Dynamic Storage and Transportation Control Method and Device for Intelligent Connected Handling Vehicle

Abstract

The present application discloses a dynamic storage and transportation control method and device for an intelligent networked handling vehicle. The method includes: obtaining environmental data collected by multiple types of sensors pre-set in a three-dimensional warehouse; generating real-time status information of the three-dimensional warehouse based on the environmental data; in the case where the real-time status information indicates that the storage layout of the three-dimensional warehouse has changed, dynamically adjusting the queue position of each handling task in the initial handling task queue according to the real-time status information to obtain a target handling task queue; the queue positions are adjusted in the order of the priorities of each handling task, and the order of the priorities of each handling task is determined according to the storage difficulty of each cargo in each handling task; controlling the intelligent networked handling vehicle to perform storage and transportation in sequence according to the queue order in the target handling task queue. Therefore, by adopting the embodiment of the present application, the handling tasks can be flexibly adjusted according to the real-time change of the storage layout, thereby improving the handling efficiency.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to a dynamic warehousing transportation control method and device for an intelligent connected handling vehicle. Background Art

[0002] In the warehouse of a large machinery manufacturing enterprise, storage and handling are key links in warehousing management. With the development of automation technology, automated storage and retrieval systems (AS / RS) suitable for warehousing have gradually been applied. Such a system achieves a high storage density and handling efficiency through multi-layer shelves and automated equipment.

[0003] In related technologies, AS / RS uses a preset task allocation logic to control the operation of handling equipment. For example, a handling vehicle will carry goods according to a task list. However, the existing control logic is based on fixed task allocation rules and cannot flexibly adjust handling tasks according to real-time changes in the warehousing layout (such as goods scattering or collapsing). When goods scatter or collapse, the goods of a certain handling task cannot be carried, and at this time, the intelligent connected handling vehicle will stop and wait for manual handling of the goods, thus reducing the handling efficiency. Summary of the Invention

[0004] Embodiments of this application provide a dynamic warehousing transportation control method and device for an intelligent connected handling vehicle. To provide a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary part is not a general review, nor is it intended to identify key / important constituent elements or delineate the protection scope of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the subsequent detailed description.

[0005] In a first aspect, embodiments of this application provide a dynamic warehousing transportation control method for an intelligent connected handling vehicle. The method includes:

[0006] Obtain environmental data collected by multiple types of sensors pre-set in a stereoscopic warehouse at a preset period;

[0007] Generate real-time status information of the stereoscopic warehouse based on the environmental data;

[0008] When the real-time status information indicates that the warehousing layout of the stereoscopic warehouse has changed, dynamically adjust the queue position of each handling task in the initial handling task queue according to the real-time status information to obtain a target handling task queue; where

[0009] The queue positions are adjusted in the order of the priorities of each handling task. The order of the priorities of each handling task is determined according to the warehousing difficulty of each cargo in each handling task. The warehousing difficulty of each cargo is obtained by analyzing the three-dimensional map of the cargo distribution space through a pre-trained warehousing difficulty measurement model. The three-dimensional map of the cargo distribution space is constructed based on the real-time status information;

[0010] Control the intelligent connected handling vehicle to perform warehousing transportation in sequence according to the queue order in the target handling task queue.

[0011] In a second aspect, an embodiment of the present application provides a dynamic warehousing transportation control device for an intelligent connected handling vehicle. The device includes:

[0012] An environmental data acquisition module, configured to acquire environmental data collected by multiple types of sensors pre-set in the stereoscopic warehouse according to a preset period;

[0013] A status information generation module, configured to generate real-time status information of the stereoscopic warehouse through the environmental data;

[0014] A queue position adjustment module, configured to, when the real-time status information indicates that the warehousing layout of the stereoscopic warehouse has changed, dynamically adjust the queue positions of each handling task in the initial handling task queue according to the real-time status information to obtain a target handling task queue; wherein,

[0015] The queue positions are adjusted in the order of the priorities of each handling task. The order of the priorities of each handling task is determined according to the warehousing difficulty of each cargo in each handling task. The warehousing difficulty of each cargo is obtained by analyzing the three-dimensional map of the cargo distribution space through a pre-trained warehousing difficulty measurement model. The three-dimensional map of the cargo distribution space is constructed based on the real-time status information;

[0016] A warehousing transportation control module, configured to control the intelligent connected handling vehicle to perform warehousing transportation in sequence according to the queue order in the target handling task queue.

[0017] The technical solution provided by the embodiment of the present application may include the following beneficial effects:

[0018] In an embodiment of the present application, on the one hand, by deploying multiple types of sensors in a three-dimensional warehouse to collect environmental data in real time, the system can generate real-time status information of the warehouse using the data collected by these sensors. When the real-time status information indicates a change in the warehouse layout, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional map of the goods distribution space by a pre-trained warehouse difficulty measurement model, ensuring that the handling vehicle always gives priority to handling the goods that are easiest to handle, enabling flexible adjustment of handling tasks according to real-time changes in the warehouse layout, thereby significantly improving the handling efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained warehouse difficulty measurement model, the system can evaluate the warehouse difficulty of each piece of goods and dynamically adjust the priority of handling tasks accordingly. This can not only ensure that the goods are handled efficiently but also ensure that the shelf space is reasonably utilized, avoiding space waste.

[0019] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application and used together with the specification to explain the principles of the present application.

[0021] Figure 1 is a schematic flowchart of a method for dynamic warehouse transportation control of an intelligent connected handling vehicle provided by an embodiment of the present application;

[0022] Figure 2 is a model structure diagram of a pre-trained warehouse difficulty measurement model provided by an embodiment of the present application;

[0023] Figure 3 is a model processing flowchart of a pre-trained warehouse difficulty measurement model provided by an embodiment of the present application;

[0024] Figure 4 is an intelligent connected handling vehicle provided by an embodiment of the present application;

[0025] Figure 5 is a schematic diagram of an index result provided by an embodiment of the present application;

[0026] Figure 6 is a schematic flowchart of a model training method for a warehouse difficulty measurement model provided by an embodiment of the present application;

[0027] Figure 7 is a schematic structural diagram of a dynamic warehouse transportation control device of an intelligent connected handling vehicle provided by an embodiment of the present application;

[0028] Figure 8It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0029] The following description and the accompanying drawings fully disclose specific implementation manners of the present application, enabling those skilled in the art to practice them.

