AGV automatic conveying management method and system based on cloud computing

Abstract

The application discloses an AGV automatic conveying management method and system based on cloud computing, relates to the technical field of AGV conveying management, and comprises the following steps: integrating cross-warehouse tasks and AGV resources by constructing a cloud resource constraint model, and realizing global optimization scheduling in combination with order attribute rules, cargo attribute rules and dynamic constraint rules; generating a dynamic avoidance path by using a sliding straight line tracking algorithm, quantitatively evaluating collision risks by means of an obstacle heat map marker and a safety area model, and introducing an auction algorithm to realize dynamic scoring and distribution of task matching degree and power information; and integrating sensors to collect environmental data in real time, and constructing an obstacle avoidance model containing multi-dimensional information such as safety boundaries and sudden obstacles. The technology significantly improves the AGV scheduling efficiency and path safety, can be widely applied to the fields of intelligent manufacturing and warehouse logistics, and provides an efficient and reliable solution for intelligent conveying management in complex industrial scenes.

Description

Technical Field

[0001] This invention relates to the field of AGV conveying management technology, specifically to a cloud computing-based AGV automatic conveying management method and system. Background Technology

[0002] Against the backdrop of intelligent transformation in manufacturing, traditional AGV (Automated Guided Vehicle) conveying management suffers from low efficiency and scheduling difficulties. Cloud computing, with its powerful data processing and storage capabilities, has given rise to cloud-based AGV automated conveying management methods and systems. Its significance lies in optimizing AGV scheduling through cloud computing, improving conveying efficiency and accuracy, achieving intelligent resource allocation, reducing management costs, enhancing system flexibility and scalability, helping enterprises achieve automated and intelligent upgrades in production, and improving market competitiveness.

[0003] Existing cloud-based AGV automated transport management methods and systems lack cross-warehouse resource integration mechanisms, cannot dynamically coordinate the allocation of AGV tasks across multiple warehouses, and do not fully consider dynamic workshop scenarios (such as equipment movement and temporary obstacles), leading to frequent path conflicts. Therefore, it is necessary to provide a cloud-based AGV automated transport management method and system to solve the above-mentioned problems. Summary of the Invention

[0004] To address the aforementioned technical issues, this paper provides a cloud-based AGV automatic conveying management method and system. This technical solution solves the problems mentioned in the background section, such as the lack of a cross-warehouse resource integration mechanism in existing cloud-based AGV automatic conveying management methods and systems, the inability to dynamically coordinate the allocation of AGV tasks across multiple warehouses, and the failure to fully consider dynamic workshop scenarios, which leads to frequent path conflicts.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A cloud-based AGV automated transport management method includes:

[0007] S1. Obtain the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. Then, receive the AGV task request information of multiple warehouses through the cloud platform, and use the AGV task request information and location information of multiple warehouses to build a resource constraint model.

[0008] S2. Based on the AGV task request information from multiple warehouses, determine the candidate AGV information, and then filter the candidate AGVs according to the candidate AGV information and the resource constraint model to determine the preferred AGV;

[0009] S3. Download the workshop internal scene information and production line scene information from the cloud platform, randomly select a preferred AGV from the preferred AGVs to obtain the first preferred AGV, and then simulate and generate the first initial conveying route information of the first preferred AGV through AGV task request information, workshop internal scene information and production line scene information.

[0010] S4. Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, determine the conveying focus position information in the first initial conveying route information, and obtain the second initial conveying route information based on the conveying focus position information in the first initial conveying route information.

[0011] S5. Select at least one preferred AGV from the preferred AGVs, and import the second initial conveying route information into the selected preferred AGV. Then, based on the first initial conveying route information and the second initial conveying route information, simulate the conveying of the preferred AGV to obtain the AGV conveying safety constraint model.

[0012] S6. Import the AGV conveying safety constraint model into all the preferred AGVs to obtain the conveying safety AGV information. Then, import the AGV task request information into the conveying safety AGV and use the sensors deployed on the AGV to collect real-time environmental data to generate an obstacle avoidance model, thereby realizing automatic AGV conveying management.

[0013] In an optional embodiment, step S1 specifically includes:

[0014] The location information of multiple warehouses is obtained by collecting the latitude and longitude information and area information of each warehouse through GPS devices, and then importing the location information of multiple warehouses into the cloud platform.

[0015] The system receives AGV task request information from various warehouses through a cloud platform. The AGV task request information includes the starting point of the transport, the ending point of the transport, order attribute information, cargo attribute information, and time information.

[0016] Define order attribute rules and goods attribute rules based on order attribute information and goods attribute information;

[0017] Obtain AGV load information and warehouse busyness from the cloud platform, and define resource constraint rules based on AGV load information and warehouse busyness;

[0018] Based on the delivery start point, delivery end point, time information, and the location information of multiple warehouses, dynamic constraint rules are defined.

[0019] A resource constraint model is constructed based on order attribute rules, goods attribute rules, resource constraint rules, and dynamic constraint rules.

[0020] In an optional embodiment, step S2 specifically includes:

[0021] Based on AGV task request information from multiple warehouses, obtain the location information of all AGVs within the range of the conveying start point and conveying end point;

[0022] Based on the location information of all AGVs within the range of the starting point and ending point of the transport, a pool of candidate AGVs is constructed to determine the candidate AGV information, which includes candidate AGV location information, candidate AGV power information, and candidate AGV type information.

