A Cooperative Scheduling Method and System for AGV and Workstations in a Bufferless Pipeline Assembly System

Abstract

The present invention relates to a method and system for collaborative scheduling of an AGV and workstations in a bufferless pipeline assembly system. The method includes: S1: In a bufferless pipeline production system, taking the minimum working time spent as the objective of optimized scheduling, a mathematical model of the AGV set scheduling optimization problem is constructed; S2: The mathematical model is solved by using a heuristic algorithm to obtain the optimal solution for the AGV task assignment scheduling in a multi-AGV pipeline production system. The method provided by the present invention optimizes the AGV scheduling management and improves the operation efficiency.

Description

Technical Field

[0001] The present invention relates to the field of automatic control, and more particularly to a method and system for collaborative scheduling of an automated guided vehicle (AGV) and workstations in a bufferless pipeline assembly system. Background Art

[0002] An automated guided vehicle (AGV) is an intelligent automatic handling device that has a wide range of applications in various industries. In particular, it plays an important role in transporting workpieces in a manufacturing system, and its scheduling and operation efficiency will directly affect the production efficiency of the entire manufacturing system. Therefore, establishing an efficient and robust AGV scheduling system for a workshop is in line with the workshop production environment and is an important prerequisite for achieving high-efficiency production in the workshop.

[0003] Existing technical methods basically only involve scheduling in an ideal state. For example, the production conditions considered in AGV scheduling are that there is sufficient input and output buffer space at each machine, that is, products can queue at the machine for processing. Such conditions enable the decoupling of AGV scheduling and machine scheduling, reducing the scheduling difficulty. However, this setting does not conform to the actual production situation. First, no machine can have an infinite number of buffers. Second, the scheduling of AGVs and machines is generally coupled because the processing of large products is generally carried out on a carrier as a processing platform, and the AGV transports the carrier and moves between various points. Especially in a dark factory with unmanned production, any movement of products, including from the queue to the processing station, requires the participation of AGVs. Although the existing workshop AGV scheduling model greatly simplifies the manufacturing system, there is still a coupled impact between AGV and machine scheduling.

[0004] Therefore, it is necessary to design a collaborative scheduling of multiple AGVs and workstations in a bufferless discrete automated pipeline system to meet the actual workshop scheduling environment and adapt to the actual needs of real workshop production. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method and system for collaborative scheduling of an AGV and workstations in a bufferless pipeline assembly system.

[0006] The technical solution of the present invention is as follows: A method for collaborative scheduling of an AGV and workstations in a bufferless pipeline assembly system includes:

[0007] Step S1: In a bufferless pipeline production system, taking the minimum working time spent as the objective of optimizing scheduling, constructing a mathematical model for the AGV set scheduling optimization problem;

[0008] Step S2: Using a heuristic algorithm to solve the mathematical model to obtain an optimal solution for the AGV task assignment scheduling in a multi-AGV pipeline production system.

[0009] Compared with the prior art, the present invention has the following advantages:

[0010] The present invention discloses a collaborative scheduling method for AGV and workstations in a bufferless pipeline assembly system, which optimizes the scheduling management of the AGV system in a bufferless discrete automated assembly system, solves the problem of collaborative scheduling between the AGV system and workstations (machines) under high coupling degree, and greatly improves its operating efficiency compared with the existing methods, so as to achieve high-efficiency production in the workshop. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a flowchart of a collaborative scheduling method for AGV and workstations in a bufferless pipeline assembly system according to an embodiment of the present invention;

[0012] Figure 2 is a schematic diagram of the scheduling process according to an embodiment of the present invention;

[0013] Figure 3 is a schematic diagram of the state transition process according to an embodiment of the present invention;

[0014] Figure 4 is a schematic diagram of the simulation path according to an embodiment of the present invention;

[0015] Figure 5 is a structural block diagram of a collaborative scheduling system for AGV and workstations in a bufferless pipeline assembly system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The present invention provides a collaborative scheduling method for AGV and workstations in a bufferless pipeline assembly system, which optimizes the AGV scheduling management and improves the operating efficiency.

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following further elaborates on the present invention through specific embodiments in conjunction with the accompanying drawings.

