AGV forklift cooperative deployment operation method and system

Abstract

The invention provides an AGV forklift cooperative deployment operation method and system, and the method comprises the steps: obtaining the initial positions of all AGV forklifts in a forklift queue and a task sequence, and the task sequence comprises at least one to-be-executed task and corresponding initial task information; calculating the initial urgency degree of the task according to the initial task information, and preferentially bringing the task with the high initial urgency degree into a multi-objective optimization model; state information of all AGV forklifts is obtained, a multi-objective function including total completion time, energy consumption cost and load balance degree is established based on the state information, and a distribution scheme is solved through an improved NSGA-III algorithm; the distribution scheme comprises AGV forklifts and corresponding task execution sequences; and according to a set algorithm, a global path from the initial position to the target point is planned for each AGV forklift, and the AGV forklifts are controlled to operate. According to the method provided by the invention, the overall operation efficiency and the intelligent scheduling level of the AGV forklift system can be effectively improved.

Description

Technical Field

[0001] This invention generally relates to the field of AGV forklift technology, and specifically to a method and system for the collaborative deployment and operation of AGV forklifts. Background Technology

[0002] As the core equipment of automated logistics systems, AGV forklifts realize automatic handling and stacking of goods through autonomous navigation and intelligent control. They are widely used in warehousing, manufacturing and other fields. Their efficient operation is of great significance to improving logistics efficiency and reducing labor costs, and has become an indispensable key equipment in modern automated production and warehouse management.

[0003] Currently, there is an increasing number of technologies for autonomous navigation and path planning of AGV forklifts, but there are few methods for task allocation among multiple forklifts and multiple tasks. In existing technologies, the task allocation and scheduling methods for AGV forklifts are simple, which has problems such as delayed response to emergency tasks, low overall system efficiency, and uneven load distribution among AGVs leading to equipment overload or idleness, making it difficult to meet the high-efficiency operation requirements in complex scenarios. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide an AGV forklift collaborative deployment and operation method and system to solve the above problems.

[0005] The first aspect of this invention provides a method for the collaborative deployment and operation of AGV forklifts, comprising: Obtain the initial positions and task sequences of all AGV forklifts in the forklift queue, wherein the task sequence includes at least one task to be executed and the corresponding initial task information; The initial urgency of the tasks is calculated based on the initial task information. The tasks are then prioritized based on the initial urgency, and tasks with high initial urgency are included in the multi-objective optimization model first. The status information of all AGV forklifts is obtained, and a multi-objective function including total completion time, energy consumption cost, and load balancing degree is established based on the status information. The allocation scheme is solved by the improved NSGA-Ⅲ algorithm. The allocation scheme includes each AGV forklift and its corresponding task execution sequence. According to the set algorithm, a global path from the initial position to the target point is planned for each AGV forklift, and the AGV forklift is controlled to run according to the allocation scheme and the global path.

[0006] According to the technical solution provided by the present invention, the multi-objective function is as follows: , in, Indicates the total completion time. Indicates task The start time of execution. Indicates task The execution time, Indicates the number of tasks; , in, Indicates energy consumption cost, Indicates task Total load, Indicates task driving distance, Indicates that the AGV forklift is performing a task. The speed at which it travels; , in, Indicates load balancing degree. Indicates the number of AGV forklifts. Indicates the first The total load of tasks in the current task execution sequence of the forklift. express The standard deviation of the load of the forklift. This indicates the average load.

[0007] According to the technical solution provided by the present invention, after planning a global path from the initial position to the target point for each AGV forklift according to a set algorithm, and controlling the AGV forklift to run according to the allocation scheme and the global path, the method further includes: Update the initial task information for each task to obtain real-time task information, and calculate the real-time urgency based on the real-time task information; Tasks with a real-time urgency level greater than or equal to a first preset threshold and less than a second preset threshold are selected to establish a first adjustment set; the first adjustment set includes at least one first target task to be prioritized. The first target tasks are sorted from highest to lowest according to their real-time urgency to update the allocation scheme.

[0008] According to the technical solution provided by the present invention, the step of sorting the first target tasks from high to low according to their real-time urgency to update the allocation scheme includes: Filter AGV forklifts that are idle, are executing tasks with a real-time urgency level less than the first preset threshold, or are not locked and are executing tasks with a urgency level greater than or equal to the first preset threshold, and establish a first candidate set; Perform executability verification on the AGV forklifts in the first candidate set; Select all AGV forklifts that are executable, and determine the first target forklift to be executed based on the first target task after sorting by the preset optimization target; The first target task is assigned to the first target forklift, and the task execution sequence of the first target forklift is adjusted to update the assignment scheme.

[0009] According to the technical solution provided by the present invention, the executability verification of the AGV forklifts in the first candidate set includes: Determine whether the AGV forklift meets the set execution conditions; the set execution conditions include the maximum load capacity being greater than or equal to the load weight of the task, the current remaining power being greater than or equal to the estimated power consumption of the task, and the estimated time being less than or equal to the set percentage of the remaining planned time of the task.

[0010] According to the technical solution provided by the present invention, all AGV forklifts with executable capabilities are selected, and a first target forklift for execution is determined by combining the first target task after sorting based on a preset optimization objective, including: Select feasible AGV forklifts and calculate the matching objective function between each AGV forklift and the first target task: , in, Represents the objective function value. This represents the normalized total time taken to complete the first objective task. This represents the normalized total energy consumption for completing the first objective task; The AGV forklift with the smallest objective function is selected as the first target forklift.

