An industrial mobile AGV cluster control method and system applied to factories

Abstract

The present invention provides an industrial mobile AGV cluster control method and system applied to factories, aiming to improve the operation efficiency and safety of automated logistics systems. The method divides the entire factory area into multiple grid areas by uniform division at the initial moment, and each grid area allows at most one AGV to exist at the same time; according to the grid occupancy at the current moment, an improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV; then, each AGV is controlled to move according to the planned path, and the residence time of the AGV in each grid is collected; when the residence time of the AGV in a certain grid exceeds a preset threshold, the grid division of the area is replanned, and the optimal path is replanned; the present invention has the advantages of high path planning efficiency, high space utilization, strong system security and stability, and flexible task priority management, and has broad application prospects.

Description

Technical Field

[0001] The present invention belongs to the technical field of AGV cluster control, and in particular relates to an industrial mobile AGV cluster control method and system applied to a factory. Background Art

[0002] In modern industrial production, industrial mobile AGV (Automated Guided Vehicle) is the core component of the automated logistics system, and its cluster control technology has always been a hot topic and difficulty in research; traditional AGV cluster control methods mainly rely on fixed path planning and scheduling strategies. These methods often show problems such as lack of flexibility and low efficiency when dealing with complex and changeable factory environments.

[0003] In the prior art, traditional AGV cluster control methods usually use fixed paths or preset routes in path planning. This method may be competent when the factory layout is simple and the task requirements are stable, but in a complex factory environment, due to changes in various factors such as equipment layout, material flow, and production tasks, fixed path planning is often difficult to adapt to actual needs. In addition, fixed path planning may also cause unnecessary congestion and waiting for AGVs during transportation, thereby reducing overall transportation efficiency; in addition to path planning, traditional AGV cluster control methods also have the problem of single scheduling strategies. These methods usually only consider the current position and target position of the AGV, while ignoring other factors that may affect transportation efficiency, such as the load capacity of the AGV, the priority of the transportation task, etc. Therefore, in practical applications, these methods often find it difficult to achieve effective coordination and scheduling between AGVs.

[0004] In addition, with the continuous development of industrial automation technology, the factory environment has become more and more complex and changeable. Traditional AGV cluster control methods often have difficulty coping with such changes due to their lack of adaptive capabilities. For example, when the factory layout changes or the production task requirements change, these methods often require manual re-planning of paths and scheduling strategies, which not only increases the workload but also may affect production efficiency.

[0005] Therefore, how to design an AGV cluster control method that can adapt to complex environments, have efficient path planning and scheduling strategies, and have adaptive and fault handling capabilities has become a problem that needs to be solved urgently. An efficient AGV cluster control method is needed to meet the actual needs of modern industrial production. Summary of the invention

[0006] The present invention provides an industrial mobile AGV cluster control method and system applied to factories, and the specific scheme is as follows:

[0007] In a first aspect, the present invention provides an industrial mobile AGV cluster control method applied to a factory, the method comprising the following steps:

[0008] Step S1, at the initial moment, the entire factory area is divided into multiple grid areas by uniform division, and each grid area is allowed to have at most one AGV at the same time.

[0009] Furthermore, the moving speed of each AGV is controlled, and each AGV is numbered sequentially. The Euclidean distance from the current position of an AGV to the target position:

[0010] ;

[0011] in, and For the The horizontal and vertical coordinates of the current position of the AGV, and For the The horizontal and vertical coordinates of the target position of the AGV;

[0012] Set The moving speed vector of an AGV: ;

[0013] in, ， and are constants, representing the weights of the target attraction and obstacle repulsion, respectively, and n is the total number of AGVs;

[0014] Update the position of the AGV. The position coordinates of the AGV at time t+1 satisfy:

[0015] ;

[0016] ;

[0017] in, and They are respectively the first The horizontal and vertical coordinates of the current position of the AGV, and They are respectively the first The vertical coordinate of the current position of the AGV, and is the component of the moving velocity vector in the horizontal and vertical coordinates, is the time interval between adjacent moments.

