CN114493181B - Multi-load AGV task scheduling method in intelligent storage environment - Google Patents
Multi-load AGV task scheduling method in intelligent storage environment Download PDFInfo
- Publication number
- CN114493181B CN114493181B CN202210006556.6A CN202210006556A CN114493181B CN 114493181 B CN114493181 B CN 114493181B CN 202210006556 A CN202210006556 A CN 202210006556A CN 114493181 B CN114493181 B CN 114493181B
- Authority
- CN
- China
- Prior art keywords
- individuals
- agv
- individual
- population
- parent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003860 storage Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000035772 mutation Effects 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 210000000349 chromosome Anatomy 0.000 claims description 86
- 108090000623 proteins and genes Proteins 0.000 claims description 56
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 85
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 101150076211 TH gene Proteins 0.000 description 2
- 230000002860 competitive effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 208000036142 Viral infection Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000002759 chromosomal effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009385 viral infection Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06316—Sequencing of tasks or work
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Biophysics (AREA)
- Operations Research (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Accounting & Taxation (AREA)
- General Health & Medical Sciences (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Finance (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Warehouses Or Storage Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-load AGV task scheduling method in an intelligent storage environment, which comprises the following steps: acquiring warehouse-in and warehouse-out information; initializing a population to obtain an initial population with N individuals; performing crossover and mutation operations on individuals in the initial population to generate offspring individuals and combining the offspring individuals with parent individuals to form a second population with 2N individuals; dividing individuals in the second population into different levels of dominable layers according to a preset fitness function; carrying out crowding degree calculation on all individuals in the Mth grade dominable layer of the second population to obtain crowding distance; the individuals with large crowding distance are reserved in the current dominating layer, and a third population with N individuals is formed by the individuals in the previous M-1 level dominating layer; and performing multiple iterations, and determining an optimal cargo scheduling scheme according to an iteration result. The method and the system are beneficial to solving task scheduling of the multi-load AGVs in the intelligent storage field, obtaining reasonable AGVs and AGVs taking order, and improving intelligent storage performance.
Description
Technical Field
The invention belongs to the technical field of intelligent storage, and particularly relates to a multi-load AGV task scheduling method in an intelligent storage environment.
Background
Automated access systems (AS/RS, automated Storage AND RETRIEVAL SYSTEM) based on AGVs (Automated Guided Vehicle, automated guided vehicles) have become an effective and competitive solution for suppliers and distributors, and AGVs have been widely used to perform storage or retrieval tasks. Specifically, an AGV designated according to the system schedule starts from the home position, reaches a designated station according to a planned route, provides operations such as dispatch or picking, and finally returns to a warehouse position or a preset position. In the competitive intelligent manufacturing industry, AGVs can provide end-to-end warehousing services with better control and execution capabilities, and have the advantages of efficient storage and retrieval performance, low error rate, low labor cost, etc., using AGVs instead of personnel to transport goods in warehouses means reducing the risk of viral infection due to the transportation of goods.
With the development of AGV-based AS/RSs, two types of AGVs have been used in AS/RSs, namely single-load AGVs and multi-load AGVs. A single load AGV can only carry one load or SKU at a time, while a multi-load AGV can pick up multiple different loads from one or more stations. The performance comparison of single-load AGVs and multi-load AGVs is further discussed, and through analysis of experimental results, it is obviously observed that the application of multi-load AGVs can significantly reduce the necessary number of AGVs and the related congestion, and improve the effectiveness of the system. Thus, in the face of increasing AS/RS size and increasing system transport efficiency requirements, the use of multi-load AGVs has become a necessary trend.
However, the optimization of the scheduling of multi-load AGVs is far from fully understood, and many interrelated elements are simplified or even omitted, for example. Route collisions, number of AGVs, and load decisions per AGV, etc., which can have a greater impact on system performance, energy consumption, economic costs, etc.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a multi-load AGV task scheduling method in an intelligent storage environment. The technical problems to be solved by the invention are realized by the following technical scheme:
The invention provides a multi-load AGV task scheduling method in an intelligent storage environment, which comprises the following steps:
s1: the method comprises the steps of obtaining warehouse-in and warehouse-out information, wherein the warehouse-in and warehouse-out information at least comprises the number of cargoes to be transported, the number of multi-load AGVs and the goods shelf position;
S2: initializing a population according to the warehouse-in and warehouse-out information, and constructing an initial population with N individuals;
S3: performing crossover and mutation operations by taking individuals in the initial population as parent individuals to generate child individuals and combining the child individuals with the parent individuals to form a second population with 2N individuals;
S4: dividing individuals in the second population into different levels of dominable layers according to a preset fitness function;
S5: carrying out crowding degree calculation on all individuals in an Mth-level dominable layer of a second population to obtain crowding distance of each individual, wherein Nth individuals classified according to different levels in the second population are positioned in the Mth-level dominable layer;
S6: maintaining individuals meeting the requirements of crowding distances in the current dominating layer, and forming a third population with N individuals with the individuals in the previous M-1 level dominating layer;
S7: repeating the steps S2-S6 for a plurality of iterations by using the third population, and determining an optimal cargo scheduling scheme according to an iteration result when the iteration number reaches a preset iteration number.
In one embodiment of the present invention, the S2 includes:
Setting the number of individuals in an initial population, wherein the genes of each individual comprise an AGV number A k and a goods number W j which are randomly generated, and the individuals in the initial population meet the following four initial constraint conditions:
Where K represents the maximum available number of current multi-load AGVs, M represents the total number of loads to be handled, Indicating whether the load W j is to be carried by the AGV a k;
Wherein, Indicating whether the AGV a k is carrying the load W j after carrying the load W i,
Where L represents the maximum load of each multi-load AGV,
In one embodiment of the present invention, the S3 includes:
S31: the individuals in the initial population are used as parent individuals to perform cross operation in any pair, and offspring individuals with the same number as the parent individuals are obtained;
S32: carrying out mutation operation on the offspring individuals subjected to the cross operation to obtain mutated offspring chromosomes;
S33: and merging the parent individuals and the variant child individuals to form a second population with the number of individuals twice that of the initial population.
