A method and apparatus for loading an object

Abstract

A method and apparatus for loading objects are disclosed. The method includes: acquiring information about a first object to be loaded and a current loading scheme; generating a first spatial position relationship map based on the information about the first object to be loaded and the current loading scheme; acquiring N recommended loading schemes and N second spatial position relationship maps, each second spatial position relationship map being generated based on a corresponding recommended loading scheme, where N is a positive integer; and determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship map and the N second spatial position relationship maps. This scheme, based on real-time arriving information about the current object and the current container loading scheme, can quickly and flexibly determine the optimal or near-optimal loading scheme from multiple recommended loading schemes to guide manual or machine loading, thereby improving the container loading rate.

Description

Technical Field

[0001] This application relates to the field of transportation technology, and in particular to a method and apparatus for loading objects. Background Technology

[0002] With the continuous development of transportation, three-dimensional loading is frequently involved in scenarios such as goods packaging and cargo vehicle loading. This involves loading objects of different sizes into one or more containers. Furthermore, three-dimensional loading is a typical nondeterministic polynomial-time hard problem (NP-hard), requiring input of information about the objects to be loaded and the containers used to load them, and outputting the globally optimal solution for the three-dimensional loading scheme. Therefore, by calculating the globally optimal or near-globally optimal solution (scheme) of the three-dimensional loading scheme, the loading rate of three-dimensional loading can be improved, thereby effectively reducing transportation costs.

[0003] In actual object loading and vehicle loading processes, the information of the objects to be loaded and the information of the containers used for loading are known, but the order in which the objects arrive is unknown. The area available for caching arriving objects is limited, making it impossible to pre-pile all objects to be loaded into the cache area. Therefore, an algorithm is needed to determine a three-dimensional loading scheme based on the objects arriving in real time, so that workers or machines can refer to this three-dimensional loading scheme to perform object loading.

[0004] Existing algorithms typically assume that all information about the objects to be loaded and the containers used to load them is input all at once to obtain a feasible solution or a globally approximate optimal solution. Therefore, they cannot guarantee a globally optimal solution (scheme) for 3D loading, and currently, no algorithm can obtain a globally optimal solution (scheme) for 3D loading of large-scale objects within an acceptable timeframe. Therefore, there is an urgent need to propose a new object loading method that can recommend optimal or near-optimal loading schemes to guide workers or machines in loading objects. Summary of the Invention

[0005] This application provides a method and apparatus for loading objects, used to recommend optimal or near-optimal loading schemes to improve the loading rate of containers.

[0006] In a first aspect, embodiments of the present invention provide a method for loading an object. This method can be executed by a loading device or by a chip within the loading device, without limitation. It includes the following steps: obtaining information about a first object to be loaded and a current loading scheme; generating a first spatial position relationship diagram based on the information about the first object to be loaded and the current loading scheme; obtaining N recommended loading schemes and N second spatial position relationship diagrams, each second spatial position relationship diagram being generated based on a corresponding recommended loading scheme, where N is a positive integer; and determining a final loading scheme for the first object to be loaded from among the N recommended loading schemes, based on the first spatial position relationship diagram and the N second spatial position relationship diagrams.

[0007] This design receives the first object to be loaded in real time, obtains its information and the current loading scheme, and generates a first spatial position relationship map based on this information and the current loading scheme. Further, it acquires N recommended loading schemes and N second spatial position relationship maps, each generated based on a corresponding recommended loading scheme. From these N recommended schemes, the final loading scheme for the first object to be loaded is determined. Since these N recommended loading schemes can be generated and saved offline, this design can quickly and flexibly determine the optimal or near-optimal loading scheme from multiple recommended schemes based on the real-time arrival of the current object information and the current container loading scheme. This guides manual or machine loading, thereby improving the container's loading rate.

[0008] In one possible design, before obtaining the N recommended loading schemes and the spatial relationship diagrams of the N recommended loading schemes, the method further includes: obtaining information about T objects and loading information about the T objects, as well as container information for loading the objects; where T is a positive integer; the T objects include the first object to be loaded, and the loading information includes the loading order and stacking layer number of the T objects; based on the information about the T objects, the loading information of the T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, wherein the loading rate of the M loading schemes is greater than or equal to a set first threshold; where M is a positive integer greater than or equal to N; based on the M original loading schemes, M spatial relationship diagrams are generated, with each original loading scheme corresponding to one spatial relationship diagram; based on the M spatial relationship diagrams, the N recommended loading schemes are determined from the M original loading schemes.

[0009] This design allows for the acquisition of information on all objects, their loading information, and container information from historical loading records before obtaining N recommended loading schemes and their spatial relationship diagrams. Then, based on this information, a tree search algorithm is used to obtain all possible original loading schemes and generate a spatial relationship diagram for each scheme. Furthermore, the loading rates of these original loading schemes are all above a set threshold, thus providing sufficient loading options for online object loading while ensuring a high loading rate. Finally, recommended loading schemes are selected from multiple original schemes for online object loading, further ensuring loading diversity and maximizing the loading rate.

[0010] In one possible design, obtaining M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information includes: obtaining a first set of original loading schemes for each object using a tree search algorithm based on the information of each of the T objects, the loading information of each object, and the container information; and summarizing the T sets of first original loading schemes corresponding to the T objects as the M original loading schemes.

[0011] This design allows for the generation and aggregation of a first set of original loading schemes for each of the T objects, based on their information, loading details, and container information, using a tree search algorithm. Since a large number of original loading schemes are generated offline, sufficient recommended schemes can be provided for loading objects online, thereby improving loading diversity and loading efficiency.

[0012] In one possible design, obtaining M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information includes: obtaining information of at least one first block and / or at least one second block based on the information of the T objects and the loading information of the T objects, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; obtaining a set of second original loading schemes for each first block and / or each second block based on the information of each first block and / or the information of each second block, and the container information, using a tree search algorithm; summarizing the set of at least one second original loading schemes corresponding to the at least one first block, and / or summarizing the set of at least one second original loading schemes corresponding to the at least one second block, as the M original loading schemes.

[0013] This design allows for the initial combination of T objects based on their information and loading information, resulting in multiple simple blocks. These simple blocks are then combined into complex blocks. Each simple block and / or each complex block is treated as a single block. Finally, based on the information of these complex blocks and the container information, a tree search algorithm is used to obtain and summarize the original loading set corresponding to each complex block. This approach provides sufficient loading options for online loading while also improving the efficiency of actual loading.

[0014] It should be noted that in this application, based on the information of T objects, the loading information of T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes. Other heuristic methods can also be used to obtain the original loading schemes, and this application does not make any specific limitations.

[0015] In one possible design, determining the N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams includes: calculating the diversity function value of each of the M original loading schemes based on the M spatial location relationship diagrams; and selecting the original loading schemes corresponding to the first N diversity function values ​​in descending order as the N recommended loading schemes.

[0016] This design allows for the calculation of diversity function values ​​for each of the M original loading schemes based on the M spatial relationship diagrams corresponding to the M original loading schemes. Then, the diversity function values ​​of the M original loading schemes are sorted in descending order, and the original loading schemes corresponding to the top N diversity function values ​​are selected as recommended loading schemes. This ensures the diversity of recommended loading schemes and avoids the selection of identical or similar recommended loading schemes.

[0017] In one possible design, determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams can be implemented in the following ways:

[0018] The first implementation method is as follows: among the N second spatial position relationship diagrams, a target spatial position relationship diagram that is the same as the first spatial position relationship diagram is determined; the recommended loading scheme corresponding to the target spatial position relationship diagram is taken as the final loading scheme for the first object to be loaded.

[0019] In this way, the first spatial location relationship map is compared with N second spatial location relationship maps. From the N second spatial location relationship maps, the target spatial location relationship map that is the same as the first spatial location relationship map is found. The recommended loading scheme corresponding to the target spatial location relationship map is used as the final loading scheme for the first object to be loaded. This can provide an effective loading scheme for the first object to be loaded, while also ensuring the loading rate.

[0020] The second implementation method: When the first spatial position relationship diagram exists in the N second spatial position relationship diagrams, the step of determining the final loading scheme of the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams specifically includes: determining that the first spatial position relationship diagram is included in i of the N second spatial position relationship diagrams, where i is a positive integer less than or equal to N; and selecting the recommended loading scheme with the highest loading rate from the i recommended loading schemes corresponding to the i second spatial position relationship diagrams as the final loading scheme of the first object to be loaded.

[0021] In this way, the first spatial positional relationship diagram is included in i of the N second spatial positional relationship diagrams, that is, the first spatial positional relationship diagram is a subgraph of the i second spatial positional relationship diagrams; furthermore, among the i recommended loading schemes corresponding to the i second spatial positional relationship diagrams, the recommended loading scheme with the highest loading rate is selected as the final loading scheme for the first object to be loaded, so that while providing an effective loading scheme for the first object to be loaded, the loading rate can be guaranteed as much as possible.

[0022] The third implementation method: When the first spatial position relationship diagram does not exist among the N second spatial position relationship diagrams, the step of determining the final loading scheme of the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams specifically includes: adjusting the N second spatial position relationship diagrams according to the first spatial position relationship diagram to obtain the adjustment range value of the N second spatial position relationship diagrams; and taking the recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment range value as the final loading scheme of the first object to be loaded.

[0023] In this way, none of the first spatial relationship diagrams are subgraphs of the N second spatial positional relationship diagrams. At this time, the N second spatial positional relationship diagrams can be adjusted based on the first spatial positional relationship diagram to obtain the adjustment range values ​​of the N second spatial positional relationship diagrams. The recommended loading scheme corresponding to the second spatial positional relationship diagram with the smallest adjustment range value is taken as the final loading scheme of the first object to be loaded. This can improve the efficiency of providing loading schemes for the first object to be loaded, while also ensuring the loading rate.