[0023] Based on the candidate AGV type information, obtain the maximum load capacity of the candidate AGV, and based on the cargo attribute information, obtain the cargo weight. Then, use the ratio of the maximum load capacity of the candidate AGV to the cargo weight as the task matching degree.

[0024] Based on the resource constraint model, the AGV scoring threshold is obtained, and then the scores of the candidate AGVs are obtained according to the task matching degree and the power information of the candidate AGVs.

[0025] The candidate AGVs are screened by using their scores and AGV score thresholds. AGVs with scores greater than or equal to the AGV score threshold are extracted from the candidate AGV pool and a preferred AGV pool is constructed.

[0026] An auction algorithm is used to allocate preferred AGVs to AGV task request information from multiple warehouses. Thus, the AGV corresponding to the AGV request information of each warehouse is determined as the preferred AGV.

[0027] In an optional embodiment, step S3 specifically includes:

[0028] Download workshop interior scene information and production line scene information from the MES system on the cloud platform. The workshop interior scene information includes a workshop layout diagram, and the production line scene information includes shelf locations and equipment coordinates.

[0029] Using the workshop layout diagram, shelf locations, and equipment coordinates, a two-dimensional map of the production line is constructed, and the equipment coordinates are used to obtain information on the environmental footprint of the equipment. This information is then used to mark the obstacle heat map.

[0030] Based on the 2D map of the production line and the heat map of obstacles, the heat map of obstacles is removed from the 2D map of the production line, so that the remaining area is used as the preferred AGV transportable area map.

[0031] Obtain the edge points of the limited AGV transportable area map and determine the center point of the 2D map of the production line;

[0032] Using the center point of the two-dimensional map of the production line, construct a sliding straight line f(x)=h, where h is the dynamic sliding value, and the unit value of the dynamic sliding value is the distance between the edge points of the two nearest finite AGV transportable area maps.

[0033] By using a sliding straight line to traverse all edge points in the finite AGV transportable area map, and obtaining the transport path tracing line f(x)=j when there are at least two edge points on the sliding straight line when the dynamic sliding values ​​are the same;

[0034] Obtain the coordinates of the two farthest edge points in the conveying path tracking line, and construct the first conveying path tracking line segment using the coordinates of the two farthest edge points in the conveying path tracking line.

[0035] Determine the center point of the first conveying path tracking segment, and obtain the two edge points closest to the center point of the first conveying path tracking segment;

[0036] Construct a second transport path tracking segment using the two edge points closest to the center point of the transport path tracking segment;

[0037] Starting from the conveying start point, connect the center points of all the second conveying path tracking segments in sequence to obtain an irregular line segment from the conveying start point to the center point of the second conveying path tracking segment to the conveying end point;

[0038] Using irregular line segments, a conveying path tracking function is constructed, and based on the conveying path tracking function, the first initial conveying route information of the first preferred AGV corresponding to all AGV task request information is simulated and generated.

[0039] The formula for the transport path tracing function is as follows:

[0040] ;

[0041] In the formula, Let n be any point in the irregular line segment. Let n be the distance from any point n in the irregular line segment to the end point of the transport. This serves as a reference value for tracing the transport path.

[0042] In an optional embodiment, step S4 specifically includes:

[0043] Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, the conveying focus location information in the first initial conveying route information is determined. The conveying focus location information includes AGV conveying turning points and AGV conveying intersection points.

[0044] Randomly select a first initial transport route from the first initial transport route information of the first preferred AGV corresponding to all AGV task request information generated in the simulation, and designate the remaining first initial transport routes as the third initial transport routes;

[0045] Traverse the third initial transport route and define the third initial transport route containing the transport focus position from the first initial transport route information as the second initial transport route information.

[0046] In an optional embodiment, step S5 specifically includes:

[0047] Based on the first and second initial transport route information, the preferred AGV is simulated to obtain the second initial transport route information of the collision that occurs during the simulated transport of the preferred AGV.

[0048] The total number of second initial transport routes that collided is obtained. Then, based on the candidate AGV type information, the footprint of the preferred AGV is determined. The footprint of the preferred AGV is multiplied by the total number of second initial transport routes that collided to obtain the AGV transport safety area information.

[0049] The safe area information of AGV transportation is used as the safety constraint model for AGV transportation.

[0050] In an optional embodiment, step S6 specifically includes:

[0051] The AGV transport safety constraint model is imported into all the preferred AGVs to obtain transport safety AGV information. Then, the AGV task request information is imported into the transport safety AGV, and the sensors deployed on the AGV are used to collect real-time environmental data to generate an obstacle avoidance model, thereby realizing automatic transport management of AGVs.

[0052] The AGV transport safety constraint model is imported into all the preferred AGVs. The edge of the AGV transport safety constraint model is used as the safety boundary of the preferred AGV to obtain the transport safety AGV information. The transport safety AGV information includes the safety boundary of the preferred AGV, the location information of the preferred AGV, the power information of the preferred AGV, and the type information of the preferred AGV.

