An AGV rolling scheduling method and device, electronic equipment and medium

Abstract

The application discloses an AGV rolling scheduling method and device, electronic equipment and medium, the method comprises the following steps: obtaining port layout information, container task information and waiting allocation information of all AGVs from a port information management system; performing grid processing on a target port based on the port layout information to obtain a port grid map; obtaining a weight matrix by performing path planning on the port grid map according to the waiting allocation information and the container task information; the weight matrix comprises a cost of each AGV corresponding to different container tasks; obtaining a task assignment result of each AGV by using the Hungarian method according to the weight matrix; obtaining a dynamic grid table by performing path planning on the port grid map according to the task assignment result; and generating a target path planning of each AGV by performing obstacle avoidance planning according to the dynamic grid table. The application can efficiently perform AGV scheduling planning and can be widely applied to the technical field of data processing.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an AGV rolling scheduling method, device, electronic equipment and medium. Background Art

[0002] The efficiency of automated container ports primarily depends on automated guided vehicle (AGV) scheduling. Research has shown that automated container ports can reduce manual operations, improve operational stability, reduce labor costs by 45% and operating expenses by 25% to 55%, and increase port productivity by 10% to 35%. However, the rapid expansion of automated port construction worldwide poses a significant challenge to the capacity of automated container ports. Therefore, improving operational efficiency and service quality at automated container ports is a top priority. Existing mainstream literature on task assignment fails to consider the timeliness and interpretability of container transport, and also presents difficulties in coordinating with port cranes (QCs). Furthermore, the random arrival of container tasks requires computationally efficient and conflict-free path planning algorithms.

[0003] Existing research has the following problems:

[0004] (1) Optimization goal: Existing AGV-related research on task scheduling optimization aims to minimize the total operating cost, but does not consider the urgency of each container transportation task, which may easily lead to poor timeliness of some container transportation tasks.

[0005] (2) Path planning: Most existing studies use parallel computing to perform path planning on AGVs, which makes it difficult to consider the calculation rate; it is necessary to add more vehicle collision rules and combine them with the path planning algorithm to realize multi-AVG vehicle path planning. Summary of the Invention

[0006] The main purpose of the embodiments of the present invention is to provide an AGV rolling scheduling method, device, electronic device and medium to solve at least one problem of the prior art. The present invention can efficiently perform AGV scheduling.

[0007] To achieve the above objectives, an embodiment of the present invention provides an AGV rolling scheduling method, which includes:

[0008] Obtain port layout information, container task information, and all AGV waiting allocation information from the port information management system;

[0009] Perform rasterization processing on the target port based on the port layout information to obtain a port grid map;

[0010] Based on the waiting allocation information and container task information, a weight matrix is ​​obtained by performing path planning on the port grid map; the weight matrix includes the cost of each AGV corresponding to different container tasks;

[0011] The task assignment result of each AGV is obtained by the Hungarian method according to the weight matrix;

[0012] According to the task assignment results, path planning is performed on the port grid map to construct a dynamic grid table;

[0013] Obstacle avoidance planning is performed based on the dynamic grid table to generate the target path planning for each AGV.

[0014] In some embodiments, rasterizing the target port based on the port layout information to obtain a port grid map includes:

[0015] Determine the regional distribution scenarios of the target port based on the port layout information; the types of regional distribution scenarios include task receiving areas, container areas, roads, and AGVs;

[0016] Different types of regional distribution scenes are distinguished and marked, and then the regional distribution scenes are binary replaced to obtain the port grid map.

[0017] In some embodiments, based on the waiting allocation information and the container task information, a weight matrix is ​​obtained by performing path planning on a port grid map, including:

[0018] Based on the waiting allocation information and container task information, determine the moving starting point and moving end point of each AGV corresponding to different container tasks, as well as the task starting point, task start time and expected earliest completion time of the container task; the moving starting point represents the initial position of the AGV, and the moving end point represents the task end point of the container task;

[0019] The moving starting point is used as the target starting point, and the task starting point is used as the target end point. The first path evaluation is performed using the improved A-star algorithm to obtain the first path evaluation result. Based on the second path evaluation result, the initial weight matrix corresponding to each AGV for different container tasks is determined.

[0020] The task start point is used as the target start point, and the task end point is used as the target end point. The second path evaluation is performed using the improved A-star algorithm to obtain the second path evaluation result. The task time is determined based on the second path evaluation result. The weight factor is obtained by subtracting the task start time and the task time from the expected earliest completion time of the task.

[0021] Based on the sum of the weight factor and the initial weight matrix, a weight matrix is ​​obtained;

[0022] Among them, the improved A-star algorithm weights the second cost by the Manhattan distance from the target starting point to the target end point; the expression of the improved A-star algorithm is:

[0023] F(P[i])=G(P[i])+H(P[i])*(0.5+H(P[i]) / W)

[0024] Where F(P[i]) represents the path evaluation result; G(P[i]) represents the first cost from the target starting point to the intermediate state; H(P[i]) represents the second cost from the intermediate state to the target end point; W represents the Manhattan distance.

