Robot multi-machine cooperation storage yard management method for prefabricated part factory based on 5G network

Abstract

The technical scheme of the invention provides a robot multi-machine cooperation storage yard management method for a prefabricated part factory based on a 5G network. The invention aims to provide a multi-robot cooperative management method for a building material (prefabricated part) unmanned carrying storage yard of a prefabricated part factory based on a field 5G environment, and the method achieves the cooperative management of a mobile recognition robot, an AGV carrying robot and an inspection robot through a data standard work package in a 5G network environment. And the execution progress and efficiency of the robot work package are monitored and managed, the overall efficiency of building material transportation of a prefabricated part factory is improved, the related labor cost of the prefabricated part factory is reduced, the material dynamic balance of field prefabricated part production is further improved, and the overall productivity and benefits of the prefabricated part factory are improved.

Description

Technical Field

[0001] The present invention relates to a robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network, and belongs to the technical field of intelligent yard management application of a building prefabricated component factory. Background Art

[0002] With the large number of new prefabricated construction projects, the demand for prefabricated components is also increasing. In order to effectively improve the production and transportation efficiency of prefabricated component factories, some component factories have begun to explore the combined application of technologies such as "BIM+", "GIS+", and "robots" to improve the overall material transportation efficiency on site.

[0003] The Chinese invention patent with publication number CN114547866A, which was published on May 27, 2022, discloses a prefabricated component intelligent detection method based on BIM-UAV-Robot Dog, and provides a prefabricated component intelligent detection method based on BIM-UAV-Robot Dog. The method uses the full-scene scanning results of the prefabricated component factory by the drone to establish a hierarchical node model corresponding to the full scene; the target scanning path of the prefabricated component factory is determined according to the hierarchical node model, so that the prefabricated component point cloud data is collected by the robot dog traveling along the target scanning path; the component size corresponding to each prefabricated component in the prefabricated component factory is extracted from the prefabricated component point cloud data; the size of each component is compared with the corresponding geometric size information in the building three-dimensional model BIM to determine the quality inspection report corresponding to each prefabricated component. In this method, the data between devices is only interactively circulated in the form of a single information flow, and is not processed and aggregated to form a standard data work package instruction, which will greatly affect the data transmission and processing speed, and the data between each device is relatively independent, and the work packages formed between different devices cannot be sorted by standard tasks to form a complete set of multi-machine collaboration processes, which affects the accuracy and efficiency of collaborative work between devices. This method does not take into account the initial site layout technology of the project. It is necessary to scan and enter the site according to the existing task area, so as to use the layout, and the coverage content has limitations. It also does not describe the correlation method with GPS positioning and other coordinates, and there is a certain accuracy deviation in the transportation route guidance.

[0004] The Chinese invention patent application with publication number CN112598320A, published on April 2, 2021, discloses a method for managing prefabricated building components based on a building information model, and provides a method for managing prefabricated building components based on a building information model. By implanting and identifying RFID chips, the information of each component and the progress information of each construction stage are updated into the prefabricated component element corresponding to the building information model database. However, the method still requires relevant personnel to perform the association between material identification and RFID, and cannot achieve unmanned intelligent operation, and the corresponding related costs cannot be optimized.

[0005] The Chinese invention patent application with publication number CN110837541A, which was published on February 25, 2020, discloses a "prefabricated component factory production management system based on GIS+BIM system". It provides a prefabricated component factory production management system based on GIS+BIM, which provides a massive amount of geographic information data and spatial information for the beam yard through GIS technology, helping the prefabricated component factory to quickly simulate various production scene terrain data on the 3D facade, and display the spatial position relationship between each component object and the environment in real time. However, this invention patent application only considers model factors for material identification content. This model only refers to the external dimension data of the prefabricated components and the site BIM model, and does not comprehensively consider data such as the yard weight and the number of layers of yard components fully loaded, which can easily cause result deviations due to shape recognition errors. Summary of the invention

[0006] The purpose of the present invention is to provide a robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network, so that the yard management of the prefabricated component factory is more intelligent, accurate and efficient, and provides a new intelligent method for the yard management of the building prefabricated component factory.

