An AGV scheduling method and storage medium based on task bidding mechanism

By introducing a task bidding mechanism in the AGV scheduling system, AGV is allowed to calculate and report task costs by itself, and transfer or exchange before task binding, it solves the problem that the overload of calculation load and time costs of the upper computer in the existing technology cannot be dynamically adjusted. The problem has improved the efficiency and rationality of task allocation.

CN114330831BActive Publication Date: 2024-11-12KUNMING KSEC LOGISTIC INFORMATION IND
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202111438010.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-30
Publication Date
2024-11-12
Estimated Expiration
2041-11-30

AI Technical Summary

Technical Problem

The existing AGV scheduling system cannot dynamically adjust the time cost of AGV to the starting point of the task due to the heavy calculation load of the upper computer, resulting in the inability to optimize the task allocation.

Method used

AGV scheduling method based on the task bidding mechanism is adopted, and bidding is initiated through the superior scheduling system. Each AGV calculates the task cost by itself and informs the superior scheduling system and other AGVs. The AGV with the lowest cost is selected as the winning candidate and allows binding to the task. Before transfer or exchange tasks to dynamically adjust task allocation.

Benefits of technology

It greatly reduces the load of the upper-level scheduling system, improves the efficiency and rationality of task allocation, and ensures the speed of task completion.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114330831B_ABST
    Figure CN114330831B_ABST
Patent Text Reader

Abstract

The present invention discloses an AGV scheduling method and storage medium based on a task bidding mechanism, the method comprising: S1, a superior scheduling system sends a bidding notice to each AGV; S2, each AGV calculates its own task cost according to the bidding notice, and notifies its own task cost to the superior scheduling system and other AGVs; S3, an AGV with the lowest task cost is selected as a candidate AGV for winning the bid, and is marked as a pre-winning state; S4, the superior scheduling system gives the execution authority of the bidding notice to the candidate AGV for winning the bid according to the bidding result, and modifies its pre-winning state to a winning state; S5, the AGV in the winning state is bound to the corresponding task and executes the task; the task cost is calculated by each AGV itself, which greatly reduces the load of the superior scheduling system, and adopts a bidding method to call AGVs with lower costs to execute tasks, thereby ensuring the rationality of task allocation and facilitating improving the task completion speed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of AGV scheduling, and in particular to an AGV scheduling method and storage medium based on a task bidding mechanism. Background Art

[0002] In the field of industrial logistics, the Automated Guided Vehicle (AGV) is a highly efficient unmanned transport device that has been applied in many industries and will play an indispensable role in logistics automation systems in the future. However, the existing AGV dispatching system integrates AGV path planning, task management, traffic management and other functions, and is deployed in the host computer or server. Each AGV has no communication connection with each other and cannot exchange data. Therefore, the AGV itself is only responsible for actuator control and motion control, and the degree of intelligence is relatively low.

[0003] In the existing AGV system, the allocation method between AGVs and tasks is: the AGV scheduling system deployed in the host computer calculates the time cost of each idle AGV to reach the starting point of the task, selects an idle AGV with the lowest time cost to perform the task, and calculates and allocates a path for the AGV that performs the task. All calculations are completed by the scheduling system in the host computer, resulting in an increasingly heavy load on the host computer, especially in large-scale AGV scheduling systems. More importantly, the time cost of each AGV to the starting point of the task calculated by the host scheduling system is only an instantaneous time cost, which cannot be dynamically adjusted over time or the operating status of each AGV, resulting in the inability to optimize the allocation of tasks. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the present invention provides an AGV scheduling method based on a task bidding mechanism, in which a superior scheduling system is used to initiate bidding, and each AGV independently calculates the cost of executing its own task to bid, which greatly reduces the load on the superior computer and is conducive to improving work efficiency.

[0005] According to a first aspect, the present invention provides an AGV scheduling method based on a task bidding mechanism, comprising: a host scheduling system, and a plurality of interconnected AGVs, each AGV being connected to the host scheduling system, and the method comprising:

[0006] S1: The upper dispatching system sends the bidding notice to each AGV;

[0007] S2: Each AGV calculates its own task cost according to the bidding notice, and notifies its own task cost to the upper scheduling system and other AGVs;

[0008] S3: The AGV with the lowest task cost is selected as the candidate AGV for winning the bid and marked as the pre-winning state;

[0009] S4: The upper scheduling system gives the execution authority of the bidding notice to the winning candidate AGV according to the bidding results, and changes its pre-winning status to the winning status;

[0010] S5: After the AGV in the bid-winning state is bound to the corresponding task, it executes the task;

[0011] The bidding notice includes: task number, task attributes, task priority, required number of AGVs, task start time and operation list;

[0012] The task cost refers to the time cost required for the AGV to complete the operation list in the bidding notice.