[0011] According to the technical solution provided by the present invention, the step of assigning the first target task to the first target forklift and adjusting the task execution sequence of the first target forklift to update the assignment scheme includes: If it is determined that the first target forklift is not currently executing a task, the first target task is inserted at the beginning of its task execution sequence, and the execution status is updated. When it is determined that the first target forklift is executing a task with a real-time urgency level less than the first preset threshold, the first target task is inserted into the subsequent position of the currently executing task, and the execution status is updated.

[0012] According to the technical solution provided by the present invention, after updating the initial task information of each task to obtain real-time task information and calculating the real-time urgency based on the real-time task information, the method further includes: Tasks with a real-time urgency level greater than or equal to the second preset threshold are selected to establish a second adjustment set; the second adjustment set includes at least one second target task to be preempted. AGV forklifts that are currently executing tasks with a real-time urgency level less than the first preset threshold and a time redundancy level greater than or equal to the third preset threshold are selected to establish a second candidate set; the time redundancy level is the ratio of the remaining planned time to the remaining estimated time of the second target task. The executability of the AGV forklifts in the second candidate set is verified. The AGV forklift with executability and the shortest remaining time for the current task is selected as the second target forklift. The second target forklift is then controlled to interrupt the currently executing task and execute the second target task. When it is determined that the interrupted task has insufficient remaining planned time due to the interruption, it is included in the first adjustment set to update the allocation scheme; If it is determined that the interrupted task has sufficient remaining planned time due to the interruption, after the second target task is completed, the second target forklift is controlled to return to the interruption position of the interrupted task and resume execution based on the saved state.

[0013] According to the technical solution provided by the present invention, the method further includes: When a new task is inserted into the task sequence, the initial urgency of the new task is calculated; If the initial urgency of the new task is less than the first preset threshold, the new task is inserted at the end of the task sequence and assigned to the idle AGV forklift with the lowest current load for execution. If the initial urgency of the new task is determined to be greater than or equal to the first preset threshold and less than the second preset threshold, the new task is included in the first adjustment set to trigger the priority adjustment; When the initial urgency of the new task is determined to be greater than the second preset threshold, the new task is included in the second adjustment set to trigger task preemption. At the same time, the allocation scheme is optimized based on the remaining tasks by improving the NSGA-III algorithm.

[0014] A second aspect of the present invention provides an AGV forklift collaborative deployment and operation system for performing the AGV forklift collaborative deployment and operation method described above, the system comprising: The parameter acquisition module is configured to acquire the initial position and task sequence of all AGV forklifts in the forklift queue. The task sequence includes at least one task to be executed and the corresponding initial task information. An initial calculation module is configured to calculate the initial urgency of a task based on the initial task information, prioritize tasks based on the initial urgency, and include tasks with high initial urgency into the multi-objective optimization model first. A forklift scheduling module is configured to acquire the status information of all AGV forklifts, establish a multi-objective function based on the status information including total completion time, energy consumption cost, and load balancing degree, and solve the allocation scheme through an improved NSGA-Ⅲ algorithm; the allocation scheme includes each AGV forklift and its corresponding task execution sequence. The planning and execution module is configured to plan a global path from the initial position to the target point for each AGV forklift according to a set algorithm, and control the AGV forklift to run according to the allocation scheme and the global path.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By acquiring the initial position and task sequence of the AGV forklifts, and calculating and sorting the initial urgency of tasks based on the initial task information, high-urgency tasks are prioritized for inclusion in the optimization model, enabling priority response to critical tasks and improving the timeliness of handling urgent tasks; a multi-objective function is constructed based on AGV status information, including total completion time, energy consumption cost, and load balancing, making the objective optimization more comprehensive. This reduces the overall task completion time while lowering the total system energy consumption and balancing the load of each AGV, preventing individual devices from being overloaded or idle; an improved NSGA-Ⅲ algorithm is used to solve the allocation scheme, which can efficiently obtain the optimal task matching relationship and execution order. Combined with global path planning, this achieves the coordinated and efficient operation of the AGV forklift group, significantly improving the overall operating efficiency, economy, and coordination of the system. The above technical features work together to form a complete collaborative mechanism from task priority division and multi-objective optimization allocation to path planning execution, comprehensively improving the overall operating efficiency and intelligent scheduling level of the AGV forklift system. Attached Figure Description

[0016] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of the steps of the AGV forklift collaborative deployment and operation method provided in Example 1; Figure 2 This is a schematic diagram of the AGV forklift collaborative deployment and operation system provided in Example 2.

[0017] The reference numerals are as follows: 10, Parameter Acquisition Module; 20, Initial Calculation Module; 30, Forklift Scheduling Module; 40, Planning Execution Module. Detailed Implementation

[0018] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Example 1 Please refer to Figure 1 This embodiment provides a method for the collaborative deployment and operation of AGV forklifts, including: S100: Obtain the initial position and task sequence of all AGV forklifts in the forklift queue, wherein the task sequence includes at least one task to be executed and the corresponding initial task information.

[0021] Specifically, each forklift queue includes multiple AGV forklifts. Each AGV forklift has a unique number and is parked at its corresponding charging station when not performing a task. After receiving a task execution instruction, it moves from the charging station to the target point.