[0007] In order to achieve the above object, the technical solution of the present invention is to provide a robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network, which is characterized by comprising the following steps:

[0008] S1: Automatically arrange the BIM initial site model according to the site information;

[0009] S2: Generate a robot transportation path plan based on the BIM initial site model and robot and material information, which specifically includes the following steps:

[0010] S21: By measuring the coordinates of the starting point of the actual site point and defining the relevant points of the prefabricated components in the model, the loading point A (X Start , Y Start ) and the coordinates of the unloading points of each prefabricated component yard Wherein, i represents the area to which different types of prefabricated components belong, j represents the jth prefabricated component stacking area of ​​the same type, j = 1, 2, ..., u;

[0011] S22: Prefabricate the loading point A(X Start , Y Start ), coordinates of unloading points of each prefabricated component yard The initial BIM layout model locates the scope of equipment in the site, the overall model boundary, the yard area, and the yard interval. Based on the above data and the Dijkstra algorithm, the optimal transportation path set is obtained by fitting the curing pit loading node to the unloading node of each material area.

[0012] S23: Optimal path for transporting various materials The starting state c of the AGV task work package and the standard speed V of the AGV are combined with the bubble sort algorithm, and the limit that no more than 3 AGVs can be located within the 1m2 range of the path at the same time is given to obtain the AGV dynamic task time sorting;

[0013] S3: The mobile robot and AGV divide the standard work packages according to the path plan and perform multi-machine collaborative tasks, including the following steps:

[0014] S31: The mobile recognition robot moves to the curing cellar loading point A (X Start , Y Start ), the prefabricated component category image recognition algorithm after training is used to identify the category S of the transported component in the current task state. At the same time, when a component is first identified in the range of this point while being transported by the slide rail, this moment is defined as the work package start time t Start ;

[0015] S32: Start time t of work package Start , issue a command to the AGV to carry the work package Where m represents the weight of the transported material, t End Indicates the planned work package end time, t End =D n / V, Z represents the work package serial number, and A represents the coordinate A (X) of the initial point of material transportation described in step S22. Start , Y Start ), B represents the coordinates of the material unloading point described in step S22 Indicates the optimal material transportation path described in step S22;

[0016] S33: Based on the above work package instance, the AGV transport task is executed, and the work package cumulative resource allocation is performed. Referring to step S23, the overall transport multi-machine collaborative work package task execution plan P is obtained, where P = {O1, O2, ..., O u};

[0017] S34: Obtain the actual AGV task end time Kt through the AGV task end trigger point End , where Kt End It means that the AGV enters the unloading point range and is in a non-displacement state for more than 5 seconds, which is considered as the end time of the work package;

[0018] S35: The difference Q between the planned time and the actual time is regarded as the actual plan correction value, and Q is defined as the time deviation, Q = |Kt End -t End|: When Q ≥ 20s, the existing task plan is dynamically updated and the next task starts t 2 Start Time correction, t 2 start =t Start + Q, and obtain the revised overall multi-machine collaborative work package task execution plan P′={O′1,O′2,…,O′ u};

[0019] S4: The inspection robot collects data and provides feedback based on the on-site yard performance, and dynamically adjusts the transportation path plan.

[0020] Preferably, the step S1 comprises the following steps:

[0021] S11: Survey the boundary coordinates of the site through a GPS locator to obtain GPS coordinate information of the boundary positioning of the transportation starting area including the yard boundary and the curing pit;

[0022] S12: The GPS coordinates obtained in the previous step are converted into spatial coordinates through the Bursa formula to obtain the converted spatial coordinate information, which is expressed as (X1, Y1), (X2, Y2), (X3, Y3), ..., (X i , Y i );

[0023] S13: Based on the converted spatial coordinate information, refer to the prefabricated component processing drawing to add the following related content definitions:

[0024] Define the model starting area and yard boundary range, input into the system, and obtain the 2D plane space coordinate model;

[0025] Define the prefabricated component type T for the current task batch Type = a, b, ..., n and the maximum size of each type of component N (m) = (N Length , N Width , N Height ) and KT Type Key yards affect the overall dimensions, the maximum stacking quantity M of each component, and the maximum size and weight information M(t), among which the prefabricated wall KT Type =N Length , N Width , prefabricated panel KT Type =N Length , N Width , Prefabricated Stairs KT Type =N Length , N Width , Prefabricated Bay Window KT Type =N Length , N Height , N LengthIndicates the maximum length of the prefabricated component, N Width Indicates the maximum width of the prefabricated component, N Height Indicates the maximum height dimension of the prefabricated component;

