Automatic guided vehicle scheduling method and system based on equipment process relationship

Through the combination of graph attention network and genetic algorithm, the AGV scheduling solution is optimized, and the problem of traditional AGV scheduling algorithm lacks predictability and adaptability in complex production environments is solved, and more efficient and flexible AGV scheduling is achieved, which improves the response speed. and resource utilization efficiency.

CN119026829APending Publication Date: 2024-11-26CHINA TELECOM CORP LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410983200.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Traditional AGV scheduling algorithms lack predictability and adaptability in complex production environments, and cannot maximize the operation efficiency of AGV, especially in the initial population quality, optimization of fitness function, adjustment of selection mechanism, cross-operation and improvement of mutation strategies. There are shortcomings in the aspects and it is difficult to respond quickly to changes in market demand.

Method used

By constructing a graph attention network, extracting device characteristics and spatial relationships, constructing initial populations based on attention weights, determining fitness function, selection probability, intersection point and mutation probability, combining genetic algorithms to optimize the AGV scheduling scheme, and improving the predictiveness of scheduling and Adaptability.

Benefits of technology

Improve the predictability and adaptability of the AGV scheduling scheme, optimize the initial population quality, dynamically adjust the fitness function and selection mechanism, enhance cross-operation and mutation strategies, improve scheduling response speed and resource utilization efficiency, and reduce operational costs. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026829A_ABST
    Figure CN119026829A_ABST
Patent Text Reader

Abstract

The invention discloses an automatic guided vehicle scheduling method and system based on an equipment process relationship, and the method comprises the steps: obtaining a graph attention network according to the process relationship of equipment served by an automatic guided vehicle; through a graph attention mechanism, extracting an attention weight among equipment features, a spatial relationship and equipment nodes from a graph attention network; constructing an initial population according to the equipment characteristics and the spatial relationship; determining a first fitness function based on the attention weight; determining a first individual selection probability based on the attention weight; determining an intersection point of the intersection operation based on the attention weight; based on the attention weight, the mutation probability is adjusted; and solving the initial population by adopting a genetic algorithm based on the first fitness function, the first individual selection probability, the intersection point and the mutation probability to obtain a scheduling scheme of the automated guided vehicle. The predictability of the automatic guided vehicle scheduling scheme is improved, the adaptability of the scheme to future changes is improved, and the method can be widely applied to the technical field of computers.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an automatic guided vehicle scheduling method and system based on equipment process relationship. Background Art

[0002] In the field of industrial automation, especially in the scheduling problem of automatic guided vehicles (AGVs), traditional algorithms often start scheduling calculations only after an order is placed, which limits the response speed and flexibility of the scheduling system. Although existing technologies can handle certain scheduling tasks, these methods often fail to maximize the operating efficiency of AGVs in complex production environments due to their lack of predictability and adaptability, especially in terms of initial population quality, optimization of fitness functions, adjustment of selection mechanisms, and improvement of crossover operations and mutation strategies. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to propose an automated guided vehicle scheduling method and system based on equipment-process relationship, aiming to achieve predictability and adaptability of automated guided vehicle scheduling.

[0004] To achieve the above-mentioned purpose, one aspect of an embodiment of the present application proposes an automatic guided vehicle scheduling method based on equipment process relationship, the method comprising:

[0005] According to the process relationship of the equipment served by the automated guided vehicle, a graph attention network is obtained;

[0006] Extracting device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism;

[0007] constructing an initial population according to the device characteristics and the spatial relationship;

[0008] Based on the attention weight, determining a first fitness function; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population;

[0009] Based on the attention weight, determining a first individual selection probability; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population;

[0010] Based on the attention weight, determining a crossover point of the crossover operation;

[0011] Based on the attention weight, adjusting the mutation probability; wherein the mutation probability is used to control the mutation probability of the crossover individual;

[0012] The initial population is solved by using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to obtain a scheduling plan for the automatic guided vehicle.