[0018] Embodiment 1

[0019] As Figure 1 shown, a collaborative scheduling method for AGV and workstations in a bufferless pipeline assembly system provided by an embodiment of the present invention includes the following steps:

[0020] Step S1: In a bufferless pipeline production system, taking the minimum working time spent as the objective of optimizing scheduling, construct a mathematical model for the AGV set scheduling optimization problem;

[0021] Step S2: Use a heuristic algorithm to solve the mathematical model to obtain an optimal solution for AGV task assignment scheduling in a multi-AGV pipeline production system.

[0022] In one embodiment, the above step S1: In a bufferless pipeline production system, with the minimum working time spent as the optimization scheduling goal, a mathematical model for the AGV set scheduling optimization problem is constructed, specifically including:

[0023] Define the objective function:

[0024] min c w，p

[0025] The objective function represents minimizing the departure time of the last product at the last station, that is, the time when all products complete production. The constraint conditions (1)-(14) of the objective function are as follows:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040] Among them, w represents the number of workstations; p represents the number of products to be produced; b represents the number of available AGVs; d represents the distance between workstations; h represents the distance of the longitudinal road; W represents the set of workstations, W = {1, 2, …, w}; P represents the set of products, P = {1, 2, …, p}; V represents the set of AGVs, V = {1, 2, …, v}; D represents the set of tasks, D = {(i, j)|1 ≤ i < |W|, 0 ≤ j ≤ |P|, i ∈ Z, j ∈ Z}; A represents the set of virtual tasks, representing the first task of the AGV, A = {(0, b)|v ∈ V}; E represents the set of virtual tasks, representing the last task of the AGV, E = {(-1, -1)}; s i represents the processing time of workstation i; t i，m represents the shortest travel time of the AGV between any two workstations i and m on the production line; (i, j) is the task representation method, where product j moves from workstation i to workstation i + 1, the starting point of the task is workstation i, the ending point is workstation i + 1, and the product being transported is j; B1 = {(0, j)||0 ≤ j ≤ p, j ∈ Z}; B2 = {(i, 0)|0 < j < w, j ∈ Z}; C1 = {(w, j)|0 ≤ j ≤ p, j ∈ Z}; C2 = {(i, 0)||1 ≤ i ≤ w, i ∈ Z}; CD = D ∪ C1 ∪ C2; BD = D ∪ B1 ∪ B2

[0041] Among them, the decision variables are:

[0042] x i，j，m，n : The tasks (i, j) and (m, n) are two consecutive tasks of the same AGV;

[0043] b i，j : The time when the AGV leaves the end point of task (i, j) after completing the task;

[0044] c i，j : The time when task (i, j) starts to be executed, that is, the time when product j leaves workstation i;

[0045] Constraints (1) and (2) are used to determine the task execution order. A handling task can only have one previous task and one subsequent task;

[0046] Constraint (3) limits the scope of two adjacent handling tasks of the same AGV. Task (m, n) cannot be a task that must have been completed when task (i, j) is completed, nor can it be any task with the starting point at workstation i + 1;

[0047] Constraint (4) is the process handover constraint, indicating that adjacent product handling tasks on adjacent workstations cannot be assigned to the same AGV;

[0048] Constraints (5) and (6) are associated constraints: Constraint (5) requires that if the AGV consecutively executes two tasks (i, j) and (m, n), the time when the AGV leaves the end point of task (i, j), i.e., station i + 1, plus the travel time on the road, i.e., the time to reach the start point of task (m, n), shall not be earlier than the end time of task (m - 1, n), i.e., the time when other AGVs leave station m; Constraint (6) means that if the AGV consecutively executes two tasks (i, j) and (m, n), then the start time of task (m, n) is not earlier than the time when the AGV leaves the end point of task (i, j), i.e., station i + 1, plus the travel time on the road, i.e., the time to reach the start point of task (m, n).

[0049] Constraints (7) to (10) are time constraints: Constraint (7) requires that the time when product j arrives at station i + 1 shall not be earlier than the time when product j - 1 leaves station i + 1; Constraint (8) represents the time relationship when the same product leaves two adjacent stations; Constraint (9) represents the relationship between the start and end times of the same handling task; Constraint (10) requires that the time when the product processed at the first station leaves station 1 shall not be earlier than the time when the previous product leaves station 1 plus the processing time of station 1.