[0024] The fourth implementation method: If the adjustment of the N second spatial position relationship diagrams is carried out using the third implementation method, and the adjustment range of the N second spatial position relationship diagrams is greater than the set second threshold, then according to the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, a tree search algorithm is used to obtain Q loading schemes, where Q is a positive integer; the loading information of the first object to be loaded includes the loading order and stacking layer number of the first object to be loaded; the loading scheme with the highest loading rate among the Q loading schemes is taken as the final loading scheme of the first object to be loaded.

[0025] In this method, if the adjustment values ​​of N second spatial location relationship maps based on the first spatial location relationship map are all greater than the set threshold, that is, each recommended loading scheme is adjusted too much and becomes unusable, then multiple loading schemes can be obtained online using a tree search algorithm based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information. The loading scheme with the highest loading rate is taken as the final loading scheme for the first object to be loaded. This provides an effective loading scheme for the first object to be loaded to guide workers or machines to load, while also ensuring the loading rate.

[0026] In one possible design, the information of the first object to be loaded includes, but is not limited to, at least one of the following: the length of the first object to be loaded, the height of the first object to be loaded, the width of the first object to be loaded, and the weight of the first object to be loaded.

[0027] Through this design, the information of the first object to be loaded includes, but is not limited to, the length, width, height and weight of the first object to be loaded. Therefore, this information can be used as valid information for selecting a loading scheme for the first object to be loaded.

[0028] Secondly, this application also provides an object loading device, which can be applied to a device for loading articles or goods and has the function of implementing the method described in the first aspect or any possible design of the first aspect. The loading device may include: an input module, a generation module, a processing module, and an output module.

[0029] It should be understood that the aforementioned input and output modules can be communication interface circuits to realize the reception and / or transmission of data and / or information. The aforementioned input and output modules can also be transceivers to realize the reception and / or transmission of data and / or information. Therefore, this application does not specifically limit the specific form of the input and output modules.

[0030] Thirdly, this application also provides an object loading device, which can be applied to a device for loading articles or goods and has the function of implementing the method described in the first aspect or any possible design of the first aspect. This function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more units corresponding to the above-described function, such as an input unit, a generation unit, a processing unit, and an output unit.

[0031] Fourthly, embodiments of this application also provide a computer storage medium storing a software program that, when read and executed by one or more processors, can implement the method provided in the first aspect or any of the designs described above.

[0032] Fifthly, embodiments of this application also provide a computer program product containing instructions that, when run on a computer, cause the method provided in the first aspect or any of the designs described above to be executed.

[0033] Sixthly, embodiments of this application also provide a chip system including a processor for supporting clients in implementing the functions involved in the first aspect above.

[0034] In one possible design, the chip system further includes a memory for storing necessary program instructions and data to be executed by the loading device. The chip system may consist of chips or may include chips and other discrete components.

[0035] The technical effects that can be achieved by any design in any of the second to sixth aspects mentioned above can be referred to the description of the technical effects that can be achieved by any design in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0036] Figure 1 A schematic diagram illustrating an application scenario of object loading provided in an embodiment of this application;

[0037] Figure 2 A flowchart illustrating a method for loading an object, as provided in an embodiment of this application;

[0038] Figure 3A A schematic diagram illustrating the spatial coordinates of an object (block) as provided in an embodiment of this application;

[0039] Figure 3B A schematic diagram showing the coordinates of different blocks in spatial location provided in an embodiment of this application;

[0040] Figure 3C A schematic diagram illustrating a spatial relationship diagram provided in an embodiment of this application;

[0041] Figure 3D A schematic diagram illustrating the spatial relationship of generating a final loading scheme, provided in an embodiment of this application;

[0042] Figure 3E A schematic diagram of spatial division for three-dimensional loading provided in an embodiment of this application;

[0043] Figure 3F A schematic diagram illustrating the method of obtaining the original loading scheme using a tree search algorithm, provided as an embodiment of this application;

[0044] Figure 3G A schematic diagram illustrating the search process of a tree search algorithm provided in an embodiment of this application;

[0045] Figure 3H A schematic diagram illustrating the combination of objects into simple blocks and complex blocks, provided for embodiments of this application;

[0046] Figure 3I A schematic diagram of a spatial positional relationship diagram and sub-diagrams provided for embodiments of this application;

[0047] Figure 4A A schematic diagram illustrating an offline stage for generating the original loading scheme, provided as an embodiment of this application;

[0048] Figure 4B A schematic diagram of a process for selecting recommended loading schemes provided in an embodiment of this application;

[0049] Figure 5 A recommended flowchart of an object loading method provided in an embodiment of this application;

[0050] Figure 6 This is a schematic flowchart illustrating a method for loading an object, as provided in an embodiment of this application.

[0051] Figure 7A A histogram showing the test results of different schemes provided in the embodiments of this application on the BR dataset;

[0052] Figure 7B A bar chart showing the test results of different solutions provided in the embodiments of this application on a public dataset;

[0053] Figure 8This is a schematic diagram of an object loading device provided in an embodiment of this application;

[0054] Figure 9 This is a schematic diagram of an object loading device provided in an embodiment of this application. Detailed Implementation

[0055] This application provides a method and apparatus for loading an object. The method and apparatus are based on the same or similar technical concepts. Since the principles of the method and apparatus for solving the problem are similar, the implementation of the apparatus and method can refer to each other, and repeated parts will not be described again.

[0056] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0057] The embodiments of this application are typically applied to fields such as cargo handling or cargo loading, such as in automated logistics handling scenarios. Figure 1 This illustrates possible application scenarios for an object loading method provided in this application. See also... Figure 1 As shown, suppose a pile of objects A arrives. These objects A can be loaded into containers according to a specified loading order or stacking method, either manually or by machine. The containers are then moved to a vehicle by hand or machine. Alternatively, the pile of objects A can be directly moved to the vehicle by hand or machine, following the specified loading order or stacking method. Finally, the vehicle transports the pile of objects A to its destination.

[0058] In the above scenario, to achieve automatic loading of object A, automated 3D loading is often involved. 3D loading is a typical nondeterministic polynomial-time hard problem, and currently no algorithm can obtain a globally optimal solution within an acceptable time for large-scale item loading. Typically, the default approach is to take all items to be loaded and the container used for loading as input, and output a feasible solution or a globally approximate optimal solution. However, in actual item packing and vehicle loading processes, although all items to be loaded are known in advance, the arrival order of each item is unknown (i.e., items arrive randomly over time). In such cases, an algorithm is needed to output a 3D loading scheme in real time based on the order of object arrival. This can truly guide workers or machines to perform orderly loading, thereby improving the loading rate of a container (such as a truck bed), meaning the same container can hold more items.

[0059] The following section will first explain some of the terms used in the embodiments of this application so that those skilled in the art can understand them.

[0060] 1) The objects involved in the embodiments of this application can be physical entities that are loaded, such as goods from a factory or merchandise from a supermarket. During the loading process, the loaded objects can all be objects of the same material, or goods or merchandise of different materials; this application does not impose specific limitations on these. Furthermore, this application does not impose specific limitations on the shape, size, or weight of the loaded objects.

[0061] 2) The container involved in the embodiments of this application can be an object with a certain internal space and a physical entity that can be used to load objects. The container can be processed into a sealed state after loading objects, such as a cardboard box, wooden box, or box truck, or it can be processed into an unsealed state, such as an unsealed box or truck. Therefore, this application does not make specific limitations on the shape, size, weight, material, etc. of the container for loading objects.

[0062] 3) The loading rate involved in the embodiments of this application can be expressed as the ratio of the total volume of objects actually loaded in a container to the total volume of the container. It can reflect the allocation of space when using the container to load objects. Generally, the higher the loading rate, the higher the utilization rate of the container, that is, the more reasonable the allocation when using the container to load objects. The lower the loading rate, the lower the utilization rate of the container, that is, the unreasonable allocation of space when using the container to load objects.

[0063] 4) The loading scheme involved in the embodiments of this application can represent how to load an object into a specified container. Typically, the loading scheme can include the specific spatial placement of one or more objects in the container, the number of stacked layers of objects, and the loading order of objects. Therefore, the specific content included in the loading scheme of this application can be set according to the actual situation, and this application does not make specific limitations.

[0064] Currently, the existing loading solutions include the following:

[0065] The existing technical solution one transforms the original dynamic problem into a static problem by dividing it into time slices. For example, a batch of objects to be loaded is buffered every 10 minutes, and these objects are loaded all at once using a traditional 3D loading algorithm. Traditional 3D loading algorithms include: mathematical programming methods, such as Instant Linear Programming (ILP) and Dynamic Programming (DP); heuristic algorithms, such as Next Fit (NF), First Fit (FF), Best Fit (BF), and First Fit Decreasing (FFD); and learning optimization algorithms, such as Unsupervised Learning (RL) and Deep Learning (DL). Although this solution can simply use existing algorithms and adapt quickly, it loads objects all at once and does not obtain a loading plan in real time according to the order in which the objects arrive. That is, it only considers local information and ignores global information. Therefore, this solution cannot guarantee that the obtained loading plan is optimal.

[0066] Existing technical solution two employs the FB absolute approximation algorithm to solve the one-dimensional online loading problem. First, incomplete containers are classified. Then, a constructor is used to match the classified containers. Finally, newly arriving objects are loaded according to the construction rules. However, this solution does not consider the loading problem in real-time based on the order of object arrival; it only considers local information and ignores global information. Therefore, this solution cannot guarantee that the obtained loading solution is optimal. Furthermore, this solution is only applicable to one-dimensional loading. In multi-dimensional loading problems, it is difficult to uniformly determine the classification and matching of containers, as well as the loading rules for objects.

[0067] Existing technical solution three employs a harmonic loading algorithm to address the online loading problem. This involves categorizing arriving objects into different size ranges and placing objects from each range into the same container. If a container is full, a new container is used for loading. However, this solution requires manual determination of the object size ranges, making it highly customized. Furthermore, the solution has poor applicability and ultimately cannot guarantee that the resulting loading scheme is optimal.