[0013] In some embodiments, obtaining a graph attention network according to the process relationship of the equipment served by the automated guided vehicle comprises the following steps:

[0014] According to the process relationship of the equipment, the equipment correlation is obtained;

[0015] Determining a graph attention value between devices according to the device correlation; wherein the graph attention value ranges from 0 to 1;

[0016] An adjacency matrix reflecting the device relevance is constructed according to the graph attention values ​​to obtain a graph attention network.

[0017] In some embodiments, constructing the initial population according to the device characteristics and the spatial relationship comprises the following steps:

[0018] Using a word vector model, performing graph embedding processing according to the device features and the spatial relationship, and generating a low-dimensional vector representation of the node;

[0019] Chromosomes are generated according to the low-dimensional word vector representation to obtain an initial population.

[0020] In some embodiments, the step of calculating the first fitness function comprises the following steps:

[0021] Perform weighted processing on the total production time, the production line load balance index, the total equipment standby time and the attention weight to obtain intermediate data;

[0022] The sum of the intermediate data is calculated to obtain a fitness value.

[0023] In some embodiments, determining the first individual selection probability based on the attention weight comprises the following steps:

[0024] Calculate the sum of the attention weights according to the population size and the attention weight of each individual in the population to obtain a weight sum;

[0025] The first individual selection probability is determined based on the ratio of the current individual's attention weight to the total weight.

[0026] In some embodiments, determining the intersection point of the crossover operation based on the attention weight comprises the following steps:

[0027] The node with the highest attention weight among the individuals to be crossed is taken as the crossing point.

[0028] In some embodiments, adjusting the mutation probability based on the attention weight comprises the following steps:

[0029] Determine the position of the node with the lowest attention weight in the individual to be mutated as the mutation position;

[0030] Based on the attention weight of the node, the mutation probability of the mutation position is calculated.

[0031] To achieve the above purpose, another aspect of the embodiment of the present application provides an automatic guided vehicle scheduling system based on equipment process relationship, the system comprising:

[0032] The first module is used to obtain a graph attention network according to the process relationship of the equipment served by the automated guided vehicle;

[0033] The second module is used to extract device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism;

[0034] A third module is used to construct an initial population according to the device characteristics and the spatial relationship;

[0035] A fourth module is used to determine a first fitness function based on the attention weight; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population;

[0036] A fifth module is used to determine a first individual selection probability based on the attention weight; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population;

[0037] A sixth module, configured to determine a crossover point of the crossover operation based on the attention weight;

[0038] A seventh module is used to adjust the mutation probability based on the attention weight; wherein the mutation probability is used to control the mutation probability of the crossover individual;

[0039] The eighth module is used to solve the initial population using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to obtain a scheduling plan for the automatic guided vehicle.

[0040] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above-mentioned method when executing the computer program.

[0041] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0042] The embodiments of the present application include at least the following beneficial effects: The present application provides an automatic guided vehicle scheduling method and system based on equipment process relationships. The scheme determines a graph attention network based on the equipment process relationships, combines the attention weights of the graph attention network with the genetic algorithm, analyzes the complex dependencies between the equipment processes through the graph attention network, and determines the first fitness function, the first individual selection probability, the crossover point and the mutation probability based on the attention weights. This can not only reflect current needs, but also optimize the quality of the population, accurately evaluate population individuals, predict and adapt to future changes, which is conducive to improving the predictability of the automatic guided vehicle scheduling scheme and improving the adaptability of the scheme to future changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide a further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 It is a step diagram of an automatic guided vehicle scheduling method based on equipment process relationship provided in an embodiment of the present application;

[0045] Figure 2 is a flowchart of step S100 provided in an embodiment of the present application;

[0046] Figure 3 is a flowchart of step S300 provided in an embodiment of the present application;

[0047] Figure 4 is a flowchart of step S400 provided in an embodiment of the present application;

[0048] Figure 5 is a flowchart of step S500 provided in an embodiment of the present application;

[0049] Figure 6 is a flowchart of step S600 provided in an embodiment of the present application;

[0050] Figure 7 is a flowchart of step S700 provided in an embodiment of the present application;

[0051] Figure 8 It is a schematic diagram of the improvement of the genetic algorithm provided in the embodiment of the present application;