[0050] Constraints (11) to (14) are variable definition constraints, where the variable x is a 0 - 1 variable, and b and c are continuous variables.

[0051] In one embodiment, the above step S2: Use the heuristic algorithm to solve the mathematical model to obtain the optimal solution for the AGV task assignment scheduling in the multi - AGV pipeline production system, as Figure 2 shown, specifically including:

[0052] Step S21: The first product enters the first station, and an unused AGV is assigned to the first station.

[0053] Step S22: Determine whether there is a loaded AGV. If not, go to step S23. If there is, after state conversion, go to step S23, where the state conversion, as Figure 3 shown, specifically includes the following steps:

[0054] Step S221: Check all AGVs. If an AGV is currently in the loaded operation state, add it to the set of loaded AGVs.

[0055] Step S222: Determine whether the destination station w of each AGV in the set of loaded AGVs is in the idle state:

[0056] If the station w is idle, the AGV and the product it is carrying can be directly moved to station w, and the remaining processing time of the product at station w and the tardiness time of the AGV are modified;

[0057] If the station w is in a non-idle state, the first idle station w' after station w needs to be found f , and the products on all stations between station w f and station w' are all moved back one station in sequence to make station w idle, and the AGV and the product it is carrying are moved to station w;

[0058] Step S23: Determine the set of AGVs to be allocated, which specifically includes the following steps:

[0059] Step S231: Determine the status of the AGV. Among them, the status of the AGV includes: idle AGV, loaded AGV, and allocated AGV; an idle AGV is an AGV that can be allocated, a loaded AGV can be allocated after a status conversion, and an allocated AGV includes a waiting AGV and an empty-load moving AGV, which refers to an AGV that has been assigned a task but has not started carrying the product yet;

[0060] Step S232: Determine whether the AGV is allocated according to the remaining start time RST and status of the AGV. Among them, the remaining start time RST refers to the interval time from a certain moment when the i-th AGV that has been assigned a task starts to execute the task, that is, carry the product and move. The calculation formula (13) is as follows:

[0061] RST i = max{restDeadheadTripTime i ,

[0062] restProcessingTime w , restTimeToBeIdle w+1 - d} (15)

[0063] Among them, restDeadheadTripTime i represents the time required for the AGV i to reach the task start point, restProcessingTime w represents the remaining processing time of the work station w, and restTimeToBeIdle w+1 represents the shortest time required for the work station w + 1 to become idle;

[0064] Step S233: If the AGV is in a waiting state and its RST is greater than or equal to 2h unit time, add this AGV to the set of AGVs to be allocated;

[0065] If the AGV is in the no-load moving state and the AGV is moving from the right side to the left side of the production line, it is necessary to determine that when the remaining moving time of the AGV is greater than 1 hour unit time, the AGV is added to the set of AGVs to be allocated, otherwise it does not participate in the task allocation;

[0066] Step S24: Determine the set of workstations to be allocated, specifically including:

[0067] According to the idle workstations, the production line after equivalent conversion is divided into N segments, so that each segment only contains non-idle workstations; calculate the number l of workstations in each segment n , calculate the number q of workstations participating in AGV allocation in each segment n , and thus determine the largest q in terms of number in each segment n workstations are added to the set of workstations to be allocated, q n The calculation formula (16) is as follows:

[0068]

[0069] Among them, assignableAgvNum represents the number of AGVs that can be allocated in the system;

[0070] Step S25: Assign the AGVs to be allocated to the workstations to be allocated, specifically including:

[0071] Starting from the left side of the production line, AGV assignment is carried out in each segment respectively. According to the current state of the production line, select the th to the th AGVs to be allocated and assign them to the nth segment; the workstations that need to allocate AGVs in each segment are the largest q n unallocated AGV workstations in terms of number in the segment; there may also be already matched AGVs and workstations in each segment. Use Fixed to represent the set of tuples of already matched AGVs and workstations, and add the already matched AGVs and workstations to the AGV set and the workstation set respectively; among them, the method model for assigning AGVs to workstations is as shown in formula (17) and its constraint conditions (18) - (22):

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078] Among them, the decision variables are as follows:

[0079] x v，w represents whether the AGV v is assigned to station w, where v ∈ AV and w ∈ WA;

[0080] f w represents the time when the product on station w can move, where w ∈ WA.