[0022] In step S100, when the system starts execution, the initial position of each AGV forklift in the forklift queue can be determined by its number. The system receives a task sequence, which may include one or more tasks. Each task carries corresponding initial task information, including: task ID, task type, task weight, travel distance, operation time, and deadline. The task type includes at least outbound, inbound, inventory, and urgent outbound orders. Different task types correspond to different task weights, which characterize the degree of impact of different task types on urgency. The travel distance represents the Euclidean distance between the starting point and the target point of the task. The operation time represents the actual time the forklift spends operating the task, i.e., the time occupied by the mechanical operation process during loading, unloading, and handling. The deadline represents the final time node by which the task must be completed. In this embodiment, the initial task information is illustrated in Table 1 below: Table 1

[0023] S200: Calculate the initial urgency of the task based on the initial task information, prioritize the tasks based on the initial urgency, and include tasks with high initial urgency into the multi-objective optimization model first.

[0024] Specifically, in step S200, after obtaining the initial task information, the initial urgency of each task is calculated based on the initial task information; the initial urgency is calculated using the following formula (I): Formula (1); in, Indicates the initial urgency level; This indicates the estimated time for the task, which is the total estimated time from the start to the completion of the task (including travel time and operation time. The travel time is calculated based on the travel distance and the preset benchmark travel speed. In this embodiment, the benchmark travel speed is 1.22m / s). This represents the total planned time, which is the maximum allowed completion period for the task (deadline - planned start time). Indicates task weight, and In this embodiment, the weighting coefficient is represented. It is 0.6. It is 0.4.

[0025] Using the data in Table 1 as an example, the initial urgency levels of the eight tasks are obtained sequentially. These eight tasks are then sorted according to their initial urgency levels, resulting in: (0.958) > (0.877) > (0.698) > (0.525) > (0.500) > (0.492) > (0.421) > (0.410); The sorted tasks are incorporated into the multi-objective optimization model in order of their initial urgency, so that the model prioritizes the allocation of tasks with high initial urgency.

[0026] S300: Obtain the status information of all AGV forklifts, establish a multi-objective function based on the status information including total completion time, energy consumption cost and load balancing degree, and solve the allocation scheme by improving the NSGA-Ⅲ algorithm; the allocation scheme includes each AGV forklift and its corresponding task execution sequence.

[0027] Specifically, in step S300, the status information of all AGV forklifts in the forklift queue is first obtained. This status information includes: load capacity, travel speed, and battery level. In this embodiment, the status information of the AGV forklifts is illustrated in Table 2 below: Table 2

[0028] After obtaining the status information, a multi-objective function is created based on the total completion time, energy consumption cost, and load balancing. The multi-objective function is represented as follows: Formula (II) in, Indicates the total completion time. Indicates task The start time of execution. Indicates task The execution time, Indicates the number of tasks; ; Formula (3) in, Indicates energy consumption cost, Indicates task Total load, Indicates task driving distance, Indicates that the AGV forklift is performing a task. The speed at which it travels; Formula (IV) in, Indicates load balancing degree. Indicates the number of AGV forklifts. Indicates the first The total load of tasks in the current task execution sequence of the forklift. express The standard deviation of the load of the forklift. This indicates the average load.

[0029] Using the initial task information given in Table 1 and the status information of the AGV forklift given in Table 2 as examples, the total completion time, energy consumption cost and load balance of the AGV forklift for each task are calculated through a multi-objective function.

[0030] The allocation scheme is obtained by solving the multi-objective function using the improved NSGA-III algorithm. The specific execution steps of the improved NSGA-III algorithm are not detailed here. The execution task sequence and total task load for each AGV forklift are obtained sequentially to generate the allocation scheme, which is denoted as the initial allocation scheme. The initial allocation scheme is shown in Table 3 below: Table 3

[0031] S400: Based on the set algorithm, plan a global path from the initial position to the target point for each AGV forklift, and control the AGV forklift to run according to the allocation scheme and the global path.

[0032] Specifically, in step S400, after obtaining the initial allocation scheme, path planning is performed based on the initial and target points of the assigned tasks for each AGV forklift to obtain a global path. In this embodiment, an improved A* algorithm (or a similar heuristic path search algorithm) is used for path planning. After obtaining the global path, multiple AGV forklifts are controlled to start executing tasks according to the global path.

[0033] Furthermore, after step S400, the following steps are also included: S500: Update the initial task information of each task to obtain real-time task information, and calculate the real-time urgency based on the real-time task information.

[0034] Specifically, after a task begins execution, the initial task information for each task changes over time, which in turn may cause the urgency of each task to change. Therefore, it is necessary to update the initial task information.

[0035] Specifically, in step S500, the initial task information for each task is updated according to a set time interval. This includes updating data such as whether the task is completed, the task start time, and the remaining time until the task deadline. The set time interval can be 5s, 10s, or other durations. After the update, the real-time task information for each task is obtained. After obtaining the real-time task information, the real-time urgency of each task is calculated based on the real-time task information. The calculation of the real-time urgency is as follows (Five): Formula (5); in, Indicates the real-time urgency level. This indicates the remaining estimated time for the task, that is, the estimated time for the unfinished portion of the task. (Subtract the execution time) This indicates the remaining planned time, that is, the remaining allowed time for the task. (Subtract the running time) , and Same as the initial urgency calculation formula.

[0036] The real-time urgency of each unfinished task can be calculated using formula (5) and real-time task information. The real-time urgency reflects the current urgency of each task and is used to monitor situations where the urgency of a task spikes due to a fault in an AGV forklift or other issues, thus preventing the task from being completed before its deadline.