In one embodiment of the present invention, the S31 includes:
S311: randomly selecting two individuals from the initial population as a first parent individual and a second parent individual, wherein the genes are divided into a cargo number W j and an AGV number A k, randomly selecting one cargo number or the AGV number from the first parent individual as a first parent designated gene, copying all genes before the first parent designated gene to the corresponding position of a first offspring chromosome, inheriting the cargo numbers of the rest of the first offspring chromosome according to the cargo number sequence of the second parent individual, deleting the cargo numbers already existing in the first offspring chromosome, randomly selecting one cargo number from the first parent chromosome and the AGV number interval in the second parent chromosome, randomly inserting the cargo numbers into the first offspring chromosome, and reserving the individuals meeting the initial constraint condition to form the first offspring chromosome;
S312: randomly selecting one gene from the second parent individuals as a second parent designated gene, copying all genes before the second parent designated gene to the corresponding positions of second offspring chromosomes, inheriting the rest cargo numbers in the second offspring chromosomes according to the gene sequence of the first parent individuals and deleting the cargo numbers existing in the second offspring chromosomes, wherein the number of AGV numbers of the second offspring chromosomes is a number randomly selected in a number interval of AGV numbers in the first parent chromosomes and the second parent chromosomes, randomly inserting the AGV numbers into the cargo numbers, and reserving individuals meeting the initial constraint conditions so as to form the second offspring chromosomes;
s313: and repeating the steps S311-S312, and intersecting the rest parent individuals according to the preset intersecting probability, so as to obtain child individuals with the same number as the parent individuals.
In one embodiment of the present invention, the S32 includes:
Selecting offspring individuals after the crossover operation, carrying out position exchange on different genes to carry out mutation operation, then checking whether the exchanged individuals meet the initial constraint condition, if so, reserving the offspring individuals after mutation, and if not, discarding and carrying out mutation operation again.
In one embodiment of the present invention, the S4 includes:
S41: respectively setting a first fitness function, a second fitness function and a third fitness function according to the number of AGVs, the maximum running time of the AGVs and the conflict among the AGVs;
s42: and sorting the individuals in the middle population by using a rapid non-dominant sorting algorithm according to the first fitness function, the second fitness function and the third fitness function, and dividing the individuals in the middle population into the dominant layers with different grades.
In one embodiment of the present invention, the first fitness function is targeted to optimize the number of AGVs, expressed as:
wherein A k represents the kth AGV, N (A k) represents whether the kth AGV is selected to transport the load, K represents the maximum number of currently selectable AGVs;
the second fitness function objective is to minimize the maximum run time of the AGV, expressed as:
F2＝min(max(G(Ak)))，k＝1,…,K
Where A k represents the kth AGV, G (A k) represents the total time for the kth AGV to transport the load, and max (G (A k)) represents the maximum load time for a single AGV from among all AGVs.
The third fitness function is to minimize collisions between different AGVs, expressed as:
Wherein A i represents the ith AGV, A j represents the jth AGV, Indicating the number of collisions between the ith and jth AGVs.
In one embodiment of the present invention, the S5 includes:
Sequencing from large to small according to a first function value of each individual in a current dominable layer, setting an initial value of a crowding distance of an individual with the smallest first function value and an individual with the largest first function value to infinity, and setting initial values of crowding distances of other individuals in the dominable layer to zero;
calculating a first crowding distance of each individual according to the initial value of the crowding distance and the first function value of each individual;
sorting the second function value of each individual in the current dominance layer from big to small;
Calculating a second crowding distance for each individual based on the first crowding distance and the second function value for each individual;
sorting the third function value of each individual in the current dominance layer from big to small;
and calculating a third crowding distance of each individual according to the second crowding distance and the third function value of each individual, and determining the third crowding distance as the crowding distance of the individual.
In one embodiment of the present invention, the first congestion distance is calculated according to the following formula:
n1＝n0+(f1(i+1)-f1(i-1))/(f1 max-f1 min)
Wherein, for each rank of the dominatable layer, n 0 represents an initial value of the congestion distance of the individual, f 1 (i+1) represents a first function value of the individuals ranked i+1, f 1 (i-1) represents a first function value of the individuals ranked i-1, f 1 max represents a maximum value of the first function value, and f 1 min represents a minimum value of the first function value;
The calculation formula of the second crowding distance is as follows:
Wherein, for each rank of the dominatable layer, n 1 represents a first crowding distance for the individual, f 2 (i+1) represents a second function value for the i+1th individual, f 2 (i-1) represents a second function value for the i-1th individual, Representing the maximum value of the second function value,/>Representing a minimum value of the second function value;
The calculation formula of the third crowding distance is as follows:
wherein, for each rank of the dominatable layer, n 2 represents the second crowding distance of the individual, f 3 (i+1) represents the third function value for the i+1th individual, f 3 (i-1) represents the third function value for the i-1th individual, Representing the maximum value of the third function value,/>Representing the minimum of the third function value.
Another aspect of the present invention provides a storage medium having stored therein a computer program for executing the steps of the multi-load AGV task scheduling method in the smart warehouse environment as described in any one of the above embodiments.
Compared with the prior art, the invention has the beneficial effects that:
1. According to the multi-load AGV task scheduling method in the intelligent storage environment, the number of the multi-load AGVs which are not considered in the past is taken into an algorithm, after the warehouse-out information is obtained, the initial population is obtained by initializing according to the warehouse-out information, and the individuals in the initial population are divided into the dominable layers of different grades through the preset fitness function, so that the task scheduling of the multi-load AGVs in the intelligent storage field is favorably solved, the reasonable AGV number and the AGV commodity-taking sequence are calculated, and the intelligent storage performance is improved.