[0081] The other variables are shown as follows:

[0082] VA represents the set of AGVs, including q n unassigned AGVs and the AGVs that have been assigned to the stations within this section;

[0083] WA represents the set of stations, including q n stations without assigned AGVs and all stations with assigned AGVs;

[0084] |VA| = |WA|, that is, the number of AGVs is equal to the number of stations;

[0085] Fixed means that (v f , w f ) ∈ Fixed, where v f ∈ VA and w f ∈ WA, which is the set of pairs of AGVs and stations not participating in the assignment;

[0086] rpt w represents the remaining processing time of station i at the current moment, where w ∈ WA;

[0087] t v，w represents the travel time of the AGV v from the current position to station w, where v ∈ VA and w ∈ WA;

[0088] w l represents that w l ∈ WA represents the last station within this section;

[0089] d represents the distance between stations;

[0090] Constraint (17) represents minimizing the sum of start times;

[0091] Constraints (18) and (19) are assignment constraints: Constraint (18) requires that each station in the WA set be assigned an AGV, and Constraint (19) means that each AGV in the VA set is to be assigned to a station;

[0092] Constraint (20) represents that the combinations of stations and AGVs not participating in the assignment must be assigned correspondingly together;

[0093] Constraint (21) indicates that the calculation method for the start time of the task at the last station in each segment is to take the maximum value from the remaining processing time of this station and the time required for the AGV assigned to this station to reach this station;

[0094] Constraint (22) indicates that the calculation method for the start time of the task at other stations in each segment is to take the maximum value from the remaining processing time of this station, the time required for the AGV assigned to this station to reach this station, and the difference between the start time of the task at the next station and the time distance between stations;

[0095] Step S26: Update the system state, specifically including:

[0096] Step S261: Define a parameter RTBI for each station to represent the time when this station needs to be converted to the idle state at the current moment, and the calculation formula (23) is as follows:

[0097]

[0098] where, restProcessTime w represents the remaining processing time of station w;

[0099] restDeadheadTripTime v represents the time for the AGV v to reach the task starting point;

[0100] Step S262: Calculate the time span interval of one-step state update. This time span is the shortest time to generate the next idle AGV, and the calculation formula (24) of interval is as follows:

[0101]

[0102] Step S263: Check each station from right to left, and adjust the state of the station according to the relationship between the station RTBI and the interval. The specific state update judgment conditions and rules are shown in Table 1:

[0103] Table 1 State update judgment conditions and rules

[0104]

[0105] Step S27: Judge whether all products have left the line and the AGV has returned to the garage. If so, the scheduling ends; otherwise, go to Step S22.

[0106] The present invention conducts a large number of simulation simulations based on multiple scenarios, and the running paths of the simulation simulations are as Figure 4As shown in the figure, the comparison of the simulation results of the scheduling method of the present invention with the optimization results of the traditional method is shown in Table 2 as follows:

[0107] Table 2 Comparison of the simulation results of the scheduling method of the present invention with the optimization results of the traditional method

[0108]

[0109]

[0110] It can be seen from Table 2 that the simulation results of the present invention show that the AGV scheduling management and optimization method of the present invention is feasible, and its operating efficiency has been greatly improved compared with the traditional enterprise scheduling method.

[0111] The present invention discloses a collaborative scheduling method for AGV and workstations in a bufferless pipeline assembly system, which schedules and manages the optimization of the AGV system in a bufferless discrete automated assembly system, solves the collaborative scheduling problem of the AGV system and workstations (machines) under high coupling degree, and greatly improves its operating efficiency compared with the existing methods, so as to achieve high-efficiency production in the workshop.

[0112] Embodiment 2

[0113] As Figure 5 shown, the embodiment of the present invention provides a collaborative scheduling system for AGV and workstations in a bufferless pipeline assembly system, including the following modules:

[0114] A mathematical model module 31 for constructing an AGV set scheduling optimization problem, which is used to construct a mathematical model of an AGV set scheduling optimization problem with the minimum working time spent as the optimization scheduling goal in a bufferless pipeline production system;

[0115] A mathematical model module 32 for solving the AGV set scheduling optimization problem, which is used to solve the mathematical model by using a heuristic algorithm to obtain the optimal solution of the AGV task assignment scheduling in a multi-AGV pipeline production system.

