Multi-AGV collaborative scheduling method and device for digital twin platform of intelligent factory

Abstract

The present invention provides a multi-AGV collaborative scheduling method and device for a digital twin platform of a smart factory. The method includes: receiving the status information sent by each AGV and the environmental information of the workshop; determining a traffic congestion coefficient based on the status information and the environmental information; and sending a charging instruction or an avoidance instruction to a target AGV based on the traffic congestion index. The multi-AGV collaborative scheduling method for the digital twin platform of the smart factory provided by the present invention receives the status information sent by each AGV and the environmental information of the workshop in real time to determine the current congestion degree of the workshop, and thus sends a charging instruction or an avoidance instruction to the AGV according to the congestion degree, instructing the AGV to go to the charging station for charging, or adjusting the route or speed for avoidance, realizing the collaborative scheduling of multiple AGVs, fully considering the possible problems of insufficient power or conflicts in the scheduling process, and greatly improving the working efficiency of the AGV cluster in the production workshop.

Description

Technical Field

[0001] The present invention relates to the technical fields of digital twin and robot automation, and particularly to a multi-AGV collaborative scheduling method and device for a digital twin platform of an intelligent factory. Background Art

[0002] Many mobile phone manufacturing enterprises have gradually started to apply intelligent robot automated guided vehicles (AGVs) to the mobile phone component production workshops of intelligent factories to achieve flexible automated production lines. At present, some enterprises have begun to use AGV equipment to upgrade the existing manufacturing execution system (MES) in factory workshops. In the production workshops where MES is applied, the workpiece material shelves and CNC machine tools are located on both sides of the workshop respectively, and AGVs travel between the workpiece material shelves and CNC machine tools to complete the transportation tasks of workpiece materials.

[0003] In the trend of increasing intelligence in manufacturing factories, there will be a dozen or even dozens of AGVs working simultaneously in complex large production workshops. Most of the existing technologies are for scheduling single or multiple independent robots, and there is less research on collaborative scheduling of multiple workpiece material transportation AGVs. Moreover, most of the research on AGV scheduling does not consider the situations where AGVs need to charge and dynamically avoid obstacles. Summary of the Invention

[0004] In view of the problems existing in the prior art, embodiments of the present invention provide a multi-AGV collaborative scheduling method and device for a digital twin platform of an intelligent factory.

[0005] The present invention provides a multi-AGV collaborative scheduling method for a digital twin platform of an intelligent factory, including:

[0006] Receiving the status information sent by each AGV and the environmental information of the workshop;

[0007] Based on the status information and the environmental information, determining a traffic congestion coefficient; the congestion coefficient is used to indicate the congestion degree of the workshop;

[0008] Based on the traffic congestion index, sending a charging instruction or an avoidance instruction to a target AGV.

[0009] In some embodiments, the sending a charging instruction or an avoidance instruction to a target AGV based on the traffic congestion index includes:

[0010] Based on the traffic congestion index, when it is determined that there is congestion, clustering the positions of AGVs to determine a first number of clusters;

[0011] Based on the first number of clusters, send a charging instruction or an avoidance instruction to the target AGV.

[0012] In some embodiments, the step of sending a charging instruction to the target AGV based on the first number of clusters includes:

[0013] Based on the first number of clusters, determine the cluster with the largest number of AGVs within the cluster as the target cluster;

[0014] Determine a second number of AGVs within the target cluster as the target AGV;

[0015] Send a charging instruction to the target AGV.

[0016] In some embodiments, the step of sending an avoidance instruction to the target AGV based on the first number of clusters includes:

[0017] Based on the first number of clusters, with the center of each cluster as the center of the circle, and with the first distance, the second distance, and the third distance as the radii, determine the first range, the second range, and the third range; the first distance is less than the second distance, and the second distance is less than the third distance;

[0018] Determine the AGVs within the first range, the second range, and the third range as the target AGV;

[0019] Send an avoidance instruction to the target AGV.

[0020] In some embodiments, the step of sending a charging instruction to the target AGV based on the traffic congestion index includes:

[0021] Based on the traffic congestion index, when it is determined that there is no congestion, determine the target AGV based on the current battery level of the AGV;

[0022] Send a charging instruction to the target AGV.

[0023] In some embodiments, the step of sending an avoidance instruction to the target AGV based on the traffic congestion index includes:

[0024] Based on the traffic congestion index, when it is determined that there is no congestion, determine the target AGV based on the distance between any two AGVs, a preset distance, and the task priorities corresponding to the two AGVs;

[0025] Send an avoidance instruction to the target AGV.

[0026] In some embodiments, before receiving the status information sent by each AGV and the environmental information of the workshop, it further includes:

[0027] Construct a scheduling model; the scheduling model includes: a first model for determining the traveling distance of the AGV, a second model for determining the scheduling time, and a third model for determining the utilization rate of the AGV;

[0028] Based on the analytic hierarchy process, determine the weight coefficients of the first model, the second model, and the third model respectively;

[0029] Based on the weight coefficients, merge the first model, the second model, and the third model into a target model;

[0030] Solve the target model to determine the target tasks assigned to each AGV.

[0031] The present invention also provides a multi-AGV collaborative scheduling device for a smart factory digital twin platform, including:

[0032] A receiving module for receiving the status information sent by each AGV and the environmental information of the workshop;

[0033] A determining module for determining a traffic congestion coefficient based on the status information and the environmental information; the congestion coefficient is used to indicate the congestion degree of the workshop;

[0034] A scheduling module for sending a charging instruction or an avoidance instruction to the target AGV based on the traffic congestion index.

[0035] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, it implements the multi-AGV collaborative scheduling method of any one of the above-mentioned smart factory digital twin platforms.