2. According to the multi-load AGV task scheduling method in the intelligent storage environment, the AGVs can collide with the least AGVs, so that the AGVs can take goods in a reduced time, and the economic benefit of intelligent storage is prompted.
3. According to the multi-load AGV task scheduling method, an allocation strategy of the AGVs can be designed under the condition that the number of the AGVs is insufficient, and the intelligent storage efficiency is improved as much as possible; the use efficiency of the AGV can be estimated, the idle load condition of the AGV is analyzed, and theoretical support is provided for improving the storage efficiency of the AGV.
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Drawings
FIG. 1 is a flowchart of a multi-load AGV task scheduling method in an intelligent warehouse environment provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of a model of an intelligent warehouse according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a crossover operation according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and the effects adopted by the invention to achieve the preset aim, the following describes in detail an AGV task scheduling method in an intelligent storage environment according to the invention with reference to the attached drawings and the specific embodiments.
The foregoing and other features, aspects, and advantages of the present invention will become more apparent from the following detailed description of the preferred embodiments when taken in conjunction with the accompanying drawings. The technical means and effects adopted by the present invention to achieve the intended purpose can be more deeply and specifically understood through the description of the specific embodiments, however, the attached drawings are provided for reference and description only, and are not intended to limit the technical scheme of the present invention.
It should be noted that in this document relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a good or apparatus that comprises a list of elements does not include only those elements, but may include other elements not expressly listed. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a good or apparatus comprising the element.
Referring to fig. 1, fig. 1 is a flowchart of a multi-load AGV task scheduling method in an intelligent warehouse environment according to an embodiment of the present invention. The multi-load AGV task scheduling method in the intelligent storage environment comprises the following steps:
s1: and acquiring warehouse-in information, wherein the warehouse-in information at least comprises the number of cargoes to be carried, the number of multi-load AGVs and the goods shelf position.
As shown in fig. 2, the intelligent warehouse of the present embodiment is composed of components such as a shelf, an AGV, a warehouse entry platform, and a conveyor belt. The warehouse management system monitors the running state of each component and coordinates the picking sequence of each AGV according to the ex-warehouse task list. The intelligent warehouse of this embodiment includes two columns of shelves, each column including 10 shelves, and a total of 50 multi-load AGVs. The AGV sends to the appointed goods shelf to pick up goods according to the system dispatching instruction, and after picking up goods, the AGV sends to the next place to pick up goods, and after all picking up goods, the AGV returns to the sorting table.
First, the present embodiment mathematically models task scheduling optimization to demonstrate performance metrics and optimization goals. Specifically, output and Input represent Input/Output stations of different work centers, which are destinations to which the AGVs deliver all the loads (cargoes), and both Output and Input can be either Input or Output stations. S i denotes the i-th shelf, s= { S i |i=0, 1. A k denotes a kth multi-load AGV, a= { a k |k=1, 2,..k } denotes a set of all multi-load AGVs. v denotes the speed of the AGV, in this embodiment, all the overloaded AGVs have the same speed. L is the constrained load (maximum load) of each multi-load AGV in the system. W j represents the j-th cargo to be transported, w= { W j |j=1, 2,..m } represents a set of cargoes to be transported. M represents the total number of loads that need to be handled and K represents the maximum number of available current multi-load AGVs. N (a k) = {0,1} (k=1, 2., K) indicates whether or not the AGV a k is selected to transport the load, a value of 1 indicates selection, and a value of 0 indicates non-selection.Is the distance between shelf S i and shelf S j. T (A k) represents the time that AGV A k is transporting the load. t ij represents the time each AGV transfers a load from the shelf S i to the shelf S j. /(I)Indicating the number of collisions that AGV A i and AGV A j have with during transport. G (a k) represents the time AGVA k to complete all given tasks. /(I)Indicating the current load of the AGV a k.
Indicating whether or not the AGV A k conveys the load W j,/>, after conveying the load W i Indicates AgvA k whether or not to convey the cargo W j,/>, after conveying the cargo W i Then it indicates AgvA k that the cargo W j is not to be transported after the cargo W i is transported.Indicating whether or not the load W j is to be carried by the AGV A k/(Representing the transport of cargo W j by AGV A k,/>It indicates that the load W j is not being performed by the AGV a k. /(I)The time required for the AGV to travel from the shelf where the load W i is located to the shelf where the load W j is located is shown. G (a k) represents the time at which the AGV a k delivers all the loads. R ij is the path solution of shelf S i to shelf S j and (x i,yi) is the coordinates of shelf S j. Wherein the following relationship exists:
S2: and initializing the population according to the warehouse-in and warehouse-out information, and constructing an initial population with N individuals.
First, the number of individuals in the initial population is set, and the genes in each individual include a randomly generated AGV number a k and a cargo number W j, and each bracket represents a pick-up sequence of an AGV:
[(Wa,Wb,…,Wc,A1),(Wd,…,We,A2),(…)]
Wherein in this individual chromosome W j represents the j-th cargo to be transported and A i represents the i-th AGV.
Illustratively, the following is a chromosome sequence of one individual randomly generated:
[W3,W1,W10,A1,W9,W4,W2,W5,W6,A2,A3,W7,W8,A4]
Wherein, given 10 tasks (i.e. the acquisition to be carried) are numbered 1-10, the goods to be carried are W 1-W10 respectively, and given the maximum number of AGVs is 4, these four AGVs are denoted by A 1,A2,A3,A4. This chromosomal sequence represents the selection of three AGVs (1, 2, 4) to transport the load, the order of the AGVs "1" is 3,1, 10, the AGVs "2" is 9, 4, 2, 5, 6, the AGVs "2" is routed from the origin to the 7 th, 8 th positions to the point of transport.
Further, each individual chromosome in the initial population also meets the following limitations:
First, in a batch task, each load can only be transported once by an AGV, i.e., multiple AGVs cannot transport the same load, thus satisfying (1):
Where K represents the maximum available number of current multi-load AGVs and M represents the total number of loads that need to be handled.
