and SKU attributes; and the SKU attributes include priority. In the step (4), the AGV node collects real-time state data including current position, target position, power, and load.

[0027] In the further preferred scheme, in the step (4), the equity weight value is calculated by a preset equity weight calculation formula, and the weighted voting mechanism allocates voting weights according to the equity weight values of the AGV nodes. The AGV node with the highest votes is determined as the optimal AGV node. The improved PoS consensus module calculates the equity weight values of the AGV nodes according to the task information and state data, and calculates the equity weight values as proposals for competing for the task, and submits the competing proposals to the smart contract execution module.

[0028] In the still further preferred scheme, the equity weight calculation formula is:

[0029] The equity weight W = a x E + b x (1 / D) + g x (1 / L) + d x P, wherein:

[0030] E: percentage of power (0 < E < 1);

[0031] D: distance of AGV from target position;

[0032] L: current number of tasks + 1;

[0033] P: SKU priority coefficient;

[0034] a, b, g, d: adjustable weight parameters.

[0035] Generally, a is the power weight parameter, b is the distance weight parameter, g is the task number weight parameter, and d is the SKU priority weight parameter.

[0036] In a preferred embodiment, in step (6), the optimal AGV node uses a genetic algorithm to plan an optimal path, and uses a high-frequency progress reporting mechanism to upload task progress to a blockchain network node. The optimal AGV node receives task instructions, combines the real-time warehouse map stored by the blockchain network node, uses a genetic algorithm to plan an optimal path and execute the task, and uses a high-frequency progress reporting mechanism to upload task progress to a blockchain network node.

[0037] In a further preferred embodiment, in step (6), the optimal AGV node uses real-time state data to drive an intelligent contract execution module to execute the task; after the task is completed, the AGV node submits the task result, which is verified by the intelligent contract execution module and then uploaded to the chain for storage and evidence, and jumps back to step (5).

[0038] In a further preferred embodiment, in step (6), the high-frequency progress reporting mechanism uploads task progress to a blockchain network node every 400-600 ms during the period.

[0039] In a further preferred embodiment, in step (6), the high-frequency progress reporting mechanism uploads task progress to a blockchain network node every 500 ms during the period.

[0040] In a preferred embodiment, in step (7), the multi-agent collaboration module receives a permission control request from the intelligent contract execution module, triggers a SKU dimension permission control module to filter data; after data filtering, triggers a multi-agent data interaction and permission management module to start a multi-agent data sharing mechanism, synchronizes task trajectory and state to a warehouse management server node and a consignee permission node, allocates corresponding permissions to the warehouse management server node and the consignee permission node, and according to the intelligent contract preset rules, only opens specific fields to the consignee permission node.

[0041] Compared with the prior art, the present application has the following advantages:

[0042] The present application constructs a decentralized, adaptive and trusted management system of AGV task allocation in high SKU scenarios through the deep integration of blockchain distributed architecture, improved PoS consensus mechanism, real-time data on-chain technology, intelligent contract collaboration and dynamic task optimization algorithm, realizes distributed efficient decision-making in high SKU scenarios, guarantees real-time data processing of task allocation, balances multi-agent data privacy and collaborative sharing, improves the adaptability of the system to dynamic changes of SKU and enhances the fault tolerance capability in abnormal scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 is the connection relationship diagram between each node and each module in embodiment 1 of the present application;

[0044] Figure 2 is a step schematic diagram of the AGV task allocation and management method in embodiment 1 of the present application;

[0045] Figure 3 is a processing flowchart of the intelligent contract execution module in embodiment 1 of the present application;

[0046] Figure 4 is a flowchart of the double-layer data real-time on-chain method in embodiment 1 of the present application;

[0047] Figure 5 is a flowchart of calculating the weight of each AGV node in embodiment 2 of the present application. DETAILED DESCRIPTION

[0048] The preferred embodiments of the present application will be further described below in combination with the drawings and the preferred embodiments of the present application.