[0013] Furthermore, the task cost includes static cost and dynamic cost; the static cost includes: the theoretical time cost required for the AGV driving path, the predictable acceleration and deceleration time cost of the AGV during driving, and other time costs spent on performing operations; the task dynamic cost includes the unpredictable time cost caused by obstacles, traffic jams, AGV failures and human intervention during AGV driving.

[0014] Furthermore, in step S2, the theoretical time cost C theorypath for:

[0015]

[0016] Among them, n is the total number of path segments that need to be traversed, L i is the distance of each path segment, V i is the speed of each path segment, is the calibration factor for each path segment.

[0017] Furthermore, step S3 also includes:

[0018] S31: The AGV that becomes the candidate AGV for winning the bid simultaneously broadcasts a pre-bid winning notification to other AGVs and the upper scheduling system;

[0019] S32: The AGV status is updated in real time. The AGV that receives the pre-bid notification determines whether to raise an objection based on the updated status. If an objection is raised, the upper-level scheduling system and the winning candidate AGV are notified at the same time, and the upper-level scheduling system organizes a second bidding; if there is no objection, enter step S4.

[0020] Furthermore, in step S32, the conditions for determining whether to raise an objection are:

[0021] In the current state, if the task cost of the AGV to perform the task is better than the task cost of the winning candidate AGV, an objection will be raised; otherwise, no objection will be raised.

[0022] Furthermore, the AGVs are connected to each other through communication; each AGV contains the same list of tasks that are currently active in the AGV system, and each AGV calculates the cost of executing tasks that have not yet been bound in the task list at regular intervals and notifies other AGVs;

[0023] The step S5 further comprises:

[0024] S51: Before the AGV in the bid winning state is bound to the task, the AGV in the bid winning state exchanges tasks or transfers tasks with other AGVs;

[0025] S52: After the AGV in the bid winning state is bound to the task, the AGV and the task are inseparable until the task is completed.

[0026] Further, in step S51, task exchange occurs between two AGVs in the bid-winning state. The first bid-winning AGV and the second bid-winning AGV must meet the following conditions:

[0027] C1>C 12 +C swap

[0028] C2>C 21 +C swap

[0029] In the formula, C1 represents the cost of the first winning AGV to perform its own winning task;

[0030] C 12 It represents the cost of the first winning AGV to perform the task of the second winning AGV;

[0031] C2 represents the cost of the second winning AGV to perform its own winning task;

[0032] C 21 represents the cost of the second winning AGV to perform the task of the first winning AGV;

[0033] C swap represents the cost of task exchange;

[0034] Task transfer occurs between an AGV in the bid winning state and an AGV in the bid losing state, and the following conditions must be met:

[0035] C1>C 31 +C transfer

[0036] Among them, C 31 It represents the cost of the unsuccessful AGV to perform the task of the successful AGV;

[0037] C1 represents the cost of the winning AGV to perform its own task;

[0038] C transfer Represents the cost of task transfer.

[0039] According to a second aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the program can be executed by a processor to implement the steps of the method described above.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] (1) In the AGV scheduling method based on the task bidding mechanism provided by the present invention, the task cost is calculated by each AGV itself, which greatly reduces the load of the upper scheduling system.

[0042] (2) Each AGV still calculates the real-time task cost after winning the bid. Before the task is bound, it can transfer or exchange the task to an AGV with lower cost, which ensures the rationality of task allocation and helps to improve the speed of task completion. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flow chart of the AGV scheduling method based on the task bidding mechanism provided by the present invention;

[0044] Figure 2 Flowchart of the bidding mechanism for tasks. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0046] Example 1

[0047] like Figure 1-2 As shown, the present invention provides an AGV scheduling method based on a task bidding mechanism, including: an upper scheduling system, a plurality of AGVs that are mutually communicatively connected based on 5G communication technology, each AGV being communicatively connected to the upper scheduling system, and the method comprising:

[0048] S1: The upper scheduling system sends a bidding notice to each AGV, triggering the task bidding process of each AGV. The bidding notice includes: task number, task attributes, task priority, required number of AGVs, task start time, and operation list; the operation list includes multiple operations, each of which includes: operation station, operation code, and operation code attributes; the operation code attributes include: operation code type, actuator lifting height, speed; the operation code type includes: asynchronous operation and synchronous operation.