[0030] It should be clear that the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0031] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0032] In the description of the present application, it should be understood that terms such as "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances. In addition, in the description of the present application, unless otherwise stated, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0033] Currently, for an automated stereoscopic warehouse applicable to warehousing, a preset task allocation logic is adopted to control the operation of handling equipment. For example, a forklift will carry goods according to a task list.

[0034] The inventor realized that the existing control logic is based on fixed task allocation rules and cannot flexibly adjust handling tasks according to real-time changes in the warehousing layout (such as goods scattering or collapsing). When goods scatter or collapse, it will cause the goods of a certain handling task to be unable to be carried. At this time, the intelligent connected forklift will stop and wait for manual handling of the goods, thereby reducing the handling efficiency.

[0035] To solve the above problems, the present application provides a dynamic storage and transportation control method and device for an intelligent connected transport vehicle to solve the problems existing in the above related technical problems. In an embodiment of the present application, on the one hand, by deploying multiple types of sensors in a stereoscopic warehouse to collect environmental data in real time, using the data collected by these sensors, the system can generate real-time status information of the warehouse. When the real-time status information indicates a change in the storage layout, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional map of the cargo distribution space by a pre-trained storage difficulty measurement model to ensure that the transport vehicle always gives priority to handling the easiest-to-handle goods, enabling flexible adjustment of handling tasks according to real-time changes in the storage layout, thus significantly improving the handling efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained storage difficulty measurement model, the system can evaluate the storage difficulty of each cargo and dynamically adjust the priority of the handling task accordingly. This can not only ensure that the goods are efficiently transported, but also ensure that the shelf space is reasonably utilized to avoid space waste. The following will be described in detail with exemplary embodiments.

[0036] The following will combine the attached Figure 1 - attached Figure 6 to introduce in detail the dynamic storage and transportation control method of the intelligent connected transport vehicle provided by the embodiment of the present application. This method can be implemented depending on a computer program and can run on a dynamic storage and transportation control device of an intelligent connected transport vehicle based on the von Neumann architecture. This computer program can be integrated into an application or run as an independent tool-type application.

[0037] Please refer to Figure 1 for a schematic flowchart of a dynamic storage and transportation control method of an intelligent connected transport vehicle provided by an embodiment of the present application. As Figure 1 shown, the method of the embodiment of the present application includes the following steps:

[0038] S101, obtain environmental data collected by multiple types of sensors pre-set in a stereoscopic warehouse according to a preset period;

[0039] Among them, the preset period is a fixed time interval set by the system. Within this time interval, the central processing unit will obtain environmental data once for analysis. The preset period can be every minute, every hour, or any time set according to actual needs. The multiple types of sensors are different types of sensors arranged in the stereoscopic warehouse, including cameras, lidar, etc., for monitoring different environmental parameters in the stereoscopic warehouse. A stereoscopic warehouse is a warehouse system that stores goods using multi-layer shelves.

[0040] In some embodiments of the present application, multiple types of sensors are installed in the automated storage and retrieval system (AS / RS). For example, cameras and lidar are installed on the aisles to monitor the occupancy of the aisles, and cameras and lidar are used to obtain the three-dimensional position information of the goods. The sensors automatically collect data for data storage. When the time duration between the current moment and the previous analysis moment meets the preset period, the dynamic warehousing transportation control strategy of the intelligent connected forklift is triggered. At this time, it is necessary to obtain the environmental data collected by multiple types of sensors pre-set in the automated storage and retrieval system (AS / RS).

[0041] Among them, the environmental data includes the warehouse images captured by the cameras and the lidar data.

[0042] S102, generate the real-time status information of the automated storage and retrieval system (AS / RS) through the environmental data;

[0043] In some embodiments of the present application, the specific process of generating the real-time status information of the automated storage and retrieval system (AS / RS) through the environmental data includes: through image recognition technology, analyze the warehouse images captured by the cameras, identify and determine the current position of each piece of goods; utilize the lidar data to analyze the occupancy of the aisles in the automated storage and retrieval system (AS / RS), and judge whether the aisles for handling goods are unobstructed or congested to obtain the aisle information of the automated storage and retrieval system (AS / RS); through the warehouse images and the lidar data, determine the occupied storage positions and the vacant storage positions to obtain the space information of the automated storage and retrieval system (AS / RS); take the current position of each piece of goods, the aisle information of the automated storage and retrieval system (AS / RS), and the space information of the automated storage and retrieval system (AS / RS) as the real-time status information of the automated storage and retrieval system (AS / RS).

[0044] Among them, image recognition technology refers to the use of a computer to process, analyze, and understand images to identify various different patterns of targets. Lidar is a technology that obtains high-precision three-dimensional information of the surrounding environment by emitting laser beams and measuring their reflection times. In the automated storage and retrieval system (AS / RS), lidar can be used to analyze the occupancy of the aisles and judge whether the aisles are unobstructed or congested. The space information of the automated storage and retrieval system (AS / RS) includes the occupied storage positions and the vacant storage positions. These information can be determined by analyzing the warehouse images captured by the cameras and the lidar data.

[0045] For example, deploy cameras and lidar sensors at key positions in the automated storage and retrieval system (AS / RS), such as on the shelves, aisles, and entrances / exits, etc., to capture images and 3D data in the warehouse in real time. Use image recognition technology to process the images captured by the cameras, extract the features of the goods in the images through a computer vision system, and identify the current positions of each good. Collect the data from the lidar sensors, analyze the 3D information of the aisles, and determine the occupancy of the aisles, so as to determine whether the aisles are clear or congested. Combine the data from the cameras and lidar sensors, analyze the occupancy of the shelves, and determine which storage locations are occupied and which are vacant. Integrate the information collected and analyzed above to generate the real-time status information of the AS / RS, including the current positions of each good, the aisle information, and the space information of the AS / RS.

[0046] S103. In the case where the real-time status information indicates that the storage layout of the AS / RS has changed, according to the real-time status information, dynamically adjust the queue positions of each handling task in the initial handling task queue to obtain a target handling task queue.

[0047] Among them, the queue positions are adjusted in the order of the priorities of each handling task. The order of the priorities of each handling task is determined according to the storage difficulty of each good in each handling task. The storage difficulty of each good is obtained by analyzing the 3D map of the goods distribution space through a pre-trained storage difficulty measurement model. The 3D map of the goods distribution space is constructed according to the real-time status information.