[0036] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the multi-AGV collaborative scheduling method of any one of the above-mentioned smart factory digital twin platforms.

[0037] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the multi-AGV collaborative scheduling method of any one of the above-mentioned smart factory digital twin platforms.

[0038] The multi-AGV collaborative scheduling method and device for the digital twin platform of the intelligent factory provided by the present invention can receive the status information sent by each AGV and the environmental information of the workshop in real time to determine the current congestion degree of the workshop, and then send a charging instruction or an avoidance instruction to the AGV according to the congestion degree, instructing the AGV to go to the charging station for charging, or adjusting the route or speed to avoid, realizing the collaborative scheduling of multiple AGVs, fully considering the problems of insufficient power or conflicts that may occur during the scheduling process, and greatly improving the working efficiency of the AGV cluster in the production workshop. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0040] Figure 1 is one of the flow diagrams of the multi-AGV collaborative scheduling method for the digital twin platform of the intelligent factory provided by the embodiment of the present invention;

[0041] Figure 2 is the second of the flow diagrams of the multi-AGV collaborative scheduling method for the digital twin platform of the intelligent factory provided by the embodiment of the present invention;

[0042] Figure 3 is the third of the flow diagrams of the multi-AGV collaborative scheduling method for the digital twin platform of the intelligent factory provided by the embodiment of the present invention;

[0043] Figure 4 is the fourth of the flow diagrams of the multi-AGV collaborative scheduling method for the digital twin platform of the intelligent factory provided by the embodiment of the present invention;

[0044] Figure 5 is the fifth of the flow diagrams of the multi-AGV collaborative scheduling method for the digital twin platform of the intelligent factory provided by the embodiment of the present invention;

[0045] Figure 6 is the structural diagram of the hybrid network collaborative control system provided by the embodiment of the present invention;

[0046] Figure 7 is the operation flow diagram of the hybrid network collaborative control system provided by the embodiment of the present invention;

[0047] Figure 8 is the schematic diagram of the multi-AGV collaborative scheduling architecture based on the digital twin platform provided by the embodiment of the present invention;

[0048] Figure 9It is a schematic structural diagram of a multi-AGV cooperative scheduling device of the digital twin platform of the intelligent factory provided by an embodiment of the present invention;

[0049] Figure 10 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0050] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] The terms "first", "second", etc. in the description and claims of the present invention are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such terms may be interchanged under appropriate circumstances so that the embodiments of the present invention can be implemented in an order different from those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and do not limit the number of objects. For example, the first object may be one or multiple. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / " generally means an "or" relationship between the associated objects before and after.

[0052] Figure 1 It is one of the flow schematic diagrams of the multi-AGV cooperative scheduling method of the digital twin platform of the intelligent factory provided by an embodiment of the present invention. As Figure 1 shown, the multi-AGV cooperative scheduling method of the digital twin platform of the intelligent factory provided by an embodiment of the present invention includes:

[0053] Step 101, receiving the status information sent by each AGV and the environmental information of the workshop;

[0054] Step 102, determining a traffic congestion coefficient based on the status information and the environmental information; the congestion coefficient is used to indicate the congestion degree of the workshop;

[0055] Step 103, sending a charging instruction or an avoidance instruction to a target AGV based on the traffic congestion index.

[0056] It should be noted that the execution subject of the multi-AGV collaborative scheduling method of the digital twin platform for the intelligent factory provided by the present invention can be an electronic device, a component in the electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a handheld computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a Network Attached Storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc. The present invention does not make specific limitations.

[0057] In step 101, receive the status information sent by each AGV and the environmental information of the workshop.

[0058] In the trend of the increasing intelligence of digital factories, there will be a dozen or even dozens of AGVs simultaneously carrying out production work in a complex large-scale production workshop.

[0059] In the entire AGV cluster system, during the process of each AGV executing tasks, it can receive the self-status information and workshop environmental information sent by each AGV.

[0060] The status information may include: information such as the current battery level information, location information, and tasks being executed of the AGV.

[0061] The environmental information of the workshop may include: information such as the location information of the shelves in the workshop, the location information of the machine tools, the location information of the charging stations in the workshop, and whether the charging stations in the workshop are idle.

[0062] In step 102, based on the status information and the environmental information, determine the traffic congestion coefficient; the congestion coefficient is used to indicate the congestion degree of the workshop.

[0063] The traffic congestion index, also known as the bus index or traffic cost index, measures the traffic congestion condition of a region by the ratio of the extra time to the original time used.

[0064] Each AGV in the AGV cluster will report its own status and workshop environmental information during the process of executing tasks, so that the traffic congestion index in the production workshop can be calculated based on the status information and the environmental information. The congestion coefficient can be used to indicate the congestion degree of the workshop.

[0065] Thus, based on the obtained traffic congestion index, it can be determined whether the entire AGV system belongs to the severe congestion level. The corresponding relationship between the value of the traffic congestion index and the traffic operation situation is shown in Table 1.

[0066] Table 1 Classification Table of Traffic Congestion Index

[0067] Traffic operation conditions Traffic congestion index value Classification Unobstructed 0-2 Good Basically unobstructed 2-4 Relatively good Slightly congested 4-6 Poor Moderately congested 6-8 Bad Severely congested 8-10 Very bad

[0068] In step 103, based on the traffic congestion index, a charging instruction or an avoidance instruction is sent to the target AGV.

[0069] Optionally, the congestion degree of the AGV system can be judged according to the traffic congestion index, and a charging instruction or an avoidance instruction is sent to the target AGV, which may include:

[0070] When the congestion degree is severe congestion, the k-means clustering algorithm is used to cluster the positions of AGVs. Then, several AGVs are selected from the category with the largest number of AGVs after clustering as the target AGVs, and a charging instruction is sent to the target AGVs. Thus, the target AGVs can go to the designated charging station for charging according to the indication of the charging instruction. This avoids a large number of AGVs gathering in some areas and alleviates the congestion situation to a certain extent.