[0049] Embodiment 1, such as Figures 1-4 The AGV task allocation and management method based on blockchain of the present embodiment includes the following steps:

[0050] (1) Hardware layer construction: the hardware layer includes a blockchain network node, an AGV node, a warehouse management server node, a cargo owner permission node and a 5G communication module;

[0051] (2) Software layer configuration: the software layer includes an improved PoS consensus module, an AGV state real-time acquisition and transmission module, an intelligent contract execution module, a multi-agent collaboration module, a SKU dimension permission control module and a multi-agent data interaction and permission management module;

[0052] The AGV state real-time acquisition and transmission module and the improved PoS consensus module are arranged in the AGV node; the smart contract execution module, the multi-agent collaboration module, the SKU dimension permission control module and the multi-agent data interaction and permission management module are arranged in the blockchain network node.

[0053] (3) The warehouse management server node generates an AGV task containing task information based on high SKU order demand, and uploads the task information to the blockchain network node after encryption, and the blockchain network node writes the task state ledger;

[0054] (4) The AGV node acquires its own state data in real time through the AGV state real-time acquisition and transmission module, and uploads the state data of the AGV node to the blockchain network node through the 5G communication module;

[0055] The improved PoS consensus module calculates the equity weight value of each AGV node according to the task information and the state data and submits a competition proposal, and determines the optimal AGV node through a weighted voting mechanism;

[0056] (5) The smart contract execution module presets the smart contract rules, and judges whether the SKU attribute in the task information involves a sensitive SKU field:

[0057] (5-1) If the SKU attribute involves a sensitive SKU field, the smart contract execution module issues a task instruction to the optimal AGV node and proceeds to step (6), and sends a permission control request to the multi-agent collaboration module and proceeds to step (7);

[0058] (5-2) If the SKU attribute does not involve a sensitive SKU field, the smart contract execution module issues a task instruction to the optimal AGV node and proceeds to step (6);

[0059] (6) The smart contract execution module automatically allocates tasks and updates the task state ledger; the optimal AGV node receives the instruction, plans the path in combination with the real-time warehouse map and executes the task, and reports the progress to the blockchain network node; after the task is completed, it is stored in the chain after being verified by the smart contract, and returns to step (5);

[0060] (7) The multi-agent collaboration module triggers the SKU dimension permission control module to filter data, and then starts the data sharing mechanism through the multi-agent data interaction and permission management module, synchronizes information to the warehouse management server node and the consignee permission node and allocates corresponding permissions.

[0061] In step (3), the task information includes the number of tasks and the SKU attribute; wherein the SKU attribute includes the priority; in step (4), the AGV node acquires its own state data including the current position, the target position, the power and the load.

[0062] In step (4), the equity weight value is calculated by a preset equity weight calculation formula, and the weighted voting mechanism allocates voting weights according to the equity weight values of the AGV nodes, and the AGV node with the highest votes is determined as the optimal AGV node. The improved PoS consensus module calculates the equity weight values of the AGV nodes according to the task information and state data, and according to the preset equity weight calculation formula, and takes the calculated equity weight values as proposals for competing for the task, and submits the competing proposals to the smart contract execution module.

[0063] In step (6), the optimal AGV node plans the optimal path using a genetic algorithm, and uploads the task progress to the blockchain network node using a high-frequency progress reporting mechanism. The optimal AGV node receives the task instruction, combines the real-time warehouse map stored in the blockchain network node, plans the optimal path using a genetic algorithm, and executes the task, and uploads the task progress to the blockchain network node using a high-frequency progress reporting mechanism.

[0064] In step (6), the optimal AGV node drives the smart contract execution module to execute the task in real time; after the task is completed, the AGV node submits the task result, which is verified by the smart contract execution module and then uploaded to the blockchain for storage, and then jumps back to step (5).

[0065] In step (6), the high-frequency progress reporting mechanism uploads the task progress to the blockchain network node every 500ms.

[0066] The embodiment of the present application Figure 3 The core logic of how the smart contract execution module handles task allocation, state verification, and data access permission control based on preset smart contract rules and specific SKUID is illustrated in the

[0067] The embodiment of the present application Figure 4 The double-layer real-time data uploading process of task information generation and uploading, AGV state periodic reporting, and high-frequency (500ms) progress uploading during task execution is intuitively presented in the

[0068] The following will specifically take the medical warehouse scene as an example (high SKU and high timeliness requirement) to execute this AGV task allocation and management method based on blockchain, including the following steps:

[0069] (1) Hardware layer construction and (2) software layer configuration

[0070] A. Blockchain network node building:

[0071] a. Hyperledger Fabric 2.3 is used to build the blockchain network node, which includes three types of nodes:

[0072] - AGV node (with 5G communication module);

[0073] - Warehouse management server node (running task generation and map management service);

[0074] - Consignee authority node (private node of pharmaceutical enterprises, used for SKU data authorization);

[0075] b. Network channel setting: create task-channel (task allocation) and data-channel (real-time data sharing) to achieve data isolation;

[0076] B. Deploy the smart contract using the following core contract code (chaincode):

[0077] Func AssignTask(ctx ContractContext,taskID string) error {

[0078] / / Verify task validity (SKU attributes)

[0079] task:=GetTaskFromLedger(ctx,taskID)

[0080] / / Call the improved PoS consensus module to calculate the equity weight of AGV nodes

[0081] validAGVs:=CalculateAGVWeight(task.SKU.Priority,task.TargetLocation)

[0082] / / Select the optimal AGV node through the weighted voting mechanism (the highest weight is the winner)

[0083] selectedAGV:=SelectAGVByVote(validAGVs)

[0084] / / The smart contract execution module generates task instructions and updates the task status ledger

[0085] UpdateTaskLedger(ctx,taskID,"ASSIGNED",selectedAGV)

[0086] returnnil

[0087] }

[0088] Step (3) Task generation and on-chain (warehouse management server node execution)

[0089] - When a pharmaceutical order arrives, the warehouse management server node parses SKU attributes (such as drug expiration date and temperature control requirements) and generates a task structure:

[0090] {

[0091] "TaskID":"T20250620001",

[0092] "SKU_Properties":{"ID":"MED_001","Priority":"High","ExpiryDate":"2025-12-31"},

[0093] "TargetLocation":"Aisle3-Shelf5",

[0094] "Deadline":"2025-06-20T14:30:00Z"

[0095] }

[0096] The task data is encrypted using AES-256 and written to the task-channel ledger via the gRPC interface by calling the blockchain SDK.

[0097] Step (4) AGV Status Reporting and Competition Proposal (AGV Node Execution)

[0098] - The AGV periodically collects and encrypts status data (every 1 second) and uploads it to the data-channel:

[0099] {

[0100] "AGV_ID":"AGV007",

[0101] "Battery":85, / / Unit: %

[0102] "Location":[23.5,45.2], / / Two-dimensional coordinates

[0103] "CurrentLoad":3 / / Current number of tasks

[0104] }

[0105] The equity weight of each AGV node is calculated by improving the PoS consensus module's equity weight calculation formula W=α×E+β×(1 / D)+γ×(1 / L)+δ×P.

[0106] Consensus (automatically executed by blockchain network nodes)

[0107] - PoS consensus module improves the collection of AGV competition proposals in task-channel, calculates the weight according to the above formula and sorts them;

[0108] Step (5) Task allocation (automatically executed by blockchain network nodes)

[0109] - The smart contract AssignTask is automatically triggered, and the instruction is issued to the optimal AGV node (such as AGV007), and the task status ledger is updated to:

[0110] TaskID: T20250620001 → Status: ASSIGNED → Executor: AGV007

[0111] Step (6) Task execution and progress high-frequency on-chain (AGV execution)

[0112] - AGV007 receives the instruction and calls the genetic algorithm path planning module (with dynamic obstacle avoidance):

[0113] - Input: Real-time warehouse map (get the latest version from the blockchain network node), task target position (23.7, 45.3)

[0114] - Output: Optimal path (such as: Path: [Aisle1→Aisle3→Shelf5])

[0115] - 500ms progress on-chain (encrypted and written to data-channel):

[0116] {

[0117] "TaskID":"T20250620001",

[0118] "Timestamp":"2025-06-20T14:00:00.500Z",

[0119] "Progress":{"Location":[23.7,45.3],"Completion":15%}

[0120] }

[0121] Step (7) Task verification and multi-agent data sharing (automatically executed by smart contract)

[0122] - AGV007 uploads the final result (including the SKU picking photo hash value) after completing the task;

[0123] - After the smart contract verification is passed:

[0124] - The task status is marked as COMPLETED in the ledger;

[0125] - Trigger SKU dimension data sharing rules:

[0126] - Open to consignee authority node: sensitive data such as drug expiration date, pickup time, etc.