[0048] Among them, the storage layout of the AS / RS refers to the physical layout of the goods stored in the AS / RS, such as the distribution of the goods on the shelves. The initial handling task queue refers to the list of handling tasks generated according to a pre-designed plan before considering the real-time status information. The target handling task queue refers to the new list of handling tasks with the optimized task execution order after being adjusted according to the real-time status information. The queue position refers to the arrangement order of each handling task in the handling task queue. The pre-trained storage difficulty measurement model refers to the model trained with historical data before the system is put into use, which is used to evaluate the storage difficulty of the goods. The 3D map of the goods distribution space is a 3D image of the goods distribution in the AS / RS constructed according to the real-time status information, which is used to intuitively display the positions and distributions of the goods in the warehouse.

[0049] In some embodiments of the present application, in the case where the current position of any one good is inconsistent with its historical position, it is determined that the storage layout of the AS / RS has changed; or, in the case where the current position of any one good is consistent with its historical position, it is determined that the storage layout of the AS / RS has not changed.

[0050] In a possible implementation manner, when the real-time status information indicates that the warehousing layout of the automated storage and retrieval system (AS / RS) has not changed, the intelligent connected forklift is controlled to perform warehousing transportation in sequence according to the queue order in the initial handling task queue.

[0051] In another possible implementation manner, when the real-time status information indicates that the warehousing layout of the AS / RS has changed, according to the real-time status information, the queue positions of each handling task in the initial handling task queue are dynamically adjusted to obtain a target handling task queue.

[0052] Among them, each handling task includes the historical location of each piece of goods.

[0053] Specifically, the specific process of dynamically adjusting the queue positions of each handling task in the initial handling task queue according to the real-time status information includes: when the current location of any piece of goods is inconsistent with the historical location, according to the aisle information and space information of the AS / RS, a three-dimensional space state diagram of each storage location in the AS / RS is constructed. The three-dimensional space state diagram is used to represent the idle state of each storage location in the AS / RS and the aisle congestion state of the aisles connected to each storage location; according to the current location of each piece of goods, each piece of goods is marked in the three-dimensional space state diagram to obtain a three-dimensional space diagram of the goods distribution of multiple pieces of goods; the three-dimensional space diagram of the goods distribution is input into a pre-trained warehousing difficulty measurement model to analyze the three-dimensional space diagram of the goods distribution through the pre-trained warehousing difficulty measurement model; the warehousing difficulty corresponding to each piece of goods is output; according to the magnitude of the warehousing difficulty corresponding to each piece of goods, the priority of each handling task to which each piece of goods belongs is determined to obtain the priority of each handling task; based on the high-low order of the priority of each handling task, each handling task in the initial handling task queue is sorted to adjust the queue positions of each handling task in the initial handling task queue.

[0054] In the embodiments of the present application, by deploying multiple types of sensors in the AS / RS to collect environmental data in real time and using the data collected by these sensors, the system can generate the real-time status information of the warehouse. When the real-time status information indicates that the warehousing layout has changed, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional space diagram of the goods distribution by the pre-trained warehousing difficulty measurement model, ensuring that the forklift always gives priority to handling the goods that are easiest to handle, and can realize flexible adjustment of the handling tasks according to the real-time change of the warehousing layout, thus significantly improving the handling efficiency.

[0055] For example Figure 2 As shown, the pre-trained warehousing difficulty measurement model includes a convolutional layer, a feature fusion layer, a fully connected layer, an activation function, and an output layer.

[0056] Specifically, the specific process of analyzing the three-dimensional map of the goods distribution space through a pre-trained warehousing difficulty measurement model includes: the convolutional layer performs grid division and clustering analysis on the three-dimensional map of the goods distribution space through a spatial analysis algorithm to calculate the position characteristics of each good in the stereoscopic warehouse and obtain the goods distribution characteristics of each good; the convolutional layer performs pattern recognition on the three-dimensional map of the goods distribution space to analyze the geometric characteristics and congestion characteristics of the channels connected to each storage position within the preset area of each good as the channel characteristics of each good; the convolutional layer performs image recognition on the three-dimensional map of the goods distribution space to obtain the free area characteristics of the storage positions within the preset area of each good and obtain the shelf characteristics of each good; the feature fusion layer fuses the goods distribution characteristics, channel characteristics, and shelf characteristics of each good to obtain the comprehensive feature vector of each good; the fully connected layer maps the comprehensive feature vector of each good to map the comprehensive feature vector to the prediction space of the warehousing difficulty; in the prediction space, the activation function captures the functional relationship between the comprehensive feature vector of each good and the warehousing difficulty; the output layer fits the warehousing difficulty of each good through the functional relationship.

[0057] In the embodiment of the present application, through this deep learning method, the warehousing difficulty of each good can be automatically evaluated, so as to dynamically adjust the priority of the handling tasks, optimize the warehousing layout, improve the handling efficiency, and reduce the risk of congestion and goods damage.

[0058] For example Figure 3 as shown Figure 3 is the model processing process diagram of the pre-trained warehousing difficulty measurement model. The three-dimensional map of the goods distribution space is the input of the model. The position information of each good is extracted from the three-dimensional map to form a feature matrix. The geometric characteristics and congestion conditions of the channels around each good are analyzed to extract the channel characteristics. The free area characteristics of the shelf where each good is located are obtained to extract the shelf characteristics. The above-extracted goods distribution characteristics, channel characteristics, and shelf characteristics are fused to form a comprehensive feature vector. This vector contains multi-faceted information of each good and is used for subsequent warehousing difficulty evaluation. The fused comprehensive feature vector is mapped to the prediction space of the warehousing difficulty to convert the high-dimensional feature vector into the predicted value of the warehousing difficulty. Finally, the warehousing difficulty characteristics of each good are output, and these characteristic values reflect the difficulty of handling the goods.

[0059] S104. Control the intelligent connected handling vehicle to perform warehousing transportation in sequence according to the queue order in the target handling task queue.

[0060] Among them, the intelligent connected handling vehicle is an automated handling device, usually equipped with sensors, a control system, and network connection functions, and can autonomously or semi-autonomously perform handling tasks in environments such as warehouses or distribution centers. For example Figure 4As shown in the figure, there is an acceleration sensor 4, a battery 1, a load sensor 2 and a weight sensor 3 on the intelligent networked transport vehicle. The target transport task queue is a transport task list dynamically adjusted according to the warehousing difficulty, which determines the order of tasks executed by the intelligent networked transport vehicle. The queue order refers to the arrangement order of tasks in the target transport task queue. Warehousing transportation refers to the process of transporting goods inside a warehouse or from the warehouse to a designated location.