[0026] Define the stacking methods of different types of prefabricated components and the ratio coefficients of different types of prefabricated building stacking areas, expressed as T a :T b :, ..., T n =1:x:,...,:t; Assume that the stacking mode is a row rack, the stacking range of the local construction is defined as 2N, and the stacking range of the remaining construction is N;

[0027] S14: According to the content defined in step S13, the enumeration solution algorithm is used to input the yard boundary range, aisle width, prefabricated component type and stacking method, stacking area ratio coefficient, and key yard-affecting dimensions of each type of component; at the same time, the minimum condition of the site surplus land area is added to obtain the spatial positioning boundary coordinates of each prefabricated component yard within the output yard boundary range, the maximum stacking quantity in the area, and the weight information;

[0028] S15: The initial BIM layout model is obtained by constructing the output information and the component factory equipment positioning dimension model information.

[0029] Preferably, in step S23, no more than three AGVs can be located within 1 m of the path at the same time. 2 Within the range, the determination contents are as follows:

[0030] Define the straight-line distance D between the AGV and the starting point n , where n represents the current AGV vehicle number, and the straight-line distance from the starting point is analyzed by the formula Among them, (X t0 , Yt0) represents the current coordinate, t0 represents the current moment of the robot;

[0031] At time t0, |D n1 -D n2 |≤1 and |D n3 -D n2 |≤1 triggers reordering, where D n1 represents the distance between the first AGV robot and the initial point at time t0, D n2 represents the distance between the second AGV robot and the initial point at time t0, D n3 Indicates the distance between the third AGV robot and the initial point at time t0.

[0032] Preferably, in step S32, the specific task judgment logic is as follows:

[0033] S321: Obtaining the actual number N of prefabricated components placed in the yard through component number image recognition by the on-site inspection robot;

[0034] S322: Obtaining the actual weight G of the prefabricated components placed in the yard through the weight sensors in each yard area;

[0035] S323: by material category i, associating the yard data of the field layout BIM model to obtain the yard area corresponding to the material target;

[0036] The method for determining the actual transport target based on the above stacking area range is as follows:

[0037] Step 1: Use the BIM model to calculate the number boundary N of such component yards max , weight boundary G max Compare with the actual yard data to determine the overall full load situation;

[0038] Step 2: Analyze the straight-line distance D from the starting point using the formula n Far or near, will not be fully loaded and D n The smaller area is set as the final delivery target;

[0039] Assuming that the identified component information i is a prefabricated bay window, this type of stacking area can be obtained in area 1 and area 2 through the field layout model association. Among them, the corresponding data of area 1 is D1, N1, G1, and the corresponding data of area 2 is D2, N2, G2, then:

[0040] When {N max ≤N1 or G max}≤G1} and {N max ≤N2 or G max ≤G2}, the equipment stops carrying;

[0041] When {N max >N1 and G max >G1} and {N max ≤N2 or G max ≤G2}, then transport to area 1;

[0042] When {N max ≤N1 or G max ≤G1} and {N max >N2 and G max >G2}, then transport to area 2;

[0043] When {N max >N1 and G max >G1} and {N max >N2 and G max >G2}, then compare D1 and D2:

[0044] If D1>D2, it is transported to area 2; if D1<D2, it is transported to area 1.

[0045] Preferably, step S4 comprises the following steps:

[0046] S41: By deploying 5G modules for equipment, the foundation for the wireless transport working mode of equipment within the site is established;

[0047] S42: Obtaining the number F of inspection components of each site within the area through site inspection by the inspection robot;

[0048] S43: Obtaining the yard weight Mg through the yard weight sensor;

[0049] S44: Trigger the inspection task of the inspection robot through the weight change on site, and obtain the dynamic yard data display of the system site; assuming that the weight of yard 1 changes, match the corresponding yard material category i and regional coordinates, and then send instructions to the inspection robot to inspect the number of components, and obtain the actual yard quantity change data;

[0050] S45: The strategy for modifying yard management data is as follows:

[0051] According to the material types of the model yard and the standard weight range of each type of component defined in the system, MgType, the calculated quantity of the yard is obtained as F1 = Mg / Mg Type ;

[0052] S46: By comparing the measured quantity with the inspection quantity, if |F-F1|≥1, the system alarms and triggers the multi-inspection robot yard quantity verification and manual on-site verification.