Each multi-load AGV can pick up multiple loads at a time, after picking up one load, only one load is needed to be picked up next, i.e. after picking up a certain load, the next destination of the AGV is determined and independent, so that (2) needs to be satisfied:
The number of picks per AGV run must not exceed its maximum load and therefore (3) must be satisfied:
in addition, in one batch of tasks, the total number of loads transported by all AGVs is equal to the total number of loads to be transported, so that (4) needs to be satisfied:
Therefore, according to the constraint conditions, the individuals which do not meet the conditions are removed, and the initial population can be obtained.
S3: and performing crossover and mutation operations by taking the individuals in the initial population as parent individuals to generate child individuals and combining the child individuals with the parent individuals to form a second population with 2N individuals.
In this embodiment, the S3 includes:
S31: and carrying out cross operation on the individuals in the initial population as parent individuals in any pair to obtain offspring individuals with the same number as the parent individuals.
Referring to fig. 3, fig. 3 is a schematic diagram of a cross operation provided in an embodiment of the present invention, and S31 in this embodiment includes:
S311: two individuals are randomly selected from the initial population to serve as a first parent individual and a second parent individual, the genes are divided into a cargo number W j and an AGV number A k, one cargo number or the AGV number is randomly selected from the first parent individual to serve as a first parent designated gene, all genes before the first parent designated gene are copied to corresponding positions of a first offspring chromosome, the cargo numbers of the rest of the first offspring chromosome are inherited according to the cargo number sequence of the second parent individual, the cargo numbers of the first offspring chromosome are deleted, the number of the AGV numbers of the first offspring chromosome is a number randomly selected in a number interval of the AGV numbers in the first parent chromosome and the second parent chromosome, the AGV numbers are randomly inserted into the cargo numbers, and the individuals meeting initial constraint conditions are reserved, so that the first offspring chromosome is formed.
S312: randomly selecting one gene from the second parent individuals as a second parent designated gene, copying all genes before the second parent designated gene to the corresponding positions of second offspring chromosomes, inheriting the rest cargo numbers in the second offspring chromosomes according to the gene sequence of the first parent individuals and deleting the cargo numbers existing in the second offspring chromosomes, wherein the number of AGV numbers of the second offspring chromosomes is a number randomly selected in a number interval of AGV numbers in the first parent chromosomes and the second parent chromosomes, randomly inserting the AGV numbers into the cargo numbers, and reserving individuals meeting initial constraint conditions, thereby forming the second offspring chromosomes.
S313: and repeating the steps S311-S312, and intersecting the rest parent individuals, so as to obtain offspring individuals with the same number as that of the parent individuals.
Specifically, since a task can only be completed by one AGV, i.e., a cargo can only be carried by one AGV, the AGV cannot repeatedly complete the same task, and thus the parent chromosome cannot be hybridized by the conventional method. This example proposes a new crossover method, i.e. randomly selecting one gene from one parent chromosome (including W i and A j), where all genes preceding the selected gene of the parent chromosome are inherited by the child chromosome, while another part of the inherited chromosome of the child chromosome is inherited by the second parent chromosome.
Illustratively, a specific crossover process is as follows:
Two parent chromosomes are randomly selected in the original population:
P1:W3,W1,A1,W9,W4,W2,W5,W6,A2,W7,W8,A3
P2:W9,W7,A1,W5,W3,W6,W8,A2,W1,W4,W2,A3
the two child chromosomes obtained from the two parent chromosomes are:
C1:W3,W1,A1,W9,W7,W2,W5,W6,A2,W8,W4,A3
C2:W9,W7,A1,W5,W3,W1,W4,A2,W2,W6,W8,A3
For the first parent chromosome P1, it is assumed that the 4 th gene W 9 is selected, the 1-4 genes of the child chromosome C1 are identical to those of the 1-4 genes of the first parent chromosome P1, and the other parts of the child chromosome C1 are inherited in the order of the second parent chromosome P2 and the existing gene W 3,W1,W9 is deleted. The total number of genes A j in the daughter chromosome C1 is equal to the number of either of the two parent chromosomes. In this example, the number of genes a j in the daughter chromosome C1 was three, after which genes a 2 and a 3 were randomly inserted into the genes after W 9, a 2 was randomly selected to carry 5 items, a 3 was carried 2 items, and the daughter chromosome C1 was generated.
Similarly, for the second parent chromosome P2, it is assumed that the 5 th gene W 3 is selected, the 1-5 genes of the child chromosome C2 are identical to those of the second parent chromosome P2, and the other parts of the child chromosome C2 inherit and delete the already existing genes in the order of the parent chromosome P1. The total number of genes A j in the daughter chromosome C2 is equal to the number of either of the two parent chromosomes. In this example, the number of genes a j in the daughter chromosome C2 was three, after which genes a 2 and a 3 were randomly inserted into the genes after W 9, generating the daughter chromosome C2. Two child chromosomes are obtained according to the procedure described above, i.e., from the two parent chromosomes in the original population.
It should be noted that the main inheritance in this example is the order of W i (except for the genes already present in the first parent chromosome) in the second parent chromosome, followed by random insertion of gene A j (the number of tasks between genes A i and A j is less than the load limit, and the last gene must be A j). In this example, the crossover probability is set to 0.8, i.e., crossover operation is performed on 80% of the parent chromosomes in the original population to obtain offspring chromosomes, and the genes of the remaining 20% of the parent chromosomes remain in the offspring chromosomes.
S32: and carrying out mutation operation on the offspring individuals subjected to the cross operation to obtain mutated offspring chromosomes.
Selecting offspring individuals after the crossover operation, carrying out position exchange on different genes to carry out mutation operation, then checking whether the exchanged individuals meet the initial constraint condition, if so, reserving the offspring individuals after mutation, and if not, discarding and carrying out mutation operation again.