[0052] Fig. 9 It is a module schematic diagram of an automatic guided vehicle scheduling system based on equipment process relationship provided in an embodiment of the present application;

[0053] Fig.10 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0055] Although the functional modules are divided in the system schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100", "second / S200", etc. in the specification, claims and the above drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0056] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0057] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0058] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0060] In related technologies, with the transformation and upgrading of the manufacturing industry, multi-variety small-batch production and large-scale personalized customization have become the new normal. This production model has brought about complex and changeable logistics needs. Among them, the automatic guided vehicle (AGV) system has become a key technology to solve this challenge due to its high degree of automation and flexibility. The AGV system can effectively adapt to the changing logistics tasks at the production site and improve the efficiency and flexibility of the manufacturing system. Therefore, the role of AGV in modern workshop logistics has become increasingly prominent, and the AGV scheduling problem has also become a hot research topic in the field of production scheduling.

[0061] In this context, the efficient operation of AGV in flexible workshops has become a key factor in improving production efficiency, reducing operating costs, and quickly responding to changes in market demand. Its efficient scheduling system can significantly improve the automation level and flexibility of material handling. However, traditional scheduling algorithms are often difficult to adapt to the complexity and dynamic changes of the production environment, especially in flexible workshops that need to handle multiple tasks and face frequent changes.

[0062] Specifically, in the field of industrial automation, especially in the scheduling problem of automatic guided vehicles (AGVs), traditional algorithms often start scheduling calculations only after the order is issued, which limits the response speed and flexibility of the scheduling system. Although existing technologies can handle certain scheduling tasks, these methods often fail to maximize the operating efficiency of AGVs in complex production environments due to their lack of predictability and adaptability, especially in terms of initial population quality, optimization of fitness functions, adjustment of selection mechanisms, and improvement of crossover operations and mutation strategies. In addition, the demand for AGV scheduling systems in related fields does not only stay on the ability to perform known tasks, but also emphasizes predictability and dynamic adjustment capabilities to adapt to immediate changes in the production process. The market urgently needs an AGV scheduling technology that can schedule in advance, dynamically optimize, and maintain high efficiency and flexibility in various production situations.

[0063] In view of this, an embodiment of the present application provides an automatic guided vehicle scheduling method and system based on equipment process relationship, which can be applied to the wireless communication and terminal fields in the computer field. The scheme combines the attention weight of the graph attention network with the genetic algorithm, analyzes the complex dependencies between the equipment processes through the graph attention network, determines the first fitness function, the first individual selection probability, the crossover point and the mutation probability based on the attention weight, and optimizes the genetic algorithm, so that the automatic guided vehicle scheduling scheme can be predicted when solving it, and improves the adaptability of the scheme to future changes. Overall, it includes: (1) Improving the quality of the initial population: using the ability of GAT (i.e., graph attention network) to analyze and predict the dependencies between the various processes in the factory, and based on these relationships, improve the quality of the initial population of the genetic algorithm, thereby providing a more reasonable initial solution. (2) Optimizing the fitness function: through the deep learning ability of GAT, dynamically adjust the fitness function so that it can more accurately evaluate the performance of various scheduling schemes and ensure that the selected population is most suitable for the current production needs. (3) Dynamically adjust the selection mechanism: according to the node importance and urgency information provided by GAT, dynamically adjust the selection mechanism in the genetic algorithm, and give priority to those individuals that can best improve production efficiency for reproduction. (4) Optimize crossover operations: Use the strong correlation between nodes identified by GAT to guide the crossover operations in the genetic algorithm, ensure that important features can be effectively inherited to the next generation, and increase the convergence speed and effect of the algorithm. (5) Improve mutation strategy: Based on the deep-level inter-process connections output by GAT, improve the mutation strategy to make mutations more targeted and enhance the algorithm's exploration ability and ability to adapt to complex environments.

