Robot collaborative sorting and palletizing system and method for urea bags in a disorderly stacked state

Abstract

The present application relates to the technical field of urea bag transportation and palletizing, and discloses a robot collaborative sorting and palletizing system and method for urea bags in a disorderly stacked state, comprising: identifying the type and posture of the urea bags based on pre-acquired image data of the urea bags in the disorderly stacked state to obtain a urea bag identification result; analyzing the outbound frequency of different types of urea bags according to preset order requirements and real-time warehouse inventory, and allocating a corresponding storage area for each type of urea bag; generating a palletizing plan based on the urea bag identification result and the corresponding storage area; based on the palletizing plan, dynamically adjusting the working area and sorting action of each robot to achieve multi-robot collaborative sorting and palletizing; the present application dynamically partitions the storage area of ​​each urea bag and formulates a corresponding palletizing method for the urea bags in each area, thereby improving the grasping success rate of the robot palletizing process, and collaborating sorting between multiple robots to improve the cargo palletizing efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of urea bag transportation and palletizing, and in particular to a robot collaborative sorting and palletizing system and method for urea bags in a disorderly stacked state. Background Art

[0002] Existing robot sorting and palletizing methods have low recognition accuracy for items in a messy stacked state, resulting in unsuccessful grasping or collapse during the sorting process; fixed partitions are set for storage of different types of goods, and the corresponding storage areas cannot be adjusted according to actual order conditions, resulting in long cargo handling distances and low warehouse efficiency; the same palletizing method is used for different goods, ignoring the fact that the risks of collapse vary depending on the palletizing state of different goods, resulting in the risk of collapse of the stacked goods.

[0003] For example, the Chinese patent application with publication number CN117755833A discloses an intelligent sorting and palletizing robot and control method, which includes: a sorting mechanism, a pallet distribution mechanism, a finished product conveying mechanism and a robot palletizing mechanism. It uses a truss mechanical structure and a combination of two sets of manipulators to achieve multi-degree-of-freedom operation. In the process, an intelligent palletizing algorithm is used to calculate the shortest path to the workstation after the product is picked out, and the products are intelligently sorted and palletized according to different specifications along the optimal path.

[0004] The above prior art has the problems raised by this background technology. In order to solve at least one of the above problems, the present invention proposes a robot collaborative sorting and stacking system and method for urea bags in a disorderly stacked state. Summary of the Invention

[0005] In response to the shortcomings of the prior art, the main purpose of the present invention is to provide a robot-assisted sorting and palletizing system and method for urea bags in a disorderly stacked state, which can effectively solve the problems in the background technology. The specific technical solutions of the present invention are as follows:

[0006] The robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state includes:

[0007] Based on the pre-acquired image data of the urea bags in a disorderly stacked state, the type and posture of the urea bags are identified to obtain a urea bag recognition result;

[0008] Analyze the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocate corresponding storage areas for each type of urea bag;

[0009] Generate a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence;

[0010] Based on the palletizing solution, the working area and sorting action of each robot are dynamically adjusted to achieve multi-robot collaborative sorting and palletizing.

[0011] Specifically, the method of identifying the type and posture of the urea bags based on the pre-acquired image data of the urea bags in a disorderly stacked state to obtain the urea bag identification result includes:

[0012] Feature extraction is performed on the pre-acquired image data of the urea bags in a disorderly stacked state to obtain image features;

[0013] Based on the image features, the position coordinates of each urea bag are identified through a preset urea bag detection model;

[0014] generating a bounding box for each urea bag according to the position coordinates;

[0015] Within the boundary box, the type and posture of each urea bag are identified through a preset urea bag posture recognition model to obtain a urea bag recognition result.

[0016] Specifically, within the bounding box, the type and posture of each urea bag are identified through a preset urea bag posture recognition model, and a urea bag recognition result is obtained, including:

[0017] Within the bounding box, extract the four corner points of the bounding box as key points;

[0018] Calculate the rotation vector and translation vector of each key point according to the position coordinates of the key points;

[0019] Calculating the spatial posture of the urea bag according to the rotation vector and the translation vector through a preset urea bag posture recognition model;

[0020] Based on the image features within the bounding box, the type of each urea bag is identified through the preset urea bag type recognition model;

[0021] Combining the type and spatial posture, a urea bag recognition result is obtained.

[0022] Specifically, according to the preset order requirements and the real-time inventory of the warehouse, the delivery frequency of different types of urea bags is analyzed, and corresponding storage areas are allocated to each type of urea bags, including:

[0023] Calculate the delivery frequency of each category of urea bags based on preset order requirements and real-time warehouse inventory;

[0024] The warehouse is divided into multiple locations using a preset dynamic area division model.

[0025] Calculate the priority of each location area based on its distance from the outbound location and its distance from other areas;

[0026] Matching a corresponding location area for each category of urea bags according to the delivery frequency and the priority of each location area;

[0027] According to the size and inventory information of each category of urea bags, the preset boundary optimization model is used to optimize the boundaries of each location area to obtain the corresponding storage area.

[0028] Specifically, according to the delivery frequency and the priority of each location area, a corresponding location area is matched for each type of urea bag, including:

[0029] The location areas are layered according to the priority of each location area from high to low to obtain multi-layer location areas;

[0030] Based on the urea bag category, corresponding urea bag information is searched in a preset urea bag information database, and the urea bags are layered to obtain multi-layer urea bags, wherein the multi-layer urea bags include easily deformable urea bags, high-density urea bags, and special-sized urea bags;

[0031] Combined with the delivery frequency, each layer of urea bags is matched to the location area of ​​the corresponding level through the preset location matching model;

[0032] In each layer of location areas, corresponding location areas are allocated according to the frequency of each type of urea bags in each layer of urea bags leaving the warehouse.

[0033] Specifically, a palletizing plan is generated based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence, including:

[0034] According to the layer type corresponding to each urea bag and the preset stacking rules, the corresponding initial stacking layer number is generated;

[0035] The initial number of stacking layers is optimized based on the posture of the urea bags and the size of the storage area to obtain an optimized number of stacking layers;

[0036] According to the position coordinates and posture of each urea bag, the urea bag closest to the current storage area and with the flattest posture is selected for each palletizing to obtain the palletizing order;

[0037] The optimized number of stacking layers and stacking sequence are combined to obtain a stacking solution.

[0038] Specifically, based on the palletizing solution, the working area and sorting action of each robot are dynamically adjusted to achieve multi-robot collaborative sorting and palletizing, including:

[0039] According to the palletizing plan, the corresponding working area is allocated to each robot in real time;

[0040] During the robot's operation, the preset path planning model is used to plan the movement trajectory of the robot's corresponding manipulator arm to obtain the movement path;

[0041] According to the posture of the urea bag, the gripping angle and strength of each robot are adjusted to obtain the sorting action;

[0042] Each robot sorts and palletizes the urea bags in the working area according to the sorting action and moving path until the urea bag palletizing tasks in all assigned working areas are completed, thereby realizing multi-robot collaborative sorting and palletizing.