[0053] The present invention aims to provide a multi-robot collaborative management method for an unmanned transport yard of building materials (prefabricated components) in a prefabricated component factory based on a 5G environment on site. The method collaboratively manages mobile recognition robots, AGV transport robots and inspection robots through data standard work packages in a 5G network environment, monitors and manages the execution progress and efficiency of robot work packages, improves the overall efficiency of building materials transportation in the prefabricated component factory, and reduces the relevant labor costs of the component factory, thereby further improving the dynamic balance of materials in the on-site prefabricated component production and improving the overall productivity and benefits of the prefabricated component factory.

[0054] The present invention manages the transportation of prefabricated components intelligently based on external data information (BIM, GIS information, image feature recognition, yard weight recognition) and robot multi-machine collaboration, introduces the concept of robot task work packages, connects the coordinated work of multiple devices through the opening and closing instructions of work packages between different robots, and improves the accuracy of task execution of robot collaboration for prefabricated components. The present invention also improves the intelligent dynamic balancing capabilities of on-site prefabricated component production-storage. The present invention conducts intelligent management of the entire process from prefabricated component maintenance out of the cellar to the on-site material yard, effectively reducing the manpower cost in the process.

[0055] Compared with the prior art solutions, the present invention has the following beneficial effects:

[0056] 1) The present invention provides an intelligent management system for multi-machine collaboration of robots in prefabricated component factories, which serves as an equipment command center for unmanned robot management in prefabricated component factories, and an intelligent management interactive platform, providing data processing and task dispatching for integrated management;

[0057] 2) The present invention provides a method for automatically arranging a prefabricated component factory yard and generating an automated BIM initial site model, which effectively improves the efficiency of manual arrangement and greatly improves the rationality of site arrangement according to the optimal solution algorithm;

[0058] 3) The present invention provides a robot optimal path execution planning method for a prefabricated component factory, which improves the robot transportation efficiency and intelligent planning level caused by the reduction of the transportation path, and reduces equipment transportation accidents caused by robot path conflicts;

[0059] 4) The present invention provides a standard work package task execution strategy for multi-machine robot collaboration in a prefabricated component factory, which improves the coherence of multi-machine robot collaboration tasks through a work package mode and a self-correction mode, and improves the overall transportation efficiency;

[0060] 5) The present invention provides a method for managing a yard of a prefabricated component factory using a patrol robot, which dynamically monitors on-site equipment data and performs multi-dimensional control of weight and identification, thereby improving the accuracy of project site management data and the accuracy of the overall strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0062] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the appended claims of the application equally.

[0063] An embodiment of the present invention discloses a yard intelligent management system adopted by a robot multi-machine collaborative yard management method of a prefabricated component factory based on a 5G network, comprising:

[0064] Site BIM display module: used to construct and display site-related information, including yard area division, yard component load, area component quantity, area component category, and area full load status.

[0065] Basic equipment and acquisition unit management module: used to obtain data management information related to equipment within the range, including robot type, number of robots, robot carrying status, robot movement path, robot 5G network status, robot size, AGV real-time load weight, sensor type, sensor positioning, and sensor operating status.

[0066] Basic data collection and management module: used for content collection and management of basic data collection, including production component category, transport component weight, yard real-time weight, yard component height, data corresponding equipment, and related historical data.

[0067] BIM site layout model management module: used to add or reduce BIM model information and enable or disable related model content, as well as content maintenance and input management of component site layout-related standards, and GIS positioning coordinates and range loading.

[0068] Work package management module: used to fit the data between different robots into standard work packages, monitor the progress and efficiency of work package completion, and issue instructions according to the work package completion trigger points.

[0069] Algorithm module: used to support the overall platform-related algorithms, such as prefabricated component type recognition, route planning, stacking height recognition and other algorithms, and can support the addition and deletion of algorithms, algorithm operation status control and other functions.

[0070] Based on the above system, a robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network disclosed in an embodiment of the present invention can realize the robot multi-machine collaborative management task of a prefabricated component factory, including robots and equipment categories such as mobile identification robots, transport robots (AGVs), inspection robots, weight sensors, and GPS sensors. Through the robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network, a complete set of prefabricated component task transportation yard work is completed, specifically including the following steps:

[0071] Step 1: According to the site BIM model information, enter the required range information of the area boundary, according to the design drawings and the corresponding component production ratio, enter the prefabricated component category and quantity ratio information, and according to the GPS coordinate sensor, enter the site boundary coordinate information.

