A AGV scheduling method and system based on matrix mapping

Abstract

The present invention relates to an AGV scheduling method and system based on matrix mapping. The present invention performs AGV scheduling by reordering the order of AGV handling tasks, matches the empty load formation before the AGV handling task with the subsequent handling tasks one by one, and gives priority to the handling tasks whose paths are included in the empty load paths in the subsequent handling tasks, which greatly reduces the empty load travel of the AGV transport vehicle, and at the same time reduces the waiting time of the subsequent tasks, and reduces the average waiting time of the AGV task execution. Compared with other AGV scheduling methods, the AGV scheduling method provided by the present invention is simple, does not involve task allocation between vehicles, has good robustness, is suitable for AGV handling occasions with long distances or large sites and multiple tasks, and can be combined with multiple production management systems, and has strong compatibility.

Description

Technical Field

[0001] The present invention relates to a scheduling method and system, belonging to the technical field of production automatic transport equipment and transport task scheduling, and relates to an AGV scheduling method and system based on matrix mapping. Background Art

[0002] Automated guided vehicles are also commonly called AGVs. They are transport vehicles equipped with electromagnetic or optical automatic navigation devices, which do not require a driver to drive, can travel along a pre-defined navigation path, and have safety protection and various transfer functions. They are key equipment for achieving full automation of the production process and promoting the progress of Industry 4.0. A complete AGV system consists of an AGV vehicle and a corresponding AGV scheduling system. The AGV scheduling system, as the brain of the AGV system, is used to comprehensively allocate handling tasks, plan paths, and operate and maintain the entire system. Whether the AGV scheduling system matches the production site will directly determine whether the AGV's handling is efficient. Efficient handling efficiency will effectively improve production rhythm and production efficiency, and enhance the flexible production capacity of the production line. At present, AGVs are widely used in various fields of intelligent manufacturing, such as aerospace, automobile, medicine, battery production and other production sites that require material handling.

[0003] Given the high degree of automation of AGV carts, there are only a few AGV transporters in the production process to undertake the material handling tasks of dozens to hundreds of workstations in the entire production site. The traditional "first come, first served" AGV scheduling method that executes the handling tasks in sequence according to the task issuance cannot improve the handling efficiency, but will increase the AGV's empty distance; AGV path planning based on site conditions is only applicable to closed-loop production sites; single-guide path network layout or mixed two-way guide path network layout, this scheduling method has clear ideas but poor system flexibility and robustness, and is currently only applicable to small-scale handling sites. Summary of the invention

[0004] The purpose of the present invention is to provide an AGV scheduling system based on matrix mapping, which solves the drawbacks of the traditional "first come, first served" AGV scheduling method in the prior art, which performs the handling tasks in sequence according to the task issuance. It solves the problem of low handling efficiency and increased AGV empty-load distance in this method; it overcomes the problem that the AGV shortest path planning based on site conditions is only applicable to closed-loop production sites; the single-guide path network layout or mixed two-way guide path network layout, this scheduling method has clear ideas but poor system flexibility and robustness, and is currently only applicable to small-scale handling sites. The present AGV scheduling method has strong versatility and a wide range of applications.

[0005] The AGV scheduling system based on matrix mapping of the present invention is divided into three parts: the first part is the matrix mapping of the workshop map, which converts the workshop map into a digital map; the second part is the matrix mapping of the AGV trolley transportation path, which displays the transportation path in the digital map through binary 0 and 1 transformation; the third part is to adjust the AGV task order through path matching, and execute the transportation task matching the empty path in advance, so as to improve the transportation efficiency.

[0006] The invention discloses an AGV scheduling method based on matrix mapping, which specifically comprises the following steps.

[0007] The first part is the matrix mapping of the workshop map. The specific steps are:

[0008] Step 1: Clearly mark the specific AGV transfer points, warehouse points and main road routes on the workshop map.

[0009] Step 2: Since most production plants are built with long spans, the direction of the long span on the map is defined as the Y-axis direction of the map coordinate system, and the main road perpendicular to the long span is defined as the X-axis direction of the map coordinate system.