[0053] Real-time environmental data is collected using sensors deployed on the AGV, including the location of sudden obstacles, the location of the working AGV, and the location of the edge of the production equipment.

[0054] When the location of a sudden obstacle, the location of the AGV in operation, or the edge of the production equipment coincides with the safety boundary of the preferred AGV, an obstacle avoidance model is generated based on the preferred AGV location information, preferred AGV type information, and real-time environmental data. Based on the obstacle avoidance model, automatic AGV conveying management is realized.

[0055] Furthermore, a cloud-based AGV automated conveying management system is proposed to implement any of the management methods described above, including:

[0056] The acquisition module is used to acquire the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. It is also used to collect real-time environmental data using sensors deployed on the AGV.

[0057] The main control module is used to receive AGV task request information from multiple warehouses through the cloud platform, and also to download workshop internal scene information and production line scene information from the cloud platform. It is used to construct a resource constraint model using the AGV task request information and location information from multiple warehouses, and to determine candidate AGV information based on the AGV task request information from multiple warehouses. Then, based on the candidate AGV information and the resource constraint model, the candidate AGVs are screened to determine the preferred AGV.

[0058] The management module is used to arbitrarily select one preferred AGV from the preferred AGVs to obtain a first preferred AGV. Then, based on AGV task request information, workshop internal scene information, and production line scene information, it simulates and generates a first initial conveying route information for the first preferred AGV. Based on the first initial conveying route information, workshop internal scene information, and production line scene information, it determines the conveying focus position information in the first initial conveying route information and obtains a second initial conveying route information based on the conveying focus position information in the first initial conveying route information. This second initial conveying route information is used to select at least one preferred AGV from the preferred AGVs and import the second initial conveying route information into the selected preferred AGV. Then, based on the first and second initial conveying route information, it simulates the conveying of the preferred AGV to obtain an AGV conveying safety constraint model. This model is then imported into all preferred AGVs to obtain conveying safety AGV information. Finally, it imports AGV task request information into the conveying safety AGV and generates an obstacle avoidance model to realize automatic AGV conveying management.

[0059] The display module is used to display the acquisition process and results of the acquisition module, the data processing process and results of the main control module, and the management process and results of the management module.

[0060] In an optional embodiment, the main control module includes:

[0061] The receiving unit is used to receive AGV task request information from multiple warehouses through the cloud platform, and is also used to download workshop internal scene information and production line scene information from the cloud platform.

[0062] The processing unit is used to construct a resource constraint model by utilizing AGV task request information and location information from multiple warehouses;

[0063] The AGV determination unit is used to determine candidate AGV information based on AGV task request information from multiple warehouses, and then filter the candidate AGVs according to the candidate AGV information and resource constraint model to determine the preferred AGV.

[0064] In an optional embodiment, the management module includes:

[0065] The path determination unit is used to arbitrarily select a preferred AGV from the preferred AGVs to obtain a first preferred AGV. Then, through AGV task request information, workshop internal scene information and production line scene information, it simulates and generates a first initial conveying route information of the first preferred AGV. Based on the first initial conveying route information, workshop internal scene information and production line scene information, it determines the conveying focus position information in the first initial conveying route information, and obtains a second initial conveying route information based on the conveying focus position information in the first initial conveying route information.

[0066] A safety constraint unit is used to select at least one preferred AGV from the preferred AGVs, import the second initial transport route information into the selected preferred AGV, and then simulate the transport of the preferred AGV based on the first initial transport route information and the second initial transport route information to obtain an AGV transport safety constraint model.

[0067] The obstacle avoidance management unit is used to import the AGV transport safety constraint model into all preferred AGVs to obtain transport safety AGV information, then import the AGV task request information into the transport safety AGV, and generate an obstacle avoidance model to realize automatic AGV transport management.

[0068] Compared with the prior art, the beneficial effects of the present invention are:

[0069] 1. This solution proposes a cloud-based AGV automatic delivery management method and system. By constructing a cloud resource constraint model, it can uniformly manage task requests and AGV resources in multiple warehouses, realize global optimization scheduling, and dynamically coordinate the allocation of AGV tasks in multiple warehouses.

[0070] 2. The proposed solution is a cloud computing-based AGV automatic conveying management method and system. By constructing a two-dimensional map of the production line, marking obstacle heat maps, and using a sliding straight line tracking algorithm, a dynamic avoidance path is generated, reducing path intersections.

[0071] 3. The proposed solution is a cloud computing-based AGV automatic conveying management method and system. It collects environmental data in real time through sensors and dynamically updates the safety boundary model. When an obstacle is detected to coincide with the AGV's safety boundary, an obstacle avoidance mechanism is triggered to avoid lag in the AGV's response when encountering sudden obstacles. Attached Figure Description

[0072] Figure 1 This is a flowchart of an AGV automatic conveying management method based on cloud computing proposed in this invention;

[0073] Figure 2 This is a flowchart illustrating the process of obtaining the preferred AGV in this invention.

[0074] Figure 3 This is a flowchart illustrating the process of obtaining the first initial transport route information in this invention.