[0064] An automatic guided vehicle scheduling method based on equipment process relationship provided in an embodiment of the present application relates to the field of computer technology. An automatic guided vehicle scheduling method based on equipment process relationship provided in an embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a vehicle-mounted terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services. The server can also be a node server in a blockchain network; the software can be an application that implements an automatic guided vehicle scheduling method based on equipment process relationship, etc., but is not limited to the above forms.

[0065] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0066] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0067] Figure 1 is an optional flow chart of an automatic guided vehicle scheduling method based on equipment process relationship provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S100 to S800.

[0068] Step S100, obtaining a graph attention network according to the process relationship of the equipment served by the automated guided vehicle.

[0069] Step S200, extracting device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism.

[0070] Step S300: constructing an initial population according to the device characteristics and the spatial relationship.

[0071] Step S400: determining a first fitness function based on the attention weight; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population.

[0072] Step S500, based on the attention weight, determine a first individual selection probability; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population.

[0073] Step S600: determining a crossover point of the crossover operation based on the attention weight.

[0074] Step S700, adjusting the mutation probability based on the attention weight; wherein the mutation probability is used to control the mutation probability of the crossover individual;

[0075] Step S800: Using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to solve the initial population, so as to obtain a scheduling plan for the automatic guided vehicle.

[0076] Steps S100 to S800 shown in the embodiment of the present application combine the attention weights of the graph attention network with the genetic algorithm, analyze the complex dependencies between the equipment processes through the graph attention network, determine the first fitness function, the first individual selection probability, the crossover point and the mutation probability based on the attention weights, and optimize the genetic algorithm, so that the solution of the automatic guided vehicle scheduling plan is predictable, thereby improving the adaptability of the plan to future changes.

[0077] See also Figure 2 In some embodiments, step S100 may include but is not limited to steps S110 to S130:

[0078] Step S110, obtaining equipment correlation according to the process relationship of the equipment.

[0079] Step S120, determining a graph attention value between devices according to the device correlation; wherein the graph attention value ranges from 0 to 1.

[0080] Step S130, constructing an adjacency matrix reflecting the device relevance according to the graph attention value to obtain a graph attention network.

[0081] Through the above steps S110 to S130, the spatiotemporal data mining method and AGV scheduling can be combined, and the connection matrix of the graph attention mechanism network can be optimized through the process relationship between factory equipment, and the genetic algorithm can be improved with this attention mechanism network for AGV scheduling.

[0082] In step S200 of some embodiments, through the graph attention mechanism, device features, spatial relationships, and attention weights between device nodes can be extracted from the graph attention network according to the various values ​​of the adjacency matrix for tuning of subsequent calculation processes.

[0083] See also Figure 3 In some embodiments, step S300 includes but is not limited to the following steps S310 to S320:

[0084] Step S310, using a word vector model, performing graph embedding processing according to the device features and the spatial relationship, and generating a low-dimensional vector representation of the node.

[0085] Step S320, generating chromosomes according to the low-dimensional word vector representation to obtain an initial population.

[0086] The chromosomes in the initial population generated by the above steps S310 to S330 can better reflect the complex relationship and collaboration requirements between devices.

[0087] See also Figure 4 In some embodiments, step S400 includes but is not limited to the following steps S410 to S420:

[0088] Step S410, performing weighted processing according to the total production time, the production line load balance index, the total equipment standby time and the attention weight to obtain intermediate data.

[0089] Step S420, calculating the sum of the intermediate data to obtain a fitness value.

[0090] Through the above steps S410 to S420, the fitness function after the introduction of the graph attention mechanism not only takes into account the production time, load balance and standby time, but also combines the actual dependencies between devices, making the fitness evaluation more comprehensive and accurate.

[0091] See also Figure 5 In some embodiments, step S500 includes but is not limited to the following steps S510 to S520:

[0092] Step S510, calculating the sum of the attention weights according to the population size and the attention weight of each individual in the population to obtain the weight sum.

[0093] Step S520, determining the first individual selection probability based on the ratio of the current individual's attention weight to the total weight.

[0094] Through the above steps S510 to S520, key nodes and important tasks can be more easily selected, thereby improving the overall quality and evolution efficiency.