[0072] Step 2: Add conditional definition, KT Type The key yard affects the overall dimensions, the maximum stacking quantity M of each component, the maximum size weight information M(t), the stacking methods of different types of prefabricated components, and the proportion coefficients of the stacking areas of different types of prefabricated structures (refer to the relevant stacking standard requirements for prefabricated components).

[0073] Step 3: Convert the world coordinates to standard coordinates according to the Bursa formula and input them into the system.

[0074] Step 4: Through the enumeration solution algorithm, input the yard boundary range, aisle width (considering the parallel width of two robots), prefabricated component types and stacking methods, stacking area ratio coefficients, and key yard-affecting dimensions of each type of component. Adding the condition of minimum surplus land area (i.e., maximum site utilization), the spatial positioning boundary coordinates of each prefabricated component yard within the output yard boundary range, the maximum stacking quantity in the area, and weight information can be obtained.

[0075] Step 5: The initial BIM layout model can be obtained by constructing the output information and the component factory equipment positioning dimension model information.

[0076] Step 6: By measuring the coordinates of the starting point of the actual site and defining the relevant points of the prefabricated components in the model, the loading point A (X Start ,Y Start ), coordinates of unloading points of each prefabricated component yard Among them, i represents the area to which different types of prefabricated components belong.

[0077] Step 7: Construct loading point A (X) through prefabrication in the model Start ,Y start ), coordinates of unloading points of each prefabricated component yard The initial BIM layout model includes the location of equipment and equipment in the field, the overall model boundary, the yard area, and the yard interval. Based on the above data and the Dijkstra algorithm, the optimal transportation path set can be obtained by fitting the curing pit loading node to the unloading node of each material area.

[0078] Step 8: Based on the optimal path for transporting various materials, the starting state c of the AGV task work package, and the standard speed V of the AGV, the bubble sort algorithm is combined, and the rule that no more than 3 AGVs can be located within 1m of the path at the same time is given. 2 The range boundaries can be used to obtain the AGV dynamic task time sorting.

[0079] Step 9: Start time t by work package start , fit the work package instruction information and issue the AGV with the work package transportation instruction

[0080] Step 10: The actual number of prefabricated components placed in the yard N can be obtained through the image recognition of the number of components by the on-site inspection robot. The actual weight of the prefabricated components placed in the yard G can be obtained through the weight sensors in each yard area. By associating the yard data of the BIM model with the material category i, the yard area corresponding to the material target can be obtained.

[0081] Step 11: Use the BIM model to compare the number and weight boundaries of such component yards with the actual yard data to determine the overall full load situation. Use the formula to analyze the straight-line distance from the starting point and set the underloaded and smaller area as the final transportation target.

[0082] Step 12: The difference Q between the planned time and the actual time can be regarded as the actual plan correction value. Define Q as the time deviation, Q = |Kt End -t End |, when Q ≥ 20s, the existing task plan is dynamically updated and the next task starts t 2 Start Time correction, t 2 Start =t Start +Q. Based on the above example, the revised overall multi-machine collaborative work package task execution plan P′={O′1,O′2……,O′ u}.

[0083] Step 13: By deploying the 5G module of the equipment, the foundation of the wireless transport working mode of the equipment within the site is established. The inspection robot inspects the site and obtains the number of inspection components F in each site within the area. The site weight sensor can obtain the site yard weight Mg. The site weight change triggers the inspection robot inspection task, and the system site dynamic yard data display can be obtained.

[0084] Step 14: Define the standard weight range of each type of component in the model yard through the material type and system Type , we can get the measured quantity of the stockpile F1=Mg / Mg Type By comparing the measured quantity with the inspection quantity, if |F-F1|≥1, the system alarms, triggering the multi-inspection robot yard quantity verification and manual on-site verification. The corrected data is returned to the system, and the current work package plan is dynamically corrected and executed.