[0075] Figure 4 This is a system framework diagram of a cloud computing-based AGV automatic conveying management system proposed in this invention. Detailed Implementation

[0076] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0077] Reference Figure 1 - Figure 4 As shown, a cloud-based AGV automated transport management method includes:

[0078] S1. Obtain the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. Then, receive the AGV task request information of multiple warehouses through the cloud platform, and use the AGV task request information and location information of multiple warehouses to build a resource constraint model.

[0079] S2. Based on the AGV task request information from multiple warehouses, determine the candidate AGV information, and then filter the candidate AGVs according to the candidate AGV information and the resource constraint model to determine the preferred AGV;

[0080] S3. Download the workshop internal scene information and production line scene information from the cloud platform, randomly select a preferred AGV from the preferred AGVs to obtain the first preferred AGV, and then simulate and generate the first initial conveying route information of the first preferred AGV through AGV task request information, workshop internal scene information and production line scene information.

[0081] S4. Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, determine the conveying focus position information in the first initial conveying route information, and obtain the second initial conveying route information based on the conveying focus position information in the first initial conveying route information.

[0082] S5. Select at least one preferred AGV from the preferred AGVs, and import the second initial conveying route information into the selected preferred AGV. Then, based on the first initial conveying route information and the second initial conveying route information, simulate the conveying of the preferred AGV to obtain the AGV conveying safety constraint model.

[0083] S6. Import the AGV conveying safety constraint model into all the preferred AGVs to obtain the conveying safety AGV information. Then, import the AGV task request information into the conveying safety AGV and use the sensors deployed on the AGV to collect real-time environmental data to generate an obstacle avoidance model, thereby realizing automatic AGV conveying management.

[0084] Furthermore, step S1 specifically includes:

[0085] The location information of multiple warehouses is obtained by collecting the latitude and longitude information and area information of each warehouse through GPS devices, and then importing the location information of multiple warehouses into the cloud platform.

[0086] The cloud platform receives AGV task request information from various warehouses. The AGV task request information includes the starting point of the transport, the ending point of the transport, order attribute information, cargo attribute information, and time information.

[0087] Define order attribute rules and goods attribute rules based on order attribute information and goods attribute information;

[0088] Obtain AGV load information and warehouse busyness from the cloud platform, and define resource constraint rules based on AGV load information and warehouse busyness;

[0089] Based on the delivery start point, delivery end point, time information, and the location information of multiple warehouses, dynamic constraint rules are defined.

[0090] A resource constraint model is constructed based on order attribute rules, goods attribute rules, resource constraint rules, and dynamic constraint rules.

[0091] Specifically, for GPS data acquisition, high-precision GPS receivers (such as devices supporting GNSS multi-mode positioning) can be used to collect latitude and longitude coordinates at warehouse boundaries and key points (entrances / exits, shelving areas), combined with RTK (Real-Time Kinematic) technology to improve positioning accuracy to the centimeter level. The warehouse outline is scanned using LiDAR or UAV aerial surveying to generate polygonal area data. During data integration and uploading, the geographic coordinates and area data can be integrated into a JSON file and uploaded to a cloud database (such as AWS IoT Core or Alibaba Cloud IoT Platform) via a RESTful API interface. A spatial index is then created in PostgreSQL using PostGIS extensions to support subsequent geospatial queries.

[0092] The system receives AGV task request information from various warehouses via a cloud platform. This information includes the delivery start / end point, order attributes, cargo attributes, and time information. An MQTT message queue (such as EMQX) is deployed in the cloud to receive task request streams from ERP and WMS systems. The delivery start / end point in the AGV task request information is the warehouse's coordinates (e.g., loading / unloading area A1, production line B3); order attributes include urgency (priority code) and cargo value level; cargo attributes describe volume (length, width, and height), weight, and temperature control requirements (if cold chain is involved). Time information includes the expected delivery time range and the latest deadline.

[0093] Resource constraint rules can be implemented by collecting AGV onboard sensor data (such as weighing sensors and battery SOC values) in real time via the OPCUA protocol and updating it to a time-series database (such as InfluxDB). Warehouse busyness can be calculated using a warehouse busyness model.

[0094] When the score exceeds a threshold (e.g., 85%), a resource scheduling warning is triggered.

[0095] For dynamic constraint rules, constraint programming (CP-SAT solver) can be used to handle time conflicts.

[0096] Therefore, it can be understood that the resource constraint model is as follows:

[0097] ,

[0098] Where i represents the task (or order) number (e.g., the transportation task that the AGV needs to perform), and j represents the AGV number (e.g., the vehicle resources participating in the task). This represents the time required for task i to be executed by AGV j, including transportation time, loading and unloading time, etc. This is a binary variable (0 or 1) indicating whether task i is assigned to AGV j. Task i is executed by AGV j. Task i is not executed by AGV j. For weight parameters, This is the overload penalty coefficient, which represents the cost or capacity limitation weight of AGVk overloading. This is an overload indicator variable, also a binary variable (0 or 1), indicating whether AGV k is overloaded. The actual load on AGVk exceeds its capacity limit. The load of AGVk is within the safe range.

[0099] Furthermore, step S2 specifically includes:

[0100] Based on AGV task request information from multiple warehouses, obtain the location information of all AGVs within the range of the conveying start point and conveying end point;

[0101] Based on the location information of all AGVs within the range of the starting point and ending point of the transport, a pool of candidate AGVs is constructed to determine the candidate AGV information, which includes the candidate AGV location information, candidate AGV power information, and candidate AGV type information.