[0043] Specifically, each robot can communicate with each other; and according to the palletizing plan, a corresponding working area is allocated to each robot in real time, including:

[0044] According to the palletizing scheme and the size of the urea bags corresponding to the storage area, the storage area is divided into multiple grids by a preset grid division model, wherein the multiple grids include storage grids and idle grids;

[0045] Each robot grabs the top urea bag from the chaotically stacked urea bags and matches it to the corresponding grid according to the type of urea bag it grabs;

[0046] According to the robot's movement process, the time required to place the urea bag into the corresponding grid is calculated to obtain the transportation time;

[0047] When multiple robots are matched to the same grid, the priority of the robot with the shortest transportation time is set to the highest priority. The robot corresponding to the highest priority will first place the grabbed urea bag in the corresponding grid, and the other robots will place the urea bag in the idle grid;

[0048] After each robot completes a urea bag palletizing process, the corresponding priority is reset to zero. The benefits of palletizing on an idle grid and on a pile of messy urea bags are calculated based on the current position to obtain the first and second benefits.

[0049] According to the first benefit and the second benefit, an area corresponding to a higher benefit value is selected as the working area for the next palletizing task;

[0050] Repeat the process of dynamically selecting the work area until all palletizing tasks are completed.

[0051] Specifically, the grasping angle and strength of each robot are adjusted according to the posture of the urea bag to obtain the sorting action, including:

[0052] Calculate the initial grasping angle of each robot based on the posture of the urea bag;

[0053] Calculate the initial grasping force of each robot based on the weight and material of the urea bag;

[0054] Sorting the urea bags based on the initial gripping angle and the initial gripping force;

[0055] During the sorting process, the initial gripping angle and initial gripping force are adjusted according to the real-time status of the urea bag to obtain the sorting action.

[0056] The robot collaborative sorting and palletizing system for urea bags in a disorderly stacked state is used to implement the robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state, comprising:

[0057] The urea bag recognition module identifies the type and posture of the urea bags based on pre-acquired image data of the urea bags in a disorderly stacked state, and obtains the urea bag recognition result;

[0058] The regional allocation module analyzes the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocates corresponding storage areas for each type of urea bag;

[0059] a palletizing plan generating module, which generates a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence;

[0060] The robot collaboration module dynamically adjusts the working area and sorting action of each robot based on the palletizing solution to achieve multi-robot collaborative sorting and palletizing.

[0061] A computer-readable storage medium stores a computer program for implementing the above-mentioned robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state.

[0062] Compared with the prior art, this application has the following beneficial effects:

[0063] This application dynamically partitions the storage area of ​​each urea bag based on the real-time status of the identified urea bags and combines it with order requirements. It also formulates corresponding palletizing methods for the urea bags in each area, thereby improving the success rate of robot palletizing. By dynamic partitioning, the palletizing and outbound delivery time of goods can be shortened, and the space utilization rate of the warehouse can be improved. Different robots collaborate in sorting, and the working area and sorting actions of each robot are dynamically adjusted to achieve efficient collaborative work of multiple robots and improve the efficiency of cargo transportation and palletizing. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1This is a workflow diagram of the robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state in Example 1 of the present invention;

[0065] Figure 2 This is a schematic diagram of the storage area division in Example 1 of the present invention;

[0066] Figure 3 This is a schematic diagram of the robot work area allocation in Example 1 of the present invention;

[0067] Figure 4 Schematic diagram of the structure of the robot collaborative sorting and palletizing system in the disorderly stacking state of urea bags in Example 2 of the present invention. DETAILED DESCRIPTION

[0068] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0069] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0070] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0071] Example 1

[0072] This embodiment provides a robot collaborative sorting and stacking method for urea bags in a disorderly stacked state, such as Figure 1 As shown, the robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state includes:

[0073] S101, based on pre-acquired image data of urea bags in a disorderly stacked state, identifying the type and posture of the urea bags to obtain a urea bag recognition result;

[0074] S102. Analyze the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocate corresponding storage areas for each type of urea bag.

[0075] S103, generating a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence;

[0076] S104: Based on the palletizing solution, dynamically adjust the working area and sorting action of each robot to achieve multi-robot collaborative sorting and palletizing.

[0077] This embodiment identifies the real-time stacking status of urea bags based on image data of the disorderly stacked urea bags. Based on the identification results, combined with order requirements and real-time warehouse inventory, a corresponding storage area is allocated to each category of urea bags. This avoids the problem of the current setting of fixed storage areas, which leads to untimely delivery of urea bags with high delivery frequency, affecting delivery efficiency. Different stacking schemes are set according to different categories of urea bags, and the working area and sorting dynamics of each robot are adjusted accordingly. Different stacking methods can be set for different types of urea bags to avoid the collapse of urea bags due to the same stacking method. The robot's working area is dynamically adjusted to enable the robot to work non-stop, thereby improving stacking efficiency.

[0078] In this embodiment, first, the image data of the urea bag is acquired by an RGB-D camera, and the position and posture of the urea bag in the image data are identified based on the target detection and recognition method of deep learning, and the type of the urea bag and its posture in space are determined; the type and posture of the urea bag are accurately identified to ensure that the robot can take corresponding sorting operations for urea bags of different types and postures, thereby improving the accuracy and efficiency of sorting and stacking, and avoiding erroneous operations caused by misjudgment. Secondly, based on the number of various types of urea bags in the preset order requirements and the inventory of different types of urea bags in the real-time inventory of the warehouse, the outbound frequency of each type of urea bag is analyzed and calculated, and based on the corresponding outbound frequency and combined with the warehouse layout, the optimal storage area is allocated to each type of urea bag, so that urea bags with a high outbound frequency are placed in an area closer to the outbound port, reducing the pickup time and improving the overall logistics operation speed. At the same time, it helps to optimize the utilization of warehouse space and improve the scientific nature of inventory management.

[0079] Specifically, the urea bag recognition results are combined with the corresponding storage area, and a palletizing plan is generated taking into account the weight, size, posture of the urea bags and the space limitations of the target storage area. The obtained palletizing plan can fully consider the actual situation of the urea bags and the characteristics of the storage area, achieve efficient palletizing, improve space utilization, ensure the stability of palletizing, and provide clear task guidance for multi-robot collaborative operations; according to the generated palletizing plan, a corresponding work area is assigned to each robot through the communication network between robots, and the sorting action of each robot is dynamically planned according to the palletizing sequence and the position of the urea bag, including the moving path, grasping action, etc. At the same time, it is ensured that multiple robots work together. Through the collaborative sorting and palletizing of multiple robots, the efficiency of sorting and palletizing is greatly improved, and the sorting and palletizing tasks of a large number of urea bags can be completed more quickly, thereby improving the automation level of the warehouse.