[0037] S600: Filter tasks whose real-time urgency is greater than or equal to a first preset threshold and less than a second preset threshold, and establish a first adjustment set; the first adjustment set includes at least one first target task to be prioritized.

[0038] Specifically, in step S600, after calculating the real-time urgency of each task, the real-time urgency of each task is compared with a first preset threshold and a second preset threshold. If the real-time urgency is less than the first preset threshold, it indicates that the task has a low urgency level and belongs to a regular or low-urgency task with relatively relaxed time constraints. It can continue to be executed according to the initial allocation scheme. If the real-time urgency is greater than or equal to the first preset threshold and less than the second preset threshold, it indicates that the task belongs to a medium-urgency task and has a priority upgrade request. It needs to be executed in a timely manner through priority adjustment, that is, the "priority adjustment" mechanism is triggered. This type of task is used as the first target task and a first adjustment set is established. All tasks in the first adjustment set need to have their priorities adjusted through the "priority adjustment" mechanism. The first adjustment set may include one first target task or multiple first target tasks.

[0039] S700: Sort the first target tasks from high to low according to the real-time urgency to update the allocation scheme.

[0040] Specifically, in step S700, when the first adjustment set includes multiple first target tasks, all first target tasks are sorted according to their real-time urgency. The initial allocation scheme is then adjusted based on the re-sorting result, and the initial allocation scheme is updated to obtain the adjusted allocation scheme. By calculating the real-time urgency after task execution begins and updating the allocation scheme according to the real-time urgency, the scheme can be adjusted based on real-time anomalies occurring during task execution, achieving dynamic scheduling. This ensures that high-urgency tasks are not disturbed, while also increasing the priority of medium-urgency tasks to prevent them from timeouts. This enhances the system's adaptability to real-time changes during task execution, further optimizing scheduling flexibility and task completion timeliness.

[0041] Furthermore, step S700 specifically includes: S710: Filter AGV forklifts that are in an idle state, are executing tasks with a real-time urgency level less than the first preset threshold, or are not locked and are executing tasks with a urgency level greater than or equal to the first preset threshold, and establish a first candidate set.

[0042] Specifically, the core purpose of step S710 is to select suitable AGV forklifts for the first target task in the first adjustment set, ensuring that they can obtain sufficient execution resources during the priority adjustment process without interfering with the normal execution of high-urgency tasks. Specifically, the first target task must be executed before low-urgency tasks. Directly selecting AGVs randomly may result in: occupying AGV forklifts currently executing high-urgency tasks, causing task conflicts; or missing reusable idle AGV forklifts or AGV forklifts executing low-urgency tasks, leading to resource waste.

[0043] Specifically, in step S710, three types of AGV forklifts need to be selected. First, AGV forklifts in an idle state are selected as candidate forklifts. An idle state means that no task is currently being executed and there is no task execution sequence. Second, the real-time urgency of the task currently being executed by each AGV forklift is obtained. If the real-time urgency is less than a first preset threshold, the AGV forklift is selected as a candidate forklift. Next, the task execution sequence of each AGV forklift is read. If there is no task in the task execution sequence with a real-time urgency greater than or equal to the first preset threshold, the AGV forklift is selected as a candidate forklift. Finally, by judging each of the three conditions one by one, the candidate forklifts selected by the three conditions are summarized to form a first candidate set, which provides a basic resource pool for subsequent executability verification and target AGV forklift selection.

[0044] S720: Perform executability verification on the AGV forklifts in the first candidate set.

[0045] Specifically, step S720 is the core step in ensuring that the first target task in the first adjustment set can be effectively executed. The first candidate set has already selected AGV forklifts, but these AGV forklifts may be unable to actually execute the task due to their own hardware limitations, such as insufficient load capacity, insufficient power, or time conflicts (inability to complete the task before the deadline). Directly assigning unverified AGV forklifts may lead to timeouts or equipment failures, thus reducing the system's execution efficiency. Therefore, executability verification eliminates unsuitable AGV forklifts in advance, providing a reliable resource foundation for subsequent task allocation.

[0046] Furthermore, step S720 specifically includes: Determine whether the AGV forklift meets the set execution conditions; the set execution conditions include the maximum load capacity being greater than or equal to the load weight of the task, the current remaining power being greater than or equal to the estimated power consumption of the task, and the estimated time being less than or equal to the set percentage of the remaining planned time of the task.

[0047] Specifically, in step S720, three conditions are used to determine whether the AGV forklift can perform the first target task: For condition one, the total load of the first target task is first obtained, and the rated load of the AGV forklift is also obtained. If the maximum load capacity of the AGV forklift is greater than or equal to the total load of the first target task, then this condition is met. For condition two, the estimated power consumption of the first target task is first calculated based on the following formula (vi): Formula (VI) in, Indicates the estimated power consumption; The energy consumption coefficient is preset by the equipment parameters; While calculating the estimated power consumption using formula (VI), the battery level of the AGV forklift is also obtained. If the remaining battery level is greater than or equal to the estimated power consumption of the task, then this condition is met. Regarding condition three, firstly, the estimated time for the first target task is calculated using the aforementioned calculation method. and remaining planned time The condition is met when the estimated time is less than or equal to a set percentage of the remaining planned time. The set percentage is used to assess the sufficiency of the AGV forklifts in completing the task. The selection of the set percentage needs to take into account factors such as the urgency of the task, the stability of AGV operation, and the overall efficiency of the system. If the set percentage is too high, although it can strictly screen out AGV forklifts with almost no risk of delay in terms of time, it may result in too few AGV forklifts available, increasing the difficulty of task allocation. If the set percentage is too low, it may allow some AGVs that barely meet the time requirement but have a greater risk of delay to participate in task execution, affecting the on-time completion rate of the task.

[0048] AGV forklifts that meet all three conditions are considered feasible and are retained, while AGV forklifts that do not meet any of the conditions are excluded.

[0049] S730: Select all AGV forklifts with executable capabilities, and determine the first target forklift to be executed based on the first target task after sorting by the preset optimization target.

[0050] Specifically, the core objective of step S730 is to select the "optimally matched" AGV forklift from the AGV forklifts that have passed the executability verification for the first target task in the first adjustment set. After the executability verification in step S720, the remaining AGV forklifts in the first candidate set all meet the basic execution conditions of the task, but different AGVs have different optimization indicators such as efficiency and energy consumption when performing the task. If they are directly randomly assigned, it may lead to waste of resources (such as using a high-energy-consuming AGV to perform a simple task) or low task execution efficiency (such as using a low-speed AGV forklift to perform a time-sensitive task). Therefore, through the objective function calculation and screening in step S730, the optimal AGV forklift can be matched for the first target task, achieving multi-objective optimization.

[0051] Furthermore, step S730 specifically includes: S731: Select AGV forklifts that are feasible, and calculate the matching objective function between each AGV forklift and the first target task: , in, Represents the objective function value. This represents the normalized total time taken to complete the first objective task. This represents the normalized total energy consumption for completing the first objective task. This represents the time weighting parameter; for high-urgency scenarios, The value can be selected from 0.7 to 0.9; for medium-urgency scenarios, The value can be selected as 0.5~0.6; for low-urgency scenarios, The value can be selected as 0.3~0.4.

[0052] Specifically, in step S731, the total time to complete the first target task is first calculated. The total time is the sum of the time it takes for the AGV forklift to complete the current task and the task execution time. The sum of these values ​​is then normalized to the [0,1] interval to obtain the result. Simultaneously, the total energy consumption for completing the first objective task is calculated. The total energy consumption is the product of the total task load, the travel distance, and the square of the AGV forklift's travel speed. Then, the total energy consumption is normalized to the [0,1] interval to obtain... .

[0053] The degree of matching between each AGV forklift and the objective function of the first objective task is calculated based on the matching objective function.

[0054] S732: Select the AGV forklift with the smallest objective function as the first target forklift.

[0055] Specifically, in step S732, all calculated matching objective function values ​​are collected, and all matching objective function values ​​are numerically compared. The AGV forklift with the smallest matching objective function value is selected as the first target forklift, and it is temporarily locked to prevent it from being repeatedly assigned by other tasks. If multiple AGV forklifts have the same matching objective function value, their normalized total time is further compared, and the AGV forklift with the smaller matching objective function value is selected as the first target forklift.

[0056] S740: Assign the first target task to the first target forklift and adjust the task execution sequence of the first target forklift to update the assignment scheme.

[0057] Specifically, in step S740, after the first target forklift is determined according to the above steps, the task execution sequence of the first target forklift is adjusted according to the first target task. The priority adjustment of the medium urgency task is changed from "scheme design" to "actual execution", which directly improves the system's response capability and scheduling accuracy to dynamic tasks.

[0058] Furthermore, step S740 specifically includes: S741: If it is determined that the first target forklift is not currently executing a task, insert the first target task at the beginning of its task execution sequence and update the execution status.

[0059] Specifically, before inserting the first target task into the task execution sequence of the first target forklift, it is necessary to distinguish the state of the first target forklift in order to determine the task execution order. In step S741, when it is determined that the first target forklift is not currently executing a task, the first target task is inserted at the beginning of its task execution sequence to update the state. The first target forklift can then execute according to the updated task execution sequence.

[0060] S742: When it is determined that the first target forklift is executing a task with a real-time urgency level less than the first preset threshold, the first target task is inserted into the subsequent position of the currently executing task, and the execution status is updated.

[0061] Specifically, in step S742, when it is determined that the first target forklift is performing a task and the task being performed is a low-urgency task, the first target is inserted into its task execution sequence at the next position after the task being performed to update the status. The first target forklift can then execute according to the updated task execution sequence after completing the current task.

[0062] Furthermore, after step S500, it also includes S810: Filter tasks whose real-time urgency is greater than or equal to the second preset threshold, and establish a second adjustment set; the second adjustment set includes at least one second target task to be preempted.

[0063] Specifically, in step S810, after calculating the real-time urgency of each task, the real-time urgency of each task is compared with the first preset threshold and the second preset threshold. If the real-time urgency is greater than the second preset threshold, it indicates that the task has a high degree of urgency and belongs to the high-urgency task category. The "priority adjustment" mechanism can no longer guarantee timely completion of this type of task, so the "task preemption" mechanism is triggered. This type of task is used as the second target task and a second adjustment set is established. All tasks in the second adjustment set must be prioritized through the "task preemption" mechanism. The second adjustment set may include one second target task or multiple second target tasks.

[0064] S820: Filter AGV forklifts that are currently executing tasks with a real-time urgency level less than the first preset threshold and a time redundancy level greater than or equal to the third preset threshold, and establish a second candidate set; the time redundancy level is the ratio of the remaining planned time to the remaining estimated time of the second target task.

[0065] Specifically, the core purpose of step S820 is to select AGV resources that can be preempted for the second target task with high urgency in the second adjustment set, so as to ensure that the "task preemption" mechanism can both prioritize the response of high urgency tasks and minimize the impact on interrupted tasks.

[0066] Specifically, in step S820, firstly, the real-time urgency of the currently executed tasks of all AGV forklifts is obtained. AGV forklifts with a real-time urgency of their current tasks less than a first preset threshold are prioritized for selection, while AGV forklifts currently executing medium-urgency or high-urgency tasks are excluded to avoid interfering with critical tasks. Secondly, the time redundancy of the currently executed tasks of the selected AGV forklifts is calculated. The time redundancy is compared with a third preset threshold, and AGV forklifts with a time redundancy greater than or equal to the third preset threshold are retained to ensure that there is still sufficient time to complete the current task even if it is interrupted. A second candidate set is then established based on the AGV forklifts that meet the conditions. Optionally, the third preset threshold is set to 1.5.

[0067] By selecting AGVs for tasks with low urgency and high time redundancy, it ensures that high-urgency tasks can preempt available resources while avoiding interrupted tasks from being unable to recover due to preemption. This prioritizes responding to urgent needs while minimizing interference with the overall system scheduling, providing a precise and reliable resource foundation for the "task preemption" mechanism.

[0068] S830: Perform executability verification on the AGV forklifts in the second candidate set, select the AGV forklift with executability and the shortest remaining time for the currently executing task as the second target forklift, and control the second target forklift to interrupt the currently executing task and execute the second target task.

[0069] Specifically, in step S830, firstly, in the obtained second candidate set, the executability of each AGV forklift in the second candidate set is verified, and the verification method is the same as in step S720 above. AGV forklifts that meet the three conditions are reserved as preemptible AGV forklifts. Next, among all preemptible AGV forklifts, the one with the shortest remaining time for the current task is selected as the second target forklift. If there are AGV forklifts with the same time, their distances to the second target task are further compared, and the one closer is selected. Finally, the second target forklift is immediately interrupted from its current task execution, the task interruption position and state are saved, and an execution command for the second target task is issued to the second target forklift, prioritizing its dispatch to the starting point of the second target task. At this time, the system synchronously updates the status of the second target forklift to "execute a high-urgency task".

[0070] Execution verification ensures that high-urgency tasks can be effectively executed. By selecting the AGV forklift with the shortest remaining execution time, the delay loss of interrupted tasks is minimized. While prioritizing the response to high-urgency needs, the interference with the original scheduling pairs is reduced, achieving a balance between efficiency and overall stability for urgent tasks.

[0071] S840: When it is determined that the interrupted task has insufficient remaining planned time due to the interruption, it is included in the first adjustment set to update the allocation scheme.

[0072] Specifically, step S840 addresses scenarios where interrupted tasks have insufficient remaining planned time due to preemption, aiming to prevent complete task failure. When the second target forklift interrupts the current task to execute a high-urgency task, the remaining time of the interrupted task may be insufficient to meet the completion requirements due to delays. If not handled promptly, this can lead to task backlog or failure.

[0073] Specifically, in step S840, the remaining planned time and remaining estimated time of the interrupted task are calculated. If the remaining planned time is less than the remaining estimated time, it is determined that there is insufficient time. The interrupted task is marked as "to be adjusted" and included in the first adjustment set to participate in priority ranking along with medium-urgency tasks. By re-including interrupted tasks with insufficient time in the scheduling, task failure due to preemption is avoided, ensuring closed-loop management of all tasks by the system. At the same time, the "priority adjustment" mechanism of the first adjustment set balances the scheduling order of the resumption of interrupted tasks with other tasks.

[0074] S850: If it is determined that the remaining planned time of the interrupted task is sufficient due to the interruption, after the second target task is completed, control the second target forklift to return to the interruption position of the interrupted task and resume execution based on the saved state.

[0075] Specifically, step S850 is for scenarios where the remaining planned time for the interrupted task is sufficient. The purpose is to resume the execution of the original task after the high-urgency task is completed, so as to avoid resource waste and task duplication. If the execution is not resumed, it will lead to the waste of resources invested in the already executed part, and other AGV forklifts need to be allocated, which will increase the system burden.

[0076] Specifically, in step S850, the remaining planned time and remaining estimated time of the interrupted task are calculated. If the remaining planned time is greater than or equal to the remaining estimated time, it is determined that there is sufficient time. The interruption location, execution progress, equipment status, and other information of the interrupted task are recorded and stored in the system cache. After the second target forklift completes the second target task, the system automatically calls the cached status information to control it to return to the interruption location and continue the execution of the interrupted task from the breakpoint. By restoring interrupted tasks with sufficient time, duplicate task allocation and resource waste are avoided, the continuity of task execution is ensured, and the additional cost of system rescheduling is reduced.

[0077] Furthermore, the method also includes: S910: When a new task is inserted into the task sequence, the initial urgency of the new task is calculated.

[0078] Specifically, in step S910, when a new task is inserted into the task sequence, its initial task information is extracted. The initial task information is consistent with the task information format in Table 1. Then, the initial urgency of the new task is calculated using formula (i) in step S200, and the initial urgency of the new task is passed to subsequent steps S920-S940 to determine the scheduling priority of the new task.

[0079] S920: When the initial urgency of the new task is less than the first preset threshold, the new task is inserted at the end of the task sequence and assigned to the idle AGV forklift with the lowest current load for execution.

[0080] Specifically, in step S920, firstly, if the initial urgency of a new task is less than a first preset threshold, it is determined to be a low-urgency task; then, the new task is inserted at the end of the task sequence and waits for scheduling according to the conventional order of "first come, first served"; next, the AGV forklifts that are currently idle are screened, and the one with the lowest current load is selected from the idle AGV forklifts. Here, the load is calculated according to the total weight of the tasks already allocated in its corresponding task execution sequence; finally, the new task is assigned to the AGV forklift, and its task execution sequence is updated.

[0081] S930: When it is determined that the initial urgency of the new task is greater than or equal to the first preset threshold and less than the second preset threshold, the new task is included in the first adjustment set to trigger the priority adjustment.

[0082] Specifically, in step S930, if the initial urgency of the new task is confirmed to be greater than or equal to the first preset threshold and less than the second preset threshold, it is determined to be a medium urgency task. Then, the new task is added to the first adjustment set so that it participates in the "priority adjustment" mechanism with other medium urgency tasks. The logic of steps S720-S742 is called to screen AGV forklifts for the new task, verify its executability, determine the first target forklift, and insert the new task into the task execution sequence of the first target forklift according to the rules.

[0083] S940: When the initial urgency of the new task is determined to be greater than the second preset threshold, the new task is included in the second adjustment set to trigger task preemption, and the allocation scheme is optimized based on the remaining tasks by improving the NSGA-Ⅲ algorithm.

[0084] Specifically, in step S940, when the initial urgency of the new task is confirmed to be greater than the second preset threshold, it is determined to be a high-urgency task; then the new task is added to the second adjustment set and scheduled as the second target task "task preemption" mechanism; then the logic of steps S820-S830 is called to determine the second target forklift for executing the new task, control it to interrupt the current task and allocate the new task, and simultaneously save the state of the interrupted task.

[0085] The reason for optimizing the allocation scheme based on the remaining tasks by improving the NSGA-Ⅲ algorithm at the end of step S940 is that the preemption process triggered by new tasks will break the global optimality of the original allocation scheme, and local adjustments may cause a "chain imbalance". Full recalculation can re-optimize based on the real-time status of all remaining tasks to ensure the global objective is optimal. This step can be consistent with the multi-objective collaborative optimization objective of the initial allocation, avoiding problems caused by local adjustments. In addition, the background asynchronous running algorithm can take into account the real-time response of high-urgency new tasks and the overall system optimization, maintaining the overall multi-objective optimality of the system.

[0086] Example 2 Please refer to Figure 2 This embodiment provides an AGV forklift collaborative deployment and operation system for executing the AGV forklift collaborative deployment and operation method as described in Embodiment 1. The system includes: The parameter acquisition module 10 is configured to acquire the initial position and task sequence of all AGV forklifts in the forklift queue. The task sequence includes at least one task to be executed and the corresponding initial task information. An initial calculation module 20 is configured to calculate the initial urgency of a task based on the initial task information, prioritize tasks based on the initial urgency, and include tasks with high initial urgency into the multi-objective optimization model first. The forklift scheduling module 30 is configured to acquire the status information of all AGV forklifts, establish a multi-objective function based on the status information including total completion time, energy consumption cost, and load balancing degree, and solve the allocation scheme by an improved NSGA-Ⅲ algorithm; the allocation scheme includes each AGV forklift and its corresponding task execution sequence. The planning and execution module 40 is configured to plan a global path from the initial position to the target point for each AGV forklift according to a set algorithm, and control the AGV forklift to run according to the allocation scheme and the global path.

[0087] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention is not limited to the specific combination of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this invention.

Claims

1. A method for collaborative deployment and operation of AGV forklifts, characterized in that, include: Obtain the initial positions and task sequences of all AGV forklifts in the forklift queue, wherein the task sequence includes at least one task to be executed and the corresponding initial task information; The initial urgency of the tasks is calculated based on the initial task information. The tasks are then prioritized based on the initial urgency, and tasks with high initial urgency are included in the multi-objective optimization model first. The status information of all AGV forklifts is obtained, and a multi-objective function including total completion time, energy consumption cost, and load balancing degree is established based on the status information. The allocation scheme is solved by the improved NSGA-Ⅲ algorithm. The allocation scheme includes each AGV forklift and its corresponding task execution sequence. According to the set algorithm, a global path from the initial position to the target point is planned for each AGV forklift, and the AGV forklift is controlled to run according to the allocation scheme and the global path.

2. The AGV forklift collaborative deployment and operation method according to claim 1, characterized in that, The multi-objective function is as follows: ， in, Indicates the total completion time. Indicates task The start time of execution. Indicates task The execution time, Indicates the number of tasks; ， in, Indicates energy consumption cost, Indicates task Total load, Indicates task driving distance, This indicates that the AGV forklift is performing a task. The speed at which it travels; ， in, Indicates load balancing degree. Indicates the number of AGV forklifts. Indicates the first The total load of tasks in the current task execution sequence of the forklift. express The standard deviation of the load of the forklift. This indicates the average load.

3. The AGV forklift collaborative deployment and operation method according to claim 2, characterized in that, After planning a global path from the initial position to the target point for each AGV forklift according to a set algorithm, and controlling the AGV forklift to run according to the allocation scheme and the global path, the method further includes: Update the initial task information for each task to obtain real-time task information, and calculate the real-time urgency based on the real-time task information; Tasks with a real-time urgency level greater than or equal to a first preset threshold and less than a second preset threshold are selected to establish a first adjustment set; the first adjustment set includes at least one first target task to be prioritized. The first target tasks are sorted from highest to lowest according to their real-time urgency to update the allocation scheme.

4. The AGV forklift collaborative deployment and operation method according to claim 3, characterized in that, The step of sorting the first target tasks from high to low according to their real-time urgency to update the allocation scheme includes: Filter AGV forklifts that are idle, are executing tasks with a real-time urgency level less than the first preset threshold, or are not locked and are executing tasks with a urgency level greater than or equal to the first preset threshold, and establish a first candidate set; Perform executability verification on the AGV forklifts in the first candidate set; Select all AGV forklifts that are executable, and determine the first target forklift to be executed based on the first target task after sorting by the preset optimization target; The first target task is assigned to the first target forklift, and the task execution sequence of the first target forklift is adjusted to update the assignment scheme.

5. The AGV forklift collaborative deployment and operation method according to claim 4, characterized in that, The executability verification of the AGV forklifts in the first candidate set includes: Determine whether the AGV forklift meets the set execution conditions; the set execution conditions include the maximum load capacity being greater than or equal to the load weight of the task, the current remaining power being greater than or equal to the estimated power consumption of the task, and the estimated time being less than or equal to the set percentage of the remaining planned time of the task.

6. The AGV forklift collaborative deployment and operation method according to claim 5, characterized in that, Select all executable AGV forklifts, and determine the first target forklift for execution based on the first target task after sorting by preset optimization objectives, including: Select feasible AGV forklifts and calculate the matching objective function between each AGV forklift and the first target task: ， in, Represents the objective function value. This represents the normalized total time taken to complete the first objective task. This represents the normalized total energy consumption for completing the first objective task; The AGV forklift with the smallest objective function is selected as the first target forklift.

7. The AGV forklift collaborative deployment and operation method according to claim 6, characterized in that, The step of assigning the first target task to the first target forklift and adjusting the task execution sequence of the first target forklift to update the assignment scheme includes: If it is determined that the first target forklift is not currently executing a task, the first target task is inserted at the beginning of its task execution sequence, and the execution status is updated. When it is determined that the first target forklift is executing a task with a real-time urgency level less than the first preset threshold, the first target task is inserted into the subsequent position of the currently executing task, and the execution status is updated.

8. The AGV forklift collaborative deployment and operation method according to claim 7, characterized in that, After updating the initial task information of each task to obtain real-time task information, and calculating the real-time urgency based on the real-time task information, the method further includes: Tasks with a real-time urgency level greater than or equal to the second preset threshold are selected to establish a second adjustment set; the second adjustment set includes at least one second target task to be preempted. AGV forklifts that are currently executing tasks with a real-time urgency level less than the first preset threshold and a time redundancy level greater than or equal to the third preset threshold are selected to establish a second candidate set; the time redundancy level is the ratio of the remaining planned time to the remaining estimated time of the second target task. The executability of the AGV forklifts in the second candidate set is verified. The AGV forklift with executability and the shortest remaining time for the current task is selected as the second target forklift. The second target forklift is then controlled to interrupt the currently executing task and execute the second target task. When it is determined that the interrupted task has insufficient remaining planned time due to the interruption, it is included in the first adjustment set to update the allocation scheme; If it is determined that the interrupted task has sufficient remaining planned time due to the interruption, after the second target task is completed, the second target forklift is controlled to return to the interruption position of the interrupted task and resume execution based on the saved state.

9. The AGV forklift collaborative deployment and operation method according to claim 8, characterized in that, The method further includes: When a new task is inserted into the task sequence, the initial urgency of the new task is calculated; If the initial urgency of the new task is less than the first preset threshold, the new task is inserted at the end of the task sequence and assigned to the idle AGV forklift with the lowest current load for execution. If the initial urgency of the new task is determined to be greater than or equal to the first preset threshold and less than the second preset threshold, the new task is included in the first adjustment set to trigger the priority adjustment; When the initial urgency of the new task is determined to be greater than the second preset threshold, the new task is included in the second adjustment set to trigger task preemption. At the same time, the allocation scheme is optimized based on the remaining tasks by improving the NSGA-III algorithm.

10. An AGV forklift collaborative deployment and operation system, characterized in that, The system is used to perform the AGV forklift collaborative deployment operation method as described in any one of claims 1-9, the system comprising: The parameter acquisition module (10) is configured to acquire the initial position and task sequence of all AGV forklifts in the forklift queue. The task sequence includes at least one task to be executed and the corresponding initial task information. An initial calculation module (20) is configured to calculate the initial urgency of a task based on the initial task information, prioritize tasks based on the initial urgency, and include tasks with high initial urgency into the multi-objective optimization model. Forklift scheduling module (30) is configured to acquire the status information of all AGV forklifts, establish a multi-objective function based on the status information including total completion time, energy consumption cost and load balance, and solve the allocation scheme by improving the NSGA-Ⅲ algorithm; the allocation scheme includes each AGV forklift and the corresponding task execution sequence; The planning and execution module (40) is configured to plan a global path from the initial position to the target point for each AGV forklift according to a set algorithm, and control the AGV forklift to run according to the allocation scheme and the global path.

Images