[0116] The above embodiments are provided only to describe the purpose of the present invention, and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims. All equivalent substitutions and modifications made without departing from the spirit and principles of the present invention shall be covered within the scope of the present invention.

Claims

1. A collaborative scheduling method for an AGV and workstations in a bufferless pipeline assembly system, characterized in that, Including: Step S1: In a bufferless pipeline production system, taking the minimum working time spent as the objective of optimal scheduling, construct a mathematical model for the AGV set scheduling optimization problem, specifically including: Define the objective function: The objective function represents minimizing the departure time of the last product at the last station, that is, the time when all products complete production. The constraint conditions (1)-(14) of the objective function are as follows: （1） （2） （3） （4） （5） （6） （7） （8） （9） （10） （11） （12） （13） （14） Among them, represents the number of workstations; represents the number of products to be produced; represents the number of available AGVs; represents the distance between workstations; represents the distance of the longitudinal road; represents the set of workstations, ; represents the set of products, ; represents the set of AGVs, ; represents the set of tasks, ; represents the set of virtual tasks, indicating the first task of the AGV, }; represents the set of virtual tasks, indicating the last task of the AGV, ; represents the workstation processing time; represents any two workstations on the production line and between the shortest travel time; is the task representation method, the product moves from the workstation to the workstation , the starting point of the task is the workstation , the end point is the workstation , and the product to be carried is ; ; ; ; ; ; ; Among them, the decision variables are: : Task and Task are two consecutive tasks of the same AGV; : The time when the AGV leaves the task end point after completing the task ; : Task The time when it starts to execute, that is, the time when the product leaves the work station ; Constraint conditions (1) and (2) are used to determine the task execution order. A handling task can only have one preceding task and one succeeding task; Constraint condition (3) restricts the scope of two adjacent handling tasks of the same AGV. Task (m,n) cannot be a task that must have been completed when task (i,j) is completed, nor can it be any task with the starting point at station i+1; Constraint condition (4) is the process handover constraint, indicating that adjacent product handling tasks at two adjacent stations cannot be assigned to the same AGV; Constraint conditions (5) and (6) are association constraints: Constraint condition (5) requires that if an AGV continuously executes two tasks (i,j) and (m,n), the time when the AGV leaves the end point of task (i,j), that is, station i+1, plus the travel time on the road, that is, the time to reach the starting point of task (m,n), must not be earlier than the end time of task (m-1,n), that is, the time when other AGVs leave station m; Constraint condition (6) means that if an AGV continuously executes two tasks (i,j) and (m,n), then the start time of task (m,n) is not earlier than the time when the AGV leaves the end point of task (i,j), that is, station i+1, plus the travel time on the road, that is, the time to reach the starting point of task (m,n); Constraint conditions (7)-(10) are time constraints: Constraint condition (7) requires that the time when product j arrives at station i+1 is not earlier than the time when product j-1 leaves station i+1; Constraint condition (8) represents the time relationship when the same product leaves two adjacent stations; Constraint condition (9) represents the relationship between the start and end times of the same handling task; Constraint condition (10) requires that the time when the product processed at the first station leaves station 1 is not earlier than the time when the previous product leaves station 1 plus the processing time of station 1; Constraint conditions (11)-(14) are variable definition constraints. The variable x is a 0-1 variable, and b and c are continuous variables; Step S2: Use a heuristic algorithm to solve the mathematical model to obtain the optimal solution for the AGV task assignment scheduling in the multi-AGV pipeline production system, specifically including: Step S21: The first product enters the first station, and an unused AGV is assigned to the first station; Step S22: Determine whether there is a loaded AGV. If not, go to step S23. If there is, after a state transition, go to step S23; Step S23: Determine the set of AGVs to be assigned, specifically including: Step S231: Determine the status of the AGV. The status of the AGV includes: idle AGV, loaded AGV, and assigned AGV. The idle AGV is an AGV that can be assigned. The loaded AGV can be assigned after a status transition. The assigned AGV includes waiting AGV and unloaded moving AGV, which refers to an AGV that has been assigned a task but has not started transporting products yet. Step S232: Determine whether the AGV is assigned based on the remaining start time