[0080] This application dynamically partitions the storage area of ​​each urea bag based on the real-time status of the identified urea bags and combines it with order requirements. It also formulates corresponding palletizing methods for the urea bags in each area, thereby improving the success rate of robot palletizing. By dynamic partitioning, the palletizing and outbound delivery time of goods can be shortened, and the space utilization rate of the warehouse can be improved. Different robots collaborate in sorting, and the working area and sorting actions of each robot are dynamically adjusted to achieve efficient collaborative work of multiple robots and improve the efficiency of cargo transportation and palletizing.

[0081] Furthermore, the type and posture of the urea bags are identified based on the pre-acquired image data of the urea bags in a disorderly stacked state, and the urea bag identification result is obtained, including:

[0082] S201, extracting features from pre-acquired image data of urea bags in a disorderly stacked state to obtain image features;

[0083] S202: Based on the image features, identify the position coordinates of each urea bag using a preset urea bag detection model;

[0084] S203, generating a bounding box of each urea bag according to the position coordinates;

[0085] S204 : Within the boundary box, identify the type and posture of each urea bag using a preset urea bag posture recognition model to obtain a urea bag recognition result.

[0086] In this embodiment, recognition is performed based on pre-acquired image data of urea bags in a disorderly stacked state. The quality of the directly acquired image data is not high, so the acquired original image is grayed, filtered, denoised, and processed to remove noise interference in the image, enhance the image contrast, and improve the image quality. The pre-processed image data is processed, and digital image processing technology is used to extract image features, including texture features, shape features, etc., from the pre-processed image data. Image feature vectors are calculated based on the extracted features. By extracting image features, the differences between urea bags and backgrounds and between different urea bags can be highlighted, thereby improving the accuracy and efficiency of urea bag posture recognition.

[0087] Furthermore, based on the extracted image features, the position of each urea bag is searched and located in the image. According to a preset urea bag detection model (the urea bag detection model specifically adopts the YOLO model), the urea bag detection model is trained using a large amount of historical image data to obtain a pre-trained urea bag detection model. The extracted image features are input into the pre-trained urea bag detection model, and the model outputs the position coordinates of the urea bag in the image. A rectangular bounding box is generated based on the position coordinates, completely framing the urea bag within the bounding box. By generating each bounding box, the position and range of the urea bag in the image are intuitively displayed.

[0088] Specifically, within the generated bounding box, the type and posture of the urea bag are identified using a preset urea bag posture recognition model. The urea bag posture recognition model is trained using a large amount of historical data to obtain a pre-trained urea bag posture recognition model. The type and posture of the urea bag within each bounding box are identified using the pre-trained urea bag posture recognition model. Type and posture recognition are performed within the bounding box, effectively utilizing local image information to improve recognition accuracy and pertinence. The type and posture of the urea bag are accurately identified, enabling the robot to adopt appropriate operation strategies for grasping operations based on different types and postures, thereby improving the accuracy and efficiency of operations.

[0089] Furthermore, within the bounding box, the type and posture of each urea bag are identified through a preset urea bag posture recognition model to obtain a urea bag recognition result, including:

[0090] S301, within the bounding box, extract the four corner points of the bounding box as key points;

[0091] S302, calculating the rotation vector and translation vector of each key point according to the position coordinates of the key point;

[0092] S303, calculating the spatial posture of the urea bag according to the rotation vector and the translation vector using a preset urea bag posture recognition model;

[0093] S304, identifying the type of each urea bag using a preset urea bag type recognition model based on the image features within the bounding box;

[0094] S305 : Combining the type and the spatial posture, obtaining a urea bag recognition result.

[0095] In this embodiment, within each bounding box, the four corner points of each bounding box are extracted as key points, and the position coordinates of each key point are obtained. For each key point, the rotation vector and translation vector of each key point are calculated, and a reference coordinate system is determined. The upper left corner of the image is selected as the origin, the horizontal right direction is the positive direction of the x-axis, and the vertical downward direction is the positive direction of the y-axis. For each key point, according to its coordinates in the image coordinate system and the standard position coordinates (that is, the position of the key point of the urea bag in the reference coordinate system when the urea bag is laid flat), a coordinate transformation algorithm based on homogeneous coordinates is used to calculate the rotation vector and translation vector of the key point relative to the reference coordinate system. By calculating the rotation vector and translation vector, the position and orientation of each urea bag in the image are determined, thereby improving the accuracy and reliability of posture recognition and providing more accurate posture information for the robot to accurately grasp the urea bag.

[0096] Specifically, according to the calculated rotation vector and translation vector, the spatial posture of each urea bag in space is identified by a preset urea bag posture recognition model. The urea bag posture recognition model is specifically a PNP model. The calculated rotation vector and translation vector and the coordinate position of the key point are input into the PNP model. The model determines the rotation angle of the urea bag around each coordinate axis through the rotation vector, and determines the position of the urea bag in three-dimensional space through the translation vector. The Euler angle of the urea bag is calculated to represent the posture information of the urea bag, thereby comprehensively identifying the spatial posture of the urea bag; the spatial posture of the urea bag is calculated by the urea bag posture recognition model, which accurately reflects the actual placement status of the urea bag, helps the robot plan a more reasonable grasping path and action, and improves the success rate of grasping.

[0097] Specifically, based on the image features within the bounding box, the type of each urea bag is identified using a pre-set urea bag type recognition model. This urea bag type recognition model is specifically a CNN model, which is trained using a large amount of historical data to obtain a pre-trained urea bag type recognition model. The model matches and compares the image features with the features of various urea bags learned during training to determine the urea bag type. Accurately identifying the urea bag type helps to adopt different palletizing strategies based on different urea bag types, improving the refinement of warehouse management and the accuracy of operations, thereby improving overall palletizing efficiency. The obtained urea bag type and spatial posture information are integrated to obtain a complete urea bag recognition result, providing comprehensive and accurate information for subsequent sorting and palletizing operations, allowing robots or operators to formulate the optimal operation plan based on the type and posture of the urea bag.

[0098] Furthermore, according to the preset order requirements and the real-time inventory of the warehouse, the delivery frequency of different types of urea bags is analyzed, and a corresponding storage area is allocated to each type of urea bag, including:

[0099] S401. Calculate the delivery frequency of each category of urea bags based on preset order requirements and real-time warehouse inventory;

[0100] S402: Divide the warehouse into regions using a preset dynamic region division model to obtain multiple location regions;

[0101] S403. Calculate the priority of each location area based on the distance from the delivery location and the distance between each location area and other areas;

[0102] S404, matching a corresponding location area for each type of urea bag according to the delivery frequency and the priority of each location area;

[0103] S405 : Based on the size and inventory information of each category of urea bags, the boundaries of each location area are optimized using a preset boundary optimization model to obtain a corresponding storage area.