[0061] In some embodiments of the present application, the intelligent networked transport vehicle receives instructions from the central control system through a wireless network. The central control system sends transport instructions to the transport vehicle according to the queue order in the target transport task queue. The intelligent networked transport vehicle arrives at the designated cargo position according to the instructions. The intelligent networked transport vehicle performs a transport action to move the goods from the current position to the target position.

[0062] Furthermore, the central control system monitors the operating status and task execution of the transport vehicle in real time. If an unexpected situation occurs (such as a change in the cargo position, channel congestion, etc.), the central control system will dynamically adjust the target transport task queue according to the real-time status information and reschedule the transport vehicle. After the transport vehicle completes the task, it sends a confirmation message of task completion to the central control system. The central control system updates the task queue status and assigns the next task to the transport vehicle.

[0063] In the embodiments of the present application, through steps S101 - S104, the indicators related to the transport of the electric transport vehicle have exceeded 90%, as Figure 5 shown.

[0064] In the embodiments of the present application, on the one hand, by deploying multiple types of sensors in the stereoscopic warehouse to collect environmental data in real time, using the data collected by these sensors, the system can generate real-time status information of the warehouse. When the real-time status information indicates that the warehousing layout has changed, the system will dynamically adjust the priority of each task in the transport task queue. This adjustment is based on the analysis of the three-dimensional map of the cargo distribution space by a pre-trained warehousing difficulty measurement model, ensuring that the transport vehicle always gives priority to handling the easiest-to-transport goods, enabling flexible adjustment of transport tasks according to real-time warehousing layout changes, thereby significantly improving the transport efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained warehousing difficulty measurement model, the system can evaluate the warehousing difficulty of each cargo and dynamically adjust the priority of transport tasks accordingly. This can not only ensure the efficient transport of goods but also ensure the reasonable use of shelf space and avoid space waste.

[0065] Please refer to Figure 6 , which is a schematic flowchart of a model training method for a gravity distribution uniformity recognition model provided by the embodiments of the present application. As Figure 6 shown, the method of the embodiments of the present application may include the following steps:

[0066] S201, collect historical environmental data collected by multiple types of sensors pre-set in the three-dimensional warehouse;

[0067] Among them, the historical environmental data refers to the warehouse environmental data collected and stored over a period of time, and these data reflect the state of the warehouse at different time points.

[0068] S202, construct three-dimensional maps of the distribution spaces of historical goods at different historical moments according to the historical environmental data;

[0069] In some embodiments of the present application, the specific process of constructing three-dimensional maps of the distribution spaces of historical goods at different historical moments according to the historical environmental data includes: constructing historical state information of the three-dimensional warehouse at each historical moment according to the historical environmental data, where the historical state information includes the first position of each historical good, the historical passage information and historical space information of the three-dimensional warehouse; constructing a three-dimensional state map of the historical space of each storage location in the three-dimensional warehouse according to the historical passage information and historical space information of the three-dimensional warehouse; and marking each historical good on the three-dimensional state map of the historical space according to the first position of each historical good to obtain three-dimensional maps of the distribution spaces of historical goods at different historical moments.

[0070] Specifically, the specific process of constructing a three-dimensional state map of the historical space of each storage location in the three-dimensional warehouse according to the historical passage information and historical space information of the three-dimensional warehouse includes: performing time dimension alignment on the historical passage information and historical space information according to the historical timestamp to obtain an aligned information group; creating a basic three-dimensional model framework according to the design drawings and specification parameters of the three-dimensional warehouse; dividing the data of the aligned information group into multiple time slices in chronological order; mapping the data of each time slice into the three-dimensional model framework to update the default states of each storage location and passage in the three-dimensional model framework; and using computer graphics technology to perform rendering integration on the updated states of each storage location and passage in the three-dimensional model framework to obtain a three-dimensional state map of the historical space of each storage location in the three-dimensional warehouse.

[0071] In a possible implementation, based on the collected historical environmental data, the historical state information of the stereoscopic warehouse at each historical moment is constructed. This information includes: the first position of each historical cargo, the historical passage information of the stereoscopic warehouse, and the historical space information, such as the free and occupied states of the shelves. According to the design drawings and specification parameters of the stereoscopic warehouse, a basic three-dimensional model framework is created. This framework will serve as the benchmark for subsequent data mapping. Align the historical passage information and historical space information in the time dimension according to the historical timestamps to form an aligned information group. This ensures that data at different time points can be correctly associated and compared. Divide the aligned information group into data for multiple time slices in chronological order. Map the data for each time slice into the three-dimensional model framework and update the default states of each bin and passage in the model. Using computer graphics technology, render and integrate the updated states of each bin and passage in the three-dimensional model framework. This step involves the process of converting a three-dimensional scene or model into a two-dimensional image, which involves the coordinated work of multiple steps and components to visualize the three-dimensional data. After the rendering is completed, the historical space three-dimensional state diagrams of each bin in the stereoscopic warehouse are obtained. These diagrams show the warehouse layout and cargo distribution at different historical moments.

[0072] S203. Construct storage difficulty labels for different historical cargos in each three-dimensional diagram of the historical cargo distribution space;

[0073] In some embodiments of the present application, the specific process of constructing storage difficulty labels for different historical cargos in each three-dimensional diagram of the historical cargo distribution space includes: determining the historical cargo distribution characteristics, historical passage characteristics, and historical shelf characteristics of different historical cargos through each three-dimensional diagram of the historical cargo distribution space; analyzing the distribution uniformity of different historical cargos using the historical cargo distribution characteristics; analyzing the smooth passage probability of goods for different historical cargos using the historical passage characteristics; analyzing the size of the surrounding free area of different historical cargos using the historical shelf characteristics; quantifying the distribution uniformity, smooth passage probability of goods, and size of the surrounding free area of different historical cargos into a unified dimension and performing weighted summation as the storage difficulty labels for different historical cargos in each three-dimensional diagram of the historical cargo distribution space. This method has high accuracy and can improve the training accuracy of the model.

[0074] Specifically, the calculation formula for the distribution uniformity is:

[0075]

[0076] where is the number of historical cargos, is the number of moments at each historical moment, is the th historical cargo at the th historical moment, and the historical position characterized by the historical cargo distribution characteristics. is the average value of all positions, is the standard deviation of the positions.

[0077] Specifically, the calculation formula for the smooth passage probability of goods is:

[0078]

[0079] where, is the number of moments at each historical moment, is the th geometric fitness characterized by the historical channel features within the th historical moment (calculated based on the width, length, and curvature features in the channel features), is the congestion impact degree characterized by the historical channel features within the

[0080] Specifically, the calculation formula for the size of the surrounding idle area is:

[0081]

[0082] where, represents the area of the th storage position characterized by the historical shelf features, represents the idle state of the th storage position characterized by the historical shelf features, 0 indicates occupied, and 1 indicates idle. represents the distance weight between the th storage position characterized by the historical shelf features and the target goods. The closer the distance, the higher the weight.