In particular, since each task can be performed only once, i.e. each cargo can be handled only once, two nodes can be selected for exchange. Different genes W i and A i can be exchanged, and different genes A i and W i can be exchanged, after mutation is finished, whether the mutated individual meets the initial constraint condition or not needs to be checked, if yes, the mutation operation is reserved, and if not, the mutation operation is abandoned and carried out again. The variation strategy of the embodiment for the number of AGVs and the task sequence is as follows:
P1＝Wa,Wb,Wc,A1,Wd,We,A2,Wf,Wg,Wh,…
C1＝Wa,Wb,Wc,A1,Wd,Wg,A2,Wf,We,Wh,…
In the example, W e and W g are exchanged. In this embodiment, the mutation probability is set to 0.1, that is, mutation operation is performed on 10% of the parent chromosomes in the updated population, and variant offspring chromosomes are obtained.
S33: and merging the parent individuals and the variant child individuals to form a second population with the number of individuals twice that of the initial population.
In this embodiment, the original population is assumed to include N individuals, and the individuals in the original population undergo cross mutation to form N sub-individuals, thereby forming a second population having 2N individuals.
S4: and dividing the individuals in the second population into different levels of dominative layers according to a preset fitness function.
Specifically, the S4 includes:
S41: and respectively setting a first fitness function, a second fitness function and a third fitness function according to the number of AGVs, the maximum running time of the AGVs and the conflict among the AGVs.
First, a fitness function is set according to the number of AGVs, the maximum running time of the AGVs and the collision among the AGVs, wherein the fitness function comprises a first fitness function, a second fitness function and a third fitness function. The first fitness function aims at optimizing the number of AGVs and is expressed as:
Where A k represents the kth AGV, N (A k) represents whether the kth AGV is selected to transport the load, and K represents the maximum number of AGVs currently selectable.
The second fitness function objective is to minimize the maximum run time of the AGV, expressed as:
F2＝min(max(G(Ak)))，k＝1,…,K
Where A k represents the kth AGV, G (A k) represents the total time for the kth AGV to transport the load, and max (G (A k)) represents the maximum load time for a single AGV from among all AGVs.
The third fitness function is to minimize collisions between different AGVs, expressed as:
Wherein A i represents the ith AGV, A j represents the jth AGV, Indicating the number of collisions between the ith and jth AGVs.
S42: and sorting the individuals in the middle population by using a rapid non-dominant sorting algorithm according to the first fitness function, the second fitness function and the third fitness function, and dividing the individuals in the middle population into the dominant layers with different grades.
Determining a first function value of each individual, namely the value of the first fitness function, according to the ex-warehouse information and the first fitness function; determining a second function value of each individual according to the ex-warehouse information and the second fitness function; determining a third function value of each individual according to the ex-warehouse information and the third fitness function; and sorting the individuals in the initial population by using a first function value, a second function value and a third function value of each individual by using a rapid non-dominant sorting algorithm, and dividing all the individuals in the second population into the dominant layers with different grades. Specifically, the first function value, the second function value and the third function value of each individual in the second population are compared with the first function value, the second function value and the third function value of other individuals to respectively determine a first parameter and a second parameter of each individual, wherein the first parameter is a dominant individual set, and the second parameter is the number of individuals which dominant each individual in the population.
The specific steps of layering the population by the rapid non-dominant ranking algorithm are as follows:
Let i=1, for all j=1, 2,..n, and j+.i, compare the dominant versus non-dominant relationship between individuals X i and X j, if X i is better than X j at each objective function value, X i dominates X j, if for any X j there is no X j better than X i, then X i is said to be an non-dominant individual, repeat the above operations, find all non-dominant individuals, and constitute the first level non-dominant layer of the population. And then, ignoring the individual in the first-stage non-dominated layer, and repeating the steps in the rest individuals to obtain a second-stage non-dominated layer. The second non-dominant layer is ignored again, and the steps are repeated, so that the whole population is layered.
In this embodiment, two parameters n p and S p are calculated for each individual p in the second population, where n p is the number of individuals in the population that predominate the individual p and S p is the set of individuals in the population that predominate the individual p. Finding out all individuals with n p =0 in the population, storing in a set F 1 as a first-stage non-dominant layer, forming a set F 2 by the rest of population individuals, calculating parameters n p and S p of each individual p in the rest of individuals, obtaining a second-stage non-dominant layer, and the like.
For each individual i in set F i, whose dominant set of individuals is S i, each individual i in S i is traversed, n l＝nl -1 is performed, and individual i is saved in set F i+1 if n l = 0. The individual obtained in F i is the individual of a non-dominant layer, and F i+1 is taken as the set to be operated next time, and the operation is repeated until the whole population is classified.
In this embodiment, first two parameters n p and S p of each individual p in the second population are calculated, where n p is the number of individuals in the population that are dominant by the individual p, and S p is the set of individuals in the population that are dominant by the individual p. All individuals in the population with n p =0 were found and saved in set F 1 as the first level non-dominant layer. Subsequently, the individuals in the first-stage non-dominant layer are ignored, and the parameters n p and S p of each individual p are calculated in the remaining individuals, resulting in a second-stage non-dominant layer. The second non-dominant layer is ignored again, and the steps are repeated, so that the whole population is layered.
S5: and carrying out crowding degree calculation on all individuals in the second population preset dominable layer to obtain crowding distance of each individual.
In step S4, the 2N individuals in the second population are divided into different dominatable layers, each dominatable layer may include a different number of individuals, and the dominatable layers where the nth individual is located are obtained in the order of the rank of the dominatable layers from low to high.
Next, determining the crowding distance of all individuals in the dominable layer where the nth individual is located, specifically includes:
Sequencing from large to small according to a first function value of each individual in a current dominable layer, setting an initial value of a crowding distance of an individual with the smallest first function value and an individual with the largest first function value to infinity, and setting initial values of crowding distances of other individuals in the dominable layer to zero;
calculating a first crowding distance for each individual based on the initial value of the crowding distance and the first function value for each individual:
n1＝n0+(f1(i+1)-f1(i-1))/(f1 max-f1 min)
Wherein n 0 represents an initial value of the crowding distance of the individual, f 1 (i+1) represents a first function value of the i+1th individual, f 1 (i-1) represents a first function value of the i-1 th individual, f 1 max represents a maximum value of the first function value, and f 1 min represents a minimum value of the first function value.