[0104] In this embodiment, the order demand for urea bags and the real-time inventory in the warehouse are analyzed to determine the frequency of each type of urea bag shipment. Urea bags with high shipment frequencies are then allocated to convenient shipping areas, thereby improving overall warehouse efficiency. Specifically, the order quantities of different types of urea bags in the preset order demand and the changes in the inventory of each type of urea bag in the warehouse's real-time inventory are analyzed to calculate the shipment frequency of each type of urea bag within a certain time period. For example, if the time period is set to one day, the inventory reduction of each type of urea bag due to order shipments is counted within the time period. The inventory reduction is then divided by the duration of the time period to calculate the shipment frequency of each type of urea bag.

[0105] Specifically, the warehouse is divided into regions through a preset dynamic region division model to obtain multiple location regions. The warehouse is divided into regions according to the physical structure of the warehouse, such as the layout, shelf distribution, and aisle settings. For example, the warehouse is divided into a fast shipping area close to the shipping port, a regular storage area in the middle, a low-frequency storage area far from the shipping port, etc., and the boundaries and scope of each area are determined at the same time. Through reasonable region division, the utilization of warehouse space can be optimized, the efficiency of cargo storage and handling can be improved, and different areas can be managed and operated in a targeted manner.

[0106] At the same time, based on the distance of each location area from the outbound location and the distance between each location area and other areas, the convenience and importance of each location area in the entire warehouse operation process are evaluated and its priority is determined. The specific calculation formula is as follows:

[0107] ;

[0108] Where, is the priority value of the area, is the distance between the area and the outbound location, is the average distance between this area and other areas, is the connection weight between this area and the high-frequency area, It is the connection weight between this area and key operation areas (such as sorting area, processing area, etc.). The closer to the outbound location, the higher the priority. In this embodiment, the weight is set to 0.6, which can be set according to actual calculation requirements. The closer the connection with other areas, the higher the priority. and The contributions of connections to high-frequency areas and key operational areas to priority are respectively expressed, reflecting the role of connections to different important areas in priority calculation. By clarifying the priority of each location area, urea bags with high outbound frequency are preferentially placed in high-priority areas, reducing handling time and costs, improving warehouse logistics efficiency, and optimizing the overall warehouse layout and operational processes.

[0109] At the same time, the calculated outbound frequency of each category of urea bags is correlated and matched with the priority of each location area. Urea bags with high outbound frequency are preferentially matched to location areas with high priority to ensure efficient picking and delivery operations; urea bags with low outbound frequency are matched to location areas with relatively low priority, which can reasonably utilize warehouse space resources; through dynamic matching of warehouse locations, urea bags with different outbound frequencies can be placed in the most suitable location areas, which significantly improves the warehouse's operating efficiency and enables high-frequency outbound urea bags to be quickly transported to the outbound port to meet order requirements, while also avoiding warehouse congestion and operational chaos caused by unreasonable placement.

[0110] Furthermore, the initially divided location areas cannot adapt to the sizes of different types of urea bags. The preset boundary optimization model is used to optimize the boundaries of each location area to obtain the corresponding storage area. According to the size and inventory information of each category of urea bags allocated to each location area, the storage space required for each type of urea bag is calculated. Based on the calculated storage space requirements and combined with the actual physical structure and layout limitations of the warehouse, the boundary coordinates of the location area are adjusted through the boundary optimization model, and the shape and size of the location area are optimized. Without affecting the normal operation of other areas and without exceeding the physical boundaries of the warehouse, the location area is appropriately expanded or reduced so that it can maximize the space utilization of the warehouse while meeting the storage requirements of urea bags.

[0111] like Figure 2 As shown in the figure, the warehouse is divided into multiple location areas, which are represented by solid lines in the figure. After the corresponding storage urea bags are allocated to each area, the boundary of each location area is optimized based on the type and size of the urea bag and the size of the location area to obtain the corresponding storage area, which is represented by dotted lines in the figure. Through area division and boundary optimization, the space utilization of the warehouse can be maximized.

[0112] Furthermore, according to the delivery frequency and the priority of each location area, a corresponding location area is matched for each type of urea bag, including:

[0113] S501, stratify the location areas according to the priority of each location area from high to low to obtain multiple layers of location areas;

[0114] S502: Based on the urea bag category, search for corresponding urea bag information in a preset urea bag information database, and layer the urea bags to obtain multi-layer urea bags, wherein the multi-layer urea bags include easily deformable urea bags, high-density urea bags, and special-sized urea bags;

[0115] S503. Based on the delivery frequency, each layer of urea bags is matched to the location area of ​​the corresponding layer using a preset location matching model;

[0116] S504. In each layer of location areas, a corresponding location area is allocated according to the frequency of each type of urea bags in each layer of urea bags leaving the warehouse.

[0117] In this embodiment, the location areas are arranged from high to low according to the calculated priority of each location area, and location areas with similar priority ranges are classified into the same layer to construct a multi-layer location area structure. The stratification rules are set according to the actual calculation requirements. According to the level of priority scores, the top 30% location areas with high scores are divided into the first layer, the middle 40% are divided into the second layer, and the bottom 30% are divided into the third layer. The stratification of location areas gives the warehouse space a clear structure in terms of priority division. Location areas at different levels can correspond to different types of urea bags, which improves the rationality and efficiency of storage area allocation and helps optimize the overall warehouse operation process.

[0118] Specifically, based on the identified urea bag type, the corresponding urea bag characteristic information is searched in the preset urea bag information library, including whether it is easy to deform, the density, and whether the size is special. The urea bags are layered to obtain multi-layer urea bags, including easily deformed urea bags, high-density urea bags and special-sized urea bags. Different levels of urea bags have different requirements for storage areas. For example, easily deformed urea bags have higher requirements for the stability of the storage environment, high-density urea bags require more sturdy shelves, and special-sized urea bags require storage areas with specific spatial layouts.

[0119] Furthermore, by utilizing a preset location matching model, combined with the stratification of urea bags and the priority of the location area of ​​each level, and combined with the outbound frequency information, urea bags of different levels are matched with the location areas of the corresponding levels, and matching rules are set. In this embodiment, the matching rules are set as follows: the layer of easily deformable urea bags is preferentially matched to the location area layer with the lowest priority, because the easily deformable urea bags have a greater risk of collapse. Placing them in the location area with a low priority can reduce the impact of collapse on other areas; for the high-density urea bag layer, considering its large weight, in addition to the priority, it must also be matched to a location area layer with a strong shelf carrying capacity; the special-sized urea bag layer is matched to a location area layer whose spatial layout can meet its special size requirements; at the same time, in the matching process, with reference to the outbound frequency, for the urea bag category with a high outbound frequency, priority is given to selecting an area in the location area of ​​the corresponding level that is more convenient for handling and shipment.