[0083] In some other embodiments of the present application, during the process of constructing the storage difficulty labels of different historical goods in each three-dimensional map of the historical goods distribution space, the storage difficulty labels of different historical goods in each three-dimensional map of the historical goods distribution space input by the user for the client are received. The storage difficulty labels of different historical goods in each three-dimensional map of the historical goods distribution space are for empirical allocation. For example, in a handling task, if the goods are scattered, the storage difficulty of the goods is assigned the minimum value at this time. This method is fast and can improve the construction speed of the model training samples.

[0084] S204, label the storage difficulty labels of different historical goods at the positions of different historical goods in each three-dimensional map of the historical goods distribution space to obtain model training samples;

[0085] S205, create a storage difficulty measurement model using a neural network;

[0086] In some embodiments of the present application, a neural network architecture suitable for processing structured data, such as a multi-layer perceptron (MLP) or a convolutional neural network (CNN), can be adopted. Input layer: The input dimension is the dimension of the feature vector, such as the combination of cargo distribution features, aisle features, and shelf features. Hidden layer: It contains multiple neurons, and activation functions such as ReLU are used to introduce non-linearity. Output layer: The output dimension is 1, representing the predicted value of the warehousing difficulty.

[0087] S206, input the model training samples into the warehousing difficulty measurement model, and output the loss value;

[0088] S207, when the loss value reaches the minimum, generate a pre-trained warehousing difficulty measurement model; or, when the loss value does not reach the minimum, continue to execute the step of inputting the model training samples into the warehousing difficulty measurement model until the loss value reaches the minimum.

[0089] In the embodiments of the present application, on the one hand, by deploying multiple types of sensors in the stereoscopic warehouse to collect environmental data in real time, using the data collected by these sensors, the system can generate the real-time status information of the warehouse. When the real-time status information indicates that the warehousing layout has changed, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional map of the cargo distribution space by a pre-trained warehousing difficulty measurement model, ensuring that the handling vehicle always gives priority to handling the easiest-to-handle goods, and can realize flexible adjustment of the handling tasks according to the real-time changes in the warehousing layout, thus significantly improving the handling efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained warehousing difficulty measurement model, the system can evaluate the warehousing difficulty of each cargo and dynamically adjust the priority of the handling tasks accordingly. This can not only ensure that the goods are handled efficiently, but also ensure that the shelf space is reasonably utilized and space waste is avoided.

[0090] The following is an embodiment of the device of the present application, which can be used to execute the method embodiment of the present application. For the details not disclosed in the embodiment of the device of the present application, please refer to the method embodiment of the present application.

[0091] Please refer to Figure 7 , which shows a schematic structural diagram of a dynamic warehousing transportation control device for an intelligent networked handling vehicle provided by an exemplary embodiment of the present application. The dynamic warehousing transportation control device of the intelligent networked handling vehicle can be implemented as all or part of an electronic device through software, hardware, or a combination of both. The device 1 includes an environmental data acquisition module 10, a status information generation module 20, a queue position adjustment module 30, and a warehousing transportation control module 40.

[0092] The environmental data acquisition module 10 is used to acquire the environmental data collected by multiple types of sensors pre-set in the stereoscopic warehouse according to a preset period;

[0093] A status information generation module 20 for generating real-time status information of the stereoscopic warehouse based on environmental data;

[0094] A queue position adjustment module 30 for dynamically adjusting the queue position of each handling task in the initial handling task queue according to the real-time status information when the real-time status information indicates a change in the storage layout of the stereoscopic warehouse, so as to obtain a target handling task queue; wherein,

[0095] The queue position is adjusted in the order of the priority of each handling task, the order of the priority of each handling task is determined according to the storage difficulty of each cargo in each handling task, the storage difficulty of each cargo is obtained by analyzing the three-dimensional map of the cargo distribution space through a pre-trained storage difficulty measurement model, and the three-dimensional map of the cargo distribution space is constructed according to the real-time status information;

[0096] A warehousing transportation control module 40 for controlling the intelligent networked handling vehicle to perform warehousing transportation in sequence according to the queue order in the target handling task queue.

[0097] It should be noted that when the dynamic warehousing transportation control device of the intelligent networked handling vehicle provided in the above embodiment executes the dynamic warehousing transportation control method of the intelligent networked handling vehicle, only the above-mentioned division of each functional module is used for illustration. In actual application, the above functions can be assigned to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the dynamic warehousing transportation control device of the intelligent networked handling vehicle provided in the above embodiment and the embodiment of the dynamic warehousing transportation control method of the intelligent networked handling vehicle belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be elaborated here.

[0098] The serial numbers of the embodiments of the present application above are only for description and do not represent the advantages and disadvantages of the embodiments.

[0099] In an embodiment of the present application, on the one hand, by deploying multiple types of sensors in the three-dimensional warehouse to collect environmental data in real time, the system can generate real-time status information of the warehouse using the data collected by these sensors. When the real-time status information indicates a change in the warehouse layout, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional map of the goods distribution space by a pre-trained warehouse difficulty measurement model, ensuring that the handling vehicle always gives priority to handling the goods that are easiest to handle, enabling flexible adjustment of the handling tasks according to real-time changes in the warehouse layout, thus significantly improving the handling efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained warehouse difficulty measurement model, the system can evaluate the warehouse difficulty of each piece of goods and dynamically adjust the priority of the handling tasks accordingly. This can not only ensure that the goods are handled efficiently but also ensure that the shelf space is reasonably utilized, avoiding space waste.

[0100] The present application also provides a computer-readable medium, on which program instructions are stored. When the program instructions are executed by a processor, they implement the dynamic warehouse transportation control method of the intelligent connected handling vehicle provided in each of the above method embodiments.

[0101] The present application also provides a computer program product containing instructions. When it runs on a computer, it causes the computer to execute the dynamic warehouse transportation control method of the intelligent connected handling vehicle in each of the above method embodiments.

[0102] Please refer to Figure 8 , which is a schematic structural diagram of an electronic device provided in an embodiment of the present application. As Figure 8 shown, the electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0103] Among them, the communication bus 1002 is used to realize the connection and communication between these components.

[0104] Among them, the user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface.