[0025] In some embodiments, the task assignment result of each AGV is obtained by the Hungarian method according to the weight matrix, including:

[0026] Taking the total minimum weighted cost as the objective function, the minimum value of the row or column of the weight matrix is ​​subtracted from the corresponding row or column, and the weight matrix is ​​transformed into a plurality of zero elements located in different rows and columns; all the variables corresponding to the zero elements are set to 1;

[0027] Obtain the corresponding results of the AGV and container tasks corresponding to the rows and columns where the variable is 1, and then determine the task assignment results of each AGV.

[0028] In some embodiments, path planning is performed on the port grid map according to the task assignment results to construct a dynamic grid table, including:

[0029] With the time series as the horizontal coordinate and the grid sequence as the vertical coordinate, a grid barrier table is established;

[0030] Determine the calling order of AGVs based on the task priority of the container task and the task assignment results;

[0031] Based on the calling sequence, the initial position of the corresponding called AGV is used as the target starting point for path planning through the improved A-star algorithm; the improved A-star algorithm is weighted based on the Manhattan distance from the target starting point to the target end point;

[0032] The path planning results of each AGV are dynamically recorded in the grid obstacle table in time sequence to obtain a dynamic grid table.

[0033] In some embodiments, obstacle avoidance planning is performed based on the dynamic grid table to generate a target path plan for each AGV, including:

[0034] According to the dynamic grid table, the location information of the AGV is input in a rolling manner according to the priority, and the obstacle avoidance planning is performed through the IAOA-A algorithm to generate the target path planning for each AGV.

[0035] To achieve the above-mentioned purpose, another aspect of an embodiment of the present invention provides an AGV rolling scheduling device, the device comprising:

[0036] The first module is used to obtain port layout information, container task information and all AGV waiting allocation information from the port information management system;

[0037] The second module is used to perform rasterization processing on the target port based on the port layout information to obtain a port grid map;

[0038] The third module is used to obtain a weight matrix by performing path planning on the port grid map based on the waiting allocation information and container task information. The weight matrix includes the cost of each AGV corresponding to different container tasks.

[0039] The fourth module is used to obtain the task assignment result of each AGV through the Hungarian method according to the weight matrix;

[0040] The fifth module is used to construct a dynamic grid table based on the path planning in the port grid map according to the task assignment results;

[0041] The sixth module is used to perform obstacle avoidance planning based on the dynamic grid table and generate the target path planning for each AGV.

[0042] To achieve the above object, another aspect of an embodiment of the present invention provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method when executing the computer program.

[0043] To achieve the above object, another aspect of an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above method is implemented.

[0044] To achieve the above object, another aspect of an embodiment of the present invention provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0045] The embodiments of the present invention include at least the following beneficial effects: The present invention provides an AGV rolling scheduling method, device, electronic device, and medium. The scheme obtains port layout information, container task information, and waiting assignment information of all AGVs from a port information management system; performs rasterization processing on the target port based on the port layout information to obtain a port grid map; obtains a weight matrix by performing path planning on the port grid map based on the waiting assignment information and container task information; the weight matrix includes the cost of each AGV corresponding to different container tasks; obtains the task assignment result of each AGV based on the weight matrix using the Hungarian method; constructs a dynamic grid table based on the task assignment result on the port grid map; performs obstacle avoidance planning based on the dynamic grid table to generate a target path plan for each AGV. The present invention's assignment rule based on the urgency of the container considers the input of time-related information such as the position information and task status information of each AGV in real time, thereby assigning tasks to idle AGV vehicles in real time; the present invention can efficiently perform AGV scheduling planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the AGV rolling scheduling method provided by an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the port layout provided by an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of a port grid map provided by an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of setting the regional search diffusion direction provided by an embodiment of the present invention;

[0050] Figure 5 Schematic diagram of weight matrix output provided by an embodiment of the present invention;

[0051] Figure 6 is a weight calculation flow chart provided by an embodiment of the present invention;

[0052] Figure 7 This is a schematic diagram of updating a dynamic grid barrier table provided by an embodiment of the present invention;

[0053] Figure 8 This is a schematic diagram of the dynamic obstacle avoidance principle provided by an embodiment of the present invention;

[0054] Figure 9 This is a flowchart of real-time collaborative scheduling provided by an embodiment of the present invention;

[0055] Figure 10 This is an overall flow chart of the AGV rolling scheduling method provided by an embodiment of the present invention;

[0056] Figure 11 It is a structural diagram of the AGV rolling scheduling device provided by an embodiment of the present invention;

[0057] Figure 12 It is a schematic diagram of the hardware structure of the electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to make the objects, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present invention. They are merely examples of devices and methods consistent with some aspects of the embodiments of the present invention as detailed in the appended claims.

[0059] It will be understood that the terms "first," "second," and the like used in the present invention may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are merely used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of," "when," or "in response to a determination."

[0060] The terms "at least one", "plurality", "each", "any", etc. used in the present invention include at least one, two or more, multiple, two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention pertains. The terms used herein are for the purpose of describing embodiments of the present invention only and are not intended to limit the present invention.

[0062] The AGV rolling scheduling method provided by the embodiment of the present invention relates to the field of data processing technology. The AGV rolling scheduling method provided by the embodiment of the present invention can be applied to a terminal, can be applied to a server, or can be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a vehicle-mounted terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements the AGV rolling scheduling method, etc., but is not limited to the above forms.

[0063] The present invention can be used in a wide variety of general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0064] Figure 1 This is an optional flow chart of the AGV rolling scheduling method provided by an embodiment of the present invention. Figure 1 The method may include but is not limited to steps S100 to S600.

[0065] S100, obtaining port layout information, container task information, and waiting allocation information of all AGVs from the port information management system;

[0066] For example, in some specific implementations, it is possible to connect to an information management system for actual applications (such as an automated port information management system) to obtain port layout, container tasks, and global AGVs waiting for allocation information.

[0067] S200, performing rasterization processing on the target port based on the port layout information to obtain a port grid map;

[0068] It should be noted that, in some embodiments, step S200 may include: determining the regional distribution scene of the target port based on the port layout information; the types of regional distribution scenes include task receiving areas, container areas, roads and AGVs; distinguishing and marking different types of regional distribution scenes, and then performing binary replacement on the regional distribution scenes to obtain a port raster map.

[0069] For example, in some specific implementations, port gridding can be performed based on the automated port model according to port layout information. Path planning is prepared by gridding the port layout according to the classification of task receiving areas, container areas, roads, and AGVs. This is specifically implemented as follows:

[0070] The purpose of rasterizing the port is to facilitate path planning in the program. Typical methods for global path planning include the visualization method, the free space method, and the unit decomposition method. The grid method and the unit tree method are two commonly used unit decomposition methods. The grid method is a method of map modeling, which is to simulate obstacles into a collection of small squares, which is equivalent to replacing all objects in the scene with binary values. This embodiment proposes a typical automated port model with a wide range of applications, good scalability, and reconfigurability. The port layout diagram and the port rasterized model diagram are respectively as follows: Figure 2 、 Figure 3 shown.

[0071] The port gridding of the embodiment of the present invention mainly includes the following four parts:

[0072] 1) The task receiving area is located on the upper side of the grid map and is used to receive containers. The crane in the task receiving area places the containers on the transfer platform onto the automatic guided vehicle (AGV). The shore side is located at the edge of the port to facilitate the connection between the port edge QC's container operation and AGV handling;

[0073] 2) Box area, B={b1,b2,...,b n} (n is the number of AGV cargo extraction points in the container area), located on the lower side of the grid map, is used for AGC to deliver containers to the yard, and then the rail crane obtains the container from the AGV and places it in the designated location;

[0074] 3) Roads, Figure 3 The white grid in the middle is used for AGV movement. In order to reduce the occurrence of AVG vehicle conflicts and deadlocks on the road and improve the transportation efficiency of the AGV operation network, in this embodiment, the buffer area adopts a two-way guidance path, and the loading and unloading area adopts a single guidance path.

[0075] 4) AGV, A={a1,a2,...,a r} (r is the number of idle AGVs), transporting containers in the automated port at a certain speed.

[0076] Based on the grid model, the path search and planning tasks can be performed in the road grid according to the coordinates of the task receiving area and the target box area to which the task is executed, combined with the AGV position.

[0077] S300, obtaining a weight matrix by performing path planning on a port grid map according to the waiting allocation information and the container task information;

[0078] It should be noted that the weight matrix includes the cost of each AGV corresponding to different container tasks; in some embodiments, step S300 may include: determining the moving starting point, moving end point, and task starting point, task start time, and expected earliest completion time of each AGV corresponding to different container tasks based on the waiting allocation information and the container task information; the moving starting point represents the initial position of the AGV, and the moving end point represents the task end point of the container task; taking the moving starting point as the target starting point and the task starting point as the target end point, performing a first path evaluation through the improved A-star algorithm to obtain a first path evaluation result; based on the first The results of the second path evaluation determine the initial weight matrix corresponding to each AGV for different container tasks; the task start point is used as the target start point, and the task end point is used as the target end point. The second path evaluation is performed using the improved A-star algorithm to obtain the second path evaluation result; the task time is determined based on the second path evaluation result; the weight factor is obtained by subtracting the task start time and task time from the expected earliest completion time of the task; the weight matrix is ​​obtained based on the sum of the weight factor and the initial weight matrix; among them, the improved A-star algorithm weights the second cost by the Manhattan distance from the target start point to the target end point; the expression of the improved A-star algorithm is:

[0079] F(P[i])=G(P[i])+H(P[i])*(0.5+H(P[i]) / W)

[0080] Where F(P[i]) represents the path evaluation result; G(P[i]) represents the first cost from the target starting point to the intermediate state; H(P[i]) represents the second cost from the intermediate state to the target end point; W represents the Manhattan distance.

[0081] For example, in some specific implementations, a weight matrix can be obtained by preprocessing global AGV and container task information. The time required for an AGV to complete task n in an automated port is calculated using an improved A-star algorithm. Weight factors are calculated to assess task urgency. Weight matrices are then generated for each vehicle and each task based on the weight factors. The two matrices are combined to generate an updated weight matrix for calculation. This is specifically implemented as follows:

[0082] The method of obtaining a weight matrix by preprocessing data is proposed to improve the timeliness of task completion. It is essentially an assignment rule based on the urgency of the container. Considering that the transportation plan of the container not only focuses on the transportation time, but also needs to pay attention to the latest expected completion time during the transportation process and the possible completion time. Whether the task can be completed within the expected time is a key factor in the evaluation of the quality of task completion. Therefore, the embodiment of the present invention proposes an AGV task allocation algorithm based on the urgency of the container and introduces the urgency weight factor En. The calculation formula of the weight factor En is as follows:

[0083] E n =(T en -T sn )-T pn (1)

[0084] Where n is the task sequence number; T sn is the task start time; T en T is the earliest expected completion time of the task; pn Predict the completion time for the task; n is the starting point of the task; D n The end point of the task.

[0085] To obtain the predicted completion time of the task, this embodiment uses the A-star algorithm, which has a fast forward search speed, considering the traditional path planning algorithm and the ability of the automated container port to perceive global information. The A-star algorithm is a path finding and graph traversal algorithm. The evaluation function in the traditional A-star algorithm is:

[0086] F(P[i])=G(P[i])+H(P[i]) (2)

[0087] Where G(p[i]) is the path cost from P[start] to P[i], and H(P[i]) is the shortest distance from P[i] to P[end]. In the context of the automated port in this embodiment, the AGV moves in translation, so all distance calculations should use Manhattan distance.

[0088] During the A-star algorithm's operation, each time the node with the smallest F(P[i]) value (highest priority) is selected from the priority queue as the next node to be traversed. Furthermore, the A-star algorithm uses two sets to represent the nodes to be traversed and the nodes that have already been traversed, which are usually called open_set and close_set. The pseudo code for the traditional A-star algorithm is as follows:

[0089]

[0090]

[0091] To ensure that path optimization data can be quickly obtained in real time, the traditional A-star algorithm needs to be further optimized to improve the speed and search effect. This invention optimizes the A-star algorithm from two aspects: 1. Evaluation function; 2. Simplified search space.

[0092] 1) In terms of the evaluation function, to improve the search performance and speed of the A-star algorithm, the evaluation matrix is ​​optimized. This embodiment proposes the following equation to influence the evaluation value, ensuring that H(P[i]) equals G(P[i]) to ensure that the A-star algorithm maintains its optimal state.

[0093] F(P[i])=G(P[i])+H(P[i])*(0.5+H(P[i]) / W)

[0094] Where W is the Manhattan distance from the input target start point to the end point.

[0095] 2) In order to comply with the actual operation mode of AGV in automated ports and further improve the search rate of the A-star algorithm, a method of simplifying the search space of A-star in the environment set up in this embodiment is proposed. Specifically, unlike the previous hierarchical path finding and optimization of the Open table, this embodiment sets up two areas for the AGV operating environment: the Loading area and the Storage area. When searching for a path, it is limited to diffusing from the Loading area to the Storage area, thereby limiting the possibility of diffusion. This simplifies the search space while reducing the amount of calculation, thereby improving the search rate. The schematic diagram of one-way diffusion is shown in the figure below. Figure 4 shown.

[0096] Specifically, O n With D n Input the optimized A-star algorithm to calculate the time T required for AGV to complete task n in the automated port pn . Then combine the task data information T collected by the system an With T ln Calculate En. Where |E n |A smaller value indicates a higher urgency of the container.

[0097] Retrieve the current position and status information of all AGV vehicles and combine them with the system to obtain the information of each task, and convert the AGV starting position, O n 、D n Data input optimization A-star algorithm outputs the weight matrix of each vehicle and each task The weight matrix output diagram is as follows Figure 5 Then, the weight matrix output above is combined with the urgency weight of each task En to obtain the updated matrix. The matrix weight is The formula is as follows:

[0098]

[0099] The algorithmic approach of the container urgency-based task allocation algorithm proposed above is as follows: 1. Obtain container status information; 2. Preprocess the status information to obtain a weight matrix; 3. Finally, combine the Hungarian algorithm to output each task and the corresponding assigned AGV number with the minimum total weighted cost as the optimization goal. The main process of the algorithm steps: 1. Obtain real-time task information, AGV status, and location information; 2. Calculate the urgency weight E n 、 Output matrix. The weight calculation flow chart is as follows Figure 6 shown.

[0100] S400, obtaining the task assignment result of each AGV by using the Hungarian method according to the weight matrix;

[0101] It should be noted that, in some embodiments, step S400 may include: taking the total minimum weighted cost as the objective function, subtracting the minimum value of the row or column from the corresponding row or column of the weight matrix, and converting the weight matrix into a plurality of zero elements located in different rows and columns; setting the variables corresponding to all zero elements to 1; obtaining the corresponding results of the AGV and container tasks corresponding to the rows and columns where the variables are 1, and then determining the task assignment results of each AGV.

[0102] For example, in some specific implementations, the Hungarian method can be used to obtain an assignment result containing each task and the corresponding AGV number based on the weight matrix. The objective function is to minimize the total weighted cost. The minimum value of each row or column in the weight matrix is ​​subtracted from the row or column, transforming the weight matrix into a matrix with a zero elements located in different rows and columns. The variables corresponding to all zero elements are set to 1, indicating that the task corresponds to the AGV, and the remaining variables are set to 0. The AGV number a corresponding to each task n is obtained.

[0103] The specific implementation is as follows:

[0104] The Hungarian method is used to obtain the assignment result including each task and the corresponding assigned AGV number. Specifically, the weight is the matrix weight and the total minimum weighted cost is used.

[0105] The objective function is solved as follows:

[0106]

[0107] Subtract the minimum value of the row or column from the row and column of the weight matrix, and repeat this process to transform the weight matrix into a zero element (i.e., the number of idle AGVs) located in different rows and columns; let the variables corresponding to all zero elements be 1, indicating that the task corresponds to the AGV, and the remaining variables are 0; the sum of the weights of the positions marked as 1 is the minimum weighted cost, and the tasks n and AGV serial number a corresponding to these positions are the assigned task pairs.

[0108] S500, performing path planning on the port grid map according to the task assignment result to obtain a dynamic grid table;

[0109] It should be noted that, in some embodiments, step S500 may include: establishing a grid obstacle table with the time series as the horizontal coordinate and the grid sequence as the vertical coordinate; determining the calling order of the AGV according to the task priority of the container task and the task assignment result; based on the calling order, using the initial position of the corresponding called AGV as the target starting point to perform path planning through the improved A-star algorithm; the improved A-star algorithm is weighted based on the Manhattan distance from the target starting point to the target end point; and dynamically recording the path planning results of each AGV in a time series to the grid obstacle table to obtain a dynamic grid table.

[0110] For example, in some specific implementations, a grid obstacle table can be established with time series as the horizontal axis and grid sequence as the vertical axis; tasks are transmitted to assigned vehicles according to their priorities, and the paths are input in descending order of priority and the improved A-star algorithm is called for path planning; the paths planned by each AGV are dynamically recorded in the grid obstacle table block in a time sequence. The specific implementation is as follows:

[0111] Considering the characteristics of static search and grid evaluation by heuristic function of A-satr algorithm, this embodiment adopts the method of processing specific grids and temporarily includes a certain number of grids in the closelist. In order to record the grid numbers that need to be temporarily included, this embodiment proposes a grid obstacle table Block based on time series. The table diagram is shown as follows: Figure 7As shown in the figure, the Block table represents the time series on the horizontal axis and the grid sequence number on the vertical axis. The value of a grid sequence number within a certain time period indicates whether the grid sequence was occupied during that time period. The multi-AGV path planning process is transmitted to the assigned vehicles according to task priority. The paths are then entered sequentially from highest priority to lowest, and the improved A-star algorithm is used for path planning. After each AGV completes path planning, its path is recorded in the Block table. This table is also used for path planning by subsequent AGVs, and the paths are entered into the table simultaneously. Specifically, the optimized A-star algorithm is called for path planning, starting with the highest-priority vehicle and task. The grid position of each vehicle at each moment is marked in the dynamic grid obstacle table. During subsequent vehicle planning, the Block table is automatically called during the path search process. Each mark in the table is interpreted as an obstacle at that grid node at that moment. This is used to determine whether there are any obstacles around the node where the vehicle is located at that moment. If no obstacles exist, the algorithm is called normally to calculate the next optimal node. If so, the obstacle node is discarded and the algorithm is called again. After planning is complete, the path is entered into the Block table as described above. Unlike static search, the search for each node incorporates the concept of time, emphasizing that a certain moment corresponds to a certain location node. The corresponding moment when the planned vehicle arrives at each node will also change. At this time, the obstacles obtained by mobilizing the Block table will also change accordingly, thereby achieving the effect of dynamic obstacle avoidance.

[0112] S600: Perform obstacle avoidance planning based on the dynamic grid table to generate a target path plan for each AGV.

[0113] It should be noted that, in some embodiments, step S600 may include: scrolling the position information of the AGV according to the priority according to the dynamic grid table, performing obstacle avoidance planning through the IAOA-A algorithm, and generating a target path plan for each AGV.

[0114] For example, in some specific implementations, dynamic obstacle avoidance can be achieved by combining an improved A-star algorithm for static obstacle search with a time-series-based grid obstacle table, ultimately obtaining task assignment arrangements and conflict-free paths. Specific implementations are as follows:

[0115] The AGVs are input into the IAOA-A algorithm in a rolling manner according to their priorities to generate AGVs task assignment arrangements and conflict-free paths in order to ensure that path optimization data can be quickly obtained in real time and the dynamic obstacle avoidance effect can be achieved. IAOA-A realizes obstacle avoidance functions such as Figure 8 shown.

[0116] Conflicts such as deadlock and closed loops are prone to occur during multi-AGV path planning. To effectively resolve conflicts between multiple AGVs in automated port scenarios, an improved A-star algorithm is developed by requiring vehicles to detect conflicting points during multi-AGV path planning and then pre-stationary. The pseudo code for this algorithm is as follows:

[0117]

[0118] Finally, to achieve real-time dynamic obstacle avoidance, the proposed improved A-star algorithm is combined with a time-series grid table to obtain an improved automatic obstacle avoidance A-star algorithm. This embodiment proposes a static obstacle search combined with the IAOA-A algorithm based on a time-series grid obstacle table. This algorithm can dynamically change the way other AGVs are considered obstacles during automated port AGV operation, thereby achieving dynamic obstacle avoidance. The pseudo code of this algorithm is as follows:

[0119]

[0120]

[0121] This embodiment will consider a more reasonable assignment rule based on the urgency of the container and a conflict-free path planning algorithm that can be quickly realized in series. The AGVs are input into the IAOA-A algorithm according to their priorities, and the AGVs task assignment arrangement and conflict-free path planning can be generated. The real-time collaborative scheduling flow chart is as follows: Figure 9 shown.

[0122] In order to explain the principle of the technical solution of the present invention in detail, the overall process of the present invention is described below in combination with some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and cannot be regarded as a limitation of the present invention.

[0123] To address the challenges of existing technologies, this paper proposes a tandem collaborative scheduling method based on a dynamic barrier table. The optimized A-star algorithm proposed in this paper achieves superior optimization, significantly reducing the optimization time compared to the traditional A-star algorithm. Furthermore, the container urgency-based assignment rule proposed in this paper produces results that are more stable and more consistent with expected values ​​than two representative traditional assignment rules.

[0124] like Figure 10 As shown, the method includes the following steps:

[0125] Obtain port layout, container tasks and global AGVs waiting for allocation information based on the automated port information management system;

[0126] Port gridding is performed based on the automated port model according to port layout information;

[0127] The weight matrix is ​​obtained by preprocessing the global AGVs and container task information;

[0128] The Hungarian method is used to obtain the assignment results including each task and the corresponding assigned AGV serial number according to the weight matrix;

[0129] Path planning is performed based on the assignment results to construct a dynamic raster table;

[0130] Dynamic obstacle avoidance is achieved based on the dynamic grid obstacle table combined with the improved A-star algorithm, and finally the AGVs task assignment arrangement and conflict-free path are generated.

[0131] Some possible implementations of performing port gridding based on the automated port model according to the port layout information include:

[0132] The port layout is rasterized according to the classification of task receiving area, container area, road and AGV to prepare for path planning.

[0133] Some possible implementations include preprocessing the global AGVs and container mission information to obtain a weight matrix, including:

[0134] The time required for AGV to complete task n in the automated port is calculated by improving the A-star algorithm;

[0135] Calculate weight factors to assess task urgency;

[0136] A weight matrix is ​​formed for each vehicle and each task according to the weight factor;

[0137] The two matrices are combined to obtain the updated weight matrix for calculation.

[0138] In some achievable implementations, the assignment results including each task and the corresponding assigned AGV number are obtained by the Hungarian method according to the weight matrix, including:

[0139] The total minimum weighted cost is used as the objective function;

[0140] Subtract the minimum value of the row or column from the rows and columns of the weight matrix, so that the weight matrix has a zero elements located in different rows and columns;

[0141] Let all variables corresponding to zero elements be 1, indicating that the task corresponds to the AGV, and the rest of the variables be 0;

[0142] Get the AGV serial number a corresponding to each task n.

[0143] Some possible implementations of performing path planning based on the assignment results and constructing a dynamic grid table include:

[0144] A grid barrier table is established with the time series as the horizontal axis and the grid sequence as the vertical axis;

[0145] The task priority is passed to the assigned vehicle, and the priority is input in descending order and the improved A-star algorithm is called for path planning;

[0146] The path planned by each AGV is dynamically recorded in the grid obstacle table Block in time sequence.

[0147] Some feasible implementations, which combine the dynamic grid obstacle table with the improved A-star algorithm to achieve dynamic obstacle avoidance and ultimately generate AGVs task assignments and conflict-free paths, include:

[0148] Static obstacle search is combined with a time-series-based grid obstacle table to achieve dynamic obstacle avoidance, ultimately obtaining task assignment arrangements and conflict-free paths.

[0149] In summary, the present invention designs a multi-AGV rolling scheduling model based on the assignment rules of the container urgency and the proposed IAOA-A algorithm. The assignment rules based on the urgency of the container take into account the position information of each AGV in real time, the task status information and other time-related information for input, so as to assign tasks to idle AGV vehicles in real time. Combined with the consideration of dynamic obstacles based on time sequence and the improved optimized A-star algorithm, conflict-free path planning is performed for AGV vehicles that have been assigned tasks according to their priority, thereby realizing online collaborative scheduling. The present invention improves timeliness and has good task completion stability and task selection logic. Compared with the prior art, the present invention has the following advantages:

[0150] The present invention proposes an AGV collaborative scheduling method that combines an assignment rule based on the urgency of containers with an algorithm that can quickly achieve conflict-free path planning. The method has high timeliness and good task completion stability and task selection logic.

[0151] See also Figure 11 The embodiment of the present invention further provides an AGV rolling scheduling device 900, which can implement the above-mentioned AGV rolling scheduling method, and the device includes:

[0152] The first module 910 is used to obtain port layout information, container task information and waiting allocation information of all AGVs from the port information management system;

[0153] The second module 920 is used to perform rasterization processing on the target port based on the port layout information to obtain a port grid map;

[0154] The third module 930 is configured to obtain a weight matrix by performing path planning on the port grid map based on the waiting allocation information and the container task information. The weight matrix includes the cost of each AGV corresponding to different container tasks.

[0155] The fourth module 940 is used to obtain the task assignment result of each AGV by using the Hungarian method according to the weight matrix;

[0156] The fifth module 950 is used to perform path planning on the port grid map according to the task assignment result to obtain a dynamic grid table;

[0157] The sixth module 960 is used to perform obstacle avoidance planning based on the dynamic grid table and generate a target path plan for each AGV.

[0158] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0159] An embodiment of the present invention further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the aforementioned AGV rolling scheduling method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, an in-vehicle computer, or the like.

[0160] It can be understood that the contents of the above method embodiments are applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0161] See also Figure 12 , Figure 12 The hardware structure of an electronic device 1000 according to another embodiment is shown. The electronic device includes:

[0162] The processor 1001 may be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is configured to execute relevant programs to implement the technical solutions provided by the embodiments of the present invention.

[0163] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program codes are stored in the memory 1002 and are called by the processor 1001 to execute the AGV rolling scheduling method of the embodiment of the present invention.

[0164] Input / output interface 1003, used to implement information input and output;

[0165] Communication interface 1004, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0166] Bus 1005 , which transmits information between various components of the device (e.g., processor 1001 , memory 1002 , input / output interface 1003 , and communication interface 1004 );

[0167] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via a bus 1005 .

[0168] An embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned AGV rolling scheduling method is implemented.

[0169] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiment, the functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0170] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0171] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0172] The AGV rolling scheduling method, AGV rolling scheduling device, electronic device and storage medium provided by the embodiments of the present invention obtain port layout information, container task information and waiting assignment information of all AGVs from a port information management system; rasterize the target port based on the port layout information to obtain a port grid map; according to the waiting assignment information and container task information, a weight matrix is ​​obtained by performing path planning on the port grid map; the weight matrix includes the cost of each AGV corresponding to different container tasks; the task assignment result of each AGV is obtained by the Hungarian method based on the weight matrix; according to the task assignment result, path planning is performed on the port grid map to obtain a dynamic grid table; obstacle avoidance planning is performed based on the dynamic grid table to generate a target path plan for each AGV. The assignment rule based on the container urgency of the present invention considers the position information of each AGV in real time, task status information and other time-related information as input, thereby assigning tasks to idle AGV vehicles in real time; the present invention can efficiently perform AGV scheduling planning.

[0173] The embodiments described in the embodiments of the present invention are intended to more clearly illustrate the technical solutions of the embodiments of the present invention and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0174] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present invention, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0175] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0176] Those skilled in the art will appreciate that all or some of the steps, devices, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0177] The terms "first," "second," "third," "fourth," and the like (if any) in the description of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in orders other than those illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0178] It should be understood that in the present invention, "at least one (item)" refers to one or more, and "plurality" refers to two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can represent: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0179] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the above units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0180] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0181] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0182] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or 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 multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store programs.

[0183] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the invention is not limited thereby. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the invention should be within the scope of the invention.

Claims

1. An AGV rolling scheduling method, characterized in that: The method comprises: Obtain port layout information, container task information, and all AGV waiting allocation information from the port information management system; Performing rasterization processing on the target port based on the port layout information to obtain a port grid map; According to the waiting allocation information and the container task information, a weight matrix is ​​obtained by performing path planning on the port grid map; the weight matrix includes the cost of each AGV corresponding to different container tasks; Wherein, the weight matrix is ​​obtained by performing path planning on the port grid map according to the waiting allocation information and the container task information, including: Determine, based on the waiting allocation information and the container task information, the moving starting point and moving end point of each AGV corresponding to different container tasks, as well as the task starting point, task start time, and expected earliest task completion time of the container task; the moving starting point represents the initial position of the AGV, and the moving end point represents the task end point of the container task; The moving starting point is used as the target starting point, and the task starting point is used as the target end point. A first path evaluation is performed using the improved A-star algorithm to obtain a first path evaluation result; based on the first path evaluation result, an initial weight matrix corresponding to each of the AGVs to different container tasks is determined; Taking the task starting point as the target starting point and the task end point as the target end point, performing a second path evaluation using an improved A-star algorithm to obtain a second path evaluation result; determining the task time based on the second path evaluation result; and obtaining a weight factor based on the expected earliest completion time of the task minus the task start time and the task time; Obtaining a weight matrix based on the sum of the weight factor and the initial weight matrix; The improved A-star algorithm weights the second cost by the Manhattan distance from the target starting point to the target end point; the expression of the improved A-star algorithm is: Where, Indicates the path evaluation result; Indicates the first cost from the target starting point to the intermediate state; Represents the second cost from the intermediate state to the target end point; represents the Manhattan distance; Obtaining a task assignment result for each AGV by the Hungarian method according to the weight matrix; Performing path planning on the port grid map according to the task assignment result to construct a dynamic grid table; Obstacle avoidance planning is performed according to the dynamic grid table to generate a target path plan for each of the AGVs.

2. The method according to claim 1, characterized in that The step of performing rasterization processing on the target port based on the port layout information to obtain a port grid map includes: Determining the regional distribution scene of the target port based on the port layout information; the types of the regional distribution scene include a task receiving area, a container area, a road, and the AGV; Different types of regional distribution scenes are distinguished and marked, and then the regional distribution scenes are binary-substituted to obtain the port grid map.

3. The method according to claim 1, characterized in that Obtaining the task assignment result of each AGV by the Hungarian method according to the weight matrix includes: Taking the total minimum weighted cost as the objective function, subtracting the minimum value of the row or column from the row or column of the weight matrix, and transforming the weight matrix into a plurality of zero elements located in different rows and columns; setting the variables corresponding to all the zero elements to 1; The corresponding results of the AGV and the container tasks corresponding to the rows and columns where the variable is 1 are obtained, and then the task assignment result of each AGV is determined.

4. The method according to claim 1, wherein The path planning and constructing of a dynamic grid table on the port grid map according to the task assignment result includes: With the time series as the horizontal coordinate and the grid sequence as the vertical coordinate, a grid barrier table is established; Determine the calling order of the AGVs according to the task priority of the container task and the task assignment result; Based on the calling sequence, the initial position of the corresponding called AGV is used as the target starting point to perform path planning through the improved A-star algorithm; the improved A-star algorithm is weighted based on the Manhattan distance from the target starting point to the target end point; The path planning result of each AGV is dynamically recorded in the grid obstacle table in time sequence to obtain the dynamic grid table.

5. The method according to claim 1, wherein The obstacle avoidance planning is performed according to the dynamic grid table to generate a target path plan for each AGV, including: According to the dynamic grid table, the position information of the AGV is scrolled and input according to the priority, and the obstacle avoidance planning is performed through the IAOA-A algorithm to generate the target path planning for each AGV.

6. An AGV rolling scheduling device, characterized in that: The device comprises: The first module is used to obtain port layout information, container task information and all AGV waiting allocation information from the port information management system; The second module is used to perform rasterization processing on the target port based on the port layout information to obtain a port grid map; The third module is configured to obtain a weight matrix by performing path planning on the port grid map according to the waiting allocation information and the container task information; the weight matrix includes the cost of each AGV corresponding to different container tasks; Wherein, the weight matrix is ​​obtained by performing path planning on the port grid map according to the waiting allocation information and the container task information, including: Determine, based on the waiting allocation information and the container task information, the moving starting point and moving end point of each AGV corresponding to different container tasks, as well as the task starting point, task start time, and expected earliest task completion time of the container task; the moving starting point represents the initial position of the AGV, and the moving end point represents the task end point of the container task; The moving starting point is used as the target starting point, and the task starting point is used as the target end point. A first path evaluation is performed using the improved A-star algorithm to obtain a first path evaluation result; based on the first path evaluation result, an initial weight matrix corresponding to each of the AGVs to different container tasks is determined; Taking the task starting point as the target starting point and the task end point as the target end point, performing a second path evaluation using an improved A-star algorithm to obtain a second path evaluation result; determining the task time based on the second path evaluation result; and obtaining a weight factor based on the expected earliest completion time of the task minus the task start time and the task time; Obtaining a weight matrix based on the sum of the weight factor and the initial weight matrix; The improved A-star algorithm weights the second cost by the Manhattan distance from the target starting point to the target end point; the expression of the improved A-star algorithm is: Where, Indicates the path evaluation result; Indicates the first cost from the target starting point to the intermediate state; Represents the second cost from the intermediate state to the target end point; represents the Manhattan distance; A fourth module is configured to obtain a task assignment result for each AGV by using the Hungarian method according to the weight matrix; A fifth module is configured to construct a dynamic grid table by performing path planning on the port grid map according to the task assignment result; The sixth module is used to perform obstacle avoidance planning according to the dynamic grid table and generate a target path plan for each of the AGVs.

7. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images