[0120] Specifically, hierarchical matching can fully utilize warehouse space resources to meet the storage needs of urea bags with different characteristics, while also taking into account the frequency of outbound shipments, thereby improving warehouse operational efficiency. After completing hierarchical matching, the location areas on each layer are further refined and allocated based on the outbound shipment frequency of each type of urea bag within that layer. Urea bags with high outbound shipment frequencies are allocated to locations within that layer that are closer to aisles for easier handling and faster shipment, while urea bags with low outbound shipment frequencies are allocated to relatively remote locations. This improves cargo handling efficiency and reduces overall operation time.

[0121] Furthermore, a palletizing plan is generated based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence, including:

[0122] S601: Generate the corresponding initial number of stacking layers according to the layer type corresponding to each urea bag and the preset stacking rules;

[0123] S602: Optimizing the initial number of stacking layers based on the posture of the urea bags and the size of the storage area to obtain an optimized number of stacking layers;

[0124] S603, according to the position coordinates and posture of each urea bag, each time palletizing, select the urea bag that is closest to the current storage area and has a suitable posture to obtain a palletizing order;

[0125] S604: Combining the optimized number of stacking layers and stacking sequence to obtain a stacking solution.

[0126] In this embodiment, urea bags of different layer types have different characteristics such as weight, size, and fragility. A corresponding initial number of stacking layers is generated according to preset stacking rules. A stacking rule table is formulated based on actual calculation requirements to determine the initial number of stacking layers corresponding to urea bags of different layer types. For example, the initial number of stacking layers for high-density and heavy urea bags is set to 5, while the initial number of stacking layers for ordinary lightweight urea bags is set to 10. Based on the basic characteristics of urea bags, a preliminary number of stacking layers is quickly generated, which improves the efficiency of generating a stacking plan and also ensures the basic rationality and stability of the stacking to a certain extent.

[0127] Specifically, the initial number of stacking layers is optimized based on the posture of the urea bags and the size of the storage area to obtain the optimized number of stacking layers. Based on the posture of the urea bags and the size of the storage area, the space occupied by the urea bags in the storage area at different numbers of stacking layers is calculated. At the same time, combined with the factors of stacking stability, such as the center of gravity distribution, the height-direction space occupied by the bags when stacked in different layers and the impact on the center of gravity stability are calculated. Taking the size limit of the storage area as a constraint, the initial number of stacking layers is optimized. When the stacking height under the initial number of stacking layers exceeds the height of the storage area, the number of stacking layers is gradually reduced. When the plane space utilization is too low when stacked according to the initial number of stacking layers, the number of stacking layers is increased and the stability is re-evaluated. Through multiple iterative calculations, the optimal number of stacking layers that meets the storage area size limit and the stacking stability requirements is found to obtain the optimized number of stacking layers. The optimized number of stacking layers can make the stacking plan more suitable for the actual storage environment, avoid problems such as space waste and stacking collapse caused by unreasonable stacking, and improve the overall efficiency and quality of storage operations.

[0128] Specifically, when palletizing, priority is given to the urea bags closest to the current storage area to reduce the length of the robot's moving path and improve handling efficiency. At the same time, the posture of the urea bags is taken into consideration, and urea bags with appropriate postures are selected for palletizing, which helps to ensure the stability and neatness of the palletizing. For example, selecting a flat urea bag with a smooth surface as the bottom layer for stacking is more conducive to the subsequent stacking of other urea bags, and avoids tilting of the palletizing due to improper posture of the bottom urea bag. After each urea bag is selected, the information of the remaining urea bags is updated, the distance is recalculated and the posture is evaluated, and the urea bags closest to the current storage area and with appropriate postures are continuously selected and added to the palletizing sequence list in sequence until all urea bags that need to be palletized are selected, thereby determining the complete palletizing sequence. By determining a reasonable palletizing sequence, the efficiency of the palletizing operation can be significantly improved, and the time and energy consumption during the handling process can be reduced.

[0129] Specifically, the optimized number of stacking layers and the stacking sequence are integrated to form a complete stacking solution. The number of stacking layers determines the height structure of the stacking, and the stacking sequence determines the order in which each urea bag is placed during the stacking process. The stacking solution provides clear operating instructions for the stacking process, improves the accuracy and efficiency of robot stacking, reduces delays or collapses caused by improper operations, and thus improves the operating efficiency of the entire warehousing and logistics system.

[0130] Furthermore, based on the palletizing solution, the working area and sorting action of each robot are dynamically adjusted to achieve multi-robot collaborative sorting and palletizing, including:

[0131] S701. Allocate a corresponding working area to each robot in real time according to the palletizing plan;

[0132] S702: During the operation of the robot, the movement trajectory of the robot's corresponding manipulator arm is planned using a preset path planning model to obtain a movement path;

[0133] S703: Adjust the gripping angle and strength of each robot according to the posture of the urea bag to obtain a sorting action;

[0134] S704: Each robot sorts and palletizes the urea bags in the work area according to the sorting action and moving path until the urea bag palletizing task in all assigned work areas is completed, thereby realizing multi-robot collaborative sorting and palletizing.

[0135] like Figure 3 As shown, this embodiment allocates a work area to each robot based on the palletizing solution and the real-time status of each robot. The robot's work area can be dynamically adjusted to improve palletizing efficiency. The gripping angle and force of each robot are designed based on the corresponding work area of ​​each robot and the palletizing requirements. Urea bags are sorted and palletized according to real-time sorting actions, thereby improving the grasping success rate.

[0136] In this embodiment, a robot information database is established based on the real-time status of the robots. Various parameters of each robot, such as maximum load, moving speed, current power, current working status (idle, busy, and progress of the task being executed), and other real-time status information are recorded. The palletizing sequence and storage area in the palletizing plan are analyzed, and the storage area is further divided into multiple sub-areas, each of which serves as a potential robot working area. Based on the working capacity and current status of the robot, a task allocation algorithm, specifically the Hungarian algorithm, is used to assign one or more sub-areas to each robot as its working area.

[0137] For example, during the allocation process, priority is given to assigning adjacent sub-areas to the same robot to reduce the robot's travel distance. At the same time, it is necessary to ensure that the amount of work assigned to each robot is within its working capacity. For example, if a robot has a maximum load of 50kg, and the urea bags in a sub-area weigh 40kg, and the robot is currently idle, then this sub-area can be assigned to it. Reasonable work area allocation can fully utilize the effectiveness of each robot, avoid waste or overuse of robot resources, and by reducing the robot's travel distance, it can improve work efficiency and shorten the overall sorting and palletizing time, thereby improving the speed of warehouse logistics operations.