[0105] Among them, the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0106] Among them, the processor 1001 may include one or more processing cores. The processor 1001 connects various parts within the entire electronic device 1000 through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by invoking data stored in the memory 1005, it performs various functions of the electronic device 1000 and processes data. Optionally, the processor 1001 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 1001 may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001 and may be implemented separately by a single chip.

[0107] Among them, the memory 1005 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments. Optionally, the memory 1005 may also be at least one storage system located far from the aforementioned processor 1001. As Figure 8 shown, the memory 1005, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a dynamic warehousing and transportation control application program for an intelligent networked transport vehicle.

[0108] In Figure 8In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an interface for the user to input and obtain the data input by the user; and the processor 1001 can be used to call the dynamic storage and transportation control application program of the intelligent connected transport vehicle stored in the memory 1005, and specifically perform the following operations:

[0109] Obtain the environmental data collected by multiple types of sensors preset in the stereoscopic warehouse according to a preset period;

[0110] Generate the real-time status information of the stereoscopic warehouse through the environmental data;

[0111] In the case where the real-time status information indicates that the storage layout of the stereoscopic warehouse has changed, according to the real-time status information, dynamically adjust the queue position of each handling task in the initial handling task queue to obtain the target handling task queue; wherein,

[0112] The queue positions are adjusted in the order of the priorities of each handling task, and the order of the priorities of each handling task is determined according to the storage difficulty of each cargo in each handling task. The storage difficulty of each cargo is obtained by analyzing the three-dimensional map of the cargo distribution space through a pre-trained storage difficulty measurement model, and the three-dimensional map of the cargo distribution space is constructed according to the real-time status information;

[0113] Control the intelligent connected transport vehicle to perform storage and transportation in sequence according to the queue order in the target handling task queue.

[0114] In one embodiment, when the processor 1001 executes dynamically adjusting the queue position of each handling task in the initial handling task queue according to the real-time status information, it specifically performs the following operations:

[0115] In the case where the current position of any cargo is inconsistent with the historical position, according to the channel information and space information of the stereoscopic warehouse, construct a three-dimensional spatial status map of each bin in the stereoscopic warehouse, and the three-dimensional spatial status map is used to characterize the idle state of each bin in the stereoscopic warehouse and the channel congestion state of the channels connected to each bin;

[0116] Mark each cargo on the three-dimensional spatial status map according to the current position of each cargo to obtain a three-dimensional map of the cargo distribution space of multiple cargos;

[0117] Input the three-dimensional map of the cargo distribution space into a pre-trained storage difficulty measurement model to analyze the three-dimensional map of the cargo distribution space through the pre-trained storage difficulty measurement model;

[0118] Output the storage difficulty corresponding to each cargo;

[0119] Determine the priority of the handling task to which each cargo belongs according to the storage difficulty level corresponding to each cargo, so as to obtain the priority of each handling task;

[0120] Based on the high - low order of the priority of each handling task, sort each handling task in the initial handling task queue to adjust the queue position of each handling task in the initial handling task queue.

[0121] In one embodiment, when the processor 1001 executes the analysis of the three - dimensional map of the cargo distribution space through a pre - trained storage difficulty measurement model, the following operations are specifically performed:

[0122] The convolutional layer performs grid division and clustering analysis on the three - dimensional map of the cargo distribution space through a spatial analysis algorithm to calculate the position characteristics of each cargo in the stereoscopic warehouse, and obtains the cargo distribution characteristics of each cargo;

[0123] The convolutional layer performs pattern recognition on the three - dimensional map of the cargo distribution space to analyze the geometric characteristics and congestion characteristics of the channels connecting each position in the preset area of each cargo, and uses them as the channel characteristics of each cargo;

[0124] The convolutional layer performs image recognition on the three - dimensional map of the cargo distribution space to obtain the free area characteristics of the positions in the preset area of each cargo, and obtains the shelf characteristics of each cargo;

[0125] The feature fusion layer fuses the cargo distribution characteristics, channel characteristics, and shelf characteristics of each cargo to obtain the comprehensive feature vector of each cargo;

[0126] The fully - connected layer maps the comprehensive feature vector of each cargo to map the comprehensive feature vector to the prediction space of the storage difficulty level;

[0127] In the prediction space, the activation function captures the functional relationship between the comprehensive feature vector of each cargo and the storage difficulty level;

[0128] The output layer fits the storage difficulty level of each cargo through the functional relationship.

[0129] In one embodiment, when the processor 1001 executes the generation of the pre - trained storage difficulty measurement model, the following operations are specifically performed:

[0130] Collect the historical environmental data collected by multiple types of sensors pre - set in the stereoscopic warehouse;

[0131] According to the historical environmental data, construct the three - dimensional maps of the historical cargo distribution spaces at different historical moments;

[0132] Construct the storage difficulty level labels of different historical cargos in each three - dimensional map of the historical cargo distribution space;

[0133] Label the storage difficulty tags of different historical goods at the positions of different historical goods in the three-dimensional map of the distribution space of each historical good, obtaining model training samples;

[0134] Create a storage difficulty measurement model using a neural network;

[0135] Input the model training samples into the storage difficulty measurement model and output a loss value;

[0136] When the loss value reaches the minimum, generate a pre-trained storage difficulty measurement model; or, when the loss value does not reach the minimum, continue to execute the step of inputting the model training samples into the storage difficulty measurement model until the loss value reaches the minimum.

[0137] In one embodiment, when the processor 1001 executes to construct the three-dimensional map of the distribution space of each historical good at different historical moments according to the historical environment data, it specifically performs the following operations:

[0138] According to the historical environment data, construct the historical state information of the stereoscopic warehouse at each historical moment, where the historical state information includes the first position of each historical good, the historical channel information and historical space information of the stereoscopic warehouse;

[0139] According to the historical channel information and historical space information of the stereoscopic warehouse, construct the three-dimensional state map of the historical space of each bin in the stereoscopic warehouse;

[0140] Mark each historical good in the three-dimensional state map of the historical space according to the first position of each historical good, obtaining the three-dimensional map of the distribution space of each historical good at different historical moments.

[0141] In one embodiment, when the processor 1001 executes to construct the three-dimensional state map of the historical space of each bin in the stereoscopic warehouse according to the historical channel information and historical space information of the stereoscopic warehouse, it specifically performs the following operations:

[0142] Perform time dimension alignment on the historical channel information and historical space information according to the historical timestamp to obtain an alignment information group;

[0143] Create a basic three-dimensional model framework according to the design drawings and specification parameters of the stereoscopic warehouse;

[0144] Divide the alignment information group into data of multiple time slices in chronological order;

[0145] Map the data of each time slice into the three-dimensional model framework to update the default states of each bin and channel in the three-dimensional model framework;

[0146] Using computer graphics technology, render and integrate the updated status of each storage location and passageway in the three-dimensional model framework to obtain a three-dimensional historical spatial status map of each storage location in the automated storage and retrieval system.