[0102] Based on the candidate AGV type information, obtain the maximum load capacity of the candidate AGV, and based on the cargo attribute information, obtain the cargo weight. Then, use the ratio of the maximum load capacity of the candidate AGV to the cargo weight as the task matching degree.

[0103] Based on the resource constraint model, the AGV scoring threshold is obtained, and then the scores of the candidate AGVs are obtained according to the task matching degree and the power information of the candidate AGVs.

[0104] The candidate AGVs are screened by using their scores and AGV score thresholds. AGVs with scores greater than or equal to the AGV score threshold are extracted from the candidate AGV pool and a preferred AGV pool is constructed.

[0105] An auction algorithm is used to allocate preferred AGVs to AGV task request information from multiple warehouses. Thus, the AGV corresponding to the AGV request information of each warehouse is determined as the preferred AGV.

[0106] Specifically,

[0107] ,

[0108] ,

[0109] These are all weighting coefficients, which can be adjusted according to business needs.

[0110] Furthermore, step S3 specifically includes:

[0111] Download the workshop interior scene information and production line scene information from the MES system on the cloud platform. The workshop interior scene information includes the workshop layout diagram, and the production line scene information includes the shelf location and equipment coordinates.

[0112] Using the workshop layout diagram, shelf locations, and equipment coordinates, a two-dimensional map of the production line is constructed, and the equipment coordinates are used to obtain information on the environmental footprint of the equipment. This information is then used to mark the obstacle heat map.

[0113] Based on the 2D map of the production line and the heat map of obstacles, the heat map of obstacles is removed from the 2D map of the production line, so that the remaining area is used as the preferred AGV transportable area map.

[0114] Obtain the edge points of the limited AGV transportable area map and determine the center point of the 2D map of the production line;

[0115] Using the center point of the two-dimensional map of the production line, construct a sliding straight line f(x)=h, where h is the dynamic sliding value, and the unit value of the dynamic sliding value is the distance between the edge points of the two nearest finite AGV transportable area maps.

[0116] By using a sliding straight line to traverse all edge points in the finite AGV transportable area map, and obtaining the transport path tracing line f(x)=j when there are at least two edge points on the sliding straight line when the dynamic sliding values ​​are the same;

[0117] Obtain the coordinates of the two farthest edge points in the conveying path tracking line, and construct the first conveying path tracking line segment using the coordinates of the two farthest edge points in the conveying path tracking line.

[0118] Determine the center point of the first conveying path tracking segment, and obtain the two edge points closest to the center point of the first conveying path tracking segment;

[0119] Construct a second transport path tracking segment using the two edge points closest to the center point of the transport path tracking segment;

[0120] Starting from the conveying start point, connect the center points of all the second conveying path tracking segments in sequence to obtain an irregular line segment from the conveying start point to the center point of the second conveying path tracking segment to the conveying end point;

[0121] Using irregular line segments, a conveying path tracking function is constructed, and based on the conveying path tracking function, the first initial conveying route information of the first preferred AGV corresponding to all AGV task request information is simulated and generated.

[0122] The formula for the transport path tracing function is as follows:

[0123] ;

[0124] In the formula, Let n be any point in the irregular line segment. Let n be the distance from any point n in the irregular line segment to the end point of the transport. This serves as a reference value for tracing the transport path.

[0125] Specifically, according to the transport path tracking function, the first initial transport route information of the first preferred AGV corresponding to all AGV task request information is generated by simulation. This means that when the transport path tracking reference value is the smallest, a point n in the irregular line segment is a point on the first initial transport route. The points n corresponding to the smallest transport path tracking reference value are connected sequentially to obtain the first initial transport route.

[0126] Furthermore, step S4 specifically includes:

[0127] Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, determine the conveying focus location information in the first initial conveying route information. The conveying focus location information includes AGV conveying turning points and AGV conveying intersection points.

[0128] Randomly select a first initial transport route from the first initial transport route information of the first preferred AGV corresponding to all AGV task request information generated in the simulation, and designate the remaining first initial transport routes as the third initial transport routes;

[0129] Traverse the third initial transport route and define the third initial transport route containing the transport focus position from the first initial transport route information as the second initial transport route information.

[0130] Furthermore, step S5 specifically includes:

[0131] Based on the first and second initial transport route information, the preferred AGV is simulated to obtain the second initial transport route information of the collision that occurs during the simulated transport of the preferred AGV.

[0132] The total number of second initial transport routes that collided is obtained. Then, based on the candidate AGV type information, the footprint of the preferred AGV is determined. The footprint of the preferred AGV is multiplied by the total number of second initial transport routes that collided to obtain the AGV transport safety area information.

[0133] The safe area information of AGV transportation is used as the safety constraint model for AGV transportation.

[0134] Furthermore, step S6 specifically includes:

[0135] The AGV transport safety constraint model is imported into all the preferred AGVs to obtain transport safety AGV information. Then, the AGV task request information is imported into the transport safety AGV, and the sensors deployed on the AGV are used to collect real-time environmental data to generate an obstacle avoidance model, thereby realizing automatic transport management of AGVs.