[0018] Step S2: According to the grid occupancy at the current moment, an improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV.

[0019] Furthermore, in step S2, an improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV, including:

[0020] For AGV, from accepting a task to completing it, no matter how the path is planned, the number of grids it must pass through or stay in is , the mathematical expression of the fitness function is:

[0021] ;

[0022] in, For the i-th path planning sequence, each grid that must be passed or stopped only appears once. In the path planning sequence, the adjacent nodes to distance;

[0023] Construct a search map for the improved AGV cluster control algorithm:

[0024] ;

[0025] in, is the total number of paths, and all paths are iterated sequentially to the maximum number of times. At this time, the matrix The minimum fitness function value corresponds to That is the The optimal path for each AGV.

[0026] Furthermore, when dynamically planning the optimal path for each AGV, the smoothness of the path and the need to avoid sharp turns are also considered. The specific implementation method is as follows:

[0027] When constructing the fitness function, a path smoothness penalty term is introduced to evaluate the change in the angle between adjacent segments in the path. The mathematical expression of this penalty term is:

[0028] ;

[0029] in, It represents the angle formed by three consecutive nodes z, z+1 and z+2 in the path planning sequence. The size of the angle is calculated by the dot product formula of the vector:

[0030] ;

[0031] The fitness function that comprehensively considers path length and smoothness is modified as follows:

[0032] ;

[0033] in, is the weight coefficient of the smoothness penalty term, , when searching for the optimal path, through iterative optimization, the fitness function Reach the minimum value and get the optimal path.

[0034] Step S3: Control each AGV to move along the planned path and collect the AGV's stay time in each grid.

[0035] Step S4: When the AGV stays in a certain grid for longer than a preset threshold, the grid division of the area is replanned, and the optimal path is replanned.

[0036] Furthermore, the method also includes an adaptive grid division mechanism, specifically including:

[0037] Initially, uniform grid division is adopted;

[0038] During system operation, the AGV passing frequency and dwell time of each grid are counted, and the upper and lower thresholds of the passing frequency and the upper and lower thresholds of the dwell time are set.

[0039] For areas where the AGV passing frequency is higher than the passing frequency upper limit threshold or the residence time is longer than the residence time upper limit threshold, the grid division accuracy is increased.

[0040] For areas where the AGV passing frequency is lower than the passing frequency lower limit threshold or the residence time is less than the residence time lower limit threshold, the grid division accuracy is reduced.

[0041] Furthermore, within a monitoring cycle, the residence time of each AGV in grid j is monitored, and the residence time of all n AGVs in grid j is calculated. , For the The residence time of an AGV in grid j.

[0042] Furthermore, a safety distance is set for the AGV. When performing adaptive grid division, the distance between adjacent grid centers and the minimum side length of each grid should not be less than the safety distance.

[0043] For AGVs performing tasks, a priority is set for each AGV according to the importance of the task. The higher the task importance, the greater the priority. The distance between AGVs is calculated in real time. When the distance is less than the safe distance, the avoidance mechanism is triggered. When avoiding, the AGV with lower priority gives way. If the priorities are the same, the AGV on the right gives way. A one-way traffic rule is set in narrow passages to prevent AGVs from deadlocking when traveling in opposite directions. The narrow passage refers to an area without adjacent grids on both sides of the divided grid.

[0044] Furthermore, the grid density function of the area that needs to be re-divided is calculated:

[0045] ;

[0046] Where m is the number of grids in the region, and is the center coordinate of grid j, is the standard deviation of grid j, which represents the degree of discreteness of AGV activities or task distribution within the grid.

[0047] Use the Lloyd algorithm to iteratively optimize mesh division, and update the mesh center coordinates at each iteration:

[0048] ;

[0049] ;

[0050] Where V is the Voronoi region of grid j.

[0051] Calculate quality indicators for new meshing , ,in is the area of ​​grid j, when or When, that is, the quality index The change is less than the preset threshold Or when the maximum number of iterations is reached, the iteration is terminated.

[0052] Furthermore, when an AGV is detected to be abnormal, the task of the AGV is immediately suspended and a fault alarm is sent; the unfinished tasks of the faulty AGV are automatically reallocated to other available AGVs; for faults that can be automatically recovered, the preset recovery procedure is executed; for faults that require manual intervention, a maintenance work order is generated.

[0053] In a second aspect, the present invention provides an industrial mobile AGV cluster control system applied to a factory, the system being used to execute the control method described in the first aspect, the system comprising: an initial grid division module, a path planning module, and an AGV monitoring and adjustment module connected in sequence.

[0054] Furthermore, the grid division module is used to divide the entire factory area into multiple grid areas by uniform division at the initial moment, and each grid area is allowed to have at most one AGV at the same time.

[0055] Furthermore, the path planning module is used to dynamically plan the optimal path for each AGV using an improved AGV cluster control algorithm according to the grid occupancy at the current moment.

[0056] Furthermore, the AGV monitoring and adjustment module is used to control each AGV to move along the planned path and collect the AGV's residence time in each grid; when the AGV's residence time in a grid exceeds a preset threshold, the grid division of the area is replanned, and the optimal path is replanned.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] By adopting an improved AGV cluster control algorithm, the present invention can dynamically plan the optimal path for each AGV, significantly improving the efficiency and accuracy of path planning; introducing a path smoothness penalty term further optimizes path planning, reduces unnecessary turns and pauses of AGV during movement, and improves the overall transportation efficiency; in addition, the method has an adaptive grid division mechanism, which can dynamically adjust the grid division accuracy according to the passing frequency and residence time of AGV, thereby more effectively utilizing factory space, reducing conflicts and waiting time between AGVs, and improving the operating efficiency of the overall system, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of an industrial mobile AGV cluster control method applied to a factory according to the present invention;

[0060] Figure 2 The present invention is a schematic diagram of the composition of an industrial mobile AGV cluster control system applied to a factory. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example 1

[0062] like Figure 1 FIG. 1 is a flow chart of an industrial mobile AGV cluster control method applied to a factory according to the present invention, and the method comprises the following steps:

[0063] Step S1, at the initial moment, the entire factory area is divided into multiple grid areas by uniform division, and each grid area is allowed to have at most one AGV at the same time.

[0064] It should be noted that at the initial moment, the entire factory area is evenly divided into multiple grid areas, and each grid area is allowed to have at most one AGV (automatic guided vehicle) at the same time to ensure smooth traffic and avoid collisions. Next, for each AGV, we need to control its moving speed.

[0065] To control the moving speed of each AGV, each AGV needs to be numbered sequentially. The Euclidean distance from the current position of an AGV to the target position:

[0066] ;

[0067] in, and For the The horizontal and vertical coordinates of the current position of the AGV, and For the The horizontal and vertical coordinates of the target position of the AGV.

[0068] Set The moving speed vector of an AGV: ;

[0069] in, ， and are constants, representing the weights of the target attraction and obstacle repulsion, respectively, and n is the total number of AGVs;

[0070] Update the position of the AGV. The position coordinates of the AGV at time t+1 satisfy:

[0071] ;

[0072] ;

[0073] in, and They are respectively the first The horizontal and vertical coordinates of the current position of the AGV, and They are respectively the first The vertical coordinate of the current position of the AGV, and is the component of the moving velocity vector in the horizontal and vertical coordinates, is the time interval between adjacent moments.

[0074] Step S2: According to the grid occupancy at the current moment, an improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV.

[0075] It should be noted that the improved AGV cluster control algorithm is used in step S2 to dynamically plan the optimal path for each AGV, including:

[0076] For AGV, from accepting a task to completing it, no matter how the path is planned, the number of grids it must pass through or stay in is , the mathematical expression of the fitness function is:

[0077] ;

[0078] in, For the i-th path planning sequence, each grid that must be passed or stopped only appears once. In the path planning sequence, the adjacent nodes to distance; each grid appears only once in the path to ensure the uniqueness and continuity of the path.

[0079] Construct a search map for the improved AGV cluster control algorithm:

[0080] ;

[0081] in, is the total number of paths, and all paths are iterated sequentially to the maximum number of times. At this time, the matrix The minimum fitness function value corresponds to That is the The optimal path of an AGV, the smaller the value of the fitness function, the shorter the path, that is, the better the path.

[0082] When dynamically planning the optimal path for each AGV, the smoothness of the path and the need to avoid sharp turns are also considered. The specific implementation method is as follows:

[0083] When constructing the fitness function, a path smoothness penalty term is introduced to evaluate the change in the angle between adjacent segments in the path. The mathematical expression of this penalty term is:

[0084] ;

[0085] in, It represents the angle formed by three consecutive nodes z, z+1 and z+2 in the path planning sequence. The size of the angle is calculated by the dot product formula of the vector:

[0086] ;

[0087] The fitness function that comprehensively considers path length and smoothness is modified as follows:

[0088] ;

[0089] in, is the weight coefficient of the smoothness penalty term, , when searching for the optimal path, through iterative optimization, the fitness function Reach the minimum value and get the optimal path.

[0090] Step S3: Control each AGV to move along the planned path and collect the AGV's stay time in each grid.

[0091] Step S4: When the AGV stays in a certain grid for longer than a preset threshold, the grid division of the area is replanned, and the optimal path is replanned.

[0092] The method also includes an adaptive grid division mechanism, specifically including: using uniform grid division initially; during system operation, counting the AGV passing frequency and residence time of each grid, setting the upper and lower thresholds of the passing frequency, and setting the upper and lower thresholds of the residence time.

[0093] For areas where the AGV passing frequency is higher than the passing frequency upper limit threshold or the residence time is longer than the residence time upper limit threshold, the grid division accuracy is increased.

[0094] For areas where the AGV passing frequency is lower than the passing frequency lower limit threshold or the residence time is less than the residence time lower limit threshold, the grid division accuracy is reduced.

[0095] In one monitoring cycle, monitor the residence time of each AGV in grid j, and calculate the residence time of all n AGVs in grid j , For the The residence time of an AGV in grid j.

[0096] Set a safety distance for the AGV. When performing adaptive meshing, the distance between adjacent mesh centers and the minimum side length of each mesh should not be less than the safety distance.

[0097] For AGVs performing tasks, a priority is set for each AGV according to the importance of the task. The higher the task importance, the greater the priority. The distance between AGVs is calculated in real time. When the distance is less than the safe distance, the avoidance mechanism is triggered. When avoiding, the AGV with lower priority gives way. If the priorities are the same, the AGV on the right gives way. A one-way traffic rule is set in narrow passages to prevent AGVs from deadlocking when traveling in opposite directions. The narrow passage refers to an area without adjacent grids on both sides of the divided grid.

[0098] Calculate the mesh density function for the area that needs to be re-meshed:

[0099] ;

[0100] Where m is the number of grids in the region, and is the center coordinate of grid j, is the standard deviation of grid j, which represents the degree of discreteness of AGV activities or task distribution within the grid.

[0101] Use the Lloyd algorithm to iteratively optimize mesh division, and update the mesh center coordinates at each iteration:

[0102] ;

[0103] ;

[0104] Where V is the Voronoi region of grid j.

[0105] Calculate quality indicators for new meshing , ,in is the area of ​​grid j, when or When, that is, the quality index The change is less than the preset threshold Or when the maximum number of iterations is reached, the iteration is terminated.

[0106] When an AGV is detected to be abnormal, the task of the AGV is immediately suspended and a fault alarm is sent; the unfinished tasks of the faulty AGV are automatically reallocated to other available AGVs; for faults that can be automatically recovered, the preset recovery procedure is executed; for faults that require manual intervention, a maintenance work order is generated. Example 2

[0107] like Figure 2As shown, it is a schematic diagram of an industrial mobile AGV cluster control system applied to a factory of the present invention, the system is used to execute the control method described in Example 1, and is characterized in that the system includes: an initial grid division module, a path planning module, and an AGV monitoring and adjustment module connected in sequence.

[0108] The grid division module is used to divide the entire factory area into multiple grid areas by uniform division at the initial moment, and each grid area is allowed to have at most one AGV at the same time.

[0109] The path planning module is used to dynamically plan the optimal path for each AGV according to the grid occupancy at the current moment using an improved AGV cluster control algorithm.

[0110] The AGV monitoring and adjustment module is used to control each AGV to move along the planned path and collect the AGV's residence time in each grid; when the AGV's residence time in a grid exceeds a preset threshold, the grid division of the area is replanned, and the optimal path is replanned.

[0111] Monitor the operating status of AGV in real time, including parameters such as position, speed, and power; immediately suspend the task of the AGV and send a fault alarm when an AGV abnormality is detected; automatically reallocate the unfinished tasks of the faulty AGV to other available AGVs; execute the preset recovery procedure for faults that can be automatically recovered; generate a maintenance work order for faults that require manual intervention.

[0112] The system also includes an energy management module for real-time monitoring of the power status of each AGV. When the power of the AGV is lower than a preset threshold, it automatically plans to go to the nearest charging station. According to the task volume and the number of AGVs, the number and location of charging stations are dynamically adjusted to realize the rotation charging of AGVs and ensure the continuous operation of the system.

[0113] It should be noted that those skilled in the art should understand that various changes and equivalent substitutions may be made to the present invention without departing from the scope of the present invention. In addition, various modifications may be made to the present invention for specific situations or materials without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed, but should include all embodiments falling within the scope of the claims of the present invention.

Claims

1. An industrial mobile AGV cluster control method applied to a factory, characterized in that: The method comprises the following steps: Step S1, at the initial moment, the entire factory area is divided into multiple grid areas by uniform division, and each grid area is allowed to have at most one AGV at the same time; Control the moving speed of each AGV and number each AGV in sequence. The Euclidean distance from the current position of an AGV to the target position: ； in, and For the The horizontal and vertical coordinates of the current position of the AGV, and For the The horizontal and vertical coordinates of the target position of the AGV; Set The moving speed vector of an AGV: ; in, ， and are constants, representing the weights of the target attraction and obstacle repulsion, respectively, and n is the total number of AGVs; Update the position of the AGV. The position coordinates of the AGV at time t+1 satisfy: ； ； in, and They are respectively the first The horizontal and vertical coordinates of the current position of the AGV, and They are respectively the first The vertical coordinate of the current position of the AGV, and is the component of the moving velocity vector in the horizontal and vertical coordinates, is the time interval between adjacent moments; Step S2: According to the grid occupancy at the current moment, the improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV; In step S2, an improved AGV cluster control algorithm is used to dynamically plan the optimal path for each AGV, including: For AGV, from accepting a task to completing it, no matter how the path is planned, the number of grids it must pass through or stay in is , the mathematical expression of the fitness function is: ； in, For the i-th path planning sequence, each grid that must be passed or stopped only appears once. In the path planning sequence, the adjacent nodes to distance; Construct a search map for the improved AGV cluster control algorithm: ； in, is the total number of paths, and all paths are iterated sequentially to the maximum number of times. At this time, the matrix The minimum fitness function value in That is the The optimal path for each AGV; Step S3: Control each AGV to move along the planned path and collect the AGV’s residence time in each grid; Step S4: When the AGV stays in a certain grid for longer than a preset threshold, the grid division of the area is replanned, and the optimal path is replanned.

2. The industrial mobile AGV cluster control method applied to a factory according to claim 1, characterized in that: When dynamically planning the optimal path for each AGV, the smoothness of the path and the need to avoid sharp turns are also considered. The specific implementation method is as follows: When constructing the fitness function, a path smoothness penalty term is introduced to evaluate the change in the angle between adjacent segments in the path. The mathematical expression of this penalty term is: ； in, It represents the angle formed by three consecutive nodes z, z+1 and z+2 in the path planning sequence. The size of the angle is calculated by the dot product formula of the vector: ； The fitness function that comprehensively considers path length and smoothness is modified as follows: ； in, is the weight coefficient of the smoothness penalty term, , when searching for the optimal path, through iterative optimization, the fitness function Reach the minimum value and get the optimal path.

3. An industrial mobile AGV cluster control method applied to a factory according to any one of claims 1-2, characterized in that: The method also includes an adaptive meshing mechanism, specifically including: Initially, uniform grid division is adopted; During the system operation, the AGV passing frequency and residence time of each grid are counted, and the upper and lower thresholds of the passing frequency and the upper and lower thresholds of the residence time are set; For areas where the AGV passing frequency is higher than the passing frequency upper limit threshold or the residence time is longer than the residence time upper limit threshold, the grid division accuracy is increased; For areas where the AGV passing frequency is lower than the passing frequency lower limit threshold or the residence time is less than the residence time lower limit threshold, the grid division accuracy is reduced.

4. The industrial mobile AGV cluster control method applied to a factory according to claim 3 is characterized in that: In one monitoring cycle, monitor the residence time of each AGV in grid j, and calculate the residence time of all n AGVs in grid j , For the The residence time of an AGV in grid j.

5. The industrial mobile AGV cluster control method applied to a factory according to claim 4, characterized in that: Set a safety distance for the AGV. When performing adaptive meshing, the distance between adjacent mesh centers and the minimum side length of each mesh should not be less than the safety distance. For AGVs performing tasks, a priority is set for each AGV according to the importance of the task. The higher the task importance, the greater the priority. The distance between AGVs is calculated in real time. When the distance is less than the safe distance, the avoidance mechanism is triggered. When avoiding, the AGV with lower priority gives way. If the priorities are the same, the AGV on the right gives way. A one-way traffic rule is set in narrow passages to prevent AGVs from deadlocking when traveling in opposite directions. The narrow passage refers to an area without adjacent grids on both sides of the divided grid.

6. The industrial mobile AGV cluster control method applied to a factory according to claim 5, characterized in that: Calculate the mesh density function for the area that needs to be re-meshed: ； Where m is the number of grids in the region, and is the center coordinate of grid j, is the standard deviation of grid j, which represents the degree of discreteness of AGV activities or task distribution within the grid; Use the Lloyd algorithm to iteratively optimize mesh division, and update the mesh center coordinates at each iteration: ； ； Where V is the Voronoi region of grid j; Calculate quality indicators for new meshing , ,in is the area of ​​grid j, when or When, that is, the quality index The change is less than the preset threshold Or when the maximum number of iterations is reached, the iteration is terminated.

7. The industrial mobile AGV cluster control method applied to a factory according to claim 6, characterized in that: When an AGV is detected to be abnormal, the task of the AGV is immediately suspended and a fault alarm is sent; the unfinished tasks of the faulty AGV are automatically reallocated to other available AGVs; for faults that can be automatically recovered, the preset recovery procedure is executed; for faults that require manual intervention, a maintenance work order is generated.

8. An industrial mobile AGV cluster control system applied to a factory, the system being used to execute the control method according to any one of claims 1 to 6, characterized in that: The system includes: an initial grid division module, a path planning module, and an AGV monitoring and adjustment module connected in sequence; The grid division module is used to divide the entire factory area into multiple grid areas by uniform division at the initial moment, and each grid area is allowed to have at most one AGV at the same time; The path planning module is used to dynamically plan the optimal path for each AGV using an improved AGV cluster control algorithm according to the grid occupancy at the current moment; The AGV monitoring and adjustment module is used to control each AGV to move along the planned path and collect the residence time of the AGV in each grid; When the AGV stays in a certain grid for longer than a preset threshold, the grid division of the area is replanned and the optimal path is replanned.

Images