Claims

1. A robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network, characterized in that: The following steps are involved: S1: Automatically arrange the BIM initial site model according to the site information; S2: Generate a robot transportation path plan based on the BIM initial site model and robot and material information, which specifically includes the following steps: S21: By measuring the coordinates of the starting point of the actual site point and defining the relevant points of the prefabricated components in the model, the loading point A (X Start , Y Start ) and the coordinates of the unloading points of each prefabricated component yard Wherein, i represents the area to which different types of prefabricated components belong, j represents the jth prefabricated component stacking area of ​​the same type, j = 1, 2, ..., u; S22: Prefabricate the loading point A(X Start , Y Start ), coordinates of unloading points of each prefabricated component yard The initial BIM layout model locates the scope of equipment in the site, the overall model boundary, the yard area, and the yard interval. Based on the above data and the Dijkstra algorithm, the optimal transportation path set is obtained by fitting the curing pit loading node to the unloading node of each material area. S23: Optimal path for transporting various materials The starting state c of the AGV task work package and the standard speed V of the AGV are combined with the bubble sort algorithm, and it is given that no more than 3 AGVs can exist at the same time on the path 1m 2 Within the range, the dynamic task time order of AGV is obtained; S3: The mobile robot and AGV divide the standard work packages according to the path plan and perform multi-machine collaborative tasks, including the following steps: S31: The mobile recognition robot moves to the curing cellar loading point A (X Start , Y Start ), the prefabricated component category image recognition algorithm after training is used to identify the category S of the transported component in the current task state. At the same time, when a component is first identified in the range of this point while being transported by the slide rail, this moment is defined as the work package start time t Start ; S32: Start time t of work package Start , issue a command to the AGV to carry the work package Where m represents the weight of the transported material, t End Indicates the planned work package end time, t End =D n / V, Z represents the work package serial number, and A represents the coordinate A (X) of the initial point of material transportation described in step S22. Start , Y Start ), B represents the coordinates of the material unloading point described in step S22 Indicates the optimal material transportation path described in step S22; S33: Based on the above work package instance, the AGV transport task is executed, and the work package cumulative resource allocation is performed. Referring to step S23, the overall transport multi-machine collaborative work package task execution plan P is obtained, where P = {O1, O2, ..., O u }; S34: Obtain the actual AGV task end time Kt through the AGV task end trigger point End , where Kt End It means that the AGV enters the unloading point range and is in a non-displacement state for more than 5 seconds, which is considered as the end time of the work package; S35: The difference Q between the planned time and the actual time is regarded as the actual plan correction value, and Q is defined as the time deviation, Q = |Kt End -t End |: When Q ≥ 20s, the existing task plan is dynamically updated and the next task starts t 2 Start Time correction, t 2 Start =t Start + Q, and obtain the revised overall multi-machine collaborative work package task execution plan P′={O′1,O′2,…,O′ u }; S4: The inspection robot collects data and provides feedback based on the on-site yard performance, and dynamically adjusts the transportation path plan.

2. A robot multi-machine collaborative yard management method for a prefabricated component factory based on a 5G network as claimed in claim 1, characterized in that: The step S1 comprises the following steps: S11: Survey the boundary coordinates of the site through a GPS locator to obtain GPS coordinate information of the boundary positioning of the transportation starting area including the yard boundary and the curing pit; S12: The GPS coordinates obtained in the previous step are converted into spatial coordinates through the Bursa formula to obtain the converted spatial coordinate information, which is expressed as (X1, Y1), (X2, Y2), (X3, Y3), ..., (X i , Y i ); S13: Based on the converted spatial coordinate information, refer to the prefabricated component processing drawing to add the following related content definitions: Define the model starting area and yard boundary range, input into the system, and obtain the 2D plane space coordinate model; Define the prefabricated component type T for the current task batch Type = a, b, ..., n and the maximum size of each type of component N (m) = (N Length , N Width , N Height ) and KT Type Key yards affect the overall dimensions, the maximum stacking quantity M of each component, and the maximum size and weight information M(t), among which the prefabricated wall KT Type =N Length , N Width , prefabricated panel KT Type =N Length , N Width , Prefabricated Stairs KT Type =N Length , N Width , Prefabricated Bay Window KT Type =N Length , N Height , N Length Indicates the maximum length of the prefabricated component, N Width Indicates the maximum width of the prefabricated component, N Height Indicates the maximum height dimension of the prefabricated component; Define the stacking methods of different types of prefabricated components and the ratio coefficients of different types of prefabricated building stacking areas, expressed as T a :T b :, ..., T n =1:x:,...,:t; Assume that the stacking mode is a row rack, the stacking range of the local construction is defined as 2N, and the stacking range of the remaining construction is N; S14: According to the content defined in step S13, the enumeration solution algorithm is used to input the yard boundary range, aisle width, prefabricated component type and stacking method, stacking area ratio coefficient, and key yard-affecting dimensions of each type of component; at the same time, the minimum condition of the site surplus land area is added to obtain the spatial positioning boundary coordinates of each prefabricated component yard within the output yard boundary range, the maximum stacking quantity in the area, and the weight information; S15: The initial BIM layout model is obtained by constructing the output information and the component factory equipment positioning dimension model information.