[0136] The AGV transport safety constraint model is imported into all the preferred AGVs. The edge of the AGV transport safety constraint model is used as the safety boundary of the preferred AGV to obtain the transport safety AGV information. The transport safety AGV information includes the safety boundary of the preferred AGV, the location information of the preferred AGV, the power information of the preferred AGV, and the type information of the preferred AGV.

[0137] Sensors deployed on the AGV are used to collect real-time environmental data, including the location of sudden obstacles, the location of the working AGV, and the location of the edge of the production equipment.

[0138] When the location of a sudden obstacle, the location of the AGV in operation, or the edge of the production equipment coincides with the safety boundary of the preferred AGV, an obstacle avoidance model is generated based on the preferred AGV location information, preferred AGV type information, and real-time environmental data. Based on the obstacle avoidance model, automatic AGV conveying management is realized.

[0139] Furthermore, a cloud-based AGV automated conveying management system is proposed to implement any of the management methods described above, including:

[0140] The data acquisition module is used to acquire the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. It is also used to collect real-time environmental data using sensors deployed on the AGV.

[0141] The main control module is used to receive AGV task request information from multiple warehouses through the cloud platform. It is also used to download workshop internal scene information and production line scene information from the cloud platform. It is used to construct a resource constraint model using the AGV task request information and location information from multiple warehouses. It is used to determine the candidate AGV information based on the AGV task request information from multiple warehouses. Then, based on the candidate AGV information and the resource constraint model, it filters the candidate AGVs and determines the preferred AGV.

[0142] The management module is used to select any one preferred AGV from the preferred AGVs to obtain the first preferred AGV. Then, based on the AGV task request information, workshop internal scene information, and production line scene information, it simulates and generates the first initial conveying route information of the first preferred AGV. Based on the first initial conveying route information, the workshop internal scene information, and the production line scene information, it determines the conveying focus position information in the first initial conveying route information. Based on the conveying focus position information in the first initial conveying route information, it obtains the second initial conveying route information, which is used to select at least one preferred AGV from the preferred AGVs and import the second initial conveying route information into the selected preferred AGV. Then, based on the first initial conveying route information and the second initial conveying route information, it simulates the conveying of the preferred AGV to obtain the AGV conveying safety constraint model. This model is then imported into all preferred AGVs to obtain the conveying safety AGV information. Finally, it imports the AGV task request information into the conveying safety AGV and generates an obstacle avoidance model to realize automatic AGV conveying management.

[0143] The display module is used to display the acquisition process and results of the acquisition module, the data processing process and results of the main control module, and the management process and results of the management module.

[0144] Furthermore, the main control module includes:

[0145] The receiving unit is used to receive AGV task request information from multiple warehouses through the cloud platform, and also to download workshop internal scene information and production line scene information from the cloud platform.

[0146] The processing unit is used to construct a resource constraint model by utilizing AGV task request information and location information from multiple warehouses.

[0147] The AGV determination unit is used to determine the candidate AGV information based on AGV task request information from multiple warehouses. Then, based on the candidate AGV information and the resource constraint model, the candidate AGVs are screened to determine the preferred AGV.

[0148] Furthermore, the management module includes:

[0149] The path determination unit is used to arbitrarily select a preferred AGV from the preferred AGVs to obtain a first preferred AGV. Then, through AGV task request information, workshop internal scene information and production line scene information, it simulates and generates the first initial conveying route information of the first preferred AGV. Based on the first initial conveying route information, workshop internal scene information and production line scene information, it determines the conveying focus position information in the first initial conveying route information, and obtains the second initial conveying route information based on the conveying focus position information in the first initial conveying route information.

[0150] The safety constraint unit is used to select at least one preferred AGV from the preferred AGVs, import the second initial transport route information into the selected preferred AGV, and then simulate the transport of the preferred AGV based on the first initial transport route information and the second initial transport route information to obtain the AGV transport safety constraint model.

[0151] The obstacle avoidance management unit is used to import the AGV transport safety constraint model into all the preferred AGVs to obtain transport safety AGV information. Then, it imports the AGV task request information into the transport safety AGV and generates an obstacle avoidance model to realize automatic AGV transport management.

[0152] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A cloud computing-based AGV automatic conveying management method, characterized in that, include: S1. Obtain the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. Then, receive the AGV task request information of multiple warehouses through the cloud platform, and use the AGV task request information and location information of multiple warehouses to build a resource constraint model. S2. Based on the AGV task request information from multiple warehouses, determine the candidate AGV information, and then filter the candidate AGVs according to the candidate AGV information and the resource constraint model to determine the preferred AGV; S3. Download the workshop internal scene information and production line scene information from the cloud platform, randomly select a preferred AGV from the preferred AGVs to obtain the first preferred AGV, and then simulate and generate the first initial conveying route information of the first preferred AGV through AGV task request information, workshop internal scene information and production line scene information. S4. Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, determine the conveying focus position information in the first initial conveying route information, and obtain the second initial conveying route information based on the conveying focus position information in the first initial conveying route information, wherein the conveying focus position information includes AGV conveying turning points and AGV conveying intersection points. S5. Select at least one preferred AGV from the preferred AGVs, and import the second initial conveying route information into the selected preferred AGV. Then, based on the first initial conveying route information and the second initial conveying route information, simulate the conveying of the preferred AGV to obtain the AGV conveying safety constraint model. S6. Import the AGV conveying safety constraint model into all the preferred AGVs to obtain the conveying safety AGV information. Then, import the AGV task request information into the conveying safety AGV and use the sensors deployed on the AGV to collect real-time environmental data to generate an obstacle avoidance model, thereby realizing automatic AGV conveying management.