[0138] Specifically, the robot's arm's movement trajectory during palletizing is planned based on the robot's assigned work area. A preset path planning model, specifically the A* model, is used to calculate the shortest path for the arm's movement. The robot's arm's gripping angle and force are calculated based on the robot's assigned work area and the urea bag's posture. The appropriate gripping angle and force are calculated based on the urea bag's posture recognition results and the robot's gripping device characteristics. For example, for a flat urea bag, the robot can grasp it horizontally, with the gripping force determined based on the bag's weight and material. For a sideways urea bag, the gripping angle needs to be adjusted to a certain angle, and the gripping force must also be adjusted accordingly to ensure a secure grip. Adjusting the gripping angle and force based on the bag's posture significantly improves the robot's grasping success rate, reduces the risk of urea bags falling or being damaged due to improper grasping, ensures cargo safety, improves work efficiency, avoids wasting time and resources due to repeated grasping or handling of damaged goods, and enhances the reliability and stability of the entire sorting and palletizing system.

[0139] Specifically, after receiving instructions for the work area and sorting action, each robot sorts and palletizes the urea bags within its assigned work area according to the palletizing sequence specified in the palletizing plan. It accurately grasps the urea bags according to the preset sorting action and places them in the designated location according to the palletizing sequence, completing the palletizing task. It then moves to the next work area along the planned shortest path to continue palletizing until it completes the sorting and palletizing of all urea bags within the work area. After completing the task, the robot sends a task completion signal to the system. Upon receiving the task completion signals from all robots, the system confirms the completion of the entire sorting and palletizing task. By enabling multiple robots to collaboratively perform sorting and palletizing tasks, the system can fully leverage the automation and intelligence advantages of the robots, significantly improving the efficiency and accuracy of sorting and palletizing. Furthermore, the collaborative work of multiple robots can speed up cargo processing throughout the warehouse, improving warehouse operational efficiency and meeting the needs of the rapid development of modern logistics.

[0140] Furthermore, each robot can communicate with each other; and according to the palletizing plan, a corresponding working area is allocated to each robot in real time, including:

[0141] S801: Divide the storage area into a plurality of grids according to the palletizing plan and the size of the urea bags corresponding to the storage area using a preset grid division model, wherein the plurality of grids include storage grids and idle grids;

[0142] S802: Each robot grabs the top urea bag from the randomly stacked urea bags and matches the grabbed urea bag to the corresponding grid according to its type;

[0143] S803. Calculate the time required to place the urea bag into the corresponding grid based on the robot's movement process to obtain the transportation time;

[0144] S804. When multiple robots are matched to the same grid, the priority of the robot with the shortest transportation time is set to the highest priority. The robot corresponding to the highest priority will first place the grabbed urea bag in the corresponding grid, and the other robots will place the urea bag in the vacant grid;

[0145] S805: After each robot completes a urea bag palletizing process, the corresponding priority is reset to zero, and the benefits of palletizing on an idle grid and on a pile of messy urea bags are calculated based on the current position to obtain a first benefit and a second benefit;

[0146] S806: Selecting an area with a higher profit value as the working area for the next palletizing task based on the first profit and the second profit;

[0147] S807. Repeat the process of dynamically selecting the working area until all palletizing tasks are completed.

[0148] In this embodiment, the storage area is divided by a preset grid division model to obtain multiple grid areas, wherein each grid is set according to the size of the urea bags, and each grid can only hold one row of urea bags. The urea bags grasped by each robot are assigned to the corresponding grid according to the type of the urea bags grasped by each robot. When multiple robots are assigned to the same grid, a stacking conflict occurs. The conflicting robots follow the first-come-first-served principle. The robot that takes the shortest time to reach the grid will first place the urea bag in the grid. To reduce waiting time, other robots will place the grasped urea bags in the idle grid. Each time a urea bag is placed in the grid, the robot selects the area with higher profit to complete the next stacking task based on the profit from stacking the stacked urea bags from the current position and the profit from stacking the idle grid. By dynamically allocating the working areas of the robots, the stacking efficiency can be improved, and the stacking task can be efficiently completed in a complex and changeable storage environment.

[0149] In this embodiment, a preset grid division model is used to convert the complex storage area into discrete and regular grid units based on the location, shape and corresponding urea bag size data of the storage area. A row of urea bags can be placed in each grid. In this embodiment, the grid side length is set according to a multiple of the average size of the urea bags. For example, if the average length of a certain type of urea bag is 0.5 meters, the grid side length is set to 0.6 meters to ensure that the urea bags have sufficient operating space in the grid. The gridded storage area makes task allocation more precise and accurate. Each grid becomes an independent task unit, which facilitates the quantification of task volume and area allocation.

[0150] Specifically, the robot's robotic arm moves to the pile of randomly stacked urea bags and starts grabbing from the top layer. A visual recognition system is configured above each robot to identify the type of urea bag in real time and match the robot with the nearest grid of the corresponding type based on the urea type. Through automatic recognition and matching, the robot can accurately place different types of urea bags in the designated positions to ensure the standardization and orderliness of palletizing. The randomly stacked urea bags are disordered, and multiple robots will grab the same type of urea bags from the top layer. When placing the urea bags, grid conflicts will occur. In the event of a grid conflict, the time required for the robot to reach the grid is calculated based on the first-come-first-served principle, and the robot with the shortest transportation time is set to the highest priority, so that the robot with the highest priority is placed first.

[0151] For example, when multiple robots query the same target grid, each broadcasts its calculated current required transport time through the communication module. Each robot receives the transport time information of other robots and compares it to find the robot with the shortest transport time. The system sets its priority to the highest and notifies this robot through communication that it can go to the target grid first to place the urea bag. After receiving the priority judgment result, the other robots change their task path and transport the urea bag to an idle grid and place it there. In this way, congestion and conflict among multiple robots at the same target grid are avoided, and the overall palletizing speed is improved. At the same time, the utilization of idle grids also ensures the continuity of the task, and the robots will not be stalled due to conflicts.

[0152] After the robot completes a palletizing task, its priority is reset to zero and it re-participates in the allocation of palletizing tasks. At this time, the robot has two choices: place the urea bag in the idle grid into the corresponding grid or go to the urea bag pile for palletizing. By calculating the benefits of the two choices, the one with higher benefits is selected as the next palletizing task. The calculation of benefits mainly considers the distance factor and the probability of conflict with other robots. The closer to the idle grid or the pile of cluttered urea bags, the higher the benefit. At the same time, if the probability of conflict with other robots after reaching the target position is high, the benefit will be reduced accordingly. Combining these factors, the benefit values ​​of going to the idle grid and going to the pile of cluttered urea bags for palletizing are calculated respectively.