[0095] See also Figure 6 In some embodiments, step S600 includes but is not limited to the following steps S610:

[0096] Step S610, taking the node with the highest attention weight among the individuals to be crossed as the crossing point.

[0097] Through the above step S610 combined with the image annotation neural network, the highly relevant gene segments can be retained to generate chromosomes with better quality.

[0098] See also Figure 7 In some embodiments, step S700 includes but is not limited to the following steps S710 to S720:

[0099] Step S710, determining the position of the node with the lowest attention weight in the individual to be mutated as the mutation position.

[0100] Step S720: Calculate the mutation probability of the mutation position based on the attention weight of the node.

[0101] Through the above steps S710 to S720, the mutation probability is dynamically adjusted to improve the global search capability of the algorithm while avoiding falling into the local optimum.

[0102] Below, in conjunction with a specific application example of automatic guided vehicle scheduling, the solution of the embodiment of the present application is described in detail and explained:

[0103] In an embodiment of the present application, a method for scheduling an automated guided vehicle based on equipment process relationships is provided, which can be applied to solving the scheduling problem of automated guided vehicles in industrial automation, such as Figure 8 As shown, specifically, by combining the attention weight of the graph attention network with the genetic algorithm, the complex dependencies between the equipment processes are analyzed through the graph attention network, and the first fitness function, the first individual selection probability, the crossover point and the mutation probability are determined based on the attention weight. The genetic algorithm is optimized, so that the solution of the automatic guided vehicle scheduling plan is predictable, which improves the adaptability of the plan to future changes.

[0104] Furthermore, the adjacency matrix is ​​essential information for graph neural networks, because the adjacency matrix must exist when the graph neural network is trained. The adjacency matrix of the embodiment of the present application is set according to the actual production situation of the factory where the AGV is located (for example, the processes of two devices are closely connected, and this method of setting the adjacency matrix according to actual characteristics is called GAT (graph attention network).

[0105] like Figure 8The adjacency matrix in is one example. For example, many industrial products have to pass through a stamping machine before passing through a CNC machine tool. In the graph attention mechanism, their correlation is relatively large, which is 0.8 here, which means that 80% of the products processed by CNC machines need to pass through a stamping machine. So for AGV, this also means that the relationship between the two devices is very close, so genetic evolution must consider these characteristics. For stamped industrial products, because they have already been stamped, they basically no longer need welding, so their graph attention is very little. The graph attention mechanism here is determined according to the experience of the actual production line, mainly determined by the process characteristics, and physical distance is also a consideration.

[0106] Furthermore, the quality of the initial population has an important impact on the overall performance of the genetic algorithm. Using the device features and spatial relationships extracted by GAT, chromosomes with higher initial fitness can be generated, thereby improving the diversity and quality of the initial population. Specifically, Node2Vec is used for graph embedding to generate a low-dimensional vector representation of each node, which is expressed as:

[0107] h i =Node2Vec(A,d);

[0108] Among them, h i is the i-th node; d is the embedding dimension, and a smaller dimension (such as 64 or 128) is usually selected to ensure computational efficiency and embedding quality. By analyzing these embedding vectors, the population can be initialized to make the chromosome more representative and diverse.

[0109] For example, for the adjacency matrix A:

[0110]

[0111] Use Node2Vec to generate node embeddings, and the calculation formula is:

[0112] h i =Node2Vec(A,d);

[0113] Then, the chromosomes of the initial population are generated according to the obtained embedding vector. The specific expression is:

[0114] Chromosome i =AGV j ,Task k ,Time l ];

[0115] Among them, Chromosome i represents the i-th chromosome; AGV j represents the jth scheduling scheme; Task k represents the kth task; Timel Indicates the time point or time period of task execution, and parameter l represents the task k assigned to AGV j The execution time may be a specific time point (such as 10:00AM) or a time period (such as the first working hour). In addition, chromosomes are composed of genes, and each gene represents a scheduling information. For example, a specific gene may contain Chromosome1 = [AGV3, Task5, Time 10:00AM ], indicating that AGV3 will execute Task5 at 10:00AM. Multiple genes together constitute the overall scheduling plan for multiple vehicles, forming chromosomes.