[0071] When the congestion degree is not severe congestion, according to the real-time power of each AGV, the AGVs with the real-time power lower than the preset power threshold are used as the target AGVs, and a charging instruction is sent to them. Thus, the target AGVs can go to the designated charging station for charging according to the indication of the charging instruction.

[0072] When the congestion degree is mild congestion, moderate congestion or severe congestion, the k-means clustering algorithm is used to cluster the positions of AGVs. Then, they are sorted in descending order according to the number of AGVs in each cluster, and the top k clusters are selected. The central AGV0 of each cluster is found, and the AGVs within the target range with this AGV0 as the center are used as the target AGVs, and an avoidance instruction is sent to the target AGVs. Thus, the target AGVs can perform operations such as decelerating, re-planning the route or stopping running according to the indication of the avoidance instruction for avoidance.

[0073] When the congestion degree is unobstructed or basically unobstructed, if the distance between any two AGVs is less than the preset distance threshold, the priorities of the tasks executed by the two AGVs can be compared, and the AGV with the lower priority is determined as the target AGV, and an avoidance instruction is sent to the target AGV. Thus, the target AGV can perform operations such as decelerating, re-planning the route or stopping running according to the indication of the avoidance instruction for avoidance.

[0074] The multi-AGV collaborative scheduling method of the digital twin platform for intelligent factories provided by the embodiments of the present invention can receive the status information sent by each AGV and the environmental information of the workshop in real time to determine the current congestion degree of the workshop, and then send a charging instruction or an avoidance instruction to the AGV according to the congestion degree, instructing the AGV to go to the charging station for charging, or adjusting the route or speed for avoidance, realizing the collaborative scheduling of multiple AGVs, fully considering the problems of insufficient power or conflicts that may occur during the scheduling process, and greatly improving the working efficiency of the AGV cluster in the production workshop.

[0075] In some embodiments, the sending of a charging instruction or an avoidance instruction to the target AGV based on the traffic congestion index includes:

[0076] Based on the traffic congestion index, when it is determined that there is congestion, cluster the positions of the AGVs to determine the first number of clusters;

[0077] Based on the first number of clusters, send a charging instruction or an avoidance instruction to the target AGV.

[0078] Optionally, when it is determined that the entire AGV system is congested according to the traffic congestion index, the congestion includes: mild congestion, moderate congestion, and severe congestion.

[0079] Use the k-means clustering algorithm to cluster the positions of the AGVs to obtain the first number of clusters. According to the first number of clusters obtained by clustering, a charging instruction or an avoidance instruction can be sent to the target AGV.

[0080] In some embodiments, the sending of a charging instruction to the target AGV based on the first number of clusters includes:

[0081] Based on the first number of clusters, determine the cluster with the largest number of AGVs within the cluster as the target cluster;

[0082] Determine the second number of AGVs from within the target cluster as the target AGV;

[0083] Send a charging instruction to the target AGV.

[0084] Figure 2 It is the second flowchart of the multi-AGV collaborative scheduling method of the digital twin platform for intelligent factories provided by the embodiments of the present invention. As Figure 2 shown, when it is determined that the congestion degree is severe congestion, according to the first number of clusters obtained by clustering, count the number of AGVs in each cluster. The larger the number, the more AGVs there are in that area. Determine the cluster with the largest number of AGVs as the target cluster.

[0085] Select the second number of AGVs from the target cluster as the target AGV, and the second number can be set according to actual needs.

[0086] Among them, the second number of AGVs in the target cluster can be randomly selected, or the second number of AGVs with lower task priorities can be selected according to the priorities of the tasks executed by each AGV in the target cluster, or the second number of AGVs with lower battery levels can be selected according to the real-time battery levels of each AGV in the target cluster as the target AGVs.

[0087] Thus, the target AGVs go to the designated charging station for charging according to the instructions of the charging command, thereby reducing the number of AGVs in this area and alleviating the congestion in the workshop.

[0088] The multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the embodiments of the present invention, by selecting target AGVs from the most congested area to go to the charging station for charging when serious congestion occurs, reduces the number of AGVs in this area, alleviates the congestion in the workshop, and improves the working efficiency of AGVs in the workshop.

[0089] In some embodiments, sending a charging command to the target AGV based on the traffic congestion index includes:

[0090] Based on the traffic congestion index, when there is no congestion, determine the target AGV based on the current battery level of the AGV;

[0091] Send a charging command to the target AGV.

[0092] Essentially, an AGV belongs to an unmanned autonomous electric vehicle with the function of transporting objects. The battery level on the AGV determines its maximum transport distance and production work, and charging planning for the AGV needs to be carried out before the battery runs out or reaches the threshold.

[0093] Especially when there are a large number of AGVs carrying out transportation and production work in the production workshop of a large factory, it is necessary to consider the charging characteristics of AGVs and the status of the production workshop, and arrange a reasonable multi-AGV charging control and scheduling strategy, which is beneficial to improving the working efficiency of AGVs in the production workshop.

[0094] Figure 3 It is the third flow chart of the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the embodiments of the present invention. The flow chart of the multi-AGV charging control and scheduling strategy proposed by the present invention is as Figure 3 shown, and the specific implementation process is as follows.

[0095] In the entire AGV cluster system, during the process of each AGV executing tasks, it will constantly monitor whether its remaining battery level is less than the minimum battery level threshold set by the system.