RST and status of the AGV. The remaining start time RST refers to the interval time from a certain moment until the i-th AGV that has been assigned a task starts to execute the task, i.e., the time for transporting products and moving. The calculation formula (13) is as follows: (15) Among them, represents the time required to reach the task starting point, represents the remaining processing time of work position w, represents the shortest time required for work position w + 1 to become idle; Step S233: If the AGV is in the waiting state and its RST is greater than or equal to 2h unit time, add the AGV to the set of AGVs to be assigned. If the AGV is in the unloaded moving state and the AGV is moving from the right side to the left side of the production line, it is necessary to determine that when the remaining moving time of the AGV is greater than 1h unit time, add the AGV to the set of AGVs to be assigned, otherwise it does not participate in task assignment. Step S24: Determine the set of workstations to be assigned, specifically including: Divide the production line after equivalent conversion into N segments according to the idle working positions, so that each segment only contains non-idle workstations; calculate the number of workstations in each segment , calculate the number of workstations participating in AGV allocation in each segment , and thus determine the workstations with the largest numbers in each segment are added to the set of workstations to be allocated, The calculation formula (16) is as follows: (16) where assignableAgvNum represents the number of AGVs that can be assigned in the system. Step S25: Assign the AGVs to be assigned to the workstations to be assigned, specifically including: Start from the left side of the production line and assign AGVs to each section. Select the AGV from left to right according to the current status of the production line. To The AGVs to be assigned are assigned to the nth segment; the station to be assigned an AGV in each segment is the one with the largest number in the segment. There are workstations to which AGVs are not assigned. In each segment, there may be matched AGVs and workstations. The set of matched AGV and workstation tuples is represented by Fixed, and the matched AGVs and workstations are added to the AGV set and the workstation set respectively. The model of the AGV assignment method is shown in formula (17) and its constraints (18) to (22): (17) (18) (19) (20) (21) (22) where the decision variables are as follows: Whether it is assigned to a work station , ; Indicates the station The time when the product on it can move, ; The other variables are as follows: Denote the AGV set, including AGVs to be allocated and AGVs that have been allocated to workstations within this section; Represents a set of workstations, including workstations without assigned AGVs and all workstations with assigned AGVs; , that is, the number of AGVs is equal to the number of workstations; , the set of AGVs and workstations that do not participate in the allocation; Indicates the remaining processing time of the work station at the current moment ; ; Travel time from the current position to the work station , ; Indicates the last work station within this section; Indicates the distance between workstations; Constraint (17) represents minimizing the sum of start times. Constraints (18) and (19) are assignment constraints: Constraint (18) requires that each workstation in each WA set be assigned an AGV, and Constraint (19) means that each AGV in the VA set must be assigned to a workstation. Constraint (20) represents that the combination of workstations and AGVs that do not participate in assignment must be assigned together correspondingly. Constraint (21) represents that the calculation method of the start time of the task on the last workstation in each segment is to take the maximum value from the remaining processing time of this workstation and the time required for the AGV assigned to this workstation to reach this workstation. Constraint (22) represents that the calculation method of the start time of the task on other workstations in each segment is to take the maximum value from the remaining processing time of this workstation, the time required for the AGV assigned to this workstation to reach this workstation, and the difference between the start time of the task on the next workstation and the time distance between workstations. Step S26: Update the system status, specifically including: Step S261: Define a parameter RTBI for each workstation to represent the time when this workstation needs to be converted to the idle state at the current moment. The calculation formula (23) is as follows: (23) Among them, represents the remaining processing time of station w; Indicate Time to the task start point; Step S262: Calculate the time span of a single-step state update , which is the shortest time to generate the next idle AGV The calculation formula (24) is as follows: (24) Step S263: Check each station from right to left and adjust the status of the station according to the relationship between the station and interval; Step S27: Determine whether all products have left the production line and the AGVs have returned to the garage. If so, the scheduling ends; otherwise, go to Step S22.