[0127] - Open to warehouse management server node: operation data such as path trajectory, execution duration, etc.

[0128] - SKU dimension permission control logic:

[0129] - If SKU_Properties contains the "TempControl" field, only the consignee authority node can access temperature data.

[0130] - Path trajectory data is encrypted for the consignee authority node.

[0131] From the above data, AGV_ID=AGV007, current position=(23.5, 45.2), target position=(23.7, 45.3), battery level=85%, load=3 task number.

[0132] The weight of each AGV node is calculated by improving the PoS consensus module according to the weight calculation formula W=α×E+β×(1 / D)+γ×(1 / L)+δ×P, where the SKU priority coefficient is assumed to be high=1.5, medium=1.0, and low=0.8; α=0.3, β=0.6, γ=0.1, δ=0.1, and the node with the highest weight obtains the task.

[0133] D= m, distance coefficient 1 / D=4.46;

[0134] Battery weight: E=0.85;

[0135] Load factor: 1 / L=1 / (1+3)=0.25;

[0136] SKU attribute: priority=high→priority coefficient: P=1.5.

[0137] AGV007 weight: W=0.3*0.85+0.6*4.46+0.1*0.25+0.1*1.5=3.106.

[0138] As shown in Example 2, Figure 5 the following explains how to fuse AGV state (position, battery level, load) and task-related SKU attributes (such as priority, distance) to calculate the weight of interest, and select the optimal AGV execution node through weighted voting.

[0139] Proposal 1: Known AGV_ID = 001, current position = (10, 20), target position = (50, 60), power = 80%, load = 2 tasks;

[0140] D= m, distance coefficient 1 / D = 0.0177;

[0141] Power weight: E = 0.8;

[0142] Load factor: 1 / L = 1 / (1+2) = 0.33;

[0143] SKU attribute: priority = high → priority coefficient: P = 1.5.

[0144] Proposal 2: Known AGV_ID = 002, current position = (45, 55), target position = (50, 60), power = 60%, load = 1 task;

[0145] D= m, distance coefficient 1 / D = 0.1414;

[0146] Power weight: E = 0.6;

[0147] Load factor: 1 / L = 1 / (1+1) = 0.5;

[0148] SKU attribute: priority = medium → priority coefficient: P = 1.0.

[0149] Calculate the weight of each AGV node by improving the PoS consensus module according to the weight calculation formula W = α × E + β × (1 / D) + γ × (1 / L) + δ × P, where SKU priority coefficient: high = 1.5, medium = 1.0, low = 0.8; α = 0.4, β = 0.3, γ = 0.2, δ = 0.1, the highest weight obtains the task.

[0150] AGV001 weight: W1 = 0.4*0.8 + 0.3*0.12 + 0.2*0.33 + 0.1*1.5 = 0.542;

[0151] AGV002 weight: W2 = 0.4*0.6 + 0.3*0.14 + 0.2*0.5 + 0.1*1.0 = 0.482;

[0152] Optimal AGV = MAX(W1, W2,...) → AGV001.

[0153] Output: selected AGV node ID and task execution instruction.

[0154] {

[0155] "winner_AGV":"001",

[0156] "start_position":"(10,20)",

[0157] "target_position":"(50,60)",

[0158] "SKU_priority":"high"

[0159] }

[0160] In addition, it needs to be explained that the specific embodiments described in the specification, the name of each part, etc. can be different, and any equivalent or simple change made according to the structure, features and principles described in the patent concept of the present application is included in the protection scope of the present application. Those skilled in the art of the present application can make various modifications or supplements to the described specific embodiments or use similar ways to replace them, as long as they do not deviate from the structure of the present application or exceed the scope defined by the present claims, which shall belong to the protection scope of the present application.

Claims

1. A blockchain-based AGV task allocation and management method, characterized in that... Includes the following steps: (1) Hardware layer construction: The hardware layer includes blockchain network nodes, AGV nodes, warehouse management server nodes, cargo owner permission nodes and 5G communication modules; (2) Software layer configuration: The software layer includes an improved PoS consensus module, an AGV real-time status acquisition and transmission module, a smart contract execution module, a multi-entity collaboration module, a SKU-dimensional permission control module, and a multi-entity data interaction and permission management module; The AGV status real-time acquisition and transmission module and the improved PoS consensus module are both set in the AGV node; The smart contract execution module, the multi-entity collaboration module, the SKU-dimensional permission control module, and the multi-entity data interaction and permission management module are all located in the blockchain network nodes; (3) Based on the demand for high SKU orders, the warehouse management server node generates AGV tasks containing task information and uploads the task information to the blockchain network node in an encrypted manner, which then writes the task status ledger into the blockchain network node. (4) The AGV node collects its own status data in real time through the AGV status real-time acquisition and transmission module, and uploads the status data of the AGV node to the blockchain network node using the 5G communication module; The improved PoS consensus module calculates the equity weight value of each AGV node based on task information and state data, submits competitive proposals, and determines the optimal AGV node through a weighted voting mechanism. (5) The smart contract execution module presets smart contract rules and determines whether the SKU attribute in the task information involves sensitive SKU fields: (5-1) If the SKU attribute involves a sensitive SKU field, the smart contract execution module sends a task instruction to the optimal AGV node and performs step (6), and at the same time sends an access control request to the multi-subject collaboration module and performs step (7). (5-2) If the SKU attribute does not involve sensitive SKU fields, the smart contract execution module issues task instructions to the optimal AGV node and performs step (6). (6) The smart contract execution module automatically assigns tasks and updates the task status ledger; the optimal AGV node receives instructions, plans the path and executes the task in conjunction with the real-time warehouse map, and reports the progress to the blockchain network node; after the task is completed, it is verified by the smart contract and stored on the blockchain, and then returns to step (5). (7) The multi-entity collaboration module triggers the SKU dimension permission control module to filter data, and then starts the data sharing mechanism through the multi-entity data interaction and permission management module to synchronize information and assign corresponding permissions to the warehouse management server node and the cargo owner permission node.

2. The blockchain-based AGV task allocation and management method as described in claim 1, characterized in that: In step (3), the task information includes the number of tasks and SKU attributes; where the SKU attributes include priority; in step (4), the AGV node collects its own status data in real time, including current position, target position, power, and load.

3. The blockchain-based AGV task allocation and management method as described in claim 2, characterized in that: In step (4), the equity weight value is calculated using a preset equity weight calculation formula. The weighted voting mechanism allocates voting weights based on the equity weight values ​​of each AGV node, and the AGV node with the highest number of votes is determined as the optimal AGV node.

4. The blockchain-based AGV task allocation and management method as described in claim 3, characterized in that: The formula for calculating the equity weight is as follows: Equity weight W = α × E + β × (1 / D) + γ × (1 / L) + δ × P, where: E: Battery percentage (0≤E≤1); D: Distance of the AGV from the target location; L: Current task count +1; P: SKU priority coefficient; α, β, γ, δ: Adjustable weight parameters.

5. The blockchain-based AGV task allocation and management method as described in claim 1, characterized in that: In step (6), the optimal AGV node uses a genetic algorithm to plan the optimal path and uses a high-frequency progress reporting mechanism to upload the task progress to the blockchain network node.

6. The blockchain-based AGV task allocation and management method as described in claim 5, characterized in that: In step (6), the real-time status data of the optimal AGV node drives the smart contract execution module to execute the task; after the task is completed, the AGV node submits the task result, and after verification by the smart contract execution module, the task completion information is stored on the blockchain and the execution step (5) is returned.

7. The blockchain-based AGV task allocation and management method as described in claim 5, characterized in that: In step (6), the high-frequency progress reporting mechanism uploads the task progress to the blockchain network node every 400ms-600ms during the period.

8. The blockchain-based AGV task allocation and management method as described in claim 7, characterized in that: In step (6), the high-frequency progress reporting mechanism uploads the task progress to the blockchain network node every 500ms.

9. The blockchain-based AGV task allocation and management method as described in claim 1, characterized in that: In step (7), the multi-entity collaboration module receives the permission control request from the smart contract execution module and triggers the SKU dimension permission control module to perform data filtering. After data filtering, the multi-entity data interaction and permission management module is triggered to start the multi-entity data sharing mechanism, synchronize the task trajectory and status to the warehouse management server node and the cargo owner permission node, assign corresponding permissions to the warehouse management server node and the cargo owner permission node, and, according to the smart contract preset rules, open the cargo owner permission node only to authorized specific fields.

Images