[0153] Specifically, after the robot completes palletizing, it sets its priority to 0 in its control system. The robot then uses the positioning system to obtain its current position coordinates and calculates the reward for reaching an idle grid. The distance from the current position to each idle grid is measured. The shorter the distance, the higher the reward. In this embodiment, the reward is set to 1 / distance. Furthermore, considering that an idle grid is occupied by other robots, the robot evaluates the probability of conflict with other robots upon reaching an idle grid based on the position and task status of each robot in the current system. The higher the probability of conflict, the lower the reward, and the initial reward value is adjusted accordingly. The reward for reaching a pile of cluttered urea bags is calculated by similarly measuring the distance from the current position to the pile and calculating the initial reward. Based on the type and quantity of palletized urea bags, the probability of each type of urea bag appearing on the top layer of the pile is calculated. The reward corresponding to palletizing each type of urea bag is then calculated. The reward is then weighted by the corresponding probability of occurrence to obtain a weighted reward. Ultimately, the first reward (for reaching an idle grid) and the second reward (for reaching a pile of cluttered urea bags) are obtained.

[0154] Furthermore, a numerical comparison is made between the first benefit and the second benefit. If the first benefit is greater than the second benefit, the robot plans a path to go to the idle grid to perform the palletizing task; if the second benefit is greater than the first benefit, the robot plans a path to go to the pile of messy stacked urea bags to perform the palletizing operation. At the same time, the robot sends the selected work area information to other robots through the communication module. The other robots can make corresponding adjustments according to the overall task situation. The method of selecting the work area based on the benefit enables the robot to make the best decision according to the real-time situation, continuously optimize the work process, improve the completion speed and quality of the palletizing task, and reduce the conflict and energy consumption between robots.

[0155] At the same time, after completing a palletizing operation and selecting the next work area, each robot begins to move to the new area and grab and place the urea bags. After completing this series of operations, it returns to the steps of calculating the profit and selecting the work area. This cycle repeats itself. The system monitors the progress of the palletizing task in real time. When all urea bags are successfully stacked in the storage grid, the system determines that the palletizing task is completed and stops the robot's work. By dynamically selecting the work area, the adaptability and self-optimization ability of the robot system can be improved, thereby efficiently completing the palletizing task in a complex and changeable storage environment and improving the stability and reliability of the system.

[0156] Furthermore, the grasping angle and strength of each robot are adjusted according to the posture of the urea bag to obtain a sorting action, including:

[0157] S901. Calculate the initial grasping angle of each robot based on the posture of the urea bag;

[0158] S902. Calculate the initial grasping force of each robot based on the weight and material of the urea bag;

[0159] S903: sorting the urea bags based on the initial gripping angle and the initial gripping force;

[0160] S904: During the sorting process, the initial gripping angle and the initial gripping force are adjusted based on the real-time status of the urea bag to obtain a sorting action.

[0161] In this embodiment, the initial position and coordinate system of the robot grasping device are determined according to the posture of the urea bag. The center point of the urea bag is determined as the grasping point by combining the posture data of the urea bag and the coordinate system of the robot grasping device. The initial grasping angle of the robot is calculated according to the rotation vector in the posture of the urea bag. The grasping force required by the robot is calculated according to the weight and material of the urea bag. The weight and material information of the urea bag are obtained from the preset urea bag information library according to the urea bag type. The mechanical parameters such as the friction coefficient of urea bags of different materials and the gravity to be overcome are analyzed to calculate the initial grasping force of each robot.

[0162] Specifically, after receiving the instructions for the initial grasping angle and initial grasping force, the robot control system drives the grasping device to move to the grasping position of the urea bag according to the set angle, and applies the corresponding grasping force to grasp the urea bag. Then, according to the requirements of the palletizing plan, the robot transports the grasped urea bag to the designated position for placement, completing a sorting operation. During the sorting process, the state of the urea bag will change accordingly. By real-time monitoring of the state of the urea bag, such as position, posture, force, etc., the initial grasping angle and initial grasping force are dynamically adjusted to ensure that the urea bag can always be grasped and carried stably. According to the real-time posture of the urea bag in the stacked state, the robot's real-time grasping action is adjusted so that the urea bag can be accurately placed at the target position.

[0163] Example 2

[0164] In this embodiment, if Figure 4 , providing a robot collaborative sorting and palletizing system for urea bags in a disorderly stacked state, for implementing the robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state, comprising:

[0165] The urea bag recognition module identifies the type and posture of the urea bags based on pre-acquired image data of the urea bags in a disorderly stacked state, and obtains the urea bag recognition result;

[0166] The regional allocation module analyzes the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocates corresponding storage areas for each type of urea bag;

[0167] a palletizing plan generating module, which generates a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence;

[0168] The robot collaboration module dynamically adjusts the working area and sorting action of each robot based on the palletizing solution to achieve multi-robot collaborative sorting and palletizing.

[0169] In this embodiment, the urea bag recognition module improves the accuracy of urea bag type and posture recognition by accurately identifying the type (such as nitrogen fertilizer content, production batch) and posture (flat, tilted, and number of stacked layers) of each bag from a chaotically stacked urea bag. The area allocation module dynamically divides storage areas based on order requirements and real-time inventory, allocating areas near the exit for urea bags that are frequently shipped out, thereby improving the efficiency of urea bag transportation and shipment.

[0170] Specifically, the palletizing plan generation module generates a palletizing plan that takes into account both efficiency and safety based on the type and posture of the urea bag, including the number of stacking layers, stacking rules and handling priorities, to improve palletizing efficiency and reduce the risk of collapse; the robot collaboration module dynamically allocates the robot's working area, coordinates the coordinated actions between multiple robots, and adjusts the robot's grasping parameters in real time according to the changes in the posture of the urea bag, thereby improving the grasping success rate of the urea bag and realizing non-stop sorting and palletizing by the robot.

[0171] Example 3

[0172] A computer-readable storage medium stores a computer program for implementing the above-mentioned robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state.