[0116] In this way, the chromosomes in the initial population will better reflect the complex relationships and collaboration needs between devices.

[0117] Furthermore, the fitness function is optimized: the node features and attention weights extracted by GAT can be used to construct a more accurate fitness function. For example, the actual effect of the scheduling scheme can be evaluated by combining the device status, historical data, and dependencies between devices. The optimized fitness function of the embodiment of the present application can be expressed as:

[0118] Fitness=α×T total +β×L balance +γ×T idle +δ×A GAT ;

[0119] Among them, T total is the total production time; L balance is the production line load balance indicator; T idle is the total standby time of the device; A GAT It is the attention weight extracted based on GAT, reflecting the degree of dependence between devices and the priority of task execution; α, β, γ, δ are the weight parameters of each target.

[0120] By introducing the graph attention mechanism, A in the embodiment of the present application GAT ,The fitness function not only considers the production time, load ,balance and standby time, but also combines the actual dependencies among ,devices, making the evaluation more comprehensive and accurate.

[0121] Furthermore, the selection mechanism is adjusted: The selection mechanism is crucial in genetic algorithms, as it determines which individuals can reproduce and pass on their genes to the next generation. Using the attention weights extracted by GAT, the selection mechanism can be dynamically adjusted so that key devices and important tasks are selected first. For example, based on the attention weight distribution, the selection probability P i as follows:

[0122]

[0123] in, represents the attention weight of individual i, and N is the population size. In this way, key nodes and important tasks are more easily selected, which improves the overall quality and evolutionary efficiency of the population.

[0124] Furthermore, the crossover operation is optimized. During the crossover process, GAT can help identify device nodes with high collaboration relationships and guide the crossover operation. By retaining highly correlated gene fragments, better quality offspring chromosomes can be generated.

[0125] Specifically, the crossover operation can be expressed as:

[0126] Offspring1=[Parent1[:C],Parent M [C:]];

[0127] Offspring2=[Parent2[:C],Parent1[C:]];

[0128] Among them, Offspring1 and Offspring2 are both offspring chromosomes; Parent1 and Parent2 represent the parent chromosomes of the crossover operation; C is the location of the crossover point. When performing a crossover, the gene fragments with attention weights are retained first to ensure that the new individual inherits excellent characteristics. The crossover operation involves two chromosomes (parents) exchanging their gene fragments in some way to generate a new chromosome (offspring). Using the attention weights of GAT can help identify the optimal gene fragments that should be crossed.

[0129] In the embodiment of the present application, the intersection point c can be determined based on the following formula:

[0130] c=arg max j∈N(i) α ij

[0131] Among them, α ij represents the attention weight between node i and node j.

[0132] Furthermore, the mutation operation is improved: the mutation operation maintains the diversity and stability of the population by randomly changing certain genes of the chromosome. By using the attention mechanism of GAT, genes with low correlation can be mutated first, thereby maintaining the stability and diversity of the population.

[0133] Mutation operations are usually performed at low probability, and their execution expression is:

[0134] MutatedGene = RandomValue#;

[0135] (Gene i ,Gene j )=(Gene j ,Gene i )#;

[0136] Among them, MutatedGene represents a mutated gene; RandomValue represents a random value; Gene i Indicates the i-th gene; Gene j represents the jth gene.

[0137] By dynamically adjusting the mutation probability, the global search capability of the algorithm is ensured while avoiding falling into local optimality.

[0138] Specifically, the attention weight of GAT is used to determine the location of the mutation. A lower attention weight may indicate that the node is relatively isolated or has low dependency in the current production environment, which makes mutation at these points less risky. Therefore, the mutation location m can be determined by the following expression:

[0139] m=argmin j∈N(i) α ij

[0140] Among them, α ij represents the attention weight between node i and node j.

[0141] This means that nodes with the lowest attention weights are selected for mutation, since such genetic changes will have less impact on the overall system.