[0096] When it is determined according to the traffic congestion index that the entire AGV system does not have congestion or severe congestion, the target AGV is determined according to the current power of the AGV, and a charging instruction is sent to the target AGV.

[0097] Optionally, according to the current power of the AGV, the AGV with the current power less than the power threshold is used as the target AGV, and a charging instruction is sent to the target AGV. As a result, the target AGV suspends the execution of the task, chooses to go to the charging station for charging, and reports its own status information. The above is the AGV in-station threshold control strategy.

[0098] Optionally, according to the current power of the AGV, when the current power cannot complete the next task, or the remaining power after the AGV reaches the source node of the next task cannot support it to return to the charging station, the AGV is used as the target AGV, and a charging instruction is sent to the target AGV. As a result, the target AGV suspends the execution of the task, chooses to go to the charging station for charging, and reports its own status information. The above is the AGV future task power control strategy.

[0099] For the strategy of each AGV leaving the station after charging, the embodiments of the present invention also propose two strategies:

[0100] The first strategy is that when the AGV is fully charged, that is, when the power of the AGV reaches the maximum value of the battery capacity, it chooses to leave the station to execute the task only after being fully charged, that is, leaving the station according to the "charged power leaving the station strategy";

[0101] The second strategy is that since the control center calculates the task completion rate and urgency of the overall AGV system cluster, when the task completion rate is lower than the threshold or the task is relatively urgent, it can leave the station according to the actual power required when executing the next task, that is, the "charging on demand leaving the station strategy".

[0102] The multi-AGV cooperative scheduling method of the intelligent factory digital twin platform provided by the embodiments of the present invention, by monitoring the current power of each AGV in real time without congestion, enables the AGV that needs to be charged to go to the charging station for charging as soon as possible, avoids the situation of insufficient power, and improves the working efficiency of the AGV in the production workshop.

[0103] In some embodiments, sending an avoidance instruction to the target AGV based on the first number of clusters includes:

[0104] Based on the first number of clusters, with the center of each cluster as the center of the circle, and the first distance, the second distance, and the third distance as the radii, a first range, a second range, and a third range are determined; the first distance is less than the second distance, and the second distance is less than the third distance;

[0105] The AGVs within the first range, the second range, and the third range are determined as the target AGVs;

[0106] Send an avoidance instruction to the target AGV.

[0107] Figure 4 It is the fourth flow diagram of the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the embodiment of the present invention. As Figure 4 shown, optionally, when the congestion level is determined to be mild congestion, moderate congestion or severe congestion, according to the first number of clusters obtained by clustering, then sort them in descending order according to the number of AGVs in each cluster, select the top k clusters, and determine the central AGV0 of each cluster.

[0108] Taking the central AGV0 as the center and the first distance R1, the second distance R2, and the third distance R3 as the radii, where R1 < R2 < R3, determine the first range, the second range, and the third range.

[0109] Among them, the first range is a circular range determined with the central AGV0 as the center and R1 as the radius; the second range is an annular range determined with the central AGV0 as the center and R1 and R2 as the radii; the third range is an annular range determined with the central AGV0 as the center and R2 and R3 as the radii.

[0110] When other AGVs are detected within the third range, the AGVs within the third range are determined as target AGVs, and an avoidance instruction is sent to them, instructing the target AGVs to decelerate and run at v0 - rat, where r is the distance between the other AGV and the central AGV0, and a is the acceleration.

[0111] When other AGVs are detected within the second range, the AGVs within the second range are determined as target AGVs, and an avoidance instruction is sent to them, instructing the target AGVs to decelerate and run at v0 - r 2 at.

[0112] When AGV0 detects other AGVs within the first range, the AGVs within the first range are determined as target AGVs, and an avoidance instruction is sent to them, instructing the target AGVs to perform an emergency brake to stop running.

[0113] Optionally, then select the AGV with the highest task priority within the first range and start driving to leave the first range first. After the AGV with the highest task priority drives out of the first range, then sequentially select the new highest task priority stationary AGVs within the first range and start driving to leave the first range.

[0114] For AGVs with a distance r > R3 from the central AGV0, continue to execute tasks at the original speed.

[0115] Optionally, for AGVs in different ranges, the avoidance instruction carries information indicating the speed, instructing the target AGV to avoid at what speed. The specific formula is as follows:

[0116]

[0117] Among them, v represents the driving speed indicated by the avoidance instruction, v0 represents the driving speed of the target AGV, r represents the distance between the target AGV and the central AGV0, a represents the acceleration, and t represents the time.

[0118] The multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the embodiments of the present invention improves the robustness and working efficiency of the entire AGV system operation by clustering and sorting the AGV positions and using a stepped speed function for obstacle avoidance in case of congestion.

[0119] In some embodiments, sending an avoidance instruction to the target AGV based on the traffic congestion index includes:

[0120] Based on the traffic congestion index, when it is determined that there is no congestion, based on the distance between any two AGVs, a preset distance, and the task priorities corresponding to the two AGVs, determine the target AGV;

[0121] Send an avoidance instruction to the target AGV.

[0122] In the collaborative scheduling process of the multi-AGV cluster system in the intelligent factory, the most important problem is to be able to detect possible AGV path conflicts in a timely manner and how to make a reasonable obstacle avoidance strategy.

[0123] When operating normally in the case of determining that the congestion degree is non-congested, when an AGV is performing a task, the conditions for any two AGVs to collide are as follows:

[0124] d(C i (t), C j (t)) ≤ Δd safe

[0125] Among them, Δd safe represents the safety distance for collision between each AGV, and d(C i (t), C j (t)) represents the Euclidean distance between two AGVs C i and C j at time t.