[0173] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state, characterized in that: include: Based on the pre-acquired image data of the urea bags in a disorderly stacked state, the type and posture of the urea bags are identified to obtain a urea bag recognition result; Analyze the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocate corresponding storage areas for each type of urea bag; Generate a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence; Based on the palletizing solution, the working area and sorting action of each robot are dynamically adjusted to achieve multi-robot collaborative sorting and palletizing; According to the preset order requirements and real-time warehouse inventory, the delivery frequency of different types of urea bags is analyzed, and corresponding storage areas are allocated to each type of urea bags, including: Calculate the delivery frequency of each category of urea bags based on preset order requirements and real-time warehouse inventory; The warehouse is divided into multiple locations using a preset dynamic area division model. Calculate the priority of each location area based on its distance from the outbound location and its distance from other areas; Matching a corresponding location area for each category of urea bags according to the delivery frequency and the priority of each location area; Based on the size and inventory information of each category of urea bags, the preset boundary optimization model is used to optimize the boundaries of each location area to obtain the corresponding storage area; The method of dynamically adjusting the working area and sorting action of each robot based on the palletizing solution includes: According to the palletizing plan, the corresponding working area is allocated to each robot in real time; During the robot's operation, the preset path planning model is used to plan the movement trajectory of the robot's corresponding manipulator arm to obtain the movement path; According to the posture of the urea bag, the gripping angle and strength of each robot are adjusted to obtain the sorting action; Each robot sorts and palletizes the urea bags in the work area according to the sorting action and moving path until the urea bag palletizing task in all assigned work areas is completed, thereby realizing multi-robot collaborative sorting and palletizing; Each robot communicates with each other; and each robot is assigned a corresponding working area in real time according to the palletizing plan, including: According to the palletizing plan and the size of the urea bags corresponding to the storage area, the storage area is divided into multiple grids using a preset grid division model, wherein the multiple grids include storage grids and idle grids, and each grid can only hold one row of urea bags; Each robot grabs the top urea bag from the chaotically stacked urea bags and matches it to the corresponding grid according to the type of urea bag it grabs; According to the robot's movement process, the time required to place the urea bag into the corresponding grid is calculated to obtain the transportation time; When multiple robots are matched to the same grid, the priority of the robot with the shortest transportation time is set to the highest priority. The robot corresponding to the highest priority will first place the grabbed urea bag in the corresponding grid, and the other robots will place the urea bag in the idle grid; After each robot completes a urea bag palletizing process, the corresponding priority is reset to zero. The benefits of palletizing to an idle grid and to a pile of messy stacked urea bags are calculated based on the current position, the position distance, and the probability of robot conflict. For the first benefit of going to an idle grid, the first initial benefit is set to 1 / distance based on the distance from the current position to each idle grid. The situation where the idle grid is occupied by other robots is analyzed. The probability of conflict with other robots when arriving at the idle grid is analyzed based on the position and task status of each robot in the current system. The first initial benefit is corrected according to the conflict probability to obtain the first benefit. For the second benefit of going to a pile of messy urea bags, the second initial benefit is calculated based on the distance from the current position to the urea bag pile. The probability of different types of urea bags appearing on the top layer of the urea bag pile is calculated based on the type and quantity of palletized urea bags. The corresponding benefit of palletizing each type of urea bag is calculated separately, and the benefit and the corresponding probability of appearance are weighted to obtain the second benefit. According to the first benefit and the second benefit, an area corresponding to a higher benefit value is selected as the working area for the next palletizing task; Repeat the process of dynamically selecting the work area until all palletizing tasks are completed.

2. The robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to claim 1 is characterized in that: The method of identifying the type and posture of the urea bags based on the pre-acquired image data of the urea bags in a disorderly stacked state to obtain a urea bag identification result includes: Feature extraction is performed on the pre-acquired image data of the urea bags in a disorderly stacked state to obtain image features; Based on the image features, the position coordinates of each urea bag are identified through a preset urea bag detection model; generating a bounding box for each urea bag according to the position coordinates; Within the boundary box, the type and posture of each urea bag are identified through a preset urea bag posture recognition model to obtain a urea bag recognition result.

3. The robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to claim 2 is characterized in that: Within the bounding box, the type and posture of each urea bag are identified using a preset urea bag posture recognition model, and a urea bag recognition result is obtained, including: Within the bounding box, extract the four corner points of the bounding box as key points; Calculate the rotation vector and translation vector of each key point according to the position coordinates of the key points; Calculating the spatial posture of the urea bag according to the rotation vector and the translation vector through a preset urea bag posture recognition model; Based on the image features within the bounding box, the type of each urea bag is identified through the preset urea bag type recognition model; Combining the type and spatial posture, a urea bag recognition result is obtained.

4. The robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to claim 1 is characterized in that: According to the delivery frequency and the priority of each location area, each category of urea bags is matched with a corresponding location area, including: The location areas are layered according to the priority of each location area from high to low to obtain multi-layer location areas; Based on the urea bag category, corresponding urea bag information is searched in a preset urea bag information database, and the urea bags are layered to obtain multi-layer urea bags, wherein the multi-layer urea bags include easily deformable urea bags, high-density urea bags, and special-sized urea bags; Combined with the delivery frequency, each layer of urea bags is matched to the location area of ​​the corresponding level through the preset location matching model; In each layer of location areas, corresponding location areas are allocated according to the frequency of each type of urea bags in each layer of urea bags leaving the warehouse.

5. The robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to claim 1 is characterized in that: A palletizing plan is generated based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence, including: According to the layer type corresponding to each urea bag and the preset stacking rules, the corresponding initial stacking layer number is generated; The initial number of stacking layers is optimized based on the posture of the urea bags and the size of the storage area to obtain an optimized number of stacking layers; According to the position coordinates and posture of each urea bag, the urea bag closest to the current storage area and with the flattest posture is selected for each palletizing to obtain the palletizing order; The optimized number of stacking layers and stacking sequence are combined to obtain a stacking solution.

6. The robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to claim 1 is characterized in that: According to the posture of the urea bag, the gripping angle and strength of each robot are adjusted to obtain the sorting action, including: Calculate the initial grasping angle of each robot based on the posture of the urea bag; Calculate the initial grasping force of each robot based on the weight and material of the urea bag; Sorting the urea bags based on the initial gripping angle and the initial gripping force; During the sorting process, the initial gripping angle and initial gripping force are adjusted according to the real-time status of the urea bag to obtain the sorting action.

7. A robot-assisted sorting and palletizing system for urea bags in a disorderly stacked state, for implementing the robot-assisted sorting and palletizing method for urea bags in a disorderly stacked state according to any one of claims 1 to 6, characterized in that: include: The urea bag recognition module identifies the type and posture of the urea bags based on pre-acquired image data of the urea bags in a disorderly stacked state, and obtains the urea bag recognition result; The regional allocation module analyzes the delivery frequency of different types of urea bags based on preset order requirements and real-time warehouse inventory, and allocates corresponding storage areas for each type of urea bag; a palletizing plan generating module, which generates a palletizing plan based on the urea bag identification result and the corresponding storage area, wherein the palletizing plan includes the number of palletizing layers and the palletizing sequence; The robot collaboration module dynamically adjusts the working area and sorting action of each robot based on the palletizing solution to achieve multi-robot collaborative sorting and palletizing.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is used to implement the robot collaborative sorting and palletizing method for urea bags in a disorderly stacked state as described in any one of claims 1 to 6.

Images