Multi-load automatic tractor online scheduling method and system

Abstract

The invention relates to the technical field of multi-agent systems and intelligent logistics, and particularly provides a multi-load automatic tractor online scheduling method and system, and the method comprises the steps: constructing an e-commerce intelligent warehouse scene model; establishing a multi-load AGV online order scheduling problem model; defining an agent type, and constructing a multi-agent system by combining all agents based on an improved contract network protocol mechanism; establishing an order packaging mechanism, establishing a distribution transportation mechanism, and realizing order task distribution based on the order packaging mechanism and the distribution transportation mechanism; and obtaining a task execution sequence of each AGV intelligent agent, configuring a working priority for each AGV intelligent agent, and performing conflict-free scheduling in combination with the working priority and the task execution sequence of the AGV intelligent agents. When a plurality of AGVs execute the carrying task at the same time, conflicts are not easy to generate, and the scheduling efficiency of the e-commerce intelligent warehouse is improved.

Description

Technical Field

[0001] The present invention belongs to the technical fields of multi-agent systems and intelligent logistics, and particularly relates to an online scheduling method and system for multi-load automatic tractors. Background Art

[0002] In recent years, with the rapid development of e-commerce logistics, the demand for warehousing logistics has also been increasing continuously. Traditional manual sorting can no longer meet the requirements of high efficiency and high precision. Currently, with the continuous expansion of the application of automatic guided vehicles (AGVs), they have been widely used as intelligent logistics equipment in e-commerce intelligent warehouses, providing a new solution for warehousing logistics.

[0003] In the online order task scheduling scenario of an e-commerce intelligent warehouse, during the operation of the intelligent warehouse, users place orders through an online shopping platform. After the intelligent warehouse receives the orders, it mobilizes AGVs to perform handling tasks according to the orders.

[0004] With the gradual increase in the scale and order volume of e-commerce intelligent warehouses, the number of AGVs required gradually increases. When multiple AGVs perform handling tasks simultaneously, conflicts are likely to occur, resulting in low scheduling efficiency of e-commerce intelligent warehouses. Summary of the Invention

[0005] Aiming at the above deficiencies of the prior art, the present invention provides an online scheduling method and system for multi-load automatic tractors to solve the above technical problems.

[0006] In a first aspect, the present invention provides an online scheduling method for multi-load automatic tractors, including: S1, obtaining the layout rules of an e-commerce intelligent warehouse, and constructing an e-commerce intelligent warehouse scenario model based on the combination of the grid method and the topology method with the layout rules; S2, obtaining the online orders of the e-commerce intelligent warehouse, establishing a multi-load AGV online order scheduling problem model based on the e-commerce intelligent warehouse model and the online orders, and obtaining key parameters and an objective function based on the problem model; S3, defining the agent types as order management agents, packing station agents, AGV agents, and information management agents based on the multi-load AGV online order scheduling model, constructing a multi-agent system based on the improved contract net protocol mechanism and all agents, taking the party providing the task assignment among the agents as the tenderer, and the party receiving and executing the task as the bidder; S4, establishing an order packing mechanism with the order management agent as the tenderer and the packing station agent as the bidder, establishing an allocation and transportation mechanism with the packing station agent as the tenderer and the AGV agent as the bidder, and realizing order task assignment based on the order packing mechanism and the allocation and transportation mechanism; S5. Obtain the task execution order of each AGV agent based on key parameters and the objective function, configure the work priority for each AGV agent based on the power of the AGV and the number of tasks it is loaded with, and perform conflict-free scheduling by combining the work priorities of the AGV agents and the task execution order.

[0007] In an alternative embodiment, in step S1, it specifically includes: The layout rules include the shelf layout rules, the AGV docking area layout rules, and the packing table layout rules; Represent the roads, shelves, packing tables, and AGV docking areas in the warehouse environment based on the grid method; Based on the topology method, taking the crossroads of the warehouse as the reference, the center of the crossroads grid as the node, and the dividing line of the two-way road as the edge connecting the nodes, divide the warehouse into multiple areas.

[0008] In an alternative embodiment, step S2 specifically includes: Configure a loading point, an unloading point, and a load for each online order task; Combine all the tasks currently undertaken by each AGV into a task group; Calculate the time cost for each AGV to complete the task group, and the specific calculation is as follows:

[0009] Wherein, Represents the time cost required to complete the task group g represents the gth task group, Represents the number of tasks included in the gth task group, m represents the number of the AGV, M represents the total number of transportable AGVs, n represents the task number in the task group, Represents the time for each operation at the loading point or unloading point, Represents the waiting time generated by the AGV to avoid conflicts during task execution, , And Are respectively the loading points of the first, the nth, and the (n + 1)th tasks in the task group, , And Are respectively the unloading points of the first, the nth, and the (n + 1)th tasks in the task group, Represents the time cost between node m and node , Represents node And The time cost between, Represents node And The time cost between. Represents the time cost between nodes and ; Obtain the total path cost between the loading points and the total path cost between the unloading points of each AGV to complete the task group, and then synthesize to obtain the total path cost; Based on the time cost and the total path cost of each AGV to complete the task group, calculate the scheduling optimization objective function of each AGV. The specific calculation is as follows:

[0010] wherein, represents the operation of finding the minimum value.

[0011] In an alternative embodiment, in step S4, establishing an order packing mechanism with the order management agent as the bidder and the packing table agent as the bidder specifically includes: When there is a new order, take the order management agent as the bidder and formulate a tender document, which includes the location information of the order goods; Take the packing table agent as the bidder. After receiving the tender document information of the order management agent, first determine whether the number of current unfinished packing order tasks exceeds the threshold for accepting new orders; If the conditions for accepting the order are met, initiate a bid to the bidder order management agent, and provide the information of its current remaining unfinished packing tasks and the path length index from the location of the goods, so that the bidder order management agent calculates the bid value and evaluates the bid. The calculation formula for the bid value of the bid is as follows:

[0012] wherein, is the bid value of the tender document of the bidder packing table agent, is the number of orders currently accepted by the bidder packing table agent, is the maximum number of orders that can be accepted, is the Manhattan distance of the bidder from the location of the order goods, is the maximum distance of the packing table agent from the goods, and are used to adjust the weights of the number of orders undertaken and the distance of the goods; The order management agent evaluates all bidders, selects the bidder with the largest bid value to issue a winning bid notice, and waits for the winner to confirm. If the winner confirms, sign the contract to complete the task assignment.

[0013] In an alternative embodiment, in step S4, establishing a distribution and transportation mechanism with the packing table agent as the bidder and the AGV agent as the bidder specifically includes: After the packing station agent accepts the order task assigned by the order management agent, the packing station agent will specify the tender document as the tenderer. The tender document information is the location of the order goods and the location information of the packing station. As the bidder, the AGV agent first determines whether its battery power is sufficient to accept the new task and whether the current capacity can accommodate the new order goods after receiving the tender document information from the packing station agent. If the conditions are met, it will submit a bid to the tenderer, the packing station agent, calculate the difference in the total path planning length before and after accepting the task, and use the current accepted order task quantity as the tender document information to send to the packing station agent for calculating the bid value. The bid value calculation formula for the bidder is as follows:

[0014] Where is the bid value of the bidder AGV agent's tender document, is the number of orders currently accepted by the bidder AGV agent, is the maximum number of orders that can be accepted, is the total path planning length of the bidder AGV agent before accepting the task, is the total path planning length of the bidder AGV agent after accepting the task, is the maximum difference in the total path planning length of the AGV agent before and after accepting the task from this cargo, and are used to adjust the weight of the number of orders already received by the AGV agent and the maximum difference in the total path planning length of the AGV agent before and after accepting the task from this cargo. The packing station agent conducts a winning bid evaluation on all bidders, selects the bidder with the largest bid value to issue a winning bid notice, and waits for the winner to confirm; if the winner confirms, a contract is signed to complete the task assignment from the packing station agent to the AGV agent.

[0015] In an alternative embodiment, in step S5, obtaining the task execution order of each AGV agent based on the key parameters and the objective function specifically includes, when the number of tasks currently accepted by the AGV is no more than 6, using the two-stage branch and bound algorithm based on the priority queue to solve: S5-1, construct the adjacency matrix between the current position of the AGV and all task loading points, randomly generate an execution order of a loading segment as the bound, create a priority queue and put the current position of the AGV and the current path into the priority queue; S5-2, take the head node of the queue for expansion and calculate the current path length. If the path value of a certain expanded loading point is greater than the path of the current optimal loading segment execution order, then it is not added to the priority queue, otherwise, it is added to the priority queue. S5-3. If all loading points are already included in the current path and the path length is less than the optimal solution of the current loading segment, update the optimal solution of the loading segment; S5-4. If the priority queue is empty, record the execution order of the optimal loading segment and the last loading point. Otherwise, jump to S5-2.

[0016] S5-5: Construct the adjacency matrix between the last loading point in the execution order of the optimal loading segment and all unloading points, randomly generate an execution order of the unloading segment as the bound, and put the last loading point and the current path into the priority queue; S5-6. Take out the node at the head of the queue for expansion and calculate the current path length. If the path value of a certain expanded unloading point is greater than the path of the currently found optimal execution order of the unloading segment, then do not add it to the priority queue. Otherwise, add it to the priority queue; S5-7. If all unloading points are already included in the current path and the path length is less than the optimal solution of the current unloading segment, update the optimal solution of the unloading segment; S5-8. If the priority queue is not empty, jump to S5-6. Otherwise, record the optimal execution order of the unloading segment and merge it with the optimal execution order of the loading segment to obtain the final task execution order.

[0017] In an alternative embodiment, in step S5, obtaining the task execution order of each AGV agent based on the key parameters and the objective function specifically includes that when the number of tasks currently accepted by the AGV is greater than 6, a two-stage genetic algorithm is used to solve: Initialize the crossover probability, mutation probability, population size, and number of evolutionary generations of the genetic algorithm, and generate the initial population based on the task execution order in the initial loading stage and the task execution order in the initial unloading stage as chromosomes; Perform a selection operation on the current population based on the tournament selection method to generate a new population; Perform crossover operations and mutation operations on the individuals in the new population; Calculate the fitness value of the new population based on the parameters and the objective function in the problem model, record the execution order and its path length of the optimal individual in the population, and update the global optimal execution order and path length; If the number of evolutionary generations reaches the preset number of generations, output the optimal execution order and path length of the AGV. Otherwise, return to the steps of crossover operations and mutation operations.

[0018] In an alternative embodiment, the steps of configuring the working priority for each AGV agent based on the power of the AGV and the number of its loaded tasks in step S5 include: Obtain the working states of the AGV agents including the idle state, risk state, loading state, and unloading state according to the power of the AGV and the number of its loaded tasks; When a risk-state AGV conflicts with other AGVs, the risk-state AGV has the highest priority. When a risk-state AGV conflicts with another risk-state AGV, one of the AGVs is randomly designated to have a high priority; When an idle-state AGV conflicts with other AGVs, the idle-state AGV has the lowest priority. When an idle-state AGV conflicts with another idle-state AGV, one of the AGVs is randomly designated to have a high priority; When a loading-state AGV conflicts with an unloading-state AGV, the unloading-state AGV has a higher priority; When a loading-state AGV conflicts with another loading-state AGV, or an unloading-state AGV conflicts with another unloading-state AGV, the one with fewer remaining tasks to be completed has a higher priority. Otherwise, the AGV that is expected to complete the order task first has a higher priority. Otherwise, the AGV with a longer cumulative waiting time has a higher priority. If the above conditions are all the same, one of the AGVs is randomly designated to have a high priority.

[0019] In an optional implementation, the specific steps for conflict-free scheduling by combining the working priorities of AGV agents and the task execution order are as follows: The AGV sequentially places the nodes in its loading-phase execution order into a list.

[0020] Take the first loading node in the list as the current task point, calculate the global planned path based on the topological dimension, and select the driving route; The AGV agent detects whether a conflict occurs based on the grid dimension. If a conflict exists, it compares the priorities with the conflicting AGV agent and resolves the conflict based on the AGV agent priority according to the priority level, and selects to drive or stop and wait; If the AGV has not reached the current loading point, jump to the conflict detection step. Otherwise, remove the current loading node from the list. If the number of nodes in the list is not empty, jump to the step of taking the first loading node in the list as the current task point; If a new order task is accepted, perform re-planning of the execution order and jump to the step of sequentially placing into the list; The AGV sequentially places the nodes in its unloading-phase execution order into a list; Take the first unloading node in the list as the current task point, calculate the global planned path based on the topological dimension, and select the driving route; The AGV agent detects whether a conflict occurs based on the grid dimension. If a conflict exists, it compares the priorities with the conflicting AGV agent and resolves the conflict based on the AGV agent priority according to the priority level, and selects to drive or stop and wait; If the AGV has not reached the current unloading node, jump to the conflict detection step. Otherwise, remove the current unloading node from the list. If the number of nodes in the list is not empty, jump to the step of taking the first unloading node in the list as the current task point; If a new order task is accepted, put it into the buffer list and wait for the execution order planning after all current unloading tasks are completed.

[0021] In a second aspect, the present invention provides a multi-load automated guided vehicle online scheduling system. When the system is implemented, the above-mentioned multi-load automated guided vehicle online scheduling method is executed. The system includes: A scenario model construction module, which obtains the layout rules of the e-commerce intelligent warehouse and constructs an e-commerce intelligent warehouse scenario model based on the combination of the grid method and the topology method with the layout rules; A problem model construction module, which obtains the online orders of the e-commerce intelligent warehouse, establishes a multi-load AGV online order scheduling problem model based on the e-commerce intelligent warehouse model combined with the online orders, and obtains key parameters and objective functions based on the problem model; An agent construction module, which defines the agent types as order management agents, packing table agents, AGV agents, and information management agents based on the multi-load AGV online order scheduling model, constructs a multi-agent system based on the improved contract net protocol mechanism combined with all agents, and takes the party providing the task assignment of the agent as the tenderer and the party receiving and executing the task as the bidder; An order task assignment module, which establishes an order packing mechanism with the order management agent as the tenderer and the packing table agent as the bidder, and establishes an allocation and transportation mechanism with the packing table agent as the tenderer and the AGV agent as the bidder, and realizes order task assignment based on the order packing mechanism and the allocation and transportation mechanism; A scheduling execution module, which obtains the task execution order of each AGV agent based on the key parameters and the objective function, configures the working priority for each AGV agent based on the power of the AGV and the number of its load tasks, and performs conflict-free scheduling by combining the working priority of the AGV agent and the task execution order.

[0022] The beneficial effects of the present invention are as follows. The multi-load automated guided vehicle online scheduling method and system provided by the present invention construct a scenario model by obtaining the layout rules of the e-commerce intelligent warehouse, establish a scheduling problem model in combination with online orders and obtain key parameters and objective functions, define multiple agent types to construct a multi-agent system, establish an order packing and allocation and transportation mechanism to realize task assignment, obtain the task execution order based on key elements and configure the working priority to perform conflict-free scheduling, effectively improve the order task scheduling efficiency of the e-commerce intelligent warehouse, optimize the running path of the AGV, reasonably allocate tasks, adapt to the dynamic changes of the warehouse, enhance the robustness of the system, improve the overall operation efficiency, and provide strong support for the efficient operation of e-commerce logistics.

[0023] In addition, the design principle of the present invention is reliable, the structure is simple, and it has a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0025] Figure 1 is a schematic flowchart of an online scheduling method for a multi-load automatic tractor in an embodiment of the present invention.

[0026] Figure 2 is a schematic diagram of the construction of an e-commerce intelligent warehouse scenario based on a grid-topology two-dimensionality in an embodiment of the present invention.

[0027] Figure 3 is a schematic diagram of a multi-agent system protocol mechanism in an embodiment of the present invention.

[0028] Figure 4 is a schematic diagram of multi-agents and a multi-agent system in an embodiment of the present invention.

[0029] Figure 5 is a schematic diagram of a chromosome of a two-stage genetic algorithm in an embodiment of the present invention.

[0030] Figure 6 is a schematic diagram of a crossover strategy of a two-stage genetic algorithm in an embodiment of the present invention.

[0031] Figure 7 is a schematic diagram of a mutation strategy of a two-stage genetic algorithm in an embodiment of the present invention.

[0032] Figure 8 is a schematic block diagram of an online scheduling system for a multi-load automatic tractor in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs. The terms used in the description of the present invention herein are for the purpose of describing specific embodiments only and are not intended to limit the present invention.

[0035] The online scheduling method for multi-load automatic tractors provided by the embodiments of the present invention is executed by a computer device. Correspondingly, the online scheduling system for multi-load automatic tractors runs in the computer device.

[0036] Figure 1 It is a schematic flowchart of the online scheduling method for multi-load automatic tractors according to an embodiment of the present invention. Among them, Figure 1 The execution subject can be an online scheduling system for multi-load automatic tractors. According to different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0037] As Figure 1 shown, the method includes: S1, obtain the layout rules of the e-commerce intelligent warehouse, and construct an e-commerce intelligent warehouse scenario model based on the combination of the grid method and the topology method with the layout rules; It can accurately reflect the actual layout structure in the warehouse, including the distribution of areas such as shelves, AGV docking areas, and packing stations, and provide a basic framework for subsequent operations.

[0038] S2, obtain the online orders of the e-commerce intelligent warehouse, establish an online order scheduling problem model for multi-load AGVs based on the e-commerce intelligent warehouse model combined with the online orders, and obtain key parameters and objective functions based on the problem model; Integrate the dynamic information of the orders into the model. The key parameters (such as loading points, unloading points, AGV load capacity, etc.) and objective functions (such as minimizing the sum of the total path and time) obtained based on such a model can accurately reflect the essence of the actual scheduling problem and provide a reliable basis for formulating a scientific and reasonable scheduling strategy.

[0039] S3, define the types of agents as order management agents, packing station agents, AGV agents, and information management agents based on the online order scheduling model for multi-load AGVs, construct a multi-agent system based on the improved contract net protocol mechanism combined with all agents, and use the party providing the task assignment of the agent as the tenderer and the party receiving and executing the task as the bidder; It realizes the clear division of labor and efficient cooperation of different functional modules in the operation of the e-commerce intelligent warehouse. The system then selects the most suitable agent to execute the task, realizing the reasonable allocation of resources.

[0040] S4. Establish an order packing mechanism with the order management agent as the tenderer and the packing station agent as the bidder, and establish an allocation and transportation mechanism with the packing station agent as the tenderer and the AGV agent as the bidder. Achieve order task allocation based on the order packing mechanism and the allocation and transportation mechanism. By establishing an order packing mechanism with the order management agent as the tenderer and the packing station agent as the bidder, the reasonable allocation of order tasks in the packing link is realized.

[0041] S5. Obtain the task execution order of each AGV agent based on the key parameters and the objective function. Configure the work priority for each AGV agent based on the battery power of the AGV and the number of tasks it is loaded with. Perform conflict-free scheduling by combining the work priority of the AGV agent and the task execution order.

[0042] Obtaining the task execution order of each AGV agent based on the key parameters and the objective function enables scientific planning according to the specific requirements of the tasks (such as the locations of the loading point and unloading point, the cargo load, etc.) and the warehouse environment (such as the road conditions, the path cost between nodes, etc.).

[0043] Optionally, as an embodiment of the present invention, the specific steps of step S1 are as follows: The layout rules include the shelf layout rule, the AGV docking area layout rule, and the packing station layout rule; Represent the roads, shelves, packing stations, and AGV docking areas in the warehouse environment based on the grid method; Based on the topology method, taking the crossroads in the warehouse as the benchmark, the center of the grid at the crossroads as the node, and the dividing line of the two-way road as the edge connecting the nodes, divide the warehouse into multiple areas. The resulting scene graph after modeling is as Figure 2 shown.

[0044] Optionally, as an embodiment of the present invention, step S2 specifically includes: Take the length of each grid as , then the distance between the centers of two adjacent grids is also . Define that there are loading points in the warehouse, located in the adjacent grids of the shelf area. The set of loading points is represented by the set . There are unloading points. The set of unloading points is represented by the set .

[0045] Define that there are transportable AGVs in the system, represented by the set , , …, , The corresponding load capacity is It is indicated that the maximum load capacity is represented by a constant It is indicated that, on the premise of meeting the requirements of the maximum loading capacity, the transportation of goods at multiple loading points can be carried out simultaneously. Online orders in the e-commerce warehouse arrive in real time. The processed orders are converted into a set of task points (loading points and unloading points. The loading point is the location where the AGV picks up goods, and the unloading point is the location where the AGV unloads goods to the packing table), and are identified by the loading points and unloading points. For example, and are used to identify the task , where represents its loading point, represents its unloading point. = { , ,..., } represents the set of loading points in the task set, = { , ,..., } represents the set of unloading points in the task set. Each order task contains the following information: loading point , unloading point , load , that is, the task = { , , }.

[0046] It is stipulated that can receive new orders when the current load is less than , and re-plan the task execution sequence. After each task is accepted, all the currently received tasks are regarded as a task group represented by . The task group contains the number of tasks .

[0047] Calculate the time cost for each AGV to complete the task group. The specific calculation is as follows:

[0048] Among them, represents the time cost required to complete the task group , g represents the gth task group, represents the number of tasks contained in the gth task group, m represents the number of the AGV, M represents the total number of transportable AGVs, n represents the task number in the task group, represents the time for each operation at the loading point or unloading point, represents the waiting time generated by the AGV to avoid conflicts during task execution, , and are the loading points of the 1st, the nth, and the (n + 1)th tasks in the task group respectively, , and are the unloading points of the 1st, the nth, and the (n + 1)th tasks in the task group respectively, represents the time cost between node m and node , represents the time cost between node and , represents the time cost between node and , represents the time cost between node and ; Obtain the total path cost between the loading points and the total path cost between the unloading points of each AGV to complete the task group, and then synthesize them to obtain the total path cost; Based on the time cost and the total path cost of each AGV to complete the task group, calculate the scheduling optimization objective function of each AGV, and the specific calculation is as follows:

[0049] wherein, represents the operation of finding the minimum value.

[0050] Among them, the formula calculation method and constraint conditions of the multi-load AGV scheduling scheme model for the e-commerce intelligent warehouse are as follows:

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] The above formula means that each task can only be assigned to one AGV.

[0057]

[0058] The above formula means that the load capacity of〖AGV〗_m must be less than or equal to the sum of all cargo loads in its task group.

[0059]

[0060] The above formula means that at any moment, at most 1 AGV can be accommodated in the same grid in the warehouse.

[0061] Optionally, as an embodiment of the present invention, step S3 specifically includes: Analyze the scenario requirements and problem-solving characteristics of the e-commerce intelligent warehouse, divide the abstraction level, and divide the system into multiple types of agents. The online multi-load AGV scheduling problem in the e-commerce intelligent warehouse can be decomposed into the process of the order system receiving external orders, the order system allocating order tasks to the packing station, the packing station allocating order tasks to the AGV, and the AGV executing the transportation task. This process involves task collaboration among multiple agents. Therefore, through abstraction, the system can be divided into four types of agents: order management agent, packing station agent, AGV agent, and information management agent, which together form a multi-agent system.

[0062] Analyze and design in detail for the four types of agents: order management agent, packing station agent, AGV agent, and information management agent. The functions of the order management agent include receiving orders from external systems, adding the picking points corresponding to the order goods to the order task information, and the allocation and tracking between orders and the packing station agent. The functions of the packing station agent include accepting the order tasks assigned from the order management agent, the order task allocation among AGV agents, and receiving the goods transported by the AGV agent. The AGV agent is responsible for accepting the order tasks assigned from the packing station agent, planning the task execution sequence, conflict-free path planning, loading goods, transporting goods, and unloading goods. The information management agent is mainly responsible for recording and synchronizing the feedback data of the order management agent, packing station agent, and AGV agent when they execute tasks.

[0063] Determine the protocol mechanism of the multi-agent system, optimize the collaboration process among agents, achieve improved coordination efficiency, improve the evaluation mechanism, and complete the construction of the multi-agent system. In this step, an improved contract net protocol mechanism is adopted, and the agents in the multi-agent system are divided into two types of roles: the tenderer and the bidder. Among them, the party providing the task allocation is the tenderer, and the party receiving and executing the task is the bidder. This protocol mechanism mainly includes processes such as tendering, bidding, bid evaluation, authorization, and contract signing. When the tendering party receives a task and needs to allocate it, it will send a message to all bidders and wait for the bids sent by the bidders within a certain time; after receiving the information, the bidder will choose whether to bid according to its own situation. If it chooses to bid, it will simultaneously send a bid containing the bidding information for calculating the bid value; after the tendering party finishes receiving, it will evaluate all the received bids, select the bidder with the largest bid value, that is, the most suitable bidder to execute the task, and initiate a contract signing invitation. The two parties sign the contract to complete the task allocation from the tendering party to the bidder, such as Figure 3As shown. The multi-agent construction is completed, as Figure 4 shown.

[0064] Optionally, as an embodiment of the present invention, in step S4, establishing an order packing mechanism with the order management agent as the tenderer and the packing table agent as the bidder specifically includes: When there is a new order, the order management agent is the tenderer and formulates a tender bid, and the bid includes the location information of the order goods; The packing table agent is the bidder. After receiving the bid information of the order management agent, first judge whether the number of its current unfinished order tasks exceeds the threshold for accepting new orders; If the condition for accepting the order is met, initiate a bid to the tenderer order management agent, and provide its current remaining unfinished task information and the path length index from the location of the goods, so that the tenderer order management agent calculates the bid value and evaluates the bid. The calculation formula for the bid value of the bid is as follows:

[0065] Wherein, is the bid value of the bid of the packing table agent as the bidder, is the number of orders currently accepted by the packing table agent as the bidder, is the maximum number of orders that can be accepted, is the Manhattan distance of the bidder from the location of the order goods, is the maximum distance of the packing table agent from the goods, and are used to adjust the weights of the number of orders undertaken and the distance of the goods; The order management agent evaluates all bidders, selects the bidder with the largest bid value to issue a winning bid notice, and waits for the winner to confirm. If the winner confirms, a contract is signed to complete the task assignment.

[0066] Optionally, as an embodiment of the present invention, in step S4, establishing an allocation and transportation mechanism with the packing table agent as the tenderer and the AGV agent as the bidder specifically includes: After the packing table agent accepts the order task assigned by the order management agent, the packing table agent will be the tenderer to specify a tender bid, and the bid information is the location of the order goods and the location information of the packing table; The AGV agent is the bidder. After receiving the bid information of the packing table agent, first judge whether its power is sufficient to accept new tasks and whether its current capacity can accommodate the new order goods; If the conditions are met, a bid is launched to the packing station agent of the tenderer, and the difference in the total path planning length before and after accepting the task is calculated, as well as the current accepted order task quantity, which are sent to the packing station agent as tender information for calculating the tender value. The tender value calculation formula for the tenderer is as follows:

[0067] Where is the tender value of the tenderer's AGV agent, is the number of orders currently accepted by the tenderer's AGV agent, is the maximum number of orders that can be accepted, is the total path planning length of the tenderer's AGV agent before accepting the task, is the total path planning length of the tenderer's AGV agent after accepting the task, is the maximum difference in the total path planning length of the AGV agent before and after accepting the task from this cargo, and are used to adjust the weight of the number of orders already received by the AGV agent and the maximum difference in the total path planning length of the AGV agent before and after accepting the task from this cargo; The packing station agent conducts a winning bid evaluation on all tenderers, selects the tenderer with the largest tender value and sends a winning bid notice, and waits for the winner to confirm; if the winner confirms, a contract is signed to complete the task assignment from the packing station agent to the AGV agent.

[0068] Optionally, as an embodiment of the present invention, the task execution order planning refers to that the multi-load AGV agent needs to plan the execution order of tasks when receiving multiple tasks to optimize the total path length. In the multi-agent system, it is set that the AGV agent first executes the task of picking up and loading the order, and then executes the unloading task to prevent the goods of some order tasks from being unloaded and packed while being loaded by the AGV all the time. Therefore, the task execution order of the AGV agent is divided into two stages: the loading stage and the unloading stage. The loading stage refers to the process of the AGV traversing all task loading points from the current position, and the unloading stage refers to the process of traversing all task unloading points from the last loading point. Loading must be before unloading, and the two cannot intersect. The AGV can accept new orders and re-plan the execution order during the loading stage, but it is allowed to accept new orders during the unloading stage but will be put into the buffer list and continue to execute the current transportation task until all unloading tasks are completed. The present invention solves the following two situations according to the number of tasks currently accepted by the AGV: When the number of tasks currently accepted by the AGV is no more than 6, the two-stage branch and bound algorithm based on the priority queue is used to solve.

[0069] When the number of tasks currently accepted by the AGV is greater than 6, the two-stage genetic algorithm is used to solve the problem.

[0070] Optionally, as an embodiment of the present invention, in step S5, to obtain the task execution order of each AGV agent based on the key parameters and the objective function, specifically, when the number of tasks currently accepted by the AGV is not greater than 6, the two-stage branch and bound algorithm based on the priority queue is used to solve the problem: S5-1, construct the adjacency matrix between the current position of the AGV and all task loading points, randomly generate an execution order of the loading segment as the bound, create a priority queue, and put the current position of the AGV and the current path into the priority queue; S5-2, take the head node of the queue for expansion and calculate the current path length. If the path value of a certain expanded loading point is greater than the path of the current optimal loading segment execution order, then it is not added to the priority queue; otherwise, it is added to the priority queue; S5-3, if all loading points are already included in the current path and the path length is less than the optimal solution of the current loading segment, then update the optimal solution of the loading segment; S5-4, if the priority queue is empty, then record the optimal loading segment execution order and the last loading point; otherwise, jump to S5-2.

[0071] S5-5: Construct the adjacency matrix between the last loading point of the optimal loading segment execution order and all unloading points, randomly generate an execution order of the unloading segment as the bound, and put the last loading point and the current path into the priority queue; S5-6 Take out the head node of the queue for expansion and calculate the current path length. If the path value of a certain expanded unloading point is greater than the path of the current optimal unloading segment execution order found, then it is not added to the priority queue; otherwise, it is added to the priority queue; S5-7 If all unloading points are already included in the current path and the path length is less than the optimal solution of the current unloading segment, then update the optimal solution of the unloading segment; S5-8 If the priority queue is not empty, then jump to S5-6; otherwise, record the optimal unloading segment execution order, and merge it with the optimal loading segment execution order to obtain the final task execution order.

[0072] Optionally, as an embodiment of the present invention, in step S5, to obtain the task execution order of each AGV agent based on the key parameters and the objective function, specifically, when the number of tasks currently accepted by the AGV is greater than 6, the two-stage genetic algorithm is used to solve the problem: The two-stage genetic algorithm adopts a specific segmented coding method. The chromosome consists of two parts, A and B. Among them, A represents the task execution order (LS) in the loading stage, and B represents the task execution order (US) in the unloading stage. Their lengths are both (The number of tasks currently accepted by the AGV).

[0073] The numbers represent the subscript indices of the tasks within the task group. The LS part indicates that the AGV sequentially traverses the loading points of these tasks. The US part indicates that after the AGV completes all the loading tasks, it sequentially traverses the unloading points of these tasks.

[0074] The selection strategy is that when the number of parent individuals participating in hybridization is , the tournament selection method is used to select (1 - ) * individuals, and the elitist retention method is used to select * individuals ( is the proportion of elite individuals).

[0075] The crossover strategy is to randomly generate an array of length 2 * , which only contains 0s and 1s. For the positions where the values in the array are 1, the tasks at the corresponding positions of the two individuals are crossed over, and two offspring individuals are generated according to the mapping relationship.

[0076] The mutation operation is to randomly select one point each from the loading section and the unloading section as the mutation points, and randomly generate a new task point and swap it with the original one.

[0077] Refer to Figure 5 , initialize the crossover probability, mutation probability, population size, and number of evolutionary generations of the genetic algorithm, and generate the initial population based on the task execution order in the initial loading stage and the task execution order in the initial unloading stage as the chromosomes; Perform a selection operation on the current population based on the tournament selection method to generate a new population; Refer to Figure 6 and Figure 7 perform crossover operations and mutation operations on the individuals in the new population; Calculate the fitness values of the new population based on the parameters and objective function in the problem model, record the execution order and path length of the optimal individual in the population, and update the globally optimal execution order and path length; If the number of evolutionary generations reaches the preset number of generations, output the optimal execution order and path length of the AGV, otherwise return to the crossover operation and mutation operation steps.

[0078] Optionally, as an embodiment of the present invention, the step of configuring the working priority for each AGV agent based on the power of the AGV and the number of its loaded tasks in step S5 includes: Since the operation cycle of an e-commerce intelligent warehouse is relatively long, the power constraint of the AGV itself needs to be considered. Therefore, a power threshold for the AGV is set. When the AGV's power is lower than the power threshold, to prevent the AGV from losing power during operation and causing warehouse road congestion or deadlocks, no new tasks will be assigned to the AGV. In addition, to ensure the orderly operation of the AGV, the loading task and the unloading task of the AGV cannot be carried out crosswise. For all loading tasks, in order to more conveniently handle the task allocation problem based on the AGV's power and the current load task volume, a method for defining the working state of the AGV is proposed. According to the AGV's power and its load task quantity, the working state of the AGV is defined as the following types: (1) Idle state: The AGV is not assigned tasks, is located in the docking area or on the way to the docking area, and its power is higher than the power threshold. New order tasks can be assigned and the tasks can be executed immediately.

[0079] (2) Risk state: The AGV may be executing tasks in the loading section, unloading section, or not assigned tasks, but the current power of the AGV is lower than the power threshold. No new tasks can be assigned, and the final destination of the risk-state AGV is the docking area.

[0080] (3) Loading state: The AGV is executing tasks in the loading section, and its current power is higher than the power threshold. New tasks can be assigned based on the current capacity of the AGV, and the execution order can be re-planned.

[0081] (4) Unloading state: The AGV is executing tasks in the unloading section, and its current power is higher than the power threshold. New tasks can be assigned, and the new tasks are placed in the buffer pool. After all the unloading section tasks are completed, the execution order is planned.

[0082] Since conflicts may occur among multiple AGVs during task execution, it is necessary to set priorities according to the working state of the AGV. When a conflict occurs, obstacle avoidance can be carried out according to the AGV priority. For the above working states, the priorities of AGVs in different states are defined.

[0083] (1) When a risk-state AGV conflicts with other AGVs, the risk-state AGV has the highest priority. When a risk-state AGV conflicts with another risk-state AGV, one of the AGVs is randomly designated to have a high priority.

[0084] (2) When an idle-state AGV conflicts with other AGVs, the idle-state AGV has the lowest priority. When an idle-state AGV conflicts with another idle-state AGV, one of the AGVs is randomly designated to have a high priority.

[0085] (3) When a loading-state AGV conflicts with an unloading-state AGV, the unloading-state AGV has a higher priority.

[0086] When a loaded AGV conflicts with another loaded AGV, or an unloaded AGV conflicts with another unloaded AGV, the AGV with the smaller number of remaining tasks to be completed has a higher priority. Otherwise, the AGV that is expected to complete the order task first has a higher priority. Otherwise, the AGV with a longer accumulated waiting time has a higher priority. If the above conditions are the same, an AGV is randomly designated to have a high priority.

[0087] Optionally, as an embodiment of the present invention, the conflicts in conflict-free path planning mainly include three types: head-on conflict, encounter conflict, and following conflict. A head-on conflict means that two AGVs collide while traveling in opposite directions on the same lane; an encounter conflict means that two AGVs arrive at a certain grid at the same time; a following conflict means that one AGV stops waiting, blocking the travel of another following AGV. Since each path in the warehouse is a two-way passage and the vehicles all travel on the right side of the road, there is no head-on conflict. When two AGVs have a following conflict, the AGV at the rear needs to wait for the AGV in front to leave before it can travel. It is stipulated that when an AGV is at the starting point, the loading point, the unloading point, or only one unit away from these three positions, the AGV is allowed to directly travel one unit vertically and enter the grid in the same direction as the planned path.

[0088] For the encounter conflict, the present invention proposes a conflict resolution method based on the priority of AGVs. If the encounter conflict occurs at an intersection and the waiting times of the two AGVs are the same, the AGV with a higher priority passes first. Otherwise, the AGV with a longer waiting time passes first. This can ensure that the waiting time for AGVs to pass through the intersection due to avoiding conflicts is average and the shortest. If the encounter conflict occurs on a two-way road, when the AGV is at the starting point, the traveling direction of the grid where the task point is located is different from the direction to the next task point, or when the AGV is at an adjacent grid in the opposite direction to the traveling direction of the task point grid, when there is no other AGV occupying the target grid, the priority of the AGV is set to the highest, and the AGV is allowed to shuttle through the road and enter the target grid. If the target grid is occupied by other AGVs, the AGV needs to wait for the grid to be free, and in the case where the same AGV's next target needs to enter this grid, the priority of this AGV is the highest and it enters this grid first. Otherwise, according to the definition of the working state and priority of the above AGVs, the AGV with a higher priority passes first.

[0089] The specific steps for conflict-free scheduling by combining the working priority of the AGV agent and the task execution order are as follows: The AGV sequentially places the nodes in its loading stage execution order into a list.

[0090] Take the first loading node in the list as the current task point, calculate the global planned path based on the topological dimension, and select the traveling road; The AGV agent detects whether a conflict occurs based on the grid dimension. If there is a conflict, it compares the priorities with the conflicting AGV agents and resolves the conflict based on the priorities of the AGV agents, choosing to drive or stop and wait. If the AGV has not reached the current loading point, it jumps to the step of detecting whether a conflict occurs. Otherwise, it removes the current loading node from the list. If the number of nodes in the list is not empty, it jumps to the step of taking the first loading node in the list as the current task point. If a new order task is accepted, it performs re-planning of the execution order and jumps to the step of putting them into the list in sequence. The AGV puts the nodes in its own execution order during the unloading phase into the list in sequence. Takes the first unloading node in the list as the current task point, calculates the global planned path based on the topological dimension, and selects the driving route. The AGV agent detects whether a conflict occurs based on the grid dimension. If there is a conflict, it compares the priorities with the conflicting AGV agents and resolves the conflict based on the priorities of the AGV agents, choosing to drive or stop and wait. If the AGV has not reached the current unloading node, it jumps to the step of detecting whether a conflict occurs. Otherwise, it removes the current unloading node from the list. If the number of nodes in the list is not empty, it jumps to the step of taking the first unloading node in the list as the current task point. If a new order task is accepted, it is put into the buffer list and waits for the execution order planning after all current unloading tasks are completed.

[0091] In some embodiments, the multi-load automated guided vehicle online scheduling system may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the multi-load automated guided vehicle online scheduling system can be stored in the memory of the computer device and executed by at least one processor to execute (see Figure 1 description) the functions of the multi-load automated guided vehicle online scheduling. The system includes: A scenario model construction module, which obtains the layout rules of the e-commerce intelligent warehouse and constructs an e-commerce intelligent warehouse scenario model based on the combination of the grid method and the topological method with the layout rules. A problem model construction module, which obtains the online orders of the e-commerce intelligent warehouse, establishes a multi-load AGV online order scheduling problem model based on the e-commerce intelligent warehouse model combined with the online orders, and obtains the key parameters and the objective function based on the problem model. The agent construction module defines the agent types as order management agents, packing table agents, AGV agents, and information management agents based on the multi-load AGV online order scheduling model. A multi-agent system is constructed by combining all agents based on the improved contract net protocol mechanism. The party providing task allocation among the agents is regarded as the tenderer, and the party receiving and executing the tasks is regarded as the bidder. The order task allocation module establishes an order packing mechanism with the order management agent as the tenderer and the packing table agent as the bidder, and establishes an allocation and transportation mechanism with the packing table agent as the tenderer and the AGV agent as the bidder. Order task allocation is realized based on the order packing mechanism and the allocation and transportation mechanism. The scheduling execution module obtains the task execution order of each AGV agent based on key parameters and objective functions, configures the working priority for each AGV agent based on the battery power of the AGV and the number of tasks it is loaded with, and performs conflict-free scheduling by combining the working priority of the AGV agent and the task execution order.

[0092] In this embodiment, the multi-load automated guided vehicle online scheduling system can be divided into multiple functional modules according to the functions it performs, such as Figure 8 shown. The functional modules of the system may include: a scenario model construction module, a problem model construction module, an agent construction module, an order task allocation module, and a scheduling execution module. The module referred to in the present invention means a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and are stored in the memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0093] Through the scenario model construction module, accurately grasp the layout construction model of the e-commerce intelligent warehouse. The problem model construction module combines orders to establish a scheduling model and clarify key elements. The agent construction module defines multiple agents to construct an efficient multi-agent system. The order task allocation module establishes a reasonable mechanism to achieve accurate allocation. The scheduling execution module scientifically plans the task order and configures priorities to ensure conflict-free scheduling, comprehensively improving the order processing efficiency of the e-commerce intelligent warehouse, optimizing resource allocation, enhancing the adaptability and stability of the system, reducing operating costs, improving customer satisfaction, providing strong support for the efficient operation of e-commerce logistics, and promoting e-commerce enterprises to gain advantages in the fierce market competition.

[0094] Therefore, the technical effects that can be achieved in this embodiment can be referred to the description above and will not be elaborated here.

[0095] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disc, etc., various media that can store program codes, including several instructions to enable a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0096] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the terminal embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the descriptions in the method embodiments.

[0097] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the system or module can be in an electrical, mechanical or other form.

[0098] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place, or they can be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0099] In addition, in each embodiment of the present invention, the various functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0100] Although the present invention has been described in detail by reference to the accompanying drawings and in conjunction with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions should all be within the scope of the present invention. / Any person skilled in the art within the technical scope disclosed by the present invention can easily conceive of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for online dispatching of multi-load automatic tractors, characterized in that: The following steps are involved: S1, obtain the layout rules of the e-commerce smart warehouse, and build the e-commerce smart warehouse scenario model based on the grid method and topology method combined with the layout rules; S2, obtain online orders from e-commerce smart warehouses, establish a multi-load AGV online order scheduling problem model based on the e-commerce smart warehouse model and online orders, and obtain key parameters and objective functions based on the problem model; S3, based on the multi-load AGV online order scheduling model, defines the agent types as order management agent, packaging station agent, AGV agent and information management agent. Based on the improved contract network protocol mechanism, all agents are combined to build a multi-agent system. The party that provides task allocation as the tenderer, and the party that receives and executes the task as the bidder; S4, establish an order packaging mechanism with the order management agent as the tenderer and the packaging station agent as the bidder, establish an allocation and transportation mechanism with the packaging station agent as the tenderer and the AGV agent as the bidder, and realize order task allocation based on the order packaging mechanism and the allocation and transportation mechanism; S5, based on key parameters and objective functions, obtains the task execution order of each AGV agent, configures the work priority for each AGV agent based on the power of the AGV and the number of its load tasks, and performs conflict-free scheduling by combining the AGV agent work priority and task execution order.

2. The online dispatching method for multi-load automatic tractors according to claim 1, characterized in that: Step S1 specifically includes: Layout rules include shelf layout rules, AGV docking area layout rules, and packaging station layout rules; Represent roads, shelves, packaging tables and AGV docking areas in the warehouse environment based on the grid method; Based on the topological method, the warehouse is divided into multiple areas with the intersection of the warehouse as the benchmark, the center of the intersection grid as the node, and the dividing line of the two-way road as the edge connecting the nodes.

3. The online dispatching method for multi-load automatic tractors according to claim 1, characterized in that: The step S2 specifically includes: Configure loading points, unloading points and loads for each online order task; Combine all the tasks currently undertaken by each AGV into a task group; Calculate the time cost of each AGV to complete the task group. The specific calculation is: in, Indicates completion of task group The time cost required, g represents the g-th task group, represents the number of tasks contained in the g-th task group, m represents the number of AGVs, M represents the total number of AGVs that can be transported, n represents the number of tasks in the task group, Indicates the time of each operation at the loading point or unloading point. It indicates the waiting time of AGV to avoid conflicts when performing tasks. , and They are the loading points of the 1st, nth, and n+1th tasks in the task group, respectively. , and They are the unloading points of the 1st, nth, and n+1th tasks in the task group, respectively. Represents node m and node The time cost between Representation Node and The time cost between Representation Node and The time cost between Representation Node and The time cost between Obtain the sum of the path costs between the loading points and the unloading points of each AGV completing the task group and combine them to obtain the sum of the path costs; The scheduling optimization objective function of each AGV is calculated based on the total time cost and path cost of each AGV completing the task group. The specific calculation is: in, Represents an operation to find the minimum value.

4. The online dispatching method of multi-load automatic tractor according to claim 1, characterized in that: In step S4, the order management agent is used as the tenderer and the packaging agent is used as the bidder to establish an order packaging mechanism, which specifically includes: When there is a new order, the order management agent is set as the bidder and a tender document is prepared, which contains the location information of the ordered goods; The packaging agent is regarded as a bidder. After receiving the bidding information from the order management agent, it first determines whether the number of its currently unfinished packaging order tasks exceeds the threshold for accepting new orders; If the conditions for accepting the order are met, a bid is initiated to the tenderer's order management agent, and the information of the remaining unfinished packaging tasks and the path length indicator to the location of the goods are provided, so that the tenderer's order management agent calculates the bid value and evaluates the bid. The bid value calculation formula is as follows: in, Packaging agent bid values ​​for bidders, The number of orders currently accepted by the bidder packaging agent, is the maximum number of orders that can be received, is the Manhattan distance between the bidder and the location of the ordered goods, is the maximum distance between the packing station agent and the goods, and Used to adjust the weight of the number of orders undertaken and the distance of the goods; The order management agent evaluates all bidders, selects the bidder with the largest bid value, issues a winning notice, and waits for the winning bidder to confirm. If the winning bidder confirms, the contract is signed and the task assignment is completed.

5. The online dispatching method of multi-load automatic tractor according to claim 1, characterized in that: In step S4, the packaging station agent is used as the tenderer and the AGV agent is used as the bidder to establish a distribution and transportation mechanism, which specifically includes: When the packing station agent accepts the order task assigned by the order management agent, the packing station agent will specify the bidding document as the tenderer. The bidding document information includes the location of the ordered goods and the location of the packing station. As a bidder, after receiving the bidding information from the packaging station agent, the AGV agent first determines whether its power is sufficient to accept the new task and whether the current capacity can accommodate the new order goods; If the conditions are met, a bid is initiated to the tenderer packaging agent, and the difference between the total length of the path planning after and before accepting the task is calculated, as well as the current amount of accepted order tasks as the bid information, which is sent to the packaging agent to calculate the bid value. The bidder's bid value calculation formula is as follows: in is the bid value of the bidder's AGV agent, is the number of orders currently accepted by the bidder AGV agent, is the maximum number of orders that can be received, is the total length of the path planned by the bidder’s AGV agent before accepting the task, is the total length of the path planned by the bidder’s AGV agent after accepting the task, The maximum difference between the total length of the path planned by the AGV agent after it accepts the task and before it accepts the task. and The weight used to adjust the maximum difference between the number of orders received by the AGV agent and the total length of the path planned by the AGV agent before and after accepting the task. The packaging station agent evaluates all bidders, selects the bidder with the largest bid value, issues a winning notice, and waits for the winning bidder to confirm; if the winning bidder confirms, a contract is signed, completing the task allocation from the packaging station agent to the AGV agent.

6. The online dispatching method of multi-load automatic tractor according to claim 1, characterized in that: In step S5, the task execution order of each AGV agent is obtained based on the key parameters and the objective function. Specifically, when the number of tasks currently accepted by the AGV is not more than 6, a two-stage branch and bound algorithm based on a priority queue is used to solve the problem: S5-1, construct an adjacency matrix between the current position of the AGV and all task loading points, and randomly generate a loading segment execution order as a limit, create a priority queue and put the current position of the AGV and the current path into the priority queue; S5-2, take the first node of the queue for expansion and calculate the current path length. If the path value of a certain extended loading point is greater than the path of the current optimal loading segment execution order, then it will not be added to the priority queue. Otherwise, it will be added to the priority queue. S5-3, if the current path already contains all loading points and the path length is less than the optimal solution of the current loading section, then update the optimal solution of the loading section; S5-4, if the priority queue is empty, record the optimal loading segment execution order and the last loading point, otherwise jump to S5-2; S5-5: construct an adjacency matrix between the last loading point of the optimal loading segment execution order and all unloading points, and randomly generate an unloading segment execution order as a limit, and put the last loading point and the current path into a priority queue; S5-6 takes out the node at the head of the queue for expansion and calculates the current path length. If the path value of a certain expanded unloading point is greater than the path of the currently found optimal unloading segment execution order, it will not be added to the priority queue. Otherwise, it will be added to the priority queue. S5-7 If the current path already contains all unloading points and the path length is less than the optimal solution of the current unloading segment, then update the optimal solution of the unloading segment; S5-8 If the priority queue is not empty, jump to S5-6, otherwise, record the optimal unloading segment execution order and merge it with the optimal loading segment execution order to obtain the final task execution order.

7. The online dispatching method of multi-load automatic tractor according to claim 6, characterized in that: In step S5, the task execution order of each AGV agent is obtained based on the key parameters and the objective function, specifically, when the number of tasks currently accepted by the AGV is greater than 6, a two-stage genetic algorithm is used to solve: Initialize the genetic algorithm crossover probability, mutation probability, population number evolution generation, and generate the initial population based on the task execution order of the initial loading phase and the task execution order of the initial unloading phase as chromosomes; Based on the tournament selection method, the current population is selected to generate a new population; Perform crossover and mutation operations on individuals in the new population; Calculate the fitness value of the new population based on the parameters and objective function in the problem model, record the execution order and path length of the optimal individuals in the population, and update the global optimal execution order and path length; If the evolutionary generation reaches the preset generation, the optimal execution order and path length of AGV are output, otherwise the crossover operation and mutation operation steps are returned.

8. The online dispatching method of multi-load automatic tractor according to claim 7, characterized in that: The step of configuring the work priority for each AGV agent based on the power of the AGV and the number of its load tasks in step S5 includes: According to the power of AGV and the number of its load tasks, the working states of AGV intelligent body include idle state, risk state, loading state and unloading state; When a risky AGV conflicts with other AGVs, the risky AGV has the highest priority. When a risky AGV conflicts with another risky AGV, one of the AGVs is randomly assigned a high priority. When an idle AGV conflicts with other AGVs, the idle AGV has the lowest priority. When an idle AGV conflicts with another idle AGV, one of the AGVs is randomly assigned a high priority. When a conflict occurs between a loading AGV and an unloading AGV, the unloading AGV has a higher priority; When a loading AGV conflicts with another loading AGV, or an unloading AGV conflicts with another unloading AGV, the AGV with fewer remaining tasks to be completed has a higher priority. Otherwise, the AGV that is expected to complete the order task first has a higher priority. Otherwise, the AGV with a longer cumulative waiting time has a higher priority. If the above conditions are the same, the randomly designated AGV has a higher priority.

9. The online dispatching method of multi-load automatic tractor according to claim 8, characterized in that: The specific steps for conflict-free scheduling based on the AGV agent's work priority and task execution order are as follows: The AGV puts the nodes in the order of execution of its own loading phase into the list one by one; Take the first loading node in the list as the current task point, calculate the global planning path based on the topological dimension, and select the driving road; The AGV agent detects whether a conflict occurs based on the grid dimension. If a conflict occurs, it compares the priority with the conflicting AGV agent and resolves the conflict based on the priority of the AGV agent, choosing to drive or stop and wait; If the AGV has not reached the current loading point, it will jump to the step of whether a conflict occurs. Otherwise, the current loading node will be removed from the list. If the number of nodes in the list is not empty, it will jump to the step of taking the first loading node in the list as the current task point. If a new order task is accepted, the execution order is replanned and the process jumps to the step of putting the new order task into the list in sequence. AGV puts the nodes in the order of execution of its own unloading phase into the list one by one; Take the first unloading node in the list as the current task point, calculate the global planning path based on the topological dimension, and select the driving road; The AGV agent detects whether a conflict occurs based on the grid dimension. If a conflict occurs, it compares the priority with the conflicting AGV agent and resolves the conflict based on the priority of the AGV agent, choosing to drive or stop and wait; If the AGV has not reached the current unloading node, it will jump to the step of whether a conflict occurs. Otherwise, the current unloading node will be removed from the list. If the number of nodes in the list is not empty, it will jump to the step of taking the first unloading node in the list as the current task point. If a new order task is accepted, it is put into the buffer list and the task execution sequence is planned after all current unloading tasks are completed.

10. A method for online dispatching of multi-load automatic tractors, characterized in that: When the system is implemented, the online dispatching method for multi-load automatic tractors according to any one of claims 1 to 7 is executed, and the system comprises: The scenario model building module obtains the layout rules of the e-commerce smart warehouse and builds the e-commerce smart warehouse scenario model based on the grid method and topology method combined with the layout rules; The problem model building module obtains online orders from e-commerce smart warehouses, builds a multi-load AGV online order scheduling problem model based on the e-commerce smart warehouse model and online orders, and obtains key parameters and objective functions based on the problem model; The agent building module defines agent types as order management agent, packaging agent, AGV agent and information management agent based on the multi-load AGV online order scheduling model. Based on the improved contract network protocol mechanism, all agents are combined to build a multi-agent system. The party that provides task allocation as the tenderer, and the party that receives and executes the task as the bidder. The order task allocation module uses the order management agent as the tenderer and the packing station agent as the bidder to establish an order packaging mechanism, uses the packing station agent as the tenderer and the AGV agent as the bidder to establish an allocation and transportation mechanism, and implements order task allocation based on the order packaging mechanism and the allocation and transportation mechanism; The scheduling execution module obtains the task execution order of each AGV agent based on key parameters and objective functions, configures the work priority for each AGV agent based on the power of the AGV and the number of its load tasks, and performs conflict-free scheduling based on the work priority of the AGV agent and the task execution order.

Images