[0126] d(C i (t), C j (t)) is calculated as follows:

[0127] d(C i (t), C j (t)) = sqrt((x i (t) - x i (t)) 2 + (y i (t) - y j (t)) 2 )

[0128] where (x i (t), y i (t)) represents the coordinates of the location where C i is located, and (x j (t), y j (t)) represents the coordinates of the location where C j is located.

[0129] In a production workshop with multiple AGVs, each AGV needs to execute its own transportation task. Therefore, path conflict problems are inevitable. Generally speaking, the path conflict types in a multi-AGV system can be divided into three categories: pursuit conflict, head-on conflict, and intersection conflict.

[0130] (1) Pursuit conflict means that when two AGVs are traveling in the same direction on the same path, if the moving speed of the rear vehicle is greater than that of the front vehicle, the rear vehicle will hit the front vehicle after a period of time.

[0131] For pursuit conflict, when the sensor and lidar detect that the distance between two AGVs is less than the safety distance Δd safe , first judge the task priorities of the front and rear AGVs. If the task priority of the rear vehicle is high, the speed of the front vehicle can be increased while ensuring the safety distance between the front vehicle and other vehicles. If the safety distance between the front vehicle and other vehicles cannot be guaranteed, the speed of the rear vehicle is reduced.

[0132] If the task priority of the rear vehicle is low, in order to ensure that the front vehicle can execute the task normally and the front and rear AGVs do not collide, the running speed of the rear vehicle can be reduced at this time.

[0133] (2) Head-on conflict means that when two AGVs are running on the same path and moving towards each other, a conflict occurs. When the sensor and lidar detect that the distance between two AGVs is less than the safety distance Δd safe , select the AGV with the lower task priority to re-plan the route and bypass the AGV with the higher task priority on the opposite side.

[0134] (3) Intersection conflict means that when two AGVs are traveling on different routes and in different directions, due to the existence of intersection points between the routes, when the two AGVs reach the intersection path point at the same time, an intersection conflict will occur.

[0135] Determine whether two AGVs meet the following conditions:

[0136]

[0137] Among them, v1 and v2 respectively represent the speeds of two AGVs, l represents the distance between the two AGVs is l, and t safe represents the safety time between the two AGVs.

[0138] If satisfied, stop the operation of the low-priority AGV. Otherwise, the low-priority AGV regards the high-priority AGV as a moving obstacle and performs dynamic path planning to achieve dynamic obstacle avoidance.

[0139] The multi-AGV cooperative scheduling method of the intelligent factory digital twin platform provided by the embodiments of the present invention, when the multi-AGV system operates normally under non-congested conditions, proposes corresponding scheduling strategies including waiting, speed reduction, and detouring for chasing conflicts, head-on conflicts, and cross conflicts through the task priorities of AGVs, thereby improving the robustness and working efficiency of the entire AGV system operation.

[0140] In some embodiments, before receiving the status information sent by each AGV and the environmental information of the workshop, it further includes:

[0141] Construct a scheduling model; the scheduling model includes: a first model for determining the driving distance of the AGV, a second model for determining the scheduling time, and a third model for determining the AGV utilization rate;

[0142] Based on the analytic hierarchy process, respectively determine the weight coefficients of the first model, the second model, and the third model;

[0143] Based on the weight coefficients, merge the first model, the second model, and the third model into a target model;

[0144] Solve the target model to determine the target tasks assigned to each AGV.

[0145] Optionally, the established scheduling model is as follows:

[0146]

[0147] Among them, f1 is the first model, f2 is the second model, f2 is the third model, m represents the number of AGVs, n represents the number of shelves, and d ij represents the distance between the i-th shelf and the j-th shelf, and x ijk represents whether there is a k-th AGV between the i-th shelf and the j-th shelf, and x ijk takes a value of 0 or 1, and x ijkA value of 0 indicates that there is no k-th AGV between the i-th shelf and the j-th shelf, x ijk A value of 1 indicates that there is a k-th AGV between the i-th shelf and the j-th shelf, t represents the time when the last task is completed, a i represents the average utilization rate of each AGV.

[0148] The scheduling model is a multi-AGV task multi-objective scheduling model. The first model represents minimizing the total travel distance of the entire AGV system; the second model represents minimizing the completion time of the entire AGV system; the second model represents maximizing the average utilization rate of AGVs in the entire AGV system.

[0149] According to theoretical analysis, relevant evaluation indicators are selected, and the analytic hierarchy process is used to calculate the respective weights of the above three models, determine the weight coefficients of the first model, the second model, and the third model respectively, merge the multi-objective scheduling model into a target model, and perform planning and solution for a single-objective multi-constraint model.

[0150] Again, a grid search is performed with a step size of 0.05 for each weight coefficient, and then a genetic algorithm is used to obtain multiple groups of optimal alternative solutions.

[0151] Finally, each group of optimal alternative solutions is scored and sorted using the ideal solution method to obtain the final solution, and the tasks assigned to each AGV in the AGV cluster system and the execution order are obtained through decoding.

[0152] Optionally, Figure 5 is the fifth flowchart of the multi-AGV collaborative scheduling method of the digital twin platform for intelligent factories provided by the embodiments of the present invention. As Figure 5 shown, in order to achieve multi-AGV collaborative scheduling and planning for an intelligent factory, the present invention first establishes a new multi-AGV task multi-objective scheduling model.

[0153] Secondly, according to theoretical analysis, relevant evaluation indicators are selected, and the analytic hierarchy process is used to calculate the weights of the above three objectives to obtain weight coefficients, and the multi-objective scheduling model is merged into a single-objective multi-constraint model for planning and solution.

[0154] Again, a grid search is performed with a step size of 0.05 for each weight coefficient, and then a genetic algorithm is used to obtain multiple groups of optimal alternative solutions.