2. The AGV automatic conveying management method based on cloud computing according to claim 1, characterized in that, Step S1 specifically includes: The location information of multiple warehouses is obtained by collecting the latitude and longitude information and area information of each warehouse through GPS devices, and then importing the location information of multiple warehouses into the cloud platform. The system receives AGV task request information from various warehouses through a cloud platform. The AGV task request information includes the starting point of the transport, the ending point of the transport, order attribute information, cargo attribute information, and time information. Define order attribute rules and goods attribute rules based on order attribute information and goods attribute information; Obtain AGV load information and warehouse busyness from the cloud platform, and define resource constraint rules based on AGV load information and warehouse busyness; Based on the delivery start point, delivery end point, time information, and the location information of multiple warehouses, dynamic constraint rules are defined. A resource constraint model is constructed based on order attribute rules, goods attribute rules, resource constraint rules, and dynamic constraint rules.

3. The AGV automatic conveying management method based on cloud computing according to claim 2, characterized in that, Step S2 specifically includes: Based on AGV task request information from multiple warehouses, obtain the location information of all AGVs within the range of the conveying start point and conveying end point; Based on the location information of all AGVs within the range of the starting point and ending point of the transport, a pool of candidate AGVs is constructed to determine the candidate AGV information, which includes candidate AGV location information, candidate AGV power information, and candidate AGV type information. Based on the candidate AGV type information, obtain the maximum load capacity of the candidate AGV, and based on the cargo attribute information, obtain the cargo weight. Then, use the ratio of the maximum load capacity of the candidate AGV to the cargo weight as the task matching degree. Based on the resource constraint model, the AGV scoring threshold is obtained, and then the scores of the candidate AGVs are obtained according to the task matching degree and the power information of the candidate AGVs. The candidate AGVs are screened by using their scores and AGV score thresholds. AGVs with scores greater than or equal to the AGV score threshold are extracted from the candidate AGV pool and a preferred AGV pool is constructed. An auction algorithm is used to allocate preferred AGVs to AGV task request information from multiple warehouses, thus determining the preferred AGV for each warehouse's AGV request information.

4. The AGV automatic conveying management method based on cloud computing according to claim 3, characterized in that, Step S3 specifically includes: Download workshop interior scene information and production line scene information from the MES system on the cloud platform. The workshop interior scene information includes a workshop layout diagram, and the production line scene information includes shelf locations and equipment coordinates. Using the workshop layout diagram, shelf locations, and equipment coordinates, a two-dimensional map of the production line is constructed, and the equipment coordinates are used to obtain information on the environmental footprint of the equipment. This information is then used to mark the obstacle heat map. Based on the 2D map of the production line and the heat map of obstacles, the heat map of obstacles is removed from the 2D map of the production line, so that the remaining area is used as the preferred AGV transport area map. Obtain the edge points of the limited AGV transportable area map and determine the center point of the 2D map of the production line; Using the center point of the two-dimensional map of the production line, construct a sliding straight line f(x)=h, where h is the dynamic sliding value, and the unit value of the dynamic sliding value is the distance between the edge points of the two nearest finite AGV transportable area maps. By using a sliding straight line to traverse all edge points in the finite AGV transportable area map, and obtaining the transport path tracing line f(x)=j when there are at least two edge points on the sliding straight line when the dynamic sliding values ​​are the same; Obtain the coordinates of the two farthest edge points in the conveying path tracking line, and construct the first conveying path tracking line segment using the coordinates of the two farthest edge points in the conveying path tracking line. Determine the center point of the first conveying path tracking segment, and obtain the two edge points closest to the center point of the first conveying path tracking segment; Construct a second transport path tracking segment using the two edge points closest to the center point of the transport path tracking segment; Starting from the conveying start point, connect the center points of all the second conveying path tracking segments in sequence to obtain an irregular line segment from the conveying start point to the center point of the second conveying path tracking segment to the conveying end point; Using irregular line segments, a conveying path tracking function is constructed, and based on the conveying path tracking function, the first initial conveying route information of the first preferred AGV corresponding to all AGV task request information is simulated and generated.

5. The AGV automatic conveying management method based on cloud computing according to claim 4, characterized in that, Step S4 specifically includes: Based on the first initial conveying route information, the workshop interior scene information, and the production line scene information, determine the conveying focus location information in the first initial conveying route information; Randomly select a first initial transport route from the first initial transport route information of the first preferred AGV corresponding to all AGV task request information generated in the simulation, and designate the remaining first initial transport routes as the third initial transport routes; Traverse the third initial transport route and define the third initial transport route containing the transport focus position from the first initial transport route information as the second initial transport route information.