[0010] Step 3: In the horizontal direction, the horizontal line corresponding to each AGV transport point corresponds to the horizontal row of the matrix. Multiple AGV transport points on the same horizontal line can correspond to a horizontal row of the matrix.

[0011] Step 4, map the main roads in the horizontal direction of the map as a horizontal row in the corresponding horizontal position, and map other transportation points such as line-side warehouses and three-dimensional warehouses as a point in the corresponding horizontal position and represent them in a new matrix row.

[0012] Step 5, perform fuzzy algorithm matching between channels and transport points. If a transport point has multiple sub-transport points, the sub-transport points can be replaced by the parent transport point.

[0013] Step 6: If there is an AGV transfer point on one side of the main transport road, the intersection point a is the intersection of the vertical column of the main road and the horizontal column of the AGV transfer point. mn It is mapped to an element position of the map matrix, where m represents the row position of the AGV transport point and n represents the column position of the AGV transport point.

[0014] Step 7, map all the intersections in the above map according to the method in step 6, and generate a matrix with the number of horizontal lines, the number of horizontal channels, the number of stereoscopic warehouses and line-side warehouses corresponding to the AGV transfer points as the number of rows; the main transportation channels as the number of columns, all AGV point transfer points as the corresponding elements of the matrix, and the remaining positions are filled.

[0015] The second part is the matrix mapping of the AGV trolley transport path. The specific steps are:

[0016] Step 8: Assign all elements in the matrix to 0.

[0017] Step 9: According to the starting point, end point and actual running path of the AGV handling task, the value of the element along this path is changed from 0 to 1 in the matrix map.

[0018] Step 10: Under the same path, the matrices corresponding to the starting point to the end point and the end point to the starting point are the same, so the running direction is determined. The element positions corresponding to the starting point are taken in order. and the next passing point Make a judgment, if m 1 ≠m 2 Jump to step 11. If m 1 =m 2 Skip to step 12.

[0019] Step 11, m 1 ≠m 2 , continue to judge m 1 ,m 2 The size relationship, if m 1 <m 2 , then add a + sign before the path matrix to indicate the direction. If m 1 >m 2 , then add a - sign before the path matrix to indicate the direction.

[0020] Step 12, m 1 =m 2 , discard the starting point, and take the first and second points passed Medium 2 ,m 3 Continue to compare the sizes. If they are the same, continue to discard the previous point and add the next point. The comparison method is the same as step 12. If the m values ​​of all elements passed by this path are the same, skip the third part and adjust the AGV task order through path matching, and directly execute the task.

[0021] The third part is to adjust the AGV task order through path matching. The specific steps are as follows:

[0022] Step 13: Map all AGV handling tasks into a handling path matrix according to the matrix mapping step of the AGV trolley handling path in the second part above, execute the nth (n≥1) AGV handling task, and set the matrix position of the end element of this task The starting point for the next AGV handling task Combine and generate empty path for handling tasks The direction is determined by the matrix mapping determination method of the AGV transport path in the second part, and the matrix is ​​assigned a positive or negative sign. If a + sign is added before the matrix, jump to step 14; if a - sign is added before the matrix, jump to step 15.

[0023] Step 14: If the path is empty There is a + sign before the matrix. Let the starting point of any AGV transport path matrix be Filter all subsequent starting elements n n+1 =n n+2 And m n+1 <m n+2 Perform matrix subtraction on matrices preceded by a + sign.

[0024] Step 15: If the path is empty There is a minus sign before the matrix. Let the starting point of any AGV transport path matrix be Filter all subsequent starting elements n n+1 =n n+2 And m n+1 >m n+2 Perform matrix subtraction on matrices preceded by a minus sign.

[0025] Step 16, judge each element of the matrix after the subtraction operation. If there is no -1 in the matrix, it indicates that the path does not match and is discarded. If there is no -1, proceed to step 17.

[0026] Step 17, if the values ​​of the matrix elements do not contain -1 after the subtraction operation, the transport task will be sorted and executed after the running tasks. If there is no such task, the tasks will be executed in the original order.