[0155] Finally, each group of optimal alternative solutions is scored and sorted using the ideal solution method to obtain the final solution, and the tasks assigned to each AGV in the AGV cluster system and the execution order are obtained through decoding.

[0156] The multi-AGV collaborative scheduling method of the digital twin platform for intelligent factories provided by the embodiments of the present invention plans by establishing a new multi-AGV task multi-objective scheduling model, taking three objectives (minimizing the total driving distance of the entire AGV system; minimizing the completion time of the entire AGV system; maximizing the average utilization rate of AGVs in the entire AGV system) as collaborative scheduling objectives.

[0157] For the solution of the optimization problem of this NP-HARD problem, a brand-new solution method is proposed. That is: First, establish a new multi-AGV task multi-objective scheduling model. Secondly, according to theoretical analysis, select relevant evaluation indicators, and use the analytic hierarchy process to calculate the weights of the above three objectives to obtain weight coefficients, and merge the multi-objective scheduling model into a single-objective multi-constraint model for planning and solving. Thirdly, use 0.05 as the step size for each weight coefficient, perform grid search, and then use the genetic algorithm to obtain multiple groups of optimal alternative solutions.

[0158] Finally, score and rank each group of optimal alternative solutions using the ideal solution method to obtain the final solution, and obtain the tasks assigned to each AGV in the AGV cluster system and the execution order through decoding. This solution method can better ensure that the finally obtained solution can meet the requirements of the three planning objectives, and can better conform to people's cognition and evaluation.

[0159] The system control structure determines the communication method, control method, task allocation method and coordination method of the AGV cluster system, and is the key issue for the normal operation of large-scale AGV systems. By deeply studying the operation characteristics of the AGV cluster system for material transportation in intelligent factories and combining the existing software and hardware environment factors, a new system control structure is proposed, which can greatly improve the collaborative scheduling and operation efficiency of the multi-AGV system, and further improve the intelligence and automation efficiency of mobile phone factories.

[0160] Due to the traditional centralized system control network structure, once a single control center fails, it will cause the entire AGV cluster scheduling system to collapse. Moreover, with the increase in the number of AGVs and the number of scheduling tasks in the AGV cluster system, this will bring an exponential increase in the computing volume of the centralized control center, greatly consuming resources and reducing production scheduling efficiency.

[0161] On the other hand, due to problems in aspects such as resource allocation and network communication mechanisms in the traditional distributed system control network structure, it is also easy to cause problems such as cluster failures and network delays, resulting in a decline in factory production efficiency.

[0162] The present invention combines the advantages of the traditional centralized system structure and the traditional distributed system structure through the control idea of hierarchical and functional division, while avoiding the disadvantages of these two traditional system control structures.

[0163] Figure 6 is a schematic structural diagram of the hybrid network collaborative control system provided by the embodiments of the present invention. As Figure 6 shown, the task allocation and scheduling of the entire AGV cluster system are uniformly processed and calculated by the control center, which itself is also implemented by a distributed primary and standby computer cluster. At the same time, the control center monitors and receives the status information and data reported by each decentralized AGV system in the AGV system in real time, performs calculation processing on this information, and then sends the calculation results and instructions to each AGV that needs to be coordinated and scheduled.

[0164] For example, by building a Flink distributed real-time computing cluster, this function of the control center can be achieved. Specifically, as Figure 5 shown, the solution process of the multi-AGV task collaborative scheduling model can be implemented using the distributed real-time computing cluster of the control center.

[0165] In addition, each decentralized AGV robot in the AGV system also has its own monitoring and calculation functions, such as being equipped with lidar, cameras, and sensors, etc. At the same time, each AGV is also equipped with a CPU with a certain computing power and a storage unit to perform tasks such as its own path planning.

[0166] During the process of self-path planning and driving of each distributed AGV, it sends its own status information to the control center while receiving the status information or instructions of the overall AGV cluster sent by the control center. Each AGV combines the information sent by the control center and the status information obtained by its own sensors to real-time plan its own path and control strategy (such as obstacle avoidance strategy and charging strategy). Figure 7 is a schematic diagram of the operation process of the hybrid network collaborative control system provided by the embodiments of the present invention. The overall hybrid network collaborative scheduling operation process is as Figure 7 shown.

[0167] The hybrid network collaborative control system provided by the embodiments of the present invention combines the advantages of the traditional centralized system structure and the traditional distributed system structure through a hierarchical and functional control idea, while avoiding the disadvantages of these two traditional system control structures, and proposes a new hybrid network collaborative control system structure. The task allocation and scheduling of the entire AGV cluster system are uniformly processed and calculated by the control center. The control center itself is also implemented by a distributed primary and backup computer cluster. At the same time, the control center listens and receives the status information and data reported by each decentralized AGV system in the AGV system in real time, and performs calculation and processing on this information, and then sends the calculation results and instructions to each AGV that needs to be coordinated and scheduled. In particular, it is proposed that such a function of the control center can be achieved by building a Flink distributed real-time computing cluster. Each decentralized AGV robot in the AGV system also has its own monitoring and calculation functions. During the process of its own path planning and driving, each AGV sends its own status information to the control center while receiving the status information or instructions of the overall AGV cluster sent by the control center. Each AGV plans its own path and control strategy in real time by combining the information sent by the control center and the status information obtained by its own sensors.

[0168] Figure 8 It is a schematic diagram of the multi-AGV collaborative scheduling architecture based on the digital twin platform provided by the embodiments of the present invention. As Figure 8 shown, the physical system is a real system, including a production workshop, AGV vehicles, and various sensors and radar detection devices.

[0169] The physical system uses sensors, lidar, etc. to collect real-time operation data, physical system parameters and other data and upload them to the digital twin platform control center. At the same time, it receives the instructions transmitted by the digital twin platform control center and operates according to the instructions.