[0049] S2: Each AGV parses the bidding notice, identifies the task type, and determines the legitimacy of the order. Then, each AGV calculates the task cost of executing the task in the current state and notifies the upper scheduling system and other AGVs of its own task cost. That is, the AGV bids at an open reserve price.

[0050] The task cost of AGV is derived from the combination of static cost and dynamic cost. Specifically, the static cost includes the theoretical time cost required for AGV to travel the path after completing path planning, the predictable acceleration and deceleration time cost of AGV during driving, and the time cost of executing the operation. Among them, the theoretical time cost is:

[0051]

[0052] Among them, n is the total number of path segments that need to be traversed, L i is the distance of each path segment, V i is the speed of each path segment, is the calibration factor for each path segment.

[0053] The task dynamic cost includes the unpredictable time cost caused by obstacles, traffic jams, AGV failures, human intervention, etc. during the AGV's driving process. Most of the task dynamic costs are unpredictable costs. The task dynamic cost prediction will be accumulated with reference to the average time spent by each AGV at a certain node. For example, traffic jams take an average of two minutes at a certain node. When the AGV performing the task encounters traffic jams at this node, the task dynamic cost will increase by two minutes. Human intervention takes an average of three minutes at a certain path segment. When the AGV performing the task encounters human intervention at this path segment, the task dynamic cost will accumulate three minutes. The task dynamic cost is accumulated and predicted in this way.

[0054] When there is little interference from external conditions, the static cost is the determining factor of the task cost. The dynamic cost is an auxiliary factor. During the movement process, more reference will be made to the dynamic cost. According to the obstacles, traffic jams, AGV failures, human intervention, etc. reported by the AGV, real-time judgments will be made to predict the task execution cost, and this will be used as the bidding cost.

[0055] S3: Each AGV autonomously sorts its own task costs, and based on the principle of lowest cost winning the bid, selects the AGV with the lowest cost as the candidate AGV for winning the bid, and marks it as the pre-winning state.

[0056] S31: The AGV that becomes the candidate AGV for winning the bid broadcasts a preliminary winning notice to other AGVs and the upper scheduling system at the same time.

[0057] S32: The AGV status is updated in real time. In principle, the AGV that receives the pre-bid notification agrees to the pre-bid task. However, if the task cost of the AGV in its current state is better than the task cost of the candidate AGV, it will raise an objection and notify the upper scheduling system and the candidate AGV at the same time. Then the upper scheduling system will re-initiate a second bid for the task.

[0058] S4: If there is no objection, the upper-level dispatching system will give the execution authority of the bidding notice to the winning candidate AGV based on the bidding results, and change its pre-winning status to the winning status.

[0059] Each AGV contains the same list of active task orders in the current AGV system. Each AGV calculates the cost of executing tasks that are not yet bound in the task list at regular intervals.

[0060] Before the winning AGV is bound to the corresponding task, if the cost of any AGV to perform the task is lower, that is, if the following conditions are met, the task will be transferred:

[0061] C1>C 31 +C transfer

[0062] Among them, C 31 represents the cost of the unsuccessful AGV to perform the task of the successful AGV; C1 represents the cost of the successful AGV to perform its own task; C transfer Represents the cost of task transfer.

[0063] In addition, before the tasks are bound, if the time cost saved by two winning AGVs after exchanging their winning tasks is greater than the time cost loss caused by task exchange, the exchange will be carried out if the following conditions are met:

[0064] C1>C 12 +C swap

[0065] C2>C 21 +C swap

[0066] Where C1 represents the cost of the first winning AGV to perform its own winning task; C 12 represents the cost of the first winning AGV to perform the task of the second winning AGV; C2 represents the cost of the second winning AGV to perform its own winning task; C 21 represents the cost of the second winning AGV to perform the task of the first winning AGV; C swap Represents the cost of task exchange.

[0067] S5: After the AGV in the bid winning state is bound to the corresponding task, the AGV in the bid winning state executes the task according to the task order.

[0068] Example 2

[0069] If AGV1 wins the bid for task 1 and AGV2 wins the bid for task 4, the task list at this time is shown in Table 1.

[0070] AGV1 won the bid for task 1 and was pre-winning the bid for task 2. AGV has won the bid for at most one task and pre-winning one task. AGV3 has won the bid for task 3, and AGV2 has won the bid for task 4 and was pre-winning the bid for task 5.

[0071] Table 1

[0072]

[0073] After a period of time, when Task 1 is completed by AGV1, the pre-bid Task 2 of AGV1 is converted into a bid-winning task, as shown in Table 2 below.