[0027] Step 18: Repeat steps 13 to 17 until the task is completed.

[0028] The beneficial effects of the present invention are

[0029] The present invention performs AGV scheduling by reordering the order of AGV handling tasks, matches the empty load formation before the AGV handling task with the subsequent handling tasks one by one, and prioritizes the handling tasks whose paths are included in the empty load paths in the subsequent handling tasks, which greatly reduces the empty load travel of the AGV transport vehicle, while reducing the waiting time for subsequent tasks and the average waiting time for AGV task execution. Compared with other AGV scheduling methods, the AGV scheduling method provided by the present invention is simple, does not involve task allocation between vehicles, has good robustness, is suitable for AGV handling situations with long distances or large sites and multiple tasks, and can be combined with multiple production management systems, with strong compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a randomly generated site layout diagram of the present invention;

[0031] Figure 2 It is a diagram for establishing the AGV map coordinates shown in the present invention;

[0032] Figure 3It is a creation diagram of the matrix map of the present invention;

[0033] Figure 4 is the map matrix after mapping of the present invention;

[0034] Figure 5 is a digital matrix of the matrix-based path mapping of the present invention;

[0035] Figure 6 ac handling task a 3,1 Three-level digital matrix representation from workstation to stereoscopic warehouse

[0036] Figure 7 The empty-load path is matched and calculated with the subsequent handling task path. DETAILED DESCRIPTION

[0037] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] The present invention adopts the following technical scheme. The AGV scheduling system based on matrix mapping is divided into three parts: the first part is the matrix mapping of the workshop map, which converts the workshop map into a digital map; the second part is the matrix mapping of the AGV trolley transport path, which displays the transport path in the digital map through binary 0 and 1 transformation; the third part is to adjust the AGV task order through path matching, and execute the transport task matching the empty path in advance, so as to improve the transport efficiency.

[0040] The invention discloses an AGV scheduling method based on matrix mapping, which specifically comprises the following steps.

[0041] The first part is the matrix mapping of the workshop map. The specific steps are:

[0042] Step 1, such as Figure 1 As shown, the site construction map clearly marks the specific AGV handling points, warehouse points and main road paths.

[0043] Step 2: Considering that most production plants are built in a long-span manner, the vertical direction of the long-span trunk road on the map is defined as the Y-axis direction of the map coordinate system, i.e., the vertical direction. The number of trunk roads on the map is mapped to the number of columns in the digital map matrix, such as Figure 2 shown.

[0044] Step 3, create a matrix map, such as Figure 3 As shown, in the horizontal direction, the horizontal line corresponding to each AGV transport point corresponds to the horizontal row of the matrix, and multiple AGV transport points on the same horizontal line can correspond to a horizontal row of the matrix.

[0045] Step 4, map the main roads in the horizontal direction of the map as a horizontal row in the corresponding horizontal position, map the number of horizontal main roads as the number of rows of the digital map matrix, and map other transportation points such as line-side warehouses and three-dimensional warehouses as a point in the corresponding horizontal position.

[0046] Step 5, perform fuzzy algorithm matching between the channel and the transport point. If there is an AGV transport point on one side of the main transport road, jump to step 6; if there are AGV transport points on both sides of the main transport road, jump to step 7.

[0047] Step 6: If there is an AGV transfer point on one side of the main transport road, the intersection point a is the intersection of the vertical column of the main road and the horizontal column of the AGV transfer point. mn It is mapped to an element position of the map matrix, where m represents the row position of the AGV transport point and n represents the column position of the AGV transport point.

[0048] In this embodiment, if there is an AGV transfer point on one side of the main transport road, the AGV turns left or right (i.e., selects one direction) to enter the transfer point when it reaches the designated position on the main road. In the case of transfer points on both sides of the main transport road, the AGV can turn left or right (i.e., select two directions) when it reaches the designated position on the main road. This embodiment studies the scheduling of AGVs, which is calculated based on the path matching of AGVs. The path matching path refers to the running track of the AGV on the main road, regardless of whether it turns left or right. The matrix is ​​mapped through the main road, but the transfer point is not on the matrix.