[0170] The virtual system is a virtual model established according to the physical system and is the twin of the physical system. Its main functions include establishing a virtual system model, generating an initial AGV scheduling plan (including task allocation plan, charging strategy plan, conflict avoidance plan, etc.), and at the same time simulating the operation of the AGV scheduling plan through the virtual system of the digital twin platform and receiving the data fed back by the physical system to adjust the scheduling plan in real time.

[0171] The virtual system model is constructed and continuously updated and maintained according to the virtual system data and real-time operation data of the digital twin platform (including parameters and index data of various devices and scheduling plans). The data in the virtual system is obtained by processing the environmental and device data, parameters and index data in the physical system.

[0172] The operating states of the physical devices and virtual devices in the multi-AGV collaborative scheduling digital twin platform proposed by the present invention belong to a symbiotic and virtual-real mapping relationship.

[0173] The state changes of the devices in the physical system (including the task assignment scheme of AGVs, the battery power and charging strategy of AGVs, and the conflict state and obstacle avoidance strategy of AGVs, etc.) will affect the digital twin modeling and simulation results of the virtual system; conversely, the changes in the digital twin model and simulation results in the virtual system will also change the operating states of the devices in the physical system.

[0174] On the other hand, the simulation results of AGV collaborative scheduling in the virtual system and the actual operating results of AGVs in the physical system are mutually iterative and promotive throughout the AGV movement process.

[0175] First, the simulation results of the virtual system will guide the operation of the devices in the physical system, and update the operation instructions and collaborative scheduling planning results accordingly according to the development of the movement process, the change of time, and the change of environmental factors; second, the physical devices will continuously feedback data such as the movement state and environmental information to the virtual system, and the virtual system will simultaneously update and maintain elements such as the parameters, models, and methods of the virtual system, thereby affecting the simulation results of the multi-AGV collaborative scheduling scheme.

[0176] Since the multi-AGV system for a smart factory is a complex production system, aiming at the implementation and monitoring of the various collaborative scheduling technical solutions proposed by the present invention and the problems of unclear system performance and low optimization degree in the distributed and centralized scheduling of the multi-AGV cluster system, the present invention proposes a method of building a digital twin platform through a hybrid network collaborative control system structure and virtual-real model mapping to simulate and control the technical solutions proposed by the present invention, and simultaneously monitor the operation of various indicators and parameters.

[0177] The multi-AGV collaborative scheduling device of the smart factory digital twin platform provided by the present invention will be described below. The multi-AGV collaborative scheduling device of the smart factory digital twin platform described below can be mutually referred to the multi-AGV collaborative scheduling method of the smart factory digital twin platform described above.

[0178] Figure 9 It is a schematic structural diagram of the multi-AGV collaborative scheduling device of the smart factory digital twin platform provided by an embodiment of the present invention. As Figure 9 shown, the multi-AGV collaborative scheduling device of the smart factory digital twin platform provided by an embodiment of the present invention includes:

[0179] A receiving module 910, configured to receive the status information sent by each AGV and the environmental information of the workshop;

[0180] A determination module 920, configured to determine a traffic congestion coefficient based on the state information and the environmental information; the congestion coefficient is used to indicate the degree of congestion between vehicles.

[0181] A scheduling module 930, configured to send a charging instruction or an avoidance instruction to a target AGV based on the traffic congestion index.

[0182] It should be noted here that the multi-AGV collaborative scheduling device of the intelligent factory digital twin platform provided by the embodiments of the present invention can implement all the method steps implemented by the above-mentioned multi-AGV collaborative scheduling method embodiment of the intelligent factory digital twin platform, and can achieve the same technical effects. The same parts and beneficial effects as those in the method embodiment will not be specifically described in this embodiment.

[0183] Optionally, the scheduling module 930 is specifically configured to:

[0184] Based on the traffic congestion index, when it is determined that there is congestion, cluster the positions of the AGVs to determine a first number of clusters;

[0185] Based on the first number of clusters, send a charging instruction or an avoidance instruction to the target AGV.

[0186] Optionally, the scheduling module 930 is specifically configured to:

[0187] Based on the first number of clusters, determine the cluster with the largest number of AGVs within the cluster as the target cluster;

[0188] Determine a second number of AGVs within the target cluster as the target AGV;

[0189] Send a charging instruction to the target AGV.

[0190] Optionally, the scheduling module 930 is specifically configured to:

[0191] Based on the first number of clusters, with the center of each cluster as the center of a circle, and with a first distance, a second distance, and a third distance as the radii, determine a first range, a second range, and a third range; the first distance is less than the second distance, and the second distance is less than the third distance;

[0192] Determine the AGVs within the first range, the second range, and the third range as the target AGV;

[0193] Send an avoidance instruction to the target AGV.

[0194] In some embodiments, the scheduling module 930 is specifically configured to:

[0195] Based on the traffic congestion index, when there is no congestion, determine the target AGV based on the current power of the AGV.

[0196] Send a charging instruction to the target AGV.

[0197] In some embodiments, the scheduling module 930 is specifically configured to:

[0198] Based on the traffic congestion index, when there is no congestion, determine the target AGV based on the distance between any two AGVs, a preset distance, and the task priorities corresponding to the two AGVs.

[0199] Send an avoidance instruction to the target AGV.

[0200] In some embodiments, the multi-AGV collaborative scheduling device of the intelligent factory digital twin platform further includes: a distribution module, configured to:

[0201] Construct a scheduling model; the scheduling model includes: a first model for determining the driving distance of the AGV, a second model for determining the scheduling time, and a third model for determining the AGV utilization rate.