[0068] In summary, in actual object loading processes, the arrival order of objects is often unknown. Existing dynamic loading schemes mostly rely on heuristic loading based on the currently arriving objects, making it difficult to utilize global information about the objects. Therefore, they cannot guarantee that the final loading scheme will achieve the highest loading rate.

[0069] Therefore, this application provides an object loading method to quickly and flexibly determine the optimal or near-optimal loading scheme from multiple recommended loading schemes, so as to guide manual or machine loading and thereby improve the loading rate of containers.

[0070] The embodiments of this application involve at least one, including one or more; wherein, multiple means two or more. Furthermore, it should be understood that in the description of this application, terms such as "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance, nor as indicating or implying order.

[0071] The terminology used in the following embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “the,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the embodiments of this application, “one or more” means one or more (including two); “and / or” describes the relationship between related objects, indicating that three relationships may exist; for example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0072] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0073] The terms "article" and "object" used in this specification may be used interchangeably, meaning that "article" can refer to "object," and the two concepts have the same meaning.

[0074] This application provides a method for loading an object, which can be applied to, but is not limited to, methods such as... Figure 1 The loading scene shown. (See attached image.) Figure 2The diagram shown is a flowchart of a method for loading an object according to an embodiment of this application. Specifically, this method can be executed by a loading device or by a chip in the loading device. This application does not limit the method, but it may include the following steps:

[0075] S201: Obtain information about the first object to be loaded and the current loading scheme.

[0076] It should be understood that the first object to be loaded can be an object that arrives in real time, and that objects that have already arrived have been loaded into the container before the first object to be loaded arrives.

[0077] Optionally, the information of the first object to be loaded includes, but is not limited to, one or more of the following: the length, height, width, and weight of the first object to be loaded. This application does not limit this information.

[0078] S202: Generate a first spatial position relationship diagram based on the information of the first object to be loaded and the current loading scheme.

[0079] It should be understood that, in the embodiments of this application, the spatial position relationship diagram corresponding to each loading scheme includes, but is not limited to: left-right relationship diagram, front-back relationship diagram, and up-down relationship diagram (vertical relationship diagram).

[0080] Specifically, the spatial relationship diagram can be defined as follows:

[0081] like Figure 3A As shown, taking the truck bed as an example, the width of the truck bed can be defined as the X-axis direction (corresponding to the left-right relationship diagram), the length of the truck bed can be defined as the Y-axis direction (corresponding to the front-back relationship diagram), and the height of the truck container can be defined as the Z-axis direction (corresponding to the top-bottom relationship diagram).

[0082] Each object (block) inside the carriage is represented as a node in the spatial position relationship diagram. The number of the node can represent the number of the object (block). Therefore, based on the actual three-dimensional loading situation inside the carriage, it can be determined whether there are lines connecting the nodes in the spatial position relationship diagram.

[0083] Specifically, taking the example of placing object (block) i and object (block) j in the left and right directions respectively inside the carriage. If the projections of object (block) i and object (block) j on the X-axis overlap, refer to... Figure 3B As shown in (a), objects (blocks) i and j need to satisfy the following condition (1):

[0084] (x i 2 ≥x j 1 or x j2 ≥x i 1 )

[0085] in, and These represent the coordinates of object (block) i at its two edge vertices along the X-axis. and These represent the coordinates of the two edge vertices of object (block) j along the X-axis.

[0086] It should be understood that when drawing the front-to-back relationship diagram, the person is located at the X-axis position (the rear of the truck bed). Within the person's direct field of vision, it is determined whether object (block) i and object (block) j are mutually obstructed, that is, whether the projections of object (block) i and object (block) j on the X-axis overlap. When object (block) i and object (block) j meet the above condition (1), it indicates that the projections of object (block) i and object (block) j on the X-axis overlap. At this time, when drawing the left-to-right relationship diagram, it is necessary to connect the nodes corresponding to object (block) i and object (block) j.

[0087] If the projections of object (block) i and object (block) j on the Y-axis overlap, refer to Figure 3B As shown in (b), objects (blocks) i and j need to satisfy the following condition (2):

[0088] (y i 2 ≥y j 1 or y j 2 ≥y i 1 )

[0089] in, and These represent the coordinates of object (block) i at its two edge vertices along the Y-axis. and These represent the coordinates of the two edge vertices of object (block) j along the Y-axis.

[0090] It should be understood that when drawing the left-right relationship diagram, the person is located at the Y-axis position (the position of the side of the truck bed). Within the person's direct field of vision, it is determined whether object (block) i and object (block) j are mutually obstructed, that is, whether the projections of object (block) i and object (block) j on the Y-axis overlap. When object (block) i and object (block) j meet the above condition (2), it indicates that the projections of object (block) i and object (block) j on the Y-axis overlap. At this time, when drawing the front-back relationship diagram, it is necessary to connect the corresponding nodes of object (block) i and object (block) j.

[0091] If the projections of object (block) i and object (block) j on the Z-axis overlap, refer to Figure 3B As shown in (c), objects (blocks) i and j need to satisfy the following condition (3):

[0092] (z i 2 ≥z j 1 or z j 2 ≥z i 1 )

[0093] in, and These represent the coordinates of object (block) i at its two edge vertices along the Z-axis. and These represent the coordinates of the two edge vertices of object (block) j along the Z-axis.

[0094] It should be understood that when drawing the vertical relationship diagram, the person is in the carriage and looks at the height of the carriage to determine whether there is mutual occlusion between object (block) i and object (block) j in the vertical direction, that is, whether the projections of object (block) i and object (block) j on the Z-axis overlap; when object (block) i and object (block) j meet the above condition (3), it means that the projections of object (block) i and object (block) j on the Z-axis overlap. At this time, when drawing the vertical relationship diagram, it is necessary to connect the nodes corresponding to object (block) i and object (block) j.

[0095] It should be noted that in this application, the objects corresponding to nodes with the same serial number can be different objects with the same length, width, height, and weight.

[0096] For example, when drawing a before-and-after relationship diagram, refer to Figure 3C As shown, Figure 3C (a) shows the actual object loading scheme. Figure 3C In (a), there are two blocks 1 (i.e., the first block 1 and the second block 1). The two blocks 1 are placed side by side and their projections in the Y-axis direction overlap. Therefore, the two blocks 1 satisfy the above condition (2). Thus, in Figure 3C Connect the two nodes 1 in the before-after relationship diagram in (b); in addition, Figure 3C In (a), block 3 overlaps with the projections of both blocks 1 (i.e., the first block 1 and the second block 1) along the Y-axis. Therefore, block 3 and the two blocks 1 satisfy the above condition (2) respectively. Thus, in Figure 3C In the diagram of the relationship between the beginning and end in (b), node 3 is connected to two nodes 1 respectively; Figure 3CIn (a), the projections of block 5 and the two blocks 1 in the Y-axis direction also overlap. Therefore, block 5 satisfies the above condition (2) with the two blocks 1 respectively. Thus, in Figure 3C In the diagram of the relationship between nodes in (b), node 5 is connected to both nodes 1; additionally, Figure 3C In (a), block 5 overlaps with the projections of block 4 and the two blocks 6 along the Y-axis. Therefore, block 5 also satisfies the above condition (2) with block 4 and the two blocks 6. Thus, in (a), Figure 3C In the context diagram of (b), node 5 is connected to node 4 and the two nodes 6. Following this pattern, the complete context diagram of scheme (a) in 3C can be drawn, as shown below. Figure 3C As shown in (b).

[0097] When drawing a left-right relationship diagram: Figure 3C In (a), the first block 1 overlaps with the corresponding blocks 4 and 6 below it in the X-axis direction. Therefore, the first block 1 satisfies the above condition (1) with blocks 4 and 6 respectively. Figure 3C In the left-right relationship diagram in the middle (c), the first node 1 is connected to node 4 and the first node 6 respectively; Figure 3C In (a), the projection of the second block 1 on the X-axis also overlaps with the corresponding blocks 4 and 6 below it. Therefore, the second block 1 also satisfies the above condition (1) with blocks 4 and 6 respectively. Thus, Figure 3C In the left-right relationship diagram in the middle (c), the second node 1 is connected to node 4 and the second node 6 respectively; Figure 3C In (a), block 3 overlaps with blocks 4 and 5 in the X-axis direction, so block 3 also satisfies condition (1) above with blocks 4 and 5. Therefore, [the following is a possible interpretation:] Figure 3C In the left-right relationship diagram of 3C, node 3 is connected to nodes 4 and 5 respectively. Similarly, the complete left-right relationship diagram of scheme (a) in 3C can be drawn. For specific connection details, please refer to [reference needed]. Figure 3C (c)

[0098] It should be understood that, for the sake of simplicity, only the two-dimensional case is illustrated in the embodiments of this application, and the top-bottom relationship diagram (vertical relationship diagram) is not shown. However, in practice, the connection and drawing of the complete top-bottom relationship diagram (vertical relationship diagram) of scheme (a) in 3C can be referred to the front-back relationship diagram (b) and left-right relationship diagram (c) of scheme (a) in 3C, which will not be described in detail here.