Sorting the second function value of each individual in the current dominance layer from big to small;
calculating a second crowding distance for each individual based on the first crowding distance and the second function value for each individual:
Wherein n 1 represents a first crowding distance of the individual, f 2 (i+1) represents a second function value of the individual ranked in the (i+1) th, f 2 (i-1) represents a second function value of the individual ranked in the (i-1) th, Representing the maximum value of the second function value,/>Representing the minimum of the second function value.
Sorting the third function value of each individual in the current dominance layer from big to small;
Calculating a third crowding distance of each individual according to the second crowding distance and a third function value of each individual, and determining the third crowding distance as the crowding distance of the individual, wherein the third crowding distance is:
wherein n 2 represents the second crowding distance of the individual, f 3 (i+1) represents the third function value of the individual ranked in the (i+1) th, f 3 (i-1) represents the third function value of the individual ranked in the (i-1) th, Representing the maximum value of the third function value,/>Representing the minimum of the third function value.
S6: individuals whose crowding distance meets the requirement remain in the current dominance layer and form a third population of N individuals with the individuals in the previous M-1 rank dominance layer.
Specifically, all individuals in the dominable layer where the nth individual is located are rearranged in order of from large to small crowded distance, and the individuals in all the dominable layers of the level before the dominable layer where the nth individual is located and the individuals with larger crowded distance in the dominable layer where the nth individual is located are reserved to form a third population with the N individuals together.
Specifically, assuming that n=100, i.e., N individuals are included in the initial population, 2n=200 individuals are included in the second population after the cross mutation, and assuming that the 200 individuals are classified into six different levels according to a preset fitness function, the levels are arranged from low to high, if the first level of the available layers includes 33 individuals, the second level of the available layers includes 38 individuals, the third level of the available layers includes 40 individuals, the fourth level of the available layers includes 29 individuals, the fifth level of the available layers includes 33 individuals, and the sixth level of the available layers includes 27 individuals.
As can be seen from the above-mentioned ranking information, the first, second and third ranking layer includes 111 th individuals in total, and the 100 th is located at the third ranking layer, so that the crowding distances are calculated for all individuals in the third ranking layer and rearranged in order from the larger crowding distance to the smaller crowding distances, and 29 individuals with larger crowding distances are selected to form a third population with 100 individuals together with the remaining 71 individuals of the first and second ranking layer. It should be noted that the number of individuals, the number of dominating layers, and the number of individuals in each dominating layer described herein are exemplary descriptions and are not intended to be limiting in any way.
S7: repeating the steps S2-S6 for a plurality of iterations by using the third population, and determining an optimal cargo scheduling scheme according to an iteration result when the iteration number reaches a preset iteration number.
Specifically, after the first iteration is completed, the genetic codes of the individuals in the population, whether the individuals are whole or self, are changed, and then a new iteration is continued and the algorithm iteration number is increased by one. When the iteration times reach the preset iteration times, after the algorithm is cycled, the number of the multi-load AGVs and the picking sequence of each AGV can be obtained according to the recorded optimal solution.
According to the multi-load AGV task scheduling method in the intelligent storage environment, the number of the multi-load AGVs which are not considered in the past is taken into an algorithm, after the warehouse-out information is obtained, the initial population is obtained by initializing according to the warehouse-out information, and the individuals in the initial population are divided into the dominable layers of different grades through the preset fitness function, so that the task scheduling of the multi-load AGVs in the intelligent storage field is favorably solved, the reasonable AGV number and the AGV commodity taking sequence are calculated, and the intelligent storage performance is improved; through using as few AGVs as possible and as few as possible between the AGVs conflict, the AGV gets goods time has been reduced, the economic benefits of suggestion intelligent storage. In addition, according to the multi-load AGV task scheduling method, under the condition that the number of AGVs is insufficient, an allocation strategy of the AGVs is designed, and the intelligent storage efficiency is improved as much as possible; the use efficiency of the AGV can be estimated, the idle load condition of the AGV is analyzed, and theoretical support is provided for improving the storage efficiency of the AGV.
Yet another embodiment of the present invention provides a storage medium having a computer program stored therein, where the computer program is configured to execute the steps of the method for scheduling tasks of an AGV in the intelligent warehouse environment described in the foregoing embodiment. In still another aspect, the present invention provides an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the AGV task scheduling method in the intelligent warehouse environment according to the above embodiment when calling the computer program in the memory. In particular, the integrated modules described above, implemented in the form of software functional modules, may be stored in a computer readable storage medium. The software functional module is stored in a storage medium and includes instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.