[0202] Based on the analytic hierarchy process, respectively determine the weight coefficients of the first model, the second model, and the third model.

[0203] Based on the weight coefficients, merge the first model, the second model, and the third model into a target model.

[0204] Solve the target model to determine the target tasks assigned to each AGV.

[0205] Figure 10 Illustrates a schematic physical structure diagram of an electronic device, as Figure 10 shown. The electronic device may include: a processor 1010, a communication interface 1020, a memory 1030, and a communication bus 1040. Among them, the processor 1010, the communication interface 1020, and the memory 1030 communicate with each other through the communication bus 1040. The processor 1010 can call the logical instructions in the memory 1030 to execute the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform, and the method includes: receiving the status information sent by each AGV and the environmental information of the workshop; based on the status information and the environmental information, determining a traffic congestion coefficient; the congestion coefficient is used to indicate the congestion degree of the workshop; based on the traffic congestion index, sending a charging instruction or an avoidance instruction to the target AGV.

[0206] In addition, when the logical instructions in the above-mentioned memory 1030 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0207] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the above-mentioned various methods. The method includes: receiving the status information sent by each AGV and the environmental information of the workshop; based on the status information and the environmental information, determining a traffic congestion coefficient; the congestion coefficient is used to indicate the congestion degree of the workshop; based on the traffic congestion index, sending a charging instruction or an avoidance instruction to the target AGV.

[0208] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform provided by the above-mentioned various methods. The method includes: receiving the status information sent by each AGV and the environmental information of the workshop; based on the status information and the environmental information, determining a traffic congestion coefficient; the congestion coefficient is used to indicate the congestion degree of the workshop; based on the traffic congestion index, sending a charging instruction or an avoidance instruction to the target AGV.

[0209] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0210] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0211] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-AGV collaborative scheduling method for a digital twin platform of an intelligent factory, characterized in that, Including: Receiving the status information sent by each AGV and the environmental information of the workshop; Determining a traffic congestion index based on the status information and the environmental information; The traffic congestion index is used to indicate the congestion degree of the workshop; Based on the traffic congestion index, when congestion occurs, clustering the positions of AGVs to determine the first number of clusters; Based on the first number of clusters, sending a charging instruction or an avoidance instruction to a target AGV; The sending a charging instruction to a target AGV based on the first number of clusters includes: Based on the first number of clusters, determining the cluster with the largest number of AGVs within the cluster as the target cluster; Determining a second number of AGVs from within the target cluster as the target AGV; Sending a charging instruction to the target AGV; The sending an avoidance instruction to a target AGV based on the first number of clusters includes: Based on the first number of clusters, taking the center of each cluster as the center of a circle, and using a first distance, a second distance, and a third distance as radii to determine a first range, a second range, and a third range; the first distance is less than the second distance, and the second distance is less than the third distance; Determining the AGVs within the first range, the second range, and the third range as the target AGV; Sending an avoidance instruction to the target AGV; The sending a charging instruction to a target AGV based on the traffic congestion index includes: Based on the traffic congestion index, when no congestion occurs, determining the target AGV based on the current power of the AGV; Sending a charging instruction to the target AGV; The sending an avoidance instruction to a target AGV based on the traffic congestion index includes: Based on the traffic congestion index, when no congestion occurs, determining the target AGV based on the distance between any two AGVs, a preset distance, and the task priorities corresponding to the two AGVs; Sending an avoidance instruction to the target AGV.

2. The multi-AGV collaborative scheduling method of the digital twin platform of the intelligent factory according to claim 1, wherein Before receiving the status information sent by each AGV and the environmental information of the workshop, it further includes: Constructing a scheduling model; the scheduling model includes: a first model for determining the travel distance of an AGV, a second model for determining the scheduling time, and a third model for determining the utilization rate of an AGV; Based on the analytic hierarchy process, respectively determining the weight coefficients of the first model, the second model, and the third model; Based on the weight coefficients, combining the first model, the second model, and the third model into a target model; Solving the target model to determine the target tasks assigned to each AGV.

3. A multi-AGV collaborative scheduling device for a digital twin platform of an intelligent factory, characterized in that, Including: A receiving module for receiving the status information sent by each AGV and the environmental information of the workshop; A determining module for determining a traffic congestion index based on the status information and the environmental information; the traffic congestion index is used to indicate the congestion degree of the workshop; A scheduling module for sending a charging instruction or an avoidance instruction to a target AGV based on the traffic congestion index; The scheduling module is specifically used for: Based on the traffic congestion index, when congestion occurs, clustering the positions of AGVs to determine the first number of clusters; Based on the first number of clusters, send a charging instruction or an avoidance instruction to the target AGV; Based on the first number of clusters, determine the cluster with the largest number of AGVs within the cluster as the target cluster; Determine a second number of AGVs from within the target cluster as the target AGVs; Send a charging instruction to the target AGV; Based on the first number of clusters, with the center of each cluster as the center of a circle, and with the first distance, the second distance, and the third distance as the radii, determine the first range, the second range, and the third range; the first distance is less than the second distance, and the second distance is less than the third distance; Determine the AGVs within the first range, the second range, and the third range as the target AGVs; Send an avoidance instruction to the target AGV; Based on the traffic congestion index, when it is determined that there is no congestion, determine the target AGV based on the current battery level of the AGV; Send a charging instruction to the target AGV; Based on the traffic congestion index, when it is determined that there is no congestion, determine the target AGV based on the distance between any two AGVs, a preset distance, and the task priorities corresponding to the two AGVs; Send an avoidance instruction to the target AGV.

4. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform as described in claim 1 or 2.

5. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform as described in claim 1 or 2.

6. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the multi-AGV collaborative scheduling method of the intelligent factory digital twin platform as described in claim 1 or 2.

Images