3. The method for managing a multi-machine robot-assisted yard in a prefabricated component factory based on a 5G network as claimed in claim 1, characterized in that: In step S23, there cannot be 3 or more AGVs on the path 1m at the same time. 2 Within the range, the determination contents are as follows: Define the straight-line distance D between the AGV and the starting point n , where n represents the current AGV vehicle number, and the straight-line distance from the starting point is analyzed by the formula Among them, (X t0 , F t0 ) represents the current coordinate, t0 represents the current moment of the robot; At time t0, |D n1 -D n2 |≤1 and |D n3 -D n2 |≤1 triggers reordering, where D n1 represents the distance between the first AGV robot 1 and the initial point at time t0, D n2 represents the distance between the second AGV robot and the initial point at time t0, D n3 Indicates the distance between the third AGV robot and the initial point at time t0.

4. The method for managing a multi-machine robot-assisted yard in a prefabricated component factory based on a 5G network as claimed in claim 1, characterized in that: In step S32, the specific task judgment logic is as follows: S321: Obtaining the actual number N of prefabricated components placed in the yard through component number image recognition by the on-site inspection robot; S322: Obtaining the actual weight G of the prefabricated components placed in the yard through the weight sensors in each yard area; S323: by material category i, associating the yard data of the field layout BIM model to obtain the yard area corresponding to the material target; The method for determining the actual transport target based on the above stacking area range is as follows: Step 1: Use the BIM model to calculate the number boundary N of such component yards max , weight boundary G max Compare with the actual yard data to determine the overall full load situation; Step 2: Analyze the straight-line distance D from the starting point using the formula n Far or near, will not be fully loaded and D n The smaller area is set as the final delivery target; Assuming that the identified component information i is a prefabricated bay window, this type of stacking area can be obtained in area 1 and area 2 through the field layout model association. Among them, the corresponding data of area 1 is D1, N1, G1, and the corresponding data of area 2 is D2, N2, G2, then: When {N max ≤N1 or G max }≤G1} and {N max ≤N2 or G max ≤G2}, the equipment stops carrying; When {N max >N1 and G max >G1} and {N max ≤N2 or G max ≤G2}, then transport to area 1; When {N max ≤N1 or G max ≤G1} and {N max >N2 and G max >G2}, then transport to area 2; when {N max >N1 and G max >G1} and {N max >N2 and G max >G2}, then compare D1 and D2: If D1>D2, it is transported to area 2; if D1<D2, it is transported to area 1.

5. The method for managing a multi-machine robot-assisted yard in a prefabricated component factory based on a 5G network as claimed in claim 1, characterized in that: Step S4 includes the following steps: S41: By deploying 5G modules for equipment, the foundation for the wireless transport working mode of equipment within the site is established; S42: Obtaining the number F of inspection components of each site within the area through site inspection by the inspection robot; S43: Obtaining the yard weight Mg through the yard weight sensor; S44: Trigger the inspection task of the inspection robot through the weight change on site, and obtain the dynamic yard data display of the system site; assuming that the weight of yard 1 changes, match the corresponding yard material category i and regional coordinates, and then send instructions to the inspection robot to inspect the number of components, and obtain the actual yard quantity change data; S45: The strategy for modifying yard management data is as follows: The material types in the model yard and the standard weight range of each category of components defined in the system Mg Type , and the calculated quantity at the stockpile is F1 = Mg / Mg Type ; S46: By comparing the measured quantity with the inspection quantity, if |F-F1|≥1, the system alarms and triggers the multi-inspection robot yard quantity verification and manual on-site verification.

Images