Claims (6)
1. A multi-load AGV task scheduling method in an intelligent storage environment is characterized by comprising the following steps:
s1: the method comprises the steps of obtaining warehouse-in and warehouse-out information, wherein the warehouse-in and warehouse-out information at least comprises the number of cargoes to be transported, the number of multi-load AGVs and the goods shelf position;
S2: initializing a population according to the warehouse-in and warehouse-out information, and constructing an initial population with N individuals;
S3: performing crossover and mutation operations by taking individuals in the initial population as parent individuals to generate child individuals and combining the child individuals with the parent individuals to form a second population with 2N individuals;
S4: dividing the individuals in the second population into different levels of dominant layers by using a rapid non-dominant ranking according to a preset fitness function;
S5: carrying out crowding degree calculation on all individuals in an Mth-level dominable layer of a second population to obtain crowding distance of each individual, wherein Nth individuals classified according to different levels in the second population are positioned in the Mth-level dominable layer;
S6: maintaining individuals meeting the requirements of crowding distances in the current dominating layer, and forming a third population with N individuals with the individuals in the previous M-1 level dominating layer;
S7: repeating the steps S2-S6 for a plurality of iterations by using the third population, and determining an optimal cargo scheduling scheme according to an iteration result when the iteration number reaches a preset iteration number;
The step S4 comprises the following steps:
S41: respectively setting a first fitness function, a second fitness function and a third fitness function according to the number of AGVs, the maximum running time of the AGVs and the conflict among the AGVs;
S42: sorting individuals in the middle population by using a rapid non-dominant sorting algorithm according to the first fitness function, the second fitness function and the third fitness function, and dividing the individuals in the middle population into the dominant layers of different grades;
The first fitness function aims at optimizing the number of AGVs and is expressed as:
wherein A k represents the kth AGV, N (A k) represents whether the kth AGV is selected to transport the load, K represents the maximum number of currently selectable AGVs;
the second fitness function objective is to minimize the maximum run time of the AGV, expressed as:
F2＝min(max(G(Ak)))，k＝1,…,K
Where A k represents the kth AGV, G (A k) represents the total time for the kth AGV to transport the load, and max (G (A k)) represents the maximum transport time for a single AGV from among all AGVs;
the third fitness function is to minimize collisions between different AGVs, expressed as:
Wherein A i represents the ith AGV, A j represents the jth AGV, Indicating the number of collisions between the ith AGV and the jth AGV;
The step S5 comprises the following steps:
Sequencing from large to small according to a first function value of each individual in a current dominable layer, setting an initial value of a crowding distance of an individual with the smallest first function value and an individual with the largest first function value to infinity, and setting initial values of crowding distances of other individuals in the dominable layer to zero;
calculating a first crowding distance of each individual according to the initial value of the crowding distance and the first function value of each individual;
sorting the second function value of each individual in the current dominance layer from big to small;
Calculating a second crowding distance for each individual based on the first crowding distance and the second function value for each individual;
sorting the third function value of each individual in the current dominance layer from big to small;
calculating a third crowding distance of each individual according to the second crowding distance and a third function value of each individual, and determining the third crowding distance as the crowding distance of the individual;
The calculation formula of the first crowding distance is as follows:
Wherein, for each rank of the dominatable layer, n 0 represents the initial value of the congestion distance of the individual, f 1 (i+1) represents the first function value of the individual ranked i+1, f 1 (i-1) represents the first function value of the individual ranked i-1, Representing the maximum value of the first function value,/>Representing a minimum value of the first function value;
The calculation formula of the second crowding distance is as follows:
wherein, for each rank of the dominatable layer, n 1 represents a first crowding distance for the individual, f 2 (i+1) represents a second function value for the i+1th individual, f 2 (i-1) represents a second function value for the i-1th individual, Representing the maximum value of the second function value,/>Representing a minimum value of the second function value;
The calculation formula of the third crowding distance is as follows:
Wherein, for each rank of the dominatable layer, n 2 represents the second crowding distance of the individual, f 3 (i+1) represents the third function value for the i+1th individual, f 3 (i-1) represents the third function value for the i-1th individual, Representing the maximum value of the third function value,/>Representing the minimum of the third function value.
2. The multi-load AGV task scheduling method according to claim 1, wherein S2 includes:
Setting the number of individuals in an initial population, wherein the genes of each individual comprise an AGV number A k and a goods number W j which are randomly generated, and the individuals in the initial population meet the following four initial constraint conditions:
Where K represents the maximum available number of current multi-load AGVs, M represents the total number of loads to be handled, Indicating whether the load W j is to be carried by the AGV a k;
Wherein, Indicating whether the AGV a k is carrying the load W j after carrying the load W i,
Where L represents the maximum load of each multi-load AGV,
3. The multi-load AGV task scheduling method according to claim 2, wherein S3 includes:
S31: the individuals in the initial population are used as parent individuals to perform cross operation in any pair, and offspring individuals with the same number as the parent individuals are obtained;
S32: carrying out mutation operation on the offspring individuals subjected to the cross operation to obtain mutated offspring chromosomes;
S33: and merging the parent individuals and the variant child individuals to form a second population with the number of individuals twice that of the initial population.
4. The multi-load AGV task scheduling method according to claim 3, wherein S31 includes:
S311: randomly selecting two individuals from the initial population as a first parent individual and a second parent individual, wherein the genes are divided into a cargo number W j and an AGV number A k, randomly selecting one cargo number or the AGV number from the first parent individual as a first parent designated gene, copying all genes before the first parent designated gene to the corresponding position of a first offspring chromosome, inheriting the cargo numbers of the rest of the first offspring chromosome according to the cargo number sequence of the second parent individual, deleting the cargo numbers already existing in the first offspring chromosome, randomly selecting one cargo number from the first parent chromosome and the AGV number interval in the second parent chromosome, randomly inserting the cargo numbers into the first offspring chromosome, and reserving the individuals meeting the initial constraint condition to form the first offspring chromosome;
S312: randomly selecting one gene from the second parent individuals as a second parent designated gene, copying all genes before the second parent designated gene to the corresponding positions of second offspring chromosomes, inheriting the rest cargo numbers in the second offspring chromosomes according to the gene sequence of the first parent individuals and deleting the cargo numbers existing in the second offspring chromosomes, wherein the number of AGV numbers of the second offspring chromosomes is a number randomly selected in a number interval of AGV numbers in the first parent chromosomes and the second parent chromosomes, randomly inserting the AGV numbers into the cargo numbers, and reserving individuals meeting the initial constraint conditions so as to form the second offspring chromosomes;
s313: and repeating the steps S311-S312, and intersecting the rest parent individuals according to the preset intersecting probability, so as to obtain child individuals with the same number as the parent individuals.
5. The multi-load AGV task scheduling method according to claim 3, wherein S32 comprises:
Selecting offspring individuals after the crossover operation, carrying out position exchange on different genes to carry out mutation operation, then checking whether the exchanged individuals meet the initial constraint condition, if so, reserving the offspring individuals after mutation, and if not, discarding and carrying out mutation operation again.