[0099] When performing step S202 above, please refer to the following example:

[0100] For example, such as Figure 3D middle, Figure 3D (a) in the diagram represents the recommended loading scheme obtained during the offline phase. Figure 3D(b) in the diagram represents the current loading scheme. Figure 3D (c) in the diagram shows the before-and-after relationship of the recommended loading scheme, reflecting whether there is overlap (occlusion) in the projections of different blocks on the Y-axis. Figure 3D (d) in the diagram represents the left-right relationship of the recommended loading scheme, determining whether there is overlap (occlusion) in the projections of different blocks on the X-axis. (See reference) Figure 3D In diagram (b), the current loading plan already includes blocks 1, 4, and 5 inside the carriage. Therefore, the front-to-back relationship of the current loading plan is shown in the diagram below. Figure 3D As shown by the solid line connecting the two parts in (c), it indicates that blocks 1 and 5 have projective overlap on the Y-axis, and blocks 4 and 5 also have projective overlap on the Y-axis. The left-right relationship diagram of the current loading scheme is shown by the solid line connecting the two parts in (d) of the 3D diagram, indicating that blocks 1 and 4 have projective overlap on the X-axis. If a new block 1 arrives, since the spatial position relationship diagram of the current loading scheme is a sub-diagram of the spatial position relationship diagram of the recommended loading scheme, the position of the newly arrived block 1 can be inferred from the spatial position relationship diagram of the recommended loading scheme. Figure 3D In the middle position (e), after placing the newly arrived block 1, update the current loading scheme spatial relationship graph, which is still a subgraph of the spatial relationship graph corresponding to the recommended loading scheme. Continue this process to load the remaining blocks, resulting in the final loading scheme. Figure 3D As shown in (e) in the diagram.

[0101] It is important to note that Figure 3D In the diagram of the relationship between (c) in the middle, when two different nodes are connected by a dashed line, it means that the projections of the blocks actually corresponding to the two nodes in the Y direction may or may not overlap; when two different nodes are connected by a solid line, it means that the projections of the blocks actually corresponding to the two nodes in the Y direction must overlap. Figure 3D In the left-right relationship diagram of the current loading scheme in (d), when two different nodes are connected by a dashed line, it means that the projections of the blocks actually corresponding to the two nodes in the X direction may or may not overlap; when two different nodes are connected by a solid line, it means that the projections of the blocks actually corresponding to the two nodes in the X direction must overlap.

[0102] Similarly, in the vertical relationship diagram of the current loading scheme, the meaning of connecting different nodes with dashed lines or solid lines is the same as that in the front-back relationship diagram or left-right relationship diagram mentioned above, and will not be elaborated here.

[0103] S203: Obtain N recommended loading schemes and N second spatial position relationship diagrams. Each second spatial position relationship diagram is generated based on a corresponding recommended loading scheme, and N is a positive integer.

[0104] In one implementation, before obtaining the N recommended loading schemes and their spatial relationship diagrams, an offline stage is included. In this offline stage, the N recommended loading schemes and their spatial relationship diagrams are generated. Specifically, generating the N recommended loading schemes can be achieved through the following steps:

[0105] First step: Based on the information of T objects, the loading information of T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, where the loading rate of each of the M loading schemes is greater than or equal to a set first threshold; M is a positive integer greater than or equal to N.

[0106] The second step: Based on the M original loading schemes, generate M spatial relationship diagrams, with one spatial relationship diagram corresponding to each original loading scheme. The third step: Based on the M spatial relationship diagrams, determine N recommended loading schemes from the M original loading schemes.

[0107] The information for each object includes, but is not limited to, at least one of the following: the object's length, height, width, and weight.

[0108] Specifically, when performing the first step in the above implementation method: as follows: Figure 4A As shown, S401A: Obtain information on T objects and their loading information, as well as information on the containers used to load the objects; T is a positive integer; the T objects include the first object to be loaded, and the loading information includes the loading order and stacking layer number of the T objects; S402A: Based on the information on the T objects and their loading information, construct a series of different blocks; S403A: Adjust the priority order of the different blocks; S404A: Use a tree search algorithm to solve the original loading scheme; S405A: Determine whether the obtained original loading scheme meets the requirements for loading rate and number of schemes; S406A: If yes, save the large number (i.e., M) of the original loading schemes obtained from S304. If not, return to step S403A and continue executing the steps after S403A until the obtained original loading scheme meets the requirements for loading rate and number of schemes.

[0109] For example, when performing the first step in the above embodiments, it can be achieved in, but is not limited to, the following two ways:

[0110] In the first approach: based on the information of each of the T objects, the loading information of each object, and the container information, a tree search algorithm is used to obtain the first original loading scheme set for each object; the T first original loading scheme sets corresponding to the T objects are summarized as M original loading schemes.

[0111] Specifically, the loading order of T objects is adjusted, a tree search algorithm is used for 3D loading, and a large number of original loading schemes are output. The specific loading process is as follows:

[0112] The tree search algorithm yields a large number of original loading schemes, mainly for each node of the tree to access the available space Spaces = {space1, space2, ..., space...}. p The selection process is performed. If the initial available space is the entire container, i.e., Spaces = {whole_container}, then... Figure 3E As shown, each time an object is placed, the space where it is placed will be divided into up to three spaces in three dimensions. The placed object and its corresponding space are removed from the set of T objects and the set of available spaces, respectively. If the newly divided space can hold at least one of the remaining objects, it is added to the set of spaces. This process continues until the final set of spaces has no available space or cannot hold any object. The number of available spaces in the final set of spaces is the number of original loading schemes for each loading order. Finally, the original loading schemes for each loading order are summarized.

[0113] For example, such as Figure 3F As shown, the initial available space is the entire space, i.e., Spaces = {whole_container}. After placing the first object in the initial available space, the space occupied by the first object is cleared, and three new smaller available spaces (i.e., space 1, space 2, and space 3) are created. At this time, the initial available space Spaces = {whole_container} is updated, resulting in a new first space set Spaces = {space1, space2, space3}. When placing the second object, a corresponding space can be selected from the first space set. If space 1 is selected to place the second object, three smaller new available spaces are created: space (1,1), space (1,2), and space (1,3). If space 2 is selected to place the second object, three smaller new available spaces are created: space (2,1), space (2,2), and space (2,3). If space 3 is selected to place the second object, three smaller new available spaces are created: space (3,1), space (3,2), and space (3,3). Similarly, the remaining objects can be placed in the corresponding spaces selected from the previously updated Spaces set. Thus, in the loading process described above, all the original loading schemes can be summarized at the end.

[0114] Since the time complexity of a complete tree search is very high, it is necessary to control the breadth w (i.e., the number of spaces where objects need to be placed) and depth d (the maximum number of nodes to search downwards) of the tree search. The total number of original loading schemes that can be output is N = w. d .

[0115] For example, refer to Figure 3G As shown, the Figure 3G This demonstrates a tree search process with a breadth of 3 and a depth of 2, where 3 spaces are sampled at node l0 for each attempt. 1，1 , l 1，2 , ..., l 1，3 After reaching a depth of 2 layers, a heuristic greedy algorithm (such as First Fit, Next Fit, Best Fit, etc.) can be used to quickly load the remaining objects. Therefore, when loading two objects (object 1 and object 2), assuming there are 3 spaces, there are 3 spaces to choose from when placing object 1, meaning the first loading scheme includes 3 options (l 1，1 , l 1，2 , l 1，3 Since placing object 1 in the space corresponding to one of the schemes will create 3 new spaces, there are 3 spaces available for placing object 2. This means that each of the first loading schemes splits into 3 schemes. Therefore, the final output consists of a total of 9 complete original loading schemes.

[0116] In the second approach: Based on the information of T objects and the loading information of T objects, at least one first block and / or at least one second block are obtained, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; based on the information of each first block and / or each second block, and the container information, a tree search algorithm is used to obtain a set of second original loading schemes for each first block and / or a set of second original loading schemes for each second block; the sets of at least one second original loading schemes corresponding to at least one first block are summarized, and / or the sets of at least one second original loading schemes corresponding to at least one second block are summarized to form M original loading schemes.

[0117] For example, such as Figure 3H As shown, first, T objects are combined into a simple block (i.e., the first block). This involves placing objects of the same material in the same direction and stacking them in the length, width, and height directions to form a simple block (i.e., the first block). Then, the obtained simple blocks can be combined to obtain... Figure 3HThe complex block (i.e., the second block) is then processed. Further, all the resulting simple / complex blocks / objects are treated as blocks. Finally, the loading order of each block is adjusted, and a tree search algorithm is used for 3D loading, outputting a large number of original loading schemes. For details on how to use the tree search algorithm to output a large number of original loading schemes, please refer to the content described in the first method above; it will not be elaborated upon here.

[0118] It should be understood that if among T objects there exists an object whose material is different from the other objects and cannot be combined into a simple block, then that object is treated as a separate block; if among multiple simple blocks there exists a simple block that cannot be combined into a complex block, then that simple block is treated as a separate block.

[0119] It is important to note that after executing the above two methods, multiple original loading schemes (greater than or equal to M original loading schemes) can be obtained. Then, among these multiple original loading schemes, the M original loading schemes whose corresponding loading rates are all greater than or equal to a set first threshold are selected. If the number of the obtained M original loading schemes meets the set quantity (requirement), the above two processes for obtaining original loading schemes are stopped. If the number of the obtained M original loading schemes does not meet the set quantity (requirement), the loading order of the objects in the first method or the loading order of the blocks in the second method is adjusted, and the steps after adjusting the loading order are repeated until the number of obtained M original loading schemes meets the set quantity (requirement).

[0120] Specifically, when performing the second step in the above implementation: based on the M original loading schemes, generate M corresponding spatial position relationship diagrams, that is, each original loading scheme corresponds to a spatial position relationship diagram. For details, please refer to step S202 above, which will not be elaborated here.

[0121] Specifically, in performing the third step of the above implementation method—that is, determining N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams—can be achieved in the following way: Figure 4B As shown, S401B: Obtain M original loading schemes and M spatial location relationship diagrams; S402B: Calculate the diversity function value of each of the M original loading schemes based on the M spatial location relationship diagrams; S403B: Determine N recommended loading schemes from the M original loading schemes, that is, according to the order of the diversity function values ​​of the M original loading schemes from largest to smallest, select the original loading schemes corresponding to the diversity function values ​​of the first N schemes as the N recommended loading schemes; S404B: Save the N recommended schemes offline.

[0122] When performing step S402B above, the specific steps may include:

[0123] Step 1: Calculate the diversity function value for each original loading scheme. This involves comparing each of the M original loading schemes with the other original loading schemes using the diversity function, resulting in M-1 diversity function values ​​for each original loading scheme.