[0049] Step 7: Map all the intersections in the above map according to step 6 to generate a matrix with the number of horizontal lines, horizontal channels, stereo warehouses and line-side warehouses corresponding to the AGV transfer points as the number of rows and the main transportation channels as the number of columns. All AGV point transfer points are the corresponding elements of the matrix, and the remaining positions are filled. The generated matrix map is as follows: Figure 4 As shown, the generated digital matrix is ​​as follows Figure 5 As shown, the number 1 in the matrix indicates that this position is the actual transport point, and the number 0 indicates that this position is only a supplementary point and does not actually exist. The 10th row of the digital matrix indicates the horizontal transport channel, and the 11th row indicates the stereoscopic warehouse point.

[0050] The second part is the matrix mapping of the AGV trolley transport path. The specific steps are:

[0051] Step 8: Assign all elements in the matrix to 0.

[0052] Step 9: According to the starting point, end point and actual running path of the AGV handling task, the values ​​of the elements along this path are changed from 0 to 1 in the matrix map. Figure 6 As shown, point a 31The transportation path to the three-dimensional warehouse storage location passes through a 31 、a 41 、a 51 、a 61 、a 71 、a 81 、a 91 After passing through the horizontal channel to reach the stereo library, it can be represented in the matrix mapping map as Figure 6 Matrix, another transport path a 41 Move to a 91 Same reason.

[0053] Step 10: Under the same path, the matrices corresponding to the starting point to the end point and the end point to the starting point are the same, so the running direction is determined. The element positions corresponding to the starting point are taken in order. and the next passing point Make a judgment, if m 1 ≠m 2 Jump to step 11. If m 1 =m 2 Skip to step 12.

[0054] Step 11, m 1 ≠m 2 , continue to judge m 1 ,m 2 The size relationship, if m 1 <m 2 , then add a + sign before the path matrix to indicate the direction. If m 1 >m 2 , then add a - sign before the path matrix to indicate the direction.

[0055] Step 12, m 1 =m 2 , discard the starting point, and take the first and second points passed Medium 2 ,m 3 Continue to compare the sizes. If they are the same, continue to discard the previous point and add the next point. The comparison method is the same as step 11. If the m values ​​of all elements passed by this path are the same, skip the third part and adjust the AGV task order through path matching, and directly execute the task.

[0056] The third part is to adjust the AGV task order through path matching. The specific steps are as follows:

[0057] Step 13: Map all AGV handling tasks into a handling path matrix according to the matrix mapping step of the AGV trolley handling path in the second part above, execute the nth (n≥1) AGV handling task, and set the matrix position of the end element of this task The starting point for the next AGV handling task Combined to generate empty path for handling tasks The direction is determined by the matrix mapping determination method of the AGV transport path in the second part, and the matrix is ​​assigned a positive or negative sign. If a + sign is added before the matrix, jump to step 14; if a - sign is added before the matrix, jump to step 15.

[0058] Step 14: If the path is empty If there is a + sign before the matrix, all subsequent transport task matrices with a + sign before them will be screened for matrix subtraction.

[0059] Step 15, if no load path There is a minus sign before the matrix. Let the starting point of any AGV transport path matrix be Filter all subsequent starting elements n i+1 <n i+2 And m k+1 <m k+2 Perform matrix subtraction on matrices preceded by a minus sign.

[0060] Step 16, judge each element of the matrix after the subtraction operation. If there is no -1 in the matrix, it indicates that the path does not match and is discarded. If there is no -1, proceed to step 17.

[0061] like Figure 7 As shown, a 31 The empty path to the stereoscopic warehouse after a certain AGV task, a 41 Move to a 91 is a certain transport task, then the matching operation of these two tasks is as follows Figure 7 express.