2. The collaborative scheduling method for the AGV and workstations in the bufferless pipeline assembly system according to claim 1, wherein The status transition in Step S22 specifically includes: Step S221: Check all AGVs. If an AGV is currently in the loaded running state, add it to the set of loaded AGVs. Step S222: Determine whether the destination station w of each AGV in the load AGV set is in an idle state: If station w is idle, the AGV and the product it carries can be directly moved to station w, and the remaining processing time of the product at station w and the tardiness time of the AGV are modified; If the station w is not in the idle state, it is necessary to find the first idle station after the station w , and move the products on all the stations between the station and the station w one station backward in sequence, so that the station w becomes idle, and move the AGV and the product it carries to the station w.

3. A bufferless pipeline assembly system AGV and station collaborative scheduling system, characterized in that, Including the following modules: A module for constructing a mathematical model of the AGV set scheduling optimization problem, which is used to construct a mathematical model of the AGV set scheduling optimization problem with the goal of minimizing the working time spent in a bufferless pipeline production system, specifically including: Define the objective function: The objective function represents minimizing the departure time of the last product at the last station, that is, the time when all products complete production. The constraint conditions (1)-(14) of the objective function are as follows: （1） （2） （3） （4） （5） （6） （7） （8） （9） （10） （11） （12） （13） （14） Among them, represents the number of workstations; represents the number of products to be produced; represents the number of available AGVs; represents the distance between workstations; represents the distance of the longitudinal road; represents the set of workstations, ; represents the set of products, ; represents the set of AGVs, ; represents the set of tasks, ; represents the set of virtual tasks, indicating the first task of the AGV, }; represents the set of virtual tasks, indicating the last task of the AGV, ; represents the workstation processing time; represents any two workstations on the production line and between shortest travel time; is the task representation method, the product moves from the workstation to the workstation , the starting point of the task is the workstation , the end point is the workstation , and the product being carried is ; ; ; ; ; ; ; Among them, the decision variables are: : Task and task are two consecutive tasks of the same AGV; : The time when the AGV leaves the task end point after completing the task ; : Task The time when the execution starts, i.e., the time when the product leaves the work station ; Constraint conditions (1) and (2) are used to determine the task execution order. A handling task can only have one preceding task and one succeeding task; Constraint condition (3) restricts the range of two adjacent handling tasks of the same AGV. Task (m,n) cannot be a task that must have been completed when task (i,j) is completed, nor can it be any task starting from station i+1; Constraint condition (4) is a process handover constraint, indicating that adjacent product handling tasks on two adjacent stations cannot be assigned to the same AGV; Constraint conditions (5) and (6) are correlation constraints: Constraint condition (5) requires that if the AGV continuously executes two tasks (i,j) and (m,n), the time when the AGV leaves the end point of task (i,j), that is, station i+1, plus the travel time on the road, that is, the time to reach the start point of task (m,n), shall not be earlier than the end time of task (m-1,n), that is, the time when other AGVs leave station m; Constraint condition (6) means that if the AGV continuously executes two tasks (i,j) and (m,n), then the start time of task (m,n) is not earlier than the time when the AGV leaves the end point of task (i,j), that is, station i+1, plus the travel time on the road, that is, the time to reach the start point of task (m,n); Constraint conditions (7)-(10) are time constraints: Constraint condition (7) requires that the time when product j reaches station i+1 shall not be earlier than the time when product j-1 leaves station i+1; Constraint condition (8) represents the time relationship when the same product leaves two adjacent stations; Constraint condition (9) represents the relationship between the start and end times of the same handling task; Constraint condition (10) requires that the time when the product processed at the first station leaves station 1 shall not be earlier than the time when the previous product leaves station 1 plus the processing time of station 1; Constraint conditions (11)-(14) are variable definition constraints. The variable x is a 0-1 variable, and b and c are continuous variables; A module for solving the mathematical model of the AGV set scheduling optimization problem, which is used to solve the mathematical model using a heuristic algorithm to obtain the optimal solution for the AGV task assignment scheduling in a multi-AGV pipeline production system, specifically including: Step S21: The first product enters the first station, and an unused AGV is assigned to the first station; Step S22: Determine whether there is a loaded AGV. If not, go to Step S23. If so, after state conversion, go to Step S23; Step S23: Determine the set of AGVs to be assigned, specifically including: Step S231: Determine the status of the AGV. Among them, the status of the AGV includes: idle AGV, loaded AGV, and