[0142] The mutation probability p can be adjusted dynamically based on the importance of the node (determined by the attention weight). The lower the attention weight of the node, the higher the corresponding mutation can be. Based on this, the mutation probability p(j) can be determined according to the following formula:

[0143]

[0144] Such a strategy helps maintain the heritability of high-importance nodes while increasing the adaptability and diversity of the entire system.

[0145] Finally, the embodiment of the present application uses a genetic algorithm based on the above-mentioned optimized fitness function, individual selection probability, crossover point and mutation probability to solve the initial population and obtain a scheduling plan for the automatic guided vehicle (AGV).

[0146] In general, in the embodiments of the present application, GAT is not only a tool for providing node features, but its output (i.e., attention weight) directly affects the key decision-making process of the genetic algorithm, thereby ensuring that the algorithm adapts to complex environments while also enhancing the efficiency and effectiveness of the algorithm in dealing with practical problems. This method combines the powerful functions of GAT with the advantages of traditional genetic algorithms, providing a new way to deal with complex optimization problems. The embodiments of the present application include at least the following beneficial effects:

[0147] 1. Improve the response speed and accuracy of scheduling: Combine the graph attention network (GAT) with the genetic algorithm (GA), use GAT to analyze the complex dependencies between the processes in the factory to optimize various aspects of the genetic algorithm, predict future needs and adjust AGV tasks in advance, which can reduce response time and improve the accuracy and efficiency of scheduling. This combination not only improves the prediction ability of the algorithm, but also enhances the effectiveness of the fitness function and genetic operations, and makes the production process more continuous and seamless, reducing downtime and costs caused by scheduling delays.

[0148] 2. Enhance the adaptability and flexibility of the system: Through the output of GAT, the selection mechanism, crossover operation and mutation strategy in the genetic algorithm are dynamically adjusted, so that the algorithm can not only reflect the current production needs, but also predict and adapt to future changes. This dynamic adjustment mechanism improves the performance of the genetic algorithm in an uncertain environment. The algorithm can be dynamically adjusted according to real-time production data and prediction results to better cope with changes and emergencies in the production process, thereby improving the adaptability and flexibility of the entire manufacturing system.

[0149] 3. Optimize resource utilization and cost efficiency: Through more effective initial population and accurate fitness evaluation, using GAT's powerful data processing capabilities, accurately predict and improve the quality of the initial population, which directly affects the starting efficiency and final results of the genetic algorithm. At the same time, optimize the fitness function according to real-time data to ensure that the genetic algorithm can accurately evaluate and select the best individuals, and the algorithm can converge to the optimal solution faster, reducing unnecessary transportation and waiting, thereby maximizing resource utilization efficiency and reducing operating costs.

[0150] See also Fig. 9 The embodiment of the present application further provides an automatic guided vehicle scheduling system based on equipment process relationship, which can implement the above-mentioned automatic guided vehicle scheduling method based on equipment process relationship, and the system includes:

[0151] The first module 101 is used to obtain a graph attention network according to the process relationship of the equipment served by the automated guided vehicle;

[0152] The second module 102 is used to extract device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism;

[0153] The third module 103 is used to construct an initial population according to the device characteristics and the spatial relationship;

[0154] The fourth module 104 is used to determine a first fitness function based on the attention weight; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population;

[0155] A fifth module 105 is used to determine a first individual selection probability based on the attention weight; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population;

[0156] A sixth module 106, configured to determine a crossover point of the crossover operation based on the attention weight;

[0157] The seventh module 107 is used to adjust the mutation probability based on the attention weight; wherein the mutation probability is used to control the mutation probability of the crossover individual;

[0158] The eighth module 108 is used to solve the initial population using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to obtain a scheduling plan for the automatic guided vehicle.

[0159] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0160] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned automatic guided vehicle scheduling method based on equipment process relationship when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0161] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0162] See also Fig.10 , Fig.10 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0163] The processor 201 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0164] The memory 202 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 202 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 202, and the processor 201 calls and executes an automatic guided vehicle scheduling method based on equipment process relationship in the embodiment of this application;

[0165] Input / output interface 203, used to implement information input and output;

[0166] The communication interface 204 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0167] Bus 205 , which transmits information between various components of the device (e.g., processor 201 , memory 202 , input / output interface 203 , and communication interface 204 );

[0168] The processor 201 , the memory 202 , the input / output interface 203 and the communication interface 204 are connected to each other in communication within the device via the bus 205 .