[0062] The first one is an unloaded matrix. It is generated as follows: The path of the current task is from a 11,4 to a 3,1 , the next task is a 11,4 to a 4,3 , then the end point of the current task is a 3,1 , the starting point of the next task is a 11,4 The current task and the next task generate an unloaded matrix such as Figure 7 The leftmost matrix is ​​shown in the figure. Its running direction is from top to bottom, so it is a + sign.

[0063] Step 17, if the values ​​of the matrix elements do not contain -1 after the subtraction operation, the transport task will be sorted and executed after the running tasks. If there is no such task, the tasks will be executed in the original order.

[0064] Step 18: Repeat steps 13 to 17 until the task is completed.

[0065] At the same time, the present invention provides a method for generating a handling task matrix and an empty-load task matrix.

[0066] The method for generating the handling task matrix and the unloaded task matrix provided by the present invention adopts matrix transposition operation and pointer operation of C language. The specific generation method is:

[0067] 1. After step 8, the generated map matrix A is converted into a transposed matrix A T , that is, any element a in the matrix A m,n With A T Element a in the matrix n,m The original matrix A m,n Convert to A n,m , the AGV trajectory that originally moves along the Y-axis direction in the map matrix moves along the horizontal direction, that is, the row direction of the matrix, in the transposed matrix.

[0068] 2. The pointer operation provided in C language can conveniently perform matrix row operations. Through the C language void inta[n][m]={} statement, a transposed matrix with element value 0 is generated, where m is the number of rows of the original map digital matrix and n is the number of columns of the original map digital matrix.

[0069] 3. Generate the path in the transposed matrix whose element values ​​are all 0 through the pointer operation of C language. If the starting point and end point of the AGV transportation are The starting point in the transposed map becomes

[0070] 4. Create a pointer int(*p)[m] pointing to a one-dimensional array containing m integer elements, and in the nth matrix of the matrix where the path needs to be generated 1 line, through *(*(p+n 1 )) points the pointer to the row, and through the loop operation *(*(p+n 1 )+m 1 ) to *(*(p+n 1 )+m-1) are assigned a value of 1. The row is also converted to *(*(p+n 2 )+m 2 ) to *(*(p+n 2 )+m-1) is assigned the value 1.

[0071] 5. For matrix n 1 ~n 2 The operation is performed on the rows between the two matrices by *(*(p+n)+m-1) where the matrix n 1 ~n 2 The m-1th bit of the row between is assigned a value of 1 to generate a partial path through the lateral channel in the path matrix.

[0072] 6. After the above operation is completed, generate the transposed matrix again and transform the matrix A T Transpose generation (A T ) T , the obtained (A T ) T This is the required transport path matrix.

Claims

1. A scheduling method for automated guided vehicles based on matrix mapping, It is characterized in that include: A point matrix mapping step, mapping the area where the automated guided vehicle is located into a transport point matrix, wherein the rows and columns of the transport point matrix represent the main roads in the horizontal and vertical directions of the automated guided vehicle, respectively, and the elements in the transport point matrix represent the transport points of the automated guided vehicle; A path matrix mapping step, based on the starting point and the end point of the transport task and in combination with the transport point matrix, generates a task path matrix reflecting the transport task, wherein the non-zero elements in the task path matrix represent the points that the transport task passes through from the starting point and the end point; An empty path generation step, combining the matrix position of the end point element of the current transport task with the matrix position of the starting point element of the next transport task to generate an empty path matrix for the transport task; A path direction determination step, determining the task scheduling direction of the idle path matrix and the task path matrix; The task order scheduling step is to compare the idle path matrix with the task path matrix having the same scheduling direction along the matrix column direction. If the element in the idle path matrix is ​​greater than the element at the corresponding position of the task path matrix, the task corresponding to the task path matrix whose corresponding matrix elements are all smaller than the element at the corresponding position of the idle path matrix will be scheduled first after the current task is completed. Among them, first determine whether the lateral direction of the empty path matrix is ​​the same as that of other task transport paths, then compare it with the transport paths with the same lateral direction and task starting points in the same column, and then determine whether to adjust the order of the transport tasks.