assigned AGV. The idle AGV is an AGV that can be assigned. The loaded AGV can be assigned after state conversion. The assigned AGV includes waiting AGV and unloaded moving AGV, which refers to an AGV that has been assigned a task but has not started to carry the product yet; Step S232: Determine whether the AGV is assigned according to the remaining start time RST and status of the AGV. Among them, the remaining start time RST refers to the interval time from a certain moment when the i-th AGV that has been assigned a task starts to execute the task, that is, the time to carry the product and move. The calculation formula (13) is as follows: (15) Among them, represents the time required to reach the task starting point, represents the remaining processing time of work position w, represents the shortest time required for work position w + 1 to become idle; Step S233: If the AGV is in the waiting state and its RST is greater than or equal to 2h unit time, add the AGV to the set of AGVs to be assigned; If the AGV is in the unloaded moving state and the AGV is moving from the right side to the left side of the production line, it is necessary to determine that when the remaining moving time of the AGV is greater than 1h unit time, add the AGV to the set of AGVs to be assigned, otherwise it does not participate in task assignment; Step S24: Determine the set of stations to be assigned, specifically including: Divide the production line after equivalent conversion into N segments according to the idle work positions, so that each segment contains only non-idle workstations; calculate the number of workstations in each segment , calculate the number of workstations participating in AGV allocation in each segment , thereby determining the largest numbered workstations in each segment are added to the set of workstations to be allocated, The calculation formula (16) is as follows: (16) Among them, assignableAgvNum represents the number of AGVs that can be assigned in the system; Step S25: Assign the AGVs to be assigned to the stations to be assigned, specifically including: Starting from the left side of the production line, perform AGV assignment within each section respectively. According to the current state of the production line, select the to the th AGV to be assigned and allocate it to the nth section; the workstations that need to be assigned AGVs within each section are the workstations with the largest numbers within the section that have not been assigned AGVs; there may also be already matched AGVs and workstations within each section. Use Fixed to represent the set of tuples of already matched AGVs and workstations, and add the already matched AGVs and workstations to the AGV set and the workstation set respectively; among them, the method model for assigning AGVs to workstations is shown in formula (17) and its constraint conditions (18) to (22): (17) (18) (19) (20) (21) (22) Among them, the decision variables are as follows: Whether assigned to the work station , ; Indicates the station The time when the product on it can move, ; Other variables are as follows: Denote the AGV set, including AGVs to be assigned and AGVs that have been assigned to the workstations within this section; Denotes the set of workstations, including workstations without assigned AGVs and all workstations with assigned AGVs; , that is, the number of AGVs is equal to the number of workstations; , the set of AGVs and workstations that do not participate in the distribution; Indicates the remaining processing time of the workstation at the current moment , ; Travel time from the current position to the work station , ; Indicates the last work station within this section; Indicates the distance between workstations; Constraint (17) represents minimizing the sum of start times; Constraints (18) and (19) are assignment constraints: Constraint (18) requires that each station in each WA set be assigned an AGV, and Constraint (19) means that each AGV in the VA set is to be assigned to a station; Constraint (20) represents that the combination of stations and AGVs that do not participate in the assignment must be assigned together correspondingly; Constraint (21) represents that the calculation method of the start time of the task on the last station in each segment is to take the maximum value from the remaining processing time of the station and the time required for the AGV assigned to the station to reach the station; Constraint (22) represents that the calculation method of the start time of the task on other stations in each segment is to take the maximum value from the remaining processing time of the station, the time required for the AGV assigned to the station to reach the station, and the difference between the start time of the task on the next station and the time distance between the stations; Step S26: Update the system status, specifically including: Step S261: Define a parameter RTBI for each station to represent the time when the station needs to be converted to the idle state at the current moment. The calculation formula (23) is as follows: (23) Among them, represents the remaining processing time of station w; Indicates Time to the task starting point; Step S262: Calculate the time span of one-step state update , where the time span is the shortest time to generate the next idle AGV The calculation formula (24) is as follows: (24) Step S263: Check each station from right to left and adjust the status of the station according to the relationship between the station and interval; Step S27: Determine whether all products have been offlined and the AGV has returned to the garage. If so, the scheduling ends; otherwise, go to Step S22.

Images