[0169] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned automatic guided vehicle scheduling method based on equipment process relationship.

[0170] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0171] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0172] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0173] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0174] The system embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0175] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0176] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0177] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0178] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0179] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0180] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0181] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0182] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. An automated guided vehicle scheduling method based on equipment process relationship, characterized in that: The following steps are involved: According to the process relationship of the equipment served by the automated guided vehicle, a graph attention network is obtained; Extracting device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism; constructing an initial population according to the device characteristics and the spatial relationship; Based on the attention weight, determining a first fitness function; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population; Based on the attention weight, determining a first individual selection probability; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population; Based on the attention weight, determining a crossover point of the crossover operation; Based on the attention weight, adjusting the mutation probability; wherein the mutation probability is used to control the mutation probability of the crossover individual; The initial population is solved by using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to obtain a scheduling plan for the automatic guided vehicle.

2. The method according to claim 1, characterized in that The step of obtaining a graph attention network according to the process relationship of the equipment served by the automated guided vehicle comprises the following steps: According to the process relationship of the equipment, the equipment correlation is obtained; Determining a graph attention value between devices according to the device correlation; wherein the graph attention value ranges from 0 to 1; An adjacency matrix reflecting the device relevance is constructed according to the graph attention values ​​to obtain a graph attention network.

3. The method according to claim 1, characterized in that The step of constructing an initial population according to the device characteristics and the spatial relationship comprises the following steps: Using a word vector model, performing graph embedding processing according to the device features and the spatial relationship, and generating a low-dimensional vector representation of the node; Chromosomes are generated according to the low-dimensional word vector representation to obtain an initial population.

4. The method according to claim 1, characterized in that: The step of calculating the first fitness function comprises the following steps: Perform weighted processing on the total production time, the production line load balance index, the total equipment standby time and the attention weight to obtain intermediate data; The sum of the intermediate data is calculated to obtain a fitness value.

5. The method according to claim 1, characterized in that Determining the first individual selection probability based on the attention weight comprises the following steps: Calculate the sum of the attention weights according to the population size and the attention weight of each individual in the population to obtain a weight sum; The first individual selection probability is determined based on the ratio of the current individual's attention weight to the total weight.

6. The method according to claim 1, characterized in that Determining the intersection point of the crossover operation based on the attention weight comprises the following steps: The node with the highest attention weight among the individuals to be crossed is taken as the crossing point.

7. The method according to claim 1, characterized in that The step of adjusting the mutation probability based on the attention weight comprises the following steps: Determine the position of the node with the lowest attention weight in the individual to be mutated as the mutation position; Based on the attention weight of the node, the mutation probability of the mutation position is calculated.

8. An automated guided vehicle scheduling system based on equipment process relationship, characterized in that: include: The first module is used to obtain a graph attention network according to the process relationship of the equipment served by the automated guided vehicle; The second module is used to extract device features, spatial relationships, and attention weights between device nodes from the graph attention network through a graph attention mechanism; A third module is used to construct an initial population according to the device characteristics and the spatial relationship; A fourth module is used to determine a first fitness function based on the attention weight; wherein the first fitness function is used to evaluate the fitness value of each individual in the initial population; A fifth module is used to determine a first individual selection probability based on the attention weight; wherein the first individual selection probability is used to determine an individual to be crossovered for a crossover operation from the initial population; A sixth module, configured to determine a crossover point of the crossover operation based on the attention weight; A seventh module is used to adjust the mutation probability based on the attention weight; wherein the mutation probability is used to control the mutation probability of the crossover individual; The eighth module is used to solve the initial population using a genetic algorithm based on the first fitness function, the first individual selection probability, the crossover point and the mutation probability to obtain a scheduling plan for the automatic guided vehicle.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 7.

10. A computer storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the method according to any one of claims 1 to 7 when executed by the processor.