2. The method for scheduling an automated guided vehicle based on matrix mapping according to claim 1, It is characterized in that In the point matrix mapping step, positions without transport points are mapped to zero elements of corresponding positions in the point matrix, and positions with transport points are mapped to non-zero elements of corresponding positions in the transport point matrix.

3. The method for scheduling an automated guided vehicle based on matrix mapping according to claim 1, It is characterized in that In the path matrix mapping step, the element position corresponding to the starting point of the transport task and the next passing point are sequentially taken for judgment, and the running path is divided into a positive direction and a negative direction based on whether the running path is along the direction of increasing number of rows in the matrix. Among them, the direction along the direction of increasing number of rows in the matrix is ​​divided into a positive direction, and the direction along the direction of decreasing number of rows in the matrix is ​​divided into a negative direction.

4. The method for scheduling an automated guided vehicle based on matrix mapping according to claim 3, It is characterized in that If the no-load path matrix is ​​in the positive direction, select the task path matrix that is in the positive direction and does not run in the horizontal direction of the matrix for subtraction operation.

5. The method for scheduling an automated guided vehicle based on matrix mapping according to claim 3, It is characterized in that If the no-load path matrix is ​​in the negative direction, select the task path matrix that is in the negative direction and does not run in the horizontal direction of the matrix for subtraction operation.

6. The method for scheduling an automated guided vehicle based on matrix mapping according to claim 1, It is characterized in that In the path direction determination step, if the execution path of the task is only along the row direction of the matrix, the tasks in this direction will not be scheduled preferentially.

7. A scheduling system for automated guided vehicles based on matrix mapping, It is characterized in that include: The point matrix mapping module is used to map the area where the automated guided vehicle is located into a transport point matrix, wherein the rows and columns of the transport point matrix represent the main roads in the horizontal and vertical directions of the automated guided vehicle, respectively, and the elements in the transport point matrix represent the transport points of the automated guided vehicle; wherein, firstly determine whether the lateral direction of the empty path matrix is ​​the same as that of the transport paths of other tasks, and then compare it with the transport paths with the same lateral direction and task starting points in the same column, and then determine whether to adjust the order of the transport tasks; A path matrix mapping module, for generating a task path matrix reflecting the transport task based on the starting point and the end point of the transport task and in combination with the transport point matrix, wherein the non-zero elements in the task path matrix represent the points that the transport task passes through from the starting point and the end point; An empty path generation module, used for combining the matrix position of the end point element of the current handling task with the matrix position of the starting point element of the next handling task to generate an empty path matrix for the handling task; A path direction determination module, used to determine the task scheduling direction of the idle path matrix and the task path matrix; The task order scheduling module is used to compare the no-load path matrix with the task path matrix having the same scheduling direction along the matrix column direction. If the elements in the no-load path matrix are greater than the elements at the corresponding positions of the task path matrix, the tasks corresponding to the task path matrix whose corresponding matrix elements are all less than the elements at the corresponding positions of the no-load path matrix will be scheduled preferentially after the current task is completed.

8. The matrix-based dispatching system for automated guided vehicles according to claim 7, It is characterized in that In the point matrix mapping module, the position where there is no transport point is mapped to the zero element of the corresponding position in the point matrix, and the position where there is a transport point is mapped to the non-zero element of the corresponding position in the transport point matrix.

9. The matrix-based dispatching system for automated guided vehicles according to claim 7, It is characterized in that In the path matrix mapping module, the element position corresponding to the starting point of the transport task and the next passing point are sequentially taken for judgment, and the running path is divided into a positive direction and a negative direction based on whether the running path is along the direction of increasing number of rows in the matrix. Among them, the direction along the direction of increasing number of rows in the matrix is ​​divided into a positive direction, and the direction along the direction of decreasing number of rows in the matrix is ​​divided into a negative direction.

10. The matrix-based dispatching system for automated guided vehicles according to claim 9, It is characterized in that If the no-load path matrix is ​​in the positive direction, select the task path matrix that is in the positive direction and does not run in the horizontal direction of the matrix for subtraction operation.

Images