6. The AGV automatic conveying management method based on cloud computing according to claim 5, characterized in that, Step S5 specifically includes: Based on the first and second initial transport route information, the preferred AGV is simulated to obtain the second initial transport route information of the collision that occurs during the simulated transport of the preferred AGV. The total number of second initial transport routes that collided is obtained. Then, based on the candidate AGV type information, the footprint of the preferred AGV is determined. The footprint of the preferred AGV is multiplied by the total number of second initial transport routes that collided to obtain the AGV transport safety area information. The safe area information of AGV transportation is used as the safety constraint model for AGV transportation.

7. The AGV automatic conveying management method based on cloud computing according to claim 6, characterized in that, Step S6 specifically includes: S6. Import the AGV conveying safety constraint model into all the preferred AGVs to obtain the conveying safety AGV information. Then, import the AGV task request information into the conveying safety AGV and use the sensors deployed on the AGV to collect real-time environmental data to generate an obstacle avoidance model and realize automatic AGV conveying management. The AGV transport safety constraint model is imported into all the preferred AGVs. The edge of the AGV transport safety constraint model is used as the safety boundary of the preferred AGV to obtain the transport safety AGV information. The transport safety AGV information includes the safety boundary of the preferred AGV, the location information of the preferred AGV, the power information of the preferred AGV, and the type information of the preferred AGV. Real-time environmental data is collected using sensors deployed on the AGV, including the location of sudden obstacles, the location of the working AGV, and the location of the edge of the production equipment. When the location of a sudden obstacle, the location of the AGV in operation, or the edge of the production equipment coincides with the safety boundary of the preferred AGV, an obstacle avoidance model is generated based on the preferred AGV location information, preferred AGV type information, and real-time environmental data. Based on the obstacle avoidance model, automatic AGV conveying management is realized.

8. A cloud-based AGV automatic conveying management system, used to implement the management method as described in any one of claims 1-7, characterized in that, include: The acquisition module is used to acquire the location information of multiple warehouses and import the location information of multiple warehouses into the cloud platform. It is also used to collect real-time environmental data using sensors deployed on the AGV. The main control module is used to receive AGV task request information from multiple warehouses through the cloud platform, and also to download workshop internal scene information and production line scene information from the cloud platform. It is used to construct a resource constraint model using the AGV task request information and location information from multiple warehouses, and to determine candidate AGV information based on the AGV task request information from multiple warehouses. Then, based on the candidate AGV information and the resource constraint model, the candidate AGVs are screened to determine the preferred AGV. The management module is used to arbitrarily select one preferred AGV from the preferred AGVs to obtain a first preferred AGV. Then, based on AGV task request information, workshop internal scene information, and production line scene information, it simulates and generates a first initial conveying route information for the first preferred AGV. According to the first initial conveying route information, workshop internal scene information, and production line scene information, it determines the conveying focus position information in the first initial conveying route information. The conveying focus position information includes AGV conveying turning points and AGV conveying intersection points. Based on the conveying focus position information in the first initial conveying route information, it obtains second initial conveying route information, which is used to select at least one preferred AGV from the preferred AGVs and import the second initial conveying route information into the selected preferred AGV. Then, based on the first and second initial conveying route information, it simulates the conveying of the preferred AGV to obtain an AGV conveying safety constraint model. This model is then imported into all preferred AGVs to obtain conveying safety AGV information. Finally, it imports AGV task request information into the conveying safety AGV and generates an obstacle avoidance model to realize automatic AGV conveying management. The display module is used to display the acquisition process and results of the acquisition module, the data processing process and results of the main control module, and the management process and results of the management module.

9. The AGV automatic conveying management system based on cloud computing according to claim 8, characterized in that, The main control module includes: The receiving unit is used to receive AGV task request information from multiple warehouses through the cloud platform, and is also used to download workshop internal scene information and production line scene information from the cloud platform. The processing unit is used to construct a resource constraint model by utilizing AGV task request information and location information from multiple warehouses; The AGV determination unit is used to determine candidate AGV information based on AGV task request information from multiple warehouses, and then filter the candidate AGVs according to the candidate AGV information and resource constraint model to determine the preferred AGV.

10. The AGV automatic conveying management system based on cloud computing according to claim 8, characterized in that, The management module includes: The path determination unit is used to arbitrarily select a preferred AGV from the preferred AGVs to obtain a first preferred AGV. Then, through AGV task request information, workshop internal scene information and production line scene information, it simulates and generates a first initial conveying route information of the first preferred AGV. Based on the first initial conveying route information, workshop internal scene information and production line scene information, it determines the conveying focus position information in the first initial conveying route information, and obtains a second initial conveying route information based on the conveying focus position information in the first initial conveying route information. A safety constraint unit is used to select at least one preferred AGV from the preferred AGVs, import the second initial transport route information into the selected preferred AGV, and then simulate the transport of the preferred AGV based on the first initial transport route information and the second initial transport route information to obtain an AGV transport safety constraint model. The obstacle avoidance management unit is used to import the AGV transport safety constraint model into all preferred AGVs to obtain transport safety AGV information, then import the AGV task request information into the transport safety AGV, and generate an obstacle avoidance model to realize automatic AGV transport management.

Images