[0074] Table 2

[0075]

[0076] Before AGV1 is bound to Task 2, that is, before AGV reaches the operation point, Task 3 is not bound to AGV3. At this time, as shown in Tables 3 and 4, since the cost of AGV1 to perform Task 3 is 13, which is lower than the cost of AGV3 to perform Task 3, which is 16, and the cost of AGV3 to perform Task 2 is 8, which is also lower than the cost of AGV1 to perform Task 2, which is 15, and the cost difference between the two is greater than the time loss caused by task exchange, the tasks of AGV1 and AGV3 can be directly exchanged at this time. The updated task table is shown in Table 5.

[0077] Table 3

[0078]

[0079] Table 4

[0080]

[0081] Table 5

[0082]

Claims

1. An AGV scheduling method based on task bidding mechanism, comprising: A host scheduling system and a plurality of AGVs that are in communication connection with each other, each AGV being in communication connection with the host scheduling system; characterized in that the method comprises: S1: The upper dispatching system sends the bidding notice to each AGV; S2: Each AGV calculates its own task cost according to the bidding notice, and notifies its own task cost to the upper scheduling system and other AGVs; S3: The AGV with the lowest task cost is selected as the candidate AGV for winning the bid and marked as the pre-winning state; S4: The upper scheduling system gives the execution authority of the bidding notice to the winning candidate AGV according to the bidding results, and changes its pre-winning status to the winning status; S5: The AGV in the winning state starts to move to the target station to perform the task; The bidding notice includes: task number, task attributes, task priority, required number of AGVs, task start time and operation list; The task cost refers to the time cost required for the AGV to complete the operation list in the bidding notice; Step S3 also includes: S31: The AGV that becomes the candidate AGV for winning the bid simultaneously broadcasts a pre-bid winning notification to other AGVs and the upper scheduling system; S32: The AGV status is updated in real time. The AGV that receives the pre-bid notification determines whether to raise an objection based on the updated status. If an objection is raised, the upper scheduling system and the candidate AGV for the bid are notified at the same time, and the upper scheduling system organizes a second bidding. If there is no objection, the process proceeds to step S4. AGVs communicate with each other; each AGV contains the same list of tasks that are currently active in the AGV system. Each AGV calculates the cost of executing tasks that have not yet been bound in the task list at regular intervals and notifies other AGVs. The step S5 further comprises: S51: Before the AGV in the bid winning state is bound to the task, the AGV in the bid winning state exchanges tasks or transfers tasks with other AGVs; S52: After the AGV in the bid winning state is bound to the task, the AGV and the task are inseparable until the task is completed.

2. The method according to claim 1, characterized in that The task cost includes static cost and dynamic cost; the static cost includes: the theoretical time cost required for the AGV driving path, the foreseeable acceleration and deceleration time cost of the AGV during driving, and other time costs spent on performing operations; the dynamic cost includes the unpredictable time cost caused by obstacles, traffic jams, AGV failures and human intervention during the AGV driving process.

3. The method according to claim 2, characterized in that Theoretical time cost for: Among them, n is the total number of path segments that need to be traversed, L i is the distance of each path segment, V i is the speed of each path segment, is the calibration factor for each path segment.

4. The method according to claim 1, characterized in that In step S32, the conditions for determining whether to raise an objection are: In the current state, if the task cost of the AGV to perform the task is better than the task cost of the winning candidate AGV, an objection will be raised; otherwise, no objection will be raised.

5. The method according to claim 1, characterized in that In step S51, task exchange occurs between two AGVs in the bid-winning state. The first bid-winning AGV and the second bid-winning AGV must meet the following conditions: In the formula, It represents the cost of the first winning AGV to perform its own winning task; It represents the cost of the first winning AGV to perform the task of the second winning AGV; It represents the cost of the second winning AGV to perform its own winning task; represents the cost of the second winning AGV to perform the task of the first winning AGV; represents the cost of task exchange; Task transfer occurs between an AGV in the bid winning state and an AGV in the bid losing state, and the following conditions must be met: in, It represents the cost of the unsuccessful AGV to perform the task of the successful AGV; It represents the cost of the winning AGV to perform its own task; Represents the cost of task transfer.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program can be executed by a processor to implement the steps of the method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • AGV intelligent dynamic scheduling method in workshop environment

    CN106249738A

  • Construction method of intelligent module of manufacturing system, manufacturing system and manufacturing method

    CN109164762A

  • AGV path optimization algorithm

    CN109978232A

  • Real-time dispatching method for automatic production of various types of hybrid lines

    CN110456746A