[0147] In one embodiment, when the processor 1001 executes to construct the storage difficulty labels of different historical goods in each three-dimensional historical goods distribution space map, it specifically performs the following operations:

[0148] Through each three-dimensional historical goods distribution space map, determine the historical goods distribution characteristics, historical passageway characteristics, and historical shelf characteristics of different historical goods;

[0149] Adopt the historical goods distribution characteristics to analyze the distribution uniformity of different historical goods;

[0150] Adopt the historical passageway characteristics to analyze the probability of smooth entry and exit of goods for different historical goods;

[0151] Adopt the historical shelf characteristics to analyze the size of the surrounding free area of different historical goods;

[0152] Quantify the distribution uniformity, the probability of smooth entry and exit of goods, and the size of the surrounding free area of different historical goods into a unified dimension for weighted summation, and use it as the storage difficulty label of different historical goods in each three-dimensional historical goods distribution space map.

[0153] In one embodiment, when the processor 1001 executes to generate the real-time status information of the automated storage and retrieval system through environmental data, it specifically performs the following operations:

[0154] Through image recognition technology, analyze the warehouse images captured by the camera, and identify and determine the current position of each good;

[0155] Utilize the lidar data to analyze the occupancy of the passageways in the automated storage and retrieval system, and determine whether the passageways for transporting goods are smooth or congested, so as to obtain the passageway information of the automated storage and retrieval system;

[0156] Through the warehouse images and lidar data, determine the occupied storage locations and free storage locations, so as to obtain the spatial information of the automated storage and retrieval system;

[0157] Take the current position of each good, the passageway information of the automated storage and retrieval system, and the spatial information of the automated storage and retrieval system as the real-time status information of the automated storage and retrieval system.

[0158] In one embodiment, the processor 1001 further performs the following operations:

[0159] In the case where the current position of any one good is inconsistent with its historical position, determine that the storage layout of the automated storage and retrieval system has changed; or, in the case where the current position of any one good is consistent with its historical position, determine that the storage layout of the automated storage and retrieval system has not changed;

[0160] When the real-time status information indicates that the storage layout of the stereoscopic warehouse has not changed, control the intelligent connected forklift to perform warehousing transportation in sequence according to the queue order in the initial handling task queue.

[0161] In the embodiment of the present application, on the one hand, by deploying multiple types of sensors in the stereoscopic warehouse to collect environmental data in real time, the system can generate the real-time status information of the warehouse using the data collected by these sensors. When the real-time status information indicates that the storage layout has changed, the system will dynamically adjust the priority of each task in the handling task queue. This adjustment is based on the analysis of the three-dimensional map of the cargo distribution space by a pre-trained warehousing difficulty measurement model, ensuring that the forklift always gives priority to handling the easiest-to-handle goods, enabling flexible adjustment of the handling tasks according to the real-time changes in the storage layout, thus significantly improving the handling efficiency. On the other hand, by analyzing the three-dimensional map through a pre-trained warehousing difficulty measurement model, the system can evaluate the warehousing difficulty of each cargo and dynamically adjust the priority of the handling tasks accordingly. This can not only ensure that the goods are efficiently handled, but also ensure that the shelf space is reasonably utilized, avoiding space waste.

[0162] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program for controlling the dynamic warehousing transportation of the intelligent connected forklift can be stored in a computer-readable storage medium. When this program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium of the program for controlling the dynamic warehousing transportation of the intelligent connected forklift can be a magnetic disk, an optical disc, a read-only memory, or a random access memory, etc.

[0163] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of the rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A dynamic storage and transportation control method for an intelligent connected handling vehicle, characterized in that, The method includes: Obtaining environmental data collected by multiple types of sensors pre - set in a stereoscopic warehouse according to a preset period; Generating real - time status information of the stereoscopic warehouse based on the environmental data; When the real - time status information indicates that the warehousing layout of the stereoscopic warehouse has changed, dynamically adjusting the queue position of each handling task in the initial handling task queue according to the real - time status information to obtain a target handling task queue; where The queue position is adjusted in the order of the priority of each handling task, the order of the priority of each handling task is determined according to the warehousing difficulty of each cargo in each handling task, the warehousing difficulty of each cargo is obtained by analyzing a three - dimensional map of the cargo distribution space through a pre - trained warehousing difficulty measurement model, and the three - dimensional map of the cargo distribution space is constructed according to the real - time status information; Controlling an intelligent network - connected handling vehicle to perform warehousing transportation in sequence according to the queue order in the target handling task queue.

2. The method according to claim 1, characterized in that The real - time status information includes the current position of each cargo, the channel information and space information of the stereoscopic warehouse; each handling task includes the historical position of each cargo; The dynamically adjusting the queue position of each handling task in the initial handling task queue according to the real - time status information includes: When the current position of any cargo is inconsistent with the historical position, constructing a three - dimensional spatial status map of each bin in the stereoscopic warehouse according to the channel information and space information of the stereoscopic warehouse, and the three - dimensional spatial status map is used to represent the idle status of each bin in the stereoscopic warehouse and the channel congestion status of the channels connected to each bin; Marking each cargo on the three - dimensional spatial status map according to the current position of each cargo to obtain a three - dimensional map of the cargo distribution of multiple cargos; Inputting the three - dimensional map of the cargo distribution into a pre - trained warehousing difficulty measurement model to analyze the three - dimensional map of the cargo distribution through the pre - trained warehousing difficulty measurement model; Outputting the warehousing difficulty corresponding to each cargo; Determining the priority of the handling task to which each cargo belongs according to the magnitude of the warehousing difficulty corresponding to each cargo to obtain the priority of each handling task; Sorting each handling task in the initial handling task queue based on the order of the priority of each handling task to adjust the queue position of each handling task in the initial handling task queue.

3. The method according to claim 2, wherein The pre - trained warehousing difficulty measurement model includes a convolutional layer, a feature fusion layer, a fully - connected layer, an activation function, and an output layer; The analyzing the three - dimensional map of the cargo distribution through the pre - trained warehousing difficulty measurement model includes: The convolutional layer performs grid division and clustering analysis on the three - dimensional map of the cargo distribution through a spatial analysis algorithm to calculate the position characteristics of each cargo in the stereoscopic warehouse and obtain the cargo distribution characteristics of each cargo; The convolutional layer performs pattern recognition on the three-dimensional map of the cargo distribution space to analyze the geometric features and congestion features of the channels connected to each storage location within the preset area of each piece of cargo, as the channel features of each piece of cargo; The convolutional layer performs image recognition on the three-dimensional map of the cargo distribution space to obtain the free area features of the storage locations within the preset area of each piece of cargo, and obtains the shelf features of each piece of cargo; The feature fusion layer fuses the cargo distribution features, channel features, and shelf features of each piece of cargo to obtain the comprehensive feature vector of each piece of cargo; The fully connected layer maps the comprehensive feature vector of each piece of cargo to map the comprehensive feature vector to the prediction space of the warehousing difficulty level; In the prediction space, the activation function captures the functional relationship between the comprehensive feature vector of each piece of cargo and the warehousing difficulty level; The output layer fits the warehousing difficulty level of each piece of cargo through the functional relationship.

4. The method according to claim 1, wherein Generate a pre-trained warehousing difficulty level measurement model according to the following steps, including: Collect historical environmental data collected by multiple types of sensors preset in the automated warehouse; According to the historical environmental data, construct three-dimensional maps of the historical cargo distribution spaces at different historical times; Construct warehousing difficulty level labels for different historical pieces of cargo in each three-dimensional map of the historical cargo distribution space; Mark the warehousing difficulty level labels of the different historical pieces of cargo at the positions of the different historical pieces of cargo in each three-dimensional map of the historical cargo distribution space to obtain model training samples; Create a warehousing difficulty level measurement model using a neural network; Input the model training samples into the warehousing difficulty level measurement model and output a loss value; When the loss value reaches the minimum, generate a pre-trained warehousing difficulty level measurement model; or, when the loss value does not reach the minimum, continue to execute the step of inputting the model training samples into the warehousing difficulty level measurement model until the loss value reaches the minimum.

5. The method according to claim 4, wherein The step of constructing three-dimensional maps of the historical cargo distribution spaces at different historical times according to the historical environmental data includes: According to the historical environmental data, construct historical state information of the automated warehouse at each historical time, where the historical state information includes the first position of each historical piece of cargo, the historical channel information, and the historical space information of the automated warehouse; According to the historical channel information and the historical space information of the automated warehouse, construct a three-dimensional state map of the historical spaces of each storage location in the automated warehouse; Mark each historical piece of cargo on the three-dimensional state map of the historical space according to the first position of each historical piece of cargo to obtain three-dimensional maps of the historical cargo distribution spaces at different historical times.

6. The method according to claim 5, wherein The step of constructing a three-dimensional state map of the historical spaces of each storage location in the automated warehouse according to the historical channel information and the historical space information of the automated warehouse includes: Align the historical channel information and the historical space information in the time dimension according to the historical time stamps to obtain an alignment information group; Create a basic three-dimensional model framework according to the design drawings and specification parameters of the automated warehouse; Dividing the alignment information group into data of multiple time slices in chronological order; Mapping the data of each time slice into the three-dimensional model framework to update the default states of each storage location and channel in the three-dimensional model framework; Using computer graphics technology to render and integrate the updated states of each storage location and channel in the three-dimensional model framework to obtain the historical spatial three-dimensional state diagram of each storage location in the stereoscopic warehouse.

7. The method according to claim 4, characterized in that, Constructing the storage difficulty labels of different historical goods in each historical goods distribution space three-dimensional diagram, including: Determining the historical goods distribution characteristics, historical channel characteristics, and historical shelf characteristics of the different historical goods through the respective historical goods distribution space three-dimensional diagrams; Analyzing the distribution uniformity of the different historical goods by using the historical goods distribution characteristics; Analyzing the smooth passage probability of the goods of the different historical goods by using the historical channel characteristics; Analyzing the size of the surrounding idle area of the different historical goods by using the historical shelf characteristics; Quantifying the distribution uniformity, the smooth passage probability of the goods, and the size of the surrounding idle area of the different historical goods into a unified dimension and performing weighted summation as the storage difficulty labels of the different historical goods in each historical goods distribution space three-dimensional diagram.

8. The method according to claim 1, wherein The environmental data includes the warehouse images captured by the camera and the lidar data; Generating the real-time state information of the stereoscopic warehouse through the environmental data, including: Analyzing the warehouse images captured by the camera through image recognition technology to identify and determine the current position of each good; Analyzing the occupancy situation of the channels in the stereoscopic warehouse by using the lidar data to judge whether the channels for transporting goods are smooth or congested, and obtaining the channel information of the stereoscopic warehouse; Determining the occupied storage locations and the idle storage locations through the warehouse images and the lidar data to obtain the spatial information of the stereoscopic warehouse; Taking the current position of each good, the channel information of the stereoscopic warehouse, and the spatial information of the stereoscopic warehouse as the real-time state information of the stereoscopic warehouse.

9. The method according to any one of claims 2-8, characterized in that, The method further includes: In the case where the current position of any one good is inconsistent with the historical position, determining that the storage layout of the stereoscopic warehouse has changed; or, in the case where the current position of any one good is consistent with the historical position, determining that the storage layout of the stereoscopic warehouse has not changed; In the case where the real-time state information indicates that the storage layout of the stereoscopic warehouse has not changed, controlling the intelligent networked handling vehicle to perform storage transportation in sequence according to the queue order in the initial handling task queue.

10. A dynamic storage and transportation control device for an intelligent networked handling vehicle, characterized in that, The device includes: An environmental data acquisition module for acquiring environmental data collected by multiple types of sensors pre-set in the stereoscopic warehouse at a preset period; A state information generation module for generating the real-time state information of the stereoscopic warehouse through the environmental data; A queue position adjustment module for dynamically adjusting the queue positions of each handling task in the initial handling task queue according to the real-time state information to obtain a target handling task queue in the case where the real-time state information indicates that the storage layout of the stereoscopic warehouse has changed; wherein, The queue positions are adjusted in the order of the priorities of each handling task. The order of the priorities of each handling task is determined according to the warehousing difficulty of each cargo in each handling task. The warehousing difficulty of each cargo is obtained by analyzing the three-dimensional map of the cargo distribution space through a pre-trained warehousing difficulty measurement model. The three-dimensional map of the cargo distribution space is constructed based on the real-time status information; A warehousing and transportation control module, configured to control the intelligent networked handling vehicle to perform warehousing and transportation in sequence according to the queue order in the target handling task queue.

Images