[0124] The diversity evaluation function can be defined as follows:

[0125] For two original loading schemes (s, t), the difference between them can be evaluated using the following formula:

[0126] diff (s，t) =The total number of distinct edges in three relational graphs (Formula 1)

[0127] Step 2: For each of the M original loading schemes, the smallest diversity function value among the M-1 diversity function values ​​obtained from Step 1 is taken as the diversity function value of that original loading scheme. Therefore, the diversity function values ​​of the M original loading schemes can be obtained.

[0128] The diversity function value of a specific scheme s in the original loading scheme set M can be obtained using the following formula:

[0129]

[0130] It should be understood that diversity s The larger the value, the higher the degree of non-union.

[0131] Step 3: Sort the diversity function values ​​of the M original loading schemes from largest to smallest, and select the original loading schemes corresponding to the top N diversity function values ​​as the N recommended loading schemes.

[0132] For example, when M=5, the 5 original loading schemes include: the first original loading scheme, the second original loading scheme, the third original loading scheme, the fourth original loading scheme, and the fifth original loading scheme.

[0133] According to Formula 1 above, the four difference values ​​between the first original loading scheme and the other four original loading schemes are as follows: diff (1，2) diff (1，3) diff (1，4) diff (1，5) If the smallest difference value among the four difference values ​​is diff (1，2) Then the diff with the smallest difference value will be selected. (1，2) This serves as the diversity function value for the first original loading scheme. That is, diversity1 = diff (1，2) .

[0134] According to Formula 1 above, the four difference values ​​between the second original loading scheme and the other four original loading schemes are as follows: diff (2，1) (i.e., diff) (1，2) ), diff (2，3) diff (2，4) diff (2，5) If the smallest difference value among the four difference values ​​is diff (2，3) Then the diff with the smallest difference value will be selected. (2，3) This serves as the diversity function value for the second original loading scheme. That is, diversity2 = diff (2，3) .

[0135] According to Formula 1 above, the four difference values ​​between the third original loading scheme and the other four original loading schemes are as follows: diff (3，1) (i.e., diff) (1，3) ), diff (3，2) (i.e., diff) (2，3) ), diff (3，4) diff (3，5) If the smallest difference value among the four difference values ​​is diff (3，4) Then the diff with the smallest difference value will be selected. (3，4) This serves as the diversity function value for the third original loading scheme. That is, diversity3 = diff (3，4) .

[0136] According to Formula 1 above, the four difference values ​​between the fourth original loading scheme and the other four original loading schemes are as follows: diff (4，1) (i.e., diff) (1，4) ), diff (4，2) (i.e., diff) (2，4) ), diff (4，3) (i.e., diff) (3，4) ), diff (4，5) If the smallest difference value among the four difference values ​​is diff (4，5) Then the diff with the smallest difference value will be selected. (4，5) This serves as the diversity function value for the fourth original loading scheme. That is, diversity4 = diff (4，5) .

[0137] According to Formula 1 above, the four difference values ​​between the fifth original loading scheme and the other four original loading schemes are as follows: diff (5，1) (i.e., diff) (1，5) ), diff (5，2) (i.e., diff) (2，5) ), diff (5，3) (i.e., diff) (3，5)), diff (5，4) (i.e., diff) (4，5) If the smallest difference value among the four difference values ​​is diff, then... (5，1) Then the diff with the smallest difference value will be selected. (5，1) This serves as the diversity function value for the fifth original loading scheme. That is, diversity5 = diff (5，1) .

[0138] Sort the diversity values ​​(diversity1, diversity2, diversity3, diversity4, diversity5) of the five original loading schemes from largest to smallest. If the sorted values ​​are diversity1, diversity3, diversity2, diversity4, diversity5, and if the recommended loading scheme is set to 3, then the first, second, and third original loading schemes corresponding to diversity1, diversity3, and diversity2, respectively, will be selected as the recommended loading schemes.

[0139] It should be noted that the loading rates corresponding to the above five original loading schemes are all greater than or equal to the set first threshold; the set first threshold can be set according to actual requirements, and this application does not make specific limitations.

[0140] It should be understood that the implementation method in step S203 belongs to the offline loading stage. Therefore, step S203 can also be executed before the above steps S201 and / or S202. This application does not make specific limitations.

[0141] Furthermore, during the offline loading phase, multiple original loading schemes can be generated using all information about the objects to be loaded, available containers, and constraints. Then, multiple recommended loading schemes are determined from these original schemes. Therefore, when the first object to be loaded arrives, the final loading scheme can be determined from the multiple recommended schemes, ensuring not only the flexibility of loading the object but also maximizing the loading rate.

[0142] S204: Based on the first spatial position relationship diagram and N second spatial position relationship diagrams, determine the final loading scheme for the first object to be loaded from among the N recommended loading schemes.

[0143] When performing step S204, it can be implemented through the following four methods, but not limited to the following:

[0144] In the first implementation: among N second spatial position relationship diagrams, a target spatial position relationship diagram that is the same as the first spatial position relationship diagram is determined; the recommended loading scheme corresponding to the target spatial position relationship diagram is taken as the final loading scheme for the first object to be loaded.

[0145] In the second implementation: when a first spatial position relationship diagram exists in N second spatial position relationship diagrams; step S204 may include: determining that i of the second spatial position relationship diagrams contained in the N second spatial position relationship diagrams contain the first spatial position relationship diagram, where i is a positive integer less than or equal to N; and selecting the recommended loading scheme with the highest loading rate from the i recommended loading schemes corresponding to the i second spatial position relationship diagrams as the final loading scheme for the first object to be loaded.

[0146] It should be understood that in the embodiments of this application, the inclusion of a first spatial positional relationship diagram in the i second spatial positional relationship diagrams can be represented as the first spatial positional relationship diagram being a subgraph of the i second spatial positional relationship diagrams.

[0147] For example, such as Figure 3I As shown, taking the left-right relationship diagram included in the first and second spatial positional relationship diagrams as an example, Figure 3I In diagram (a), the second spatial location relationship diagram corresponding to the recommended loading scheme can be represented. Figure 3I Node 4 in (a) can be represented as the object to be loaded. Figure 3I (b) in the diagram can be represented as the first spatial positional relationship diagram corresponding to the current loading scheme. Obviously, Figure 3I In the diagram, (b) is a subgraph of (a) in 3I.

[0148] In the third implementation: In the N second spatial position relationship diagrams, there is no first spatial position relationship diagram; performing this step S204 may include: adjusting the N second spatial position relationship diagrams according to the first spatial position relationship diagram to obtain the adjustment range value of the N second spatial position relationship diagrams; taking the recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment range value as the final loading scheme of the first object to be loaded.

[0149] In the third implementation, when the first spatial positional relationship diagram is not a sub-diagram of any second spatial positional relationship diagram, at least one second spatial positional relationship diagram with a high similarity to the first spatial positional relationship diagram can first be selected. Then, referring to the first spatial positional relationship diagram corresponding to the current loading scheme, each second spatial positional relationship diagram is adjusted to determine the second spatial positional relationship diagram with the smallest adjustment value. The recommended loading scheme corresponding to the second spatial positional relationship diagram with the smallest adjustment value is taken as the final loading scheme to guide manual or machine loading of the first object to be loaded.

[0150] In the fourth implementation: When the third implementation is used to adjust the N second spatial position relationship diagrams, and the adjustment range of the N second spatial position relationship diagrams is greater than the set second threshold, step S204 can also be implemented in the following way: Based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, a tree search algorithm is used to obtain Q loading schemes, where Q is a positive integer; the specific implementation method of using the tree search algorithm to obtain multiple loading schemes can be referred to the specific description of using the tree search algorithm to obtain the original loading scheme in step S203 above, and will not be repeated here. Among them, the loading information of the first object to be loaded includes, but is not limited to: the loading order and stacking layer of the first object to be loaded; furthermore, the loading scheme with the largest loading rate among the Q loading schemes is taken as the final loading scheme of the first object to be loaded.

[0151] In addition, based on the information of T objects, the loading information of T objects, and the container information, the tree search algorithm described above can be used to obtain M original loading schemes. Other heuristic methods can also be used to obtain the original loading schemes, such as Next Fit Algorithm (NF), First Fit Algorithm (FF), Best Fit Algorithm (BF), etc. This application does not make any specific limitations.

[0152] It should be understood that after loading the first object to be loaded, this application needs to update the information of the object to be loaded in real time, and update the current loading scheme and the corresponding spatial position relationship diagram, so as to wait for the arrival of the next object to be loaded and execute the steps corresponding to the above steps S201-S204 to obtain the corresponding loading scheme so as to realize the loading of the object.

[0153] In summary, this application provides a method for loading objects. This method involves obtaining information about a first object to be loaded and a current loading scheme; generating a first spatial position relationship diagram based on the information of the first object to be loaded and the current loading scheme; then obtaining N recommended loading schemes and N second spatial position relationship diagrams, each second spatial position relationship diagram being generated based on a corresponding recommended loading scheme, where N is a positive integer; and determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams. This method, based on real-time arriving information about the current object and the current container loading scheme, can quickly and flexibly determine the optimal or near-optimal loading scheme from multiple recommended loading schemes to guide manual or machine loading, thereby improving the container loading rate.

[0154] The following describes in detail a method for loading an object according to specific embodiments of this application.

[0155] Based on the above Figure 2This application provides a detailed description of a method for loading objects, as proposed in the previous one. Figure 6 The specific implementation process of this embodiment is as follows:

[0156] S601: Obtain information about T objects and loading information for T objects, as well as information about the containers used to load the objects, where T is a positive integer greater than 0.

[0157] Specifically, information on T objects (i.e., all loaded objects) and their loading information, as well as the container information used to load the objects, can be obtained from all historical loading records. The information for each object includes, but is not limited to, any one or more of the object's length, width, and height; this application does not impose any limitations on this. The loading information for each object includes, but is not limited to, the loading order and stacking layer number of the objects.

[0158] It should be noted that this step may also require obtaining actual loading constraint information. The constraint information can be set according to actual needs, and this application does not impose specific limitations.

[0159] S602: Based on the information of T objects, the loading information of T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, where M is a positive integer greater than 0.

[0160] Specifically, to ensure loading efficiency, the following methods are preferred, but not limited to:

[0161] First, identical objects among the T objects can be combined to obtain multiple simple blocks. Further, these simple blocks can be combined into complex blocks. Then, the loading order of these complex blocks is adjusted. Each adjustment uses a tree search algorithm to generate a set of original loading schemes. These sets are then merged to obtain the aforementioned M original loading schemes, for example, 10,000 original loading schemes. The specific method described in step S203 above can be used to obtain the original loading schemes; details will not be repeated here.

[0162] S603: Generate M spatial position relationship diagrams based on M original loading schemes, with each original loading scheme corresponding to one spatial position relationship diagram.

[0163] For example, in step S602 above, among the 10,000 original loading schemes, a corresponding spatial position relationship diagram is generated for each original loading scheme. This spatial position relationship diagram for each original loading scheme includes, but is not limited to: front-back relationship diagram, left-right relationship diagram, and top-bottom relationship diagram (vertical relationship diagram).

[0164] For details on step S603, please refer to step S202 above; further details will not be provided here.

[0165] S604: Based on M spatial location relationship diagrams, determine N recommended loading schemes from the M original loading schemes, where N is a positive integer greater than or equal to M.

[0166] Specifically, firstly, each of the M original loading schemes is compared with the other M-1 original loading schemes for differences, resulting in M-1 difference values ​​for each original loading scheme (the calculation of difference values ​​can refer to Formula 1 in step S203 above). Among the M-1 difference values ​​for each original loading scheme, the value with the smallest difference is taken as the diversity function value of that original loading scheme (the calculation of diversity function values ​​can refer to Formula 2 in step S203 above), thus giving each original loading scheme a diversity function value. Further, the M diversity function values ​​of the M original loading schemes are sorted from largest to smallest, and the original loading schemes corresponding to the top N diversity function values ​​are selected as the N recommended loading schemes. For example, from the aforementioned 10,000 original loading schemes, this specific step can be used to select 200 offline recommended loading schemes.

[0167] It should be understood that the above steps S601-S604 are as follows: Figure 5 The first stage, the offline stage, retrieves all object and container information, loading constraints, and other data from historical loading records. Based on this information, M original offline loading schemes are generated. Further, diversity detection is performed on these M schemes, resulting in N recommended loading schemes, which are then saved to provide alternative loading options when objects arrive in real-time. Therefore, this scheme ensures both loading scheme diversity and loading efficiency.

[0168] S605: Obtain information about the first object to be loaded and the current loading scheme.

[0169] Specifically, this step is equivalent to: Figure 5 The system can detect objects arriving in real time and obtain information about them. It can also obtain information about the partial loading status of containers that have already been loaded manually or by machine, which are about to be used to load the arriving objects. Furthermore, it can determine the current loading plan based on the partial loading status of containers that have already been loaded manually or by machine.

[0170] S606: Generate a first spatial position relationship diagram based on the information of the first object to be loaded and the current loading scheme.

[0171] The first spatial position relationship diagram generated based on the information of the first object to be loaded and the current loading scheme includes, but is not limited to: front-back relationship diagram, left-right relationship diagram, and up-down relationship diagram (vertical relationship diagram). For details, please refer to step S202 above; further elaboration is not provided here.

[0172] S607: Based on the first spatial position relationship diagram and the N second spatial position relationship diagrams corresponding to the N recommended loading schemes, determine the final loading scheme for the first object to be loaded from among the N recommended loading schemes.

[0173] Specifically, when executing step S607, there are, but are not limited to, the following four implementation methods:

[0174] The first implementation method is to directly query the target spatial location relationship diagram that is the same as the first spatial location relationship diagram in the N second spatial location relationship diagrams; and take the recommended loading scheme corresponding to the target spatial location relationship diagram as the final loading scheme of the first object to be loaded.

[0175] The second implementation method is as follows: In N second spatial position relationship diagrams, if there are i second spatial position relationship diagrams that contain the first spatial position relationship diagram, where i is a positive integer less than or equal to N, that is, the first spatial position relationship diagram is a subgraph of the i second spatial position relationship diagrams, then calculate the loading rate of the loading schemes corresponding to the i second spatial position relationship diagrams, and take the recommended loading scheme with the largest loading rate as the final loading scheme of the first object to be loaded.

[0176] The third implementation method is as follows: In the N second spatial position relationship diagrams, if there is no second spatial position relationship diagram that contains the first spatial position relationship diagram, that is, the first spatial position relationship diagram is not a subgraph of the N second spatial position relationship diagrams, then based on the first spatial position relationship diagram, the N second spatial position relationship diagrams are adjusted, and the recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment is taken as the final loading scheme of the first object to be loaded.

[0177] The fourth implementation method, based on the third method, adjusts N second spatial positional relationship diagrams according to the first spatial positional relationship diagram. If the adjustment range of each second spatial positional relationship diagram exceeds a preset threshold (meaning the recommended solutions are adjusted too much and become unusable), a tree search algorithm is used to obtain multiple loading solutions online, based on the information of the first object to be loaded, its loading information, and the container information. The loading solution with the highest loading rate is then selected as the final loading solution for the first object to be loaded. This method provides an effective loading solution for the first object to be loaded, guiding workers or machines in loading while ensuring a high loading rate.

[0178] It should be noted that the above S605-S607 are as follows Figure 5In the second stage, the objects to be loaded arrive in real time. From the set of recommended loading schemes obtained in the first stage, a loading scheme is dynamically recommended to guide workers or machines to load the objects that arrive in real time, thereby improving the container loading rate. Figure 5 The second stage also includes further testing of the dynamically recommended loading scheme. That is, after loading the objects that arrive in real time using the dynamically recommended loading scheme, it is determined whether there are any unloaded objects. If there are none, the dynamically recommended loading scheme is output as the final loading scheme. If not, the dynamic recommended loading scheme continues to be tested (executing the above steps S605-S607) until there are no remaining objects. Thus, this scheme can ensure that all arriving objects are loaded and also make the loading rate of the final loading scheme high.

[0179] Furthermore, the solutions provided in this application will be compared with other conventional methods below, using both publicly available BR datasets and real-world production datasets.

[0180] For example, this study uses a tree search algorithm with a width of 5 and a depth of 2, and employs a heuristic bricklaying method as the baseline for comparative analysis. Table 1 shows the test results (e.g., final load rate) obtained using the BR dataset for the dynamic recommendation loading scheme of this application and other traditional loading schemes (in %). Figure 7A The bar chart shows the test results of different schemes using the publicly available BR dataset. Referring to Table 2, it displays the test results (e.g., final load rate) (in %) of the dynamically recommended loading scheme of this application and other traditional loading schemes using the publicly available dataset. Figure 7B A bar chart showing the test results of different schemes using a publicly available dataset.

[0181] Based on Table 1, Table 2, and the corresponding... Figure 1 and Figure 2 The experimental results shown demonstrate that the dynamic recommendation scheme adopted in this application outperforms the baseline method in all cases (i.e., under different numbers of manual placements). Furthermore, except when using the BR dataset and having 1 and 2 manual placements, the dynamic recommendation scheme adopted in this application outperforms the current tree search in most cases. In addition, the test results of the dynamic recommendation scheme adopted in this application are close to the test results of the global tree search.

[0182] Table 1

[0183] Number of manual placements Global Tree Search Current tree search Baseline Dynamic Recommendations 1 89.04 88.79 81.50 86.75 2 89.04 87.70 80.44 87.45 3 89.04 86.75 78.95 87.54 4 89.04 85.49 76.17 86.83

[0184] Table 2

[0185] Number of manual placements Global Tree Search Current tree search Baseline Dynamic Recommendations 1 85.51 78.97 74.05 79.13 2 85.51 78.97 74.06 79.13 3 85.51 78.97 74.05 79.13 4 85.51 78.96 74.04 79.12

[0186] Based on the same technical concept, embodiments of this application provide an object loading device. This object loading device may include modules or units corresponding to the methods / operations / steps / actions described in the above method embodiments. These modules or units may be hardware circuits, software, or a combination of hardware circuits and software. The device may have the following characteristics: Figure 8 The structure shown.

[0187] like Figure 8 As shown, the device 800 may include an input unit 801, a generation unit 802, a processing unit 803, and an output unit 804. The following is a detailed description of each unit.

[0188] The input unit 801 can be used to acquire information about the first object to be loaded and the current loading scheme; the generation unit 802 can be used to generate a first spatial position relationship diagram based on the information about the first object to be loaded and the current loading scheme; the input unit 801 can also be used to acquire N recommended loading schemes and N second spatial position relationship diagrams, each second spatial position relationship diagram being generated based on a corresponding recommended loading scheme, where N is a positive integer; the processing unit 803 can be used to determine the final loading scheme for the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams; the output unit 804 can be used to output the final loading scheme for the first object to be loaded.

[0189] In one possible design, the processing unit 803 can also be used to obtain information about T objects and their loading information, as well as container information for loading the objects, before the input unit 801 obtains N recommended loading schemes and their spatial relationship diagrams. Here, T is a positive integer; the T objects include the first object to be loaded; and the loading information includes the loading order and stacking layer number of the T objects. Then, based on the information about the T objects, their loading information, and the container information, a tree search algorithm is used to obtain M original loading schemes, where the loading rate of each of the M schemes is greater than or equal to a set first threshold. Here, M is a positive integer greater than or equal to N. Further, based on the M original loading schemes, M spatial relationship diagrams are generated, with each original loading scheme corresponding to one spatial relationship diagram. Finally, based on the M spatial relationship diagrams, the N recommended loading schemes are determined from the M original loading schemes.

[0190] In one possible design, when the processing unit 803 obtains M original loading schemes by using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, it can specifically be used to: obtain a first set of original loading schemes for each object by using a tree search algorithm based on the information of each of the T objects, the loading information of each object, and the container information; and then summarize the T sets of first original loading schemes corresponding to the T objects as the M original loading schemes.

[0191] In one possible design, when the processing unit 803 obtains M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, it can specifically be used as follows: First, based on the information of the T objects and the loading information of the T objects, at least one first block and / or at least one second block can be obtained, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; then, based on the information of each first block and / or each second block, and the container information, a tree search algorithm can be used to obtain a set of second original loading schemes for each first block and / or a set of second original loading schemes for each second block; finally, the set of at least one second original loading schemes corresponding to the at least one first block can be summarized, and / or the set of at least one second original loading schemes corresponding to the at least one second block can be summarized as the M original loading schemes.

[0192] In one possible design, when the processing unit 803 determines the N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams, it can specifically be used to: firstly, calculate the diversity function value of each of the M original loading schemes based on the M spatial location relationship diagrams; then, according to the order of the diversity function values ​​of the M original loading schemes from largest to smallest, select the original loading schemes corresponding to the first N diversity function values ​​as the N recommended loading schemes.

[0193] In one possible design, when the processing unit 803 determines the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it may specifically be used to: determine a target spatial position relationship diagram that is the same as the first spatial position relationship diagram from among the N second spatial position relationship diagrams; and take the recommended loading scheme corresponding to the target spatial position relationship diagram as the final loading scheme for the first object to be loaded.

[0194] In one possible design, the first spatial position relationship diagram exists among the N second spatial position relationship diagrams. When the processing unit 803 determines the final loading scheme of the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it can specifically be used to: determine that the first spatial position relationship diagram is included in i of the N second spatial position relationship diagrams, where i is a positive integer less than or equal to N; and select the recommended loading scheme with the highest loading rate from the i recommended loading schemes corresponding to the i second spatial position relationship diagrams as the final loading scheme of the first object to be loaded.

[0195] In one possible design, the first spatial position relationship diagram does not exist in the N second spatial position relationship diagrams. Specifically, when the processing unit 803 determines the final loading scheme for the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it can: firstly, adjust the N second spatial position relationship diagrams according to the first spatial position relationship diagram to obtain the adjustment range values ​​of the N second spatial position relationship diagrams; then, take the recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment range value as the final loading scheme for the first object to be loaded.

[0196] In one possible design, the processing unit 803 can also be used to: when the adjustment amplitude values ​​of the N second spatial position relationship diagrams are all greater than a set second threshold, it can use a tree search algorithm to obtain Q loading schemes based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, where Q is a positive integer; the loading information of the first object to be loaded includes the loading order and stacking layer number of the first object to be loaded; the loading scheme with the highest loading rate among the Q loading schemes can be used as the final loading scheme of the first object to be loaded.

[0197] In one possible design, the information of the first object to be loaded includes at least one or more of the following: the length of the first object to be loaded, the height of the first object to be loaded, the width of the first object to be loaded, and the weight of the first object to be loaded.

[0198] Based on the same technical concept, embodiments of this application provide an object loading device. This object loading device may include modules or units corresponding to the methods / operations / steps / actions described in the above method embodiments. These modules or units may be hardware circuits, software, or a combination of hardware circuits and software. The device may have the following characteristics: Figure 9The structure shown.

[0199] like Figure 9 As shown, the device 900 may include an input module 901, a generation module 902, a processing module 903, and an output module 904. The following is a detailed description of each unit.

[0200] It should be understood that the aforementioned input module 901 and output module 904 can be communication interface circuits to realize the reception and / or transmission of data and / or information. The aforementioned input module 901 and output module 904 can also be transceivers to realize the reception and / or transmission of data and / or information. Therefore, the specific form of the input and output modules is not limited in this application. Furthermore, the processing module in the loading device can actually be a processor.

[0201] The input module 901 can be used to acquire information about the first object to be loaded and the current loading scheme; the generation module 902 can be used to generate a first spatial position relationship diagram based on the information about the first object to be loaded and the current loading scheme; the input module 901 can also be used to acquire N recommended loading schemes and N second spatial position relationship diagrams, each second spatial position relationship diagram being generated based on a corresponding recommended loading scheme, where N is a positive integer; the processing module 903 can be used to determine the final loading scheme for the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams; the output module 904 can be used to output the final loading scheme for the first object to be loaded.

[0202] In one possible design, the processing module 903 can also be used to: before the input module 901 obtains N recommended loading schemes and the spatial relationship diagrams of the N recommended loading schemes, firstly, obtain information about T objects and loading information of the T objects, as well as container information for loading the objects, through the input module 901; where T is a positive integer; the T objects include the first object to be loaded, and the loading information includes the loading order and stacking layer number of the T objects; then, based on the information about the T objects, the loading information of the T objects, and the container information, use a tree search algorithm to obtain M original loading schemes, wherein the loading rate of the M loading schemes is greater than or equal to a set first threshold; where M is a positive integer greater than or equal to N; finally, generate M spatial relationship diagrams based on the M original loading schemes, with each original loading scheme corresponding to one spatial relationship diagram; and determine the N recommended loading schemes from the M original loading schemes based on the M spatial relationship diagrams.

[0203] In one possible design, when the processing module 903 obtains M original loading schemes by using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, it can specifically be used to: firstly, based on the information of each of the T objects, the loading information of each object, and the container information, use a tree search algorithm to obtain a first set of original loading schemes for each object; then, summarize the T first sets of original loading schemes corresponding to the T objects as the M original loading schemes.

[0204] In one possible design, when the processing module 903 obtains M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, it can specifically be used as follows: First, based on the information of the T objects and the loading information of the T objects, at least one first block and / or at least one second block can be obtained, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; further, based on the information of each first block and / or each second block, and the container information, a tree search algorithm can be used to obtain a set of second original loading schemes for each first block and / or a set of second original loading schemes for each second block; finally, the set of at least one second original loading schemes corresponding to the at least one first block can be summarized, and / or the set of at least one second original loading schemes corresponding to the at least one second block can be summarized as the M original loading schemes.

[0205] In one possible design, when the processing module 903 determines the N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams, it can specifically be used to: firstly, calculate the diversity function value of each of the M original loading schemes based on the M spatial location relationship diagrams; then, according to the order of the diversity function values ​​of the M original loading schemes from largest to smallest, select the original loading schemes corresponding to the first N diversity function values ​​as the N recommended loading schemes.

[0206] In one possible design, when the processing module 903 determines the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it can specifically be used to: determine a target spatial position relationship diagram that is the same as the first spatial position relationship diagram from among the N second spatial position relationship diagrams; and then use the recommended loading scheme corresponding to the target spatial position relationship diagram as the final loading scheme for the first object to be loaded.

[0207] In another possible design, the first spatial position relationship diagram exists in one of the N second spatial position relationship diagrams. Specifically, when the processing module 903 determines the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it can: firstly determine that the first spatial position relationship diagram is included in one of the i second spatial position relationship diagrams contained in the N second spatial position relationship diagrams, where i is a positive integer less than or equal to N; and then, among the i recommended loading schemes corresponding to the i second spatial position relationship diagrams, the recommended loading scheme with the highest loading rate can be used as the final loading scheme for the first object to be loaded.

[0208] In one possible design, the first spatial position relationship diagram does not exist in the N second spatial position relationship diagrams. Specifically, when the processing module 903 determines the final loading scheme for the first object to be loaded from the N recommended loading schemes based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, it can: first, adjust the N second spatial position relationship diagrams according to the first spatial position relationship diagram to obtain the adjustment range values ​​of the N second spatial position relationship diagrams; then, take the recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment range value as the final loading scheme for the first object to be loaded.

[0209] In one possible design, the processing module 903 can also be used to: when the adjustment amplitude values ​​of the N second spatial position relationship diagrams are all greater than a set second threshold, it can use a tree search algorithm to obtain Q loading schemes based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, where Q is a positive integer; the loading information of the first object to be loaded includes the loading order and stacking layer number of the first object to be loaded; furthermore, the loading scheme with the highest loading rate among the Q loading schemes can be used as the final loading scheme of the first object to be loaded.

[0210] In one possible design, the information of the first object to be loaded includes at least one or more of the following: the length of the first object to be loaded, the height of the first object to be loaded, the width of the first object to be loaded, and the weight of the first object to be loaded.

[0211] Based on the same concept as the above-described method embodiments, this application also provides a computer-readable storage medium storing some instructions. When these instructions are invoked and executed by a computer, the computer can perform the methods involved in any possible design of the above-described method embodiments. In this application, the computer-readable storage medium is not limited; for example, it can be RAM (random-access memory), ROM (read-only memory), etc.

[0212] Based on the same concept as the above method embodiments, this application also provides a computer program product that, when executed by a computer, can perform the methods involved in the method embodiments and any possible designs of the above method embodiments.

[0213] Based on the same concept as the above method embodiments, this application also provides a chip, which may include a processor and an interface circuit, for performing the methods involved in any possible implementation of the above method embodiments, wherein "coupling" means that two components are directly or indirectly combined with each other, and such combination may be fixed or movable, and such combination may allow fluid, electricity, electrical signals or other types of signals to communicate between the two components.

[0214] In summary, the embodiments of this application can quickly and flexibly determine the optimal or near-optimal loading scheme from multiple recommended loading schemes based on the real-time arrival information of the current object and the current container loading scheme, so as to guide manual or machine loading and thus improve the loading rate of the container.

[0215] Through the above description of the embodiments, those skilled in the art will clearly understand that the embodiments of this application can be implemented in hardware, firmware, or a combination thereof. When implemented in software, the above functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium accessible to a computer. For example, but not limited to, computer-readable media can include RAM, ROM, electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible to a computer. Furthermore, any connection can suitably be a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used in embodiments of this application, disks and discs include compact discs (CDs), laser discs, optical discs, digital video discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically magnetically copy data, while discs optically copy data using lasers. The combinations above should also be included within the scope of protection for computer-readable media.

[0216] In summary, the above descriptions are merely embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made based on the disclosure of this application should be included within the scope of protection of this application.

Claims

1. A method for loading an object, characterized in that, include: Obtain information about the first object to be loaded and the current loading plan; Based on the information of the first object to be loaded and the current loading scheme, a first spatial position relationship diagram is generated; Obtain N recommended loading schemes and N second spatial location relationship graphs. Each second spatial location relationship graph is generated based on a corresponding recommended loading scheme, and N is a positive integer. Based on the first spatial location relationship diagram and the N second spatial location relationship diagrams, the final loading scheme for the first object to be loaded is determined from the N recommended loading schemes.

2. The method according to claim 1, characterized in that, Before obtaining the N recommended loading schemes and the spatial relationship diagram of the N recommended loading schemes, the following steps are also included: Obtain information about T objects and their loading information, as well as information about the containers used to load the objects; where T is a positive integer; the T objects include the first object to be loaded, and the loading information includes the loading order and stacking layer number of the T objects; Based on the information of the T objects, the loading information of the T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, wherein the loading rate of the M loading schemes is greater than or equal to a set first threshold; and M is a positive integer greater than or equal to N. Based on the M original loading schemes, generate M spatial position relationship diagrams, with each original loading scheme corresponding to one spatial position relationship diagram; Based on the M spatial location relationship diagrams, the N recommended loading schemes are determined from the M original loading schemes.

3. The method according to claim 2, characterized in that, Based on the information of the T objects, the loading information of the T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, including: Based on the information of each of the T objects, the loading information of each object, and the container information, a tree search algorithm is used to obtain the first original loading scheme set for each object. The T sets of the first original loading schemes corresponding to the T objects are summarized as the M original loading schemes.

4. The method according to claim 2, characterized in that, Based on the information of the T objects, the loading information of the T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, including: Based on the information of the T objects and the loading information of the T objects, information of at least one first block and / or information of at least one second block is obtained, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; Based on the information of each first block and / or each second block, as well as the container information, a tree search algorithm is used to obtain a set of second original loading schemes for each first block and / or a set of second original loading schemes for each second block; The at least one set of at least one second original loading schemes corresponding to the at least one first block is summarized, and / or the at least one set of at least one second original loading schemes corresponding to the at least one second block is summarized, as the M original loading schemes.

5. The method according to claim 2, characterized in that, The step of determining the N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams includes: Based on the M spatial location relationship diagrams, calculate the diversity function value of each of the M original loading schemes; Based on the diversity function values ​​of the M original loading schemes in descending order, the original loading schemes corresponding to the first N diversity function values ​​are selected as the N recommended loading schemes.

6. The method according to claim 1, characterized in that, The step of determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes, based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, includes: Among the N second spatial position relationship diagrams, a target spatial position relationship diagram that is the same as the first spatial position relationship diagram is determined; The recommended loading scheme corresponding to the target spatial location relationship map shall be used as the final loading scheme for the first object to be loaded.

7. The method according to claim 1, characterized in that, The first spatial position relationship diagram exists in the N second spatial position relationship diagrams; The step of determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes, based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, includes: It is determined that i of the N second spatial position relationship diagrams contain the first spatial position relationship diagram, where i is a positive integer less than or equal to N; Among the i recommended loading schemes corresponding to the i second spatial location relationship diagrams, the recommended loading scheme with the highest loading rate is selected as the final loading scheme for the first object to be loaded.

8. The method according to claim 1, characterized in that, The first spatial position relationship diagram does not exist in any of the N second spatial position relationship diagrams; The step of determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes, based on the first spatial position relationship diagram and the N second spatial position relationship diagrams, includes: Based on the first spatial position relationship diagram, the N second spatial position relationship diagrams are adjusted to obtain the adjustment range values ​​of the N second spatial position relationship diagrams; The recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment value shall be taken as the final loading scheme for the first object to be loaded.

9. The method according to claim 8, characterized in that, The method further includes: When the adjustment values ​​of the N second spatial position relationship diagrams are all greater than the set second threshold, a tree search algorithm is used to obtain Q loading schemes based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, where Q is a positive integer; the loading information of the first object to be loaded includes the loading order and stacking layer of the first object to be loaded. The loading scheme with the highest loading rate among the Q loading schemes is taken as the final loading scheme for the first object to be loaded.

10. The method according to any one of claims 1 to 9, characterized in that, The information of the first object to be loaded includes at least one of the following: The length of the first object to be loaded, the height of the first object to be loaded, the width of the first object to be loaded, and the weight of the first object to be loaded.

11. A device for loading an object, characterized in that, include: Input module, generation module, processing module, and output module; The input module is used to obtain information about the first object to be loaded and the current loading scheme; The generation module is used to generate a first spatial position relationship diagram based on the information of the first object to be loaded and the current loading scheme. The input module is also used to obtain N recommended loading schemes and N second spatial position relationship diagrams, each second spatial position relationship diagram being generated based on a corresponding recommended loading scheme, where N is a positive integer; The processing module is used to determine the final loading scheme of the first object to be loaded from among the N recommended loading schemes, based on the first spatial position relationship diagram and the N second spatial position relationship diagrams. The output module is used to output the final loading plan for the first object to be loaded.

12. The apparatus according to claim 11, characterized in that, The processing module is further configured to: Before the input module obtains N recommended loading schemes and the spatial relationship diagram of the N recommended loading schemes, the input module obtains information on T objects and loading information on the T objects, as well as container information for loading the objects; T is a positive integer; the T objects include the first object to be loaded, and the loading information includes the loading order and stacking layer number of the T objects; Based on the information of the T objects, the loading information of the T objects, and the container information, a tree search algorithm is used to obtain M original loading schemes, wherein the loading rate of the M loading schemes is greater than or equal to a set first threshold; and M is a positive integer greater than or equal to N. Based on the M original loading schemes, generate M spatial position relationship diagrams, with each original loading scheme corresponding to one spatial position relationship diagram; Based on the M spatial location relationship diagrams, the N recommended loading schemes are determined from the M original loading schemes.

13. The apparatus according to claim 12, characterized in that, The processing module, when obtaining M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, is specifically used for: Based on the information of each of the T objects, the loading information of each object, and the container information, a tree search algorithm is used to obtain the first original loading scheme set for each object. The T sets of the first original loading schemes corresponding to the T objects are summarized as the M original loading schemes.

14. The apparatus according to claim 12, characterized in that, The processing module, when obtaining M original loading schemes using a tree search algorithm based on the information of the T objects, the loading information of the T objects, and the container information, is specifically used for: Based on the information of the T objects and the loading information of the T objects, information of at least one first block and / or information of at least one second block is obtained, wherein the first block is obtained by combining at least one object to be loaded, and the second block is obtained by combining at least one first block; Based on the information of each first block and / or each second block, as well as the container information, a tree search algorithm is used to obtain a set of second original loading schemes for each first block and / or a set of second original loading schemes for each second block; The at least one set of at least one second original loading schemes corresponding to the at least one first block is summarized, and / or the at least one set of at least one second original loading schemes corresponding to the at least one second block is summarized, as the M original loading schemes.

15. The apparatus according to claim 12, characterized in that, The processing module, when determining the N recommended loading schemes from the M original loading schemes based on the M spatial location relationship diagrams, specifically performs the following: Based on the M spatial location relationship diagrams, calculate the diversity function value of each of the M original loading schemes; Based on the diversity function values ​​of the M original loading schemes in descending order, the original loading schemes corresponding to the first N diversity function values ​​are selected as the N recommended loading schemes.

16. The apparatus according to claim 11, characterized in that, The processing module, when determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship map and the N second spatial position relationship maps, is specifically used for: Among the N second spatial position relationship diagrams, a target spatial position relationship diagram that is the same as the first spatial position relationship diagram is determined; The recommended loading scheme corresponding to the target spatial location relationship map shall be used as the final loading scheme for the first object to be loaded.

17. The apparatus according to claim 11, characterized in that, The first spatial position relationship diagram exists in the N second spatial position relationship diagrams; The processing module, when determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship map and the N second spatial position relationship maps, is specifically used for: It is determined that i of the N second spatial position relationship diagrams contain the first spatial position relationship diagram, where i is a positive integer less than or equal to N; Among the i recommended loading schemes corresponding to the i second spatial location relationship diagrams, the recommended loading scheme with the highest loading rate is selected as the final loading scheme for the first object to be loaded.

18. The apparatus according to claim 11, characterized in that, The first spatial position relationship diagram does not exist in any of the N second spatial position relationship diagrams; The processing module, when determining the final loading scheme for the first object to be loaded from among the N recommended loading schemes based on the first spatial position relationship map and the N second spatial position relationship maps, is specifically used for: Based on the first spatial position relationship diagram, the N second spatial position relationship diagrams are adjusted to obtain the adjustment range values ​​of the N second spatial position relationship diagrams; The recommended loading scheme corresponding to the second spatial position relationship diagram with the smallest adjustment value shall be taken as the final loading scheme for the first object to be loaded.

19. The apparatus according to claim 18, characterized in that, The processing module is further configured to: When the adjustment values ​​of the N second spatial position relationship diagrams are all greater than the set second threshold, a tree search algorithm is used to obtain Q loading schemes based on the information of the first object to be loaded, the loading information of the first object to be loaded, and the container information, where Q is a positive integer; the loading information of the first object to be loaded includes the loading order and stacking layer of the first object to be loaded. The loading scheme with the highest loading rate among the Q loading schemes is taken as the final loading scheme for the first object to be loaded.

20. The apparatus according to any one of claims 11 to 19, characterized in that, The information of the first object to be loaded includes at least one of the following: The length of the first object to be loaded, the height of the first object to be loaded, the width of the first object to be loaded, and the weight of the first object to be loaded.

21. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 10.

Images