6. A storage medium having stored therein a computer program for performing the steps of the multi-load AGV task scheduling method in an intelligent warehouse environment of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210006556.6A CN114493181B (en) | 2022-01-04 | 2022-01-04 | Multi-load AGV task scheduling method in intelligent storage environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210006556.6A CN114493181B (en) | 2022-01-04 | 2022-01-04 | Multi-load AGV task scheduling method in intelligent storage environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114493181A CN114493181A (en) | 2022-05-13 |
| CN114493181B true CN114493181B (en) | 2024-05-03 |
Family
ID=81509062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210006556.6A Active CN114493181B (en) | 2022-01-04 | 2022-01-04 | Multi-load AGV task scheduling method in intelligent storage environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114493181B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116882691B (en) * | 2023-07-19 | 2024-06-21 | 特维沃(上海)智能科技有限责任公司 | Automatic scheduling processing method, device and equipment for experiment plan and readable medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264120A (en) * | 2019-05-06 | 2019-09-20 | 盐城品迅智能科技服务有限公司 | A kind of intelligent storage route planning system and method based on more AGV |
| CN111832725A (en) * | 2019-04-15 | 2020-10-27 | 电子科技大学 | Multi-robot multi-task allocation method and device based on improved genetic algorithm |
| CN112783172A (en) * | 2020-12-31 | 2021-05-11 | 重庆大学 | AGV and machine integrated scheduling method based on discrete whale optimization algorithm |
| CN112884241A (en) * | 2021-03-12 | 2021-06-01 | 重庆大学 | Cloud edge collaborative manufacturing task scheduling method based on intelligent Agent |
| CN113420970A (en) * | 2021-06-10 | 2021-09-21 | 西安电子科技大学 | Task scheduling method under intelligent warehousing environment |
| CN113671910A (en) * | 2021-07-21 | 2021-11-19 | 华南理工大学 | Integrated multi-AGV flexible job shop scheduling method, device and medium |
-
2022
- 2022-01-04 CN CN202210006556.6A patent/CN114493181B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832725A (en) * | 2019-04-15 | 2020-10-27 | 电子科技大学 | Multi-robot multi-task allocation method and device based on improved genetic algorithm |
| CN110264120A (en) * | 2019-05-06 | 2019-09-20 | 盐城品迅智能科技服务有限公司 | A kind of intelligent storage route planning system and method based on more AGV |
| CN112783172A (en) * | 2020-12-31 | 2021-05-11 | 重庆大学 | AGV and machine integrated scheduling method based on discrete whale optimization algorithm |
| CN112884241A (en) * | 2021-03-12 | 2021-06-01 | 重庆大学 | Cloud edge collaborative manufacturing task scheduling method based on intelligent Agent |
| CN113420970A (en) * | 2021-06-10 | 2021-09-21 | 西安电子科技大学 | Task scheduling method under intelligent warehousing environment |
| CN113671910A (en) * | 2021-07-21 | 2021-11-19 | 华南理工大学 | Integrated multi-AGV flexible job shop scheduling method, device and medium |
Non-Patent Citations (2)
| Title |
|---|
| 改进多种群遗传算法的AutoStore***多AGV调度优化;王晓军 等;工业工程;20210815;第24卷(第4期);112-118 * |
| 智能仓储货位规划与AGV路径规划协同优化算法;蔺一帅 等;软件学报;20200114;第31卷(第9期);2770-2784 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114493181A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cheng et al. | Using a hybrid approach based on the particle swarm optimization and ant colony optimization to solve a joint order batching and picker routing problem | |
| CN110182527B (en) | Warehouse-in and warehouse-out control method and conveying system for shelf array | |
| CN108805316B (en) | Cargo handling method and apparatus | |
| Önüt et al. | A particle swarm optimization algorithm for the multiple-level warehouse layout design problem | |
| CN110182529B (en) | Warehouse-in and warehouse-out control method and conveying system for shelf array | |
| CN110182528B (en) | Warehouse-in and warehouse-out control method and conveying system for shelf array | |
| CN113359702B (en) | Intelligent warehouse AGV operation optimization scheduling method based on water wave optimization-tabu search | |
| Zhang et al. | A Building‐Block‐Based Genetic Algorithm for Solving the Robots Allocation Problem in a Robotic Mobile Fulfilment System | |
| Li et al. | A simulation study on the robotic mobile fulfillment system in high-density storage warehouses | |
| Vrysagotis et al. | Warehouse layout problems: Types of problems and solution algorithms | |
| CN109581987B (en) | AGV (automatic guided vehicle) scheduling path planning method and system based on particle swarm optimization | |
| CN115481897A (en) | AGV unmanned warehouse equipment optimization configuration method | |
| CN114493181B (en) | Multi-load AGV task scheduling method in intelligent storage environment | |
| CN111985683B (en) | Path optimization method for material distribution of multi-target discrete assembly workshop | |
| Joo et al. | Block transportation scheduling under delivery restriction in shipyard using meta-heuristic algorithms | |
| CN112884257A (en) | Goods taking path optimization method, device and system based on genetic algorithm | |
| CN111747008A (en) | Warehouse-out positioning method, device and system | |
| Pfrommer et al. | Autonomously organized block stacking warehouses: A review of decision problems and major challenges | |
| Li et al. | Pareto and niche genetic algorithm for storage location assignment optimization problem | |
| CN112508481A (en) | Intelligent storage multi-AGV scheduling method | |
| Arif et al. | Optimization of storage allocation for final goods warehouse using genetic algorithm | |
| CN112862212B (en) | Multi-AGV (automatic guided vehicle) scheduling method, device and equipment based on improved sparrow search algorithm | |
| Baray et al. | Design methodology for a multiple-level warehouse layout based on particle swarm optimization algorithm | |
| Yazdani et al. | Truck routing and scheduling for cross-docking in the supply chain: model and solution method | |
| CN117196263B (en) | Cargo scheduling method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |