A method, device and storage medium for processing loading data

Abstract

The present application discloses a loading data processing method, device and storage medium, the loading data processing method comprising: obtaining information of an object to be loaded, information of a carrier and information of a loading container; calling a linear programming model to determine first stacking data of an object to be stacked based on first object information and information of a carrier; calling a spatial search algorithm to simulate stacking of the object to be loaded on at least one subspace of the current remaining space of a target loading container based on the information of the object to be loaded, the first stacking data and information of a loading container, and obtaining second stacking data of the object to be loaded in the target loading container; calling a target local search algorithm and a target reinforcement learning algorithm to optimize the second stacking data and obtain target stacking data corresponding to the second stacking data. The loading data processing method disclosed in the present application can quickly and accurately obtain an optimized stacking method of the object to be loaded.

Description

Technical Field

[0001] The present application relates to the field of loading technology, and in particular to a loading data processing method, device and storage medium. Background Art

[0002] In the process of transporting objects to be loaded, the loading rate of the loading container accounts for a large proportion of the transportation cost. Therefore, reasonably matching the loading container with the objects to be loaded to improve the loading rate of the loading container becomes an important means to reduce transportation costs. However, there are many types of objects to be loaded, and different objects to be loaded correspond to a variety of loading rules. In addition, transportation includes two storage types: carriers (such as pallets) and loading containers of different sizes (such as truck boxes). These rules in actual operations will affect the transportation loading rate of the objects to be loaded.

[0003] It is difficult to quickly obtain a better loading solution overall if only manual loading is relied upon, and the loading rate of the loading container is related to manual experience. Therefore, it is necessary to quickly obtain a reasonable loading solution through intelligent algorithms. Three-dimensional loading is an NP-hard problem, and the loading scene of the object to be loaded is complex and the calculation complexity is high. Existing loading algorithms often have strong problem dependence, poor generalization performance, or too much calculation and low processing efficiency. Summary of the invention

[0004] The present application provides a loading data processing method, device and storage medium. By using preset first stacking data as the input item of the spatial search algorithm and simultaneously simulating the stacking of multiple subspaces of the remaining space of the target loading container, the efficiency of the spatial search algorithm can be improved; through the combined calculation of the local search algorithm and the reinforcement learning algorithm, the search range of the algorithm in the solution space is improved, and the algorithm performance is also improved, thereby significantly improving the loading data processing efficiency; the present application can quickly and accurately obtain the optimized stacking method of the object to be loaded.

[0005] In one aspect, the present application provides a loading data processing method, the method comprising:

[0006] Acquire information of objects to be loaded, information of carriers and information of loading containers, wherein the information of objects to be loaded includes first object information of at least one object to be assembled or second object information of at least one object not to be assembled;

[0007] In the case where the object to be assembled exists, a linear programming model is called to determine first stacking data of the object to be assembled based on the first object information and the carrier information; the first stacking data indicates a stacking mode of a single layer of the object to be assembled on the carrier;

[0008] Calling a spatial search algorithm, based on the information of the object to be loaded, the first stacking data and the loading container information, performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container, to obtain second stacking data of the object to be loaded in the target loading container; wherein the at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container;

[0009] A target local search algorithm and a target reinforcement learning algorithm are called to perform data optimization on the second stacking data to obtain target stacking data corresponding to the second stacking data, wherein the target stacking data indicates a target stacking mode of the object to be loaded in the target loading container.

[0010] Optionally, the first object information includes size information of the object to be assembled, and the object to be assembled includes a non-standard object. Before calling the linear programming model to determine the first stacking data of the object to be assembled based on the first object information and the carrier information, the method further includes:

[0011] The non-standard object is dimensionally mapped based on a preset shape to obtain dimension information of the non-standard object.

[0012] Optionally, before performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information, the loading data processing method further includes:

[0013] Obtaining stackable plane information in the current remaining space of the target loading container;

[0014] Based on the stackable plane information, space division is performed for each stackable plane in the current remaining space of the target loading container to obtain the at least one subspace.

[0015] Optionally, the performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information to obtain second stacking data of the object to be loaded in the target loading container includes:

[0016] Acquiring subspace information of the at least one subspace;

[0017] For each subspace of the target loading container, calling the space search algorithm to determine at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object from the currently remaining objects to be loaded based on the subspace information, the information of the object to be loaded and the first stacking data;

[0018] Based on at least one target loading object corresponding to each subspace and stacking parameters of the target loading object, updating the stackable plane information of the target loading container;

[0019] Repeat the steps of dividing the space for each stackable plane, acquiring the subspace information of the at least one subspace, and determining at least one target loading object corresponding to each subspace from the currently remaining objects to be loaded, until all the objects to be loaded have completed simulated stacking or there is no valid remaining space in the target loading container;

[0020] The second stacking data is generated based on the stacking parameters of the target loading objects in the target loading container when all the objects to be loaded have completed simulated stacking or there is no effective remaining space.

[0021] Optionally, calling the spatial search algorithm to determine at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object from the currently remaining objects to be loaded based on the subspace information, the information of the objects to be loaded and the first stacking data includes:

[0022] In the case where the non-supported objects exist, calling the spatial search algorithm to determine at least one target non-supported object and stacking parameters of the target non-supported object from the currently remaining non-supported objects based on the subspace information and the second object information;

[0023] When the objects to be assembled exist, the spatial search algorithm is called to determine at least one target object to be assembled, the corresponding target carrier and the stacking parameters of the target object to be assembled from the currently remaining objects to be assembled based on the subspace information, the first object information and the first stacking data.

[0024] Optionally, calling a target local search algorithm and a target reinforcement learning algorithm to perform data optimization on the second stacking data to obtain target stacking data corresponding to the second stacking data includes:

[0025] Calling the target local search algorithm, optimizing and calculating the second stacking data based on the current operation operator until an iteration termination condition is satisfied, and obtaining the target stacking data;

[0026] Among them, during the optimization calculation process, the intermediate parameters of the optimization calculation of the target local search algorithm are extracted; and the target reinforcement learning algorithm is called to update the current operation operator of the target local search algorithm based on the intermediate parameters.

[0027] Optionally, calling the target local search algorithm to optimize the second stacking data based on the current operation operator until an iteration termination condition is satisfied to obtain the target stacking data includes:

[0028] determining a current initial solution of the target local search algorithm based on the second stacked data;

[0029] Calling the target local search algorithm, searching and calculating the current initial solution based on the current operation operator, and obtaining the current local optimal solution;

[0030] If the iteration termination condition is not met, updating the current local optimal solution to the current initial solution;

[0031] Calling the target reinforcement learning algorithm to update the current operation operator;

[0032] Repeat the step of calling the target local search algorithm, searching and calculating the current initial solution based on the current operation operator to obtain the current local optimal solution, and when the iteration termination condition is met, determine the current local optimal solution that meets the iteration termination condition as the target stacking data.

[0033] Optionally, the target local search algorithm is a variable neighborhood search algorithm, and the target reinforcement learning algorithm is a Q-learning algorithm.

[0034] On the other hand, the present application provides a loading data processing device, the loading data processing device comprising:

[0035] A first acquisition module: used to acquire information about the object to be loaded, information about the carrier, and information about the loading container, wherein the information about the object to be loaded includes first object information of at least one object to be assembled or second object information of at least one object not to be assembled;

[0036] A first determining module is used for, when the object to be assembled exists, calling a linear programming model to determine first stacking data of the object to be assembled based on the first object information and the carrier information; the first stacking data indicates a stacking mode of a single layer of the object to be assembled on the carrier;

[0037] The first processing module is used to call a spatial search algorithm to perform simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information, so as to obtain second stacking data of the object to be loaded in the target loading container; wherein the at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container;

[0038] The first optimization module is used to call a target local search algorithm and a target reinforcement learning algorithm to optimize the second stacking data to obtain target stacking data corresponding to the second stacking data, wherein the target stacking data indicates a target stacking method of the object to be loaded in the target loading container.

[0039] On the other hand, the present application provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor as the loading data processing method described in an embodiment of the present application.

[0040] The present application provides a loading data processing method, device and storage medium, which have the following beneficial effects:

[0041] The present application uses the preset first stacking data as the input item of the spatial search algorithm and simulates the stacking of multiple subspaces of the remaining space of the target loading container at the same time, thereby improving the efficiency of the spatial search algorithm; through the combined calculation of the local search algorithm and the reinforcement learning algorithm, the search range of the algorithm in the solution space is improved, and the algorithm performance is also improved, thereby significantly improving the loading data processing efficiency; the present application can quickly and accurately obtain the optimized stacking method of the objects to be loaded. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A flowchart of a loading data processing method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of a flow chart of a subspace acquisition method provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of a stackable plane of the current remaining space of a target loading container in one embodiment;

[0046] Figure 4 A schematic flow chart of a method for simulating stacking of objects to be loaded in a subspace provided in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of a flow chart of a method for optimizing second stacking data provided in an embodiment of the present application;

[0048] Figure 6 A schematic flow chart of a method for calling a target local search algorithm to obtain a current local optimal solution provided in an embodiment of the present application;

[0049] Figure 7 A flowchart of a method for calling a target reinforcement learning algorithm to update a current operation operator provided in an embodiment of the present application;

[0050] Figure 8 A schematic diagram of the structure of a loading data processing device provided in an embodiment of the present application;

[0051] Fig. 9 A hardware structure block diagram of an electronic device for implementing a loading data processing method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0053] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0054] The following combination Figure 1 A loading data processing method provided in an embodiment of the present application is introduced, which can be applied to the field of loading technology, such as the packing of automobile parts.

[0055] Figure 1 For a flow chart of a loading data processing method provided in an embodiment of the present application, please refer to Figure 1 , a loading data processing method provided by an embodiment of the present application includes:

[0056] S101. Obtaining information about the object to be loaded, information about the carrier, and information about the loading container, wherein the information about the object to be loaded includes information about at least one first object to be supported by a group or information about at least one second object to be supported by a group;

[0057] In the embodiment of the present application, the object to be loaded refers to an object that needs to be loaded, such as goods that need to be loaded; the object to be loaded includes at least one object to be assembled and at least one non-assembled object; the object to be assembled refers to an object to be loaded that needs to be pre-loaded on a carrier, which may include but is not limited to a rack and a small-sized object to be loaded, and the non-assembled object refers to an object to be loaded that does not need to be pre-loaded on a carrier and is directly loaded in a loading container; the first object information of the object to be assembled includes but is not limited to the shape, size, loadable direction, weight, load-bearing, stability information of the object to be assembled and the sorting information of the object to be assembled sorted according to the first preset rule, and the second object information of the non-assembled object includes but is not limited to the shape, size, loadable direction, weight, load-bearing, stability information of the non-assembled object and the sorting information of the non-assembled object sorted according to the second preset rule; the information of the object to be loaded includes at least one first object information of the object to be assembled or at least one second object information of the non-assembled object. Among them, the first preset rule and the second preset rule include but are not limited to from large to small and from heavy to light; the target object to be loaded is selected in sequence based on the sorting information of the object to be assembled and the sorting information of the non-assembled object.

[0058] In an embodiment of the present application, a carrier refers to a loading device having a accommodating surface, such as a pallet, and the carrier information includes but is not limited to the length, width, limited loadable height of the carrier, and carrier sorting information sorted according to a third preset rule; wherein the third preset rule includes but is not limited to the area of ​​the accommodating surface of the carrier from large to small; and the target carriers are selected in sequence based on the carrier sorting information.

[0059] In the embodiment of the present application, the loading container refers to a loading device with a storage space, such as a carriage and a container, and the loading container information includes but is not limited to the length, width, and height of the storage space of the loading container and the loading container sorting information sorted according to the fourth preset rule. The fourth preset rule includes but is not limited to the volume of the storage space of the loading container from large to small; the target loading container is selected in sequence based on the loading container sorting information.

[0060] S103. In the case where there is an object to be supported by a group, a linear programming model is called to determine first stacking data of the object to be supported by a group based on the first object information and the carrier information; the first stacking data indicates a stacking method of a single-layer object to be supported by a group on a carrier;

[0061] In the embodiment of the present application, the linear programming model is a mixed integer linear programming model.

[0062] In an embodiment of the present application, the objects to be loaded include objects that need to be assembled, and a linear programming model is called to calculate and determine first stacking data of the objects that need to be assembled based on first object information of the objects that need to be assembled and carrier information, wherein the first stacking data refers to the stacking method of a single layer of objects that need to be assembled on the carrier, including but not limited to the placement direction, the total number of placements, the number of placements in the width direction of the carrier, and the number of placements in the length direction of the carrier of the single layer of objects that need to be assembled on the carrier.

[0063] In the embodiment of the present application, the objects to be assembled and supported include non-standard objects, such as material racks;

[0064] In the embodiment of the present application, there are non-standard objects, and before the linear programming model is called to determine the first stacking data of the object to be stacked based on the first object information and the carrier information, the loading data processing method further includes:

[0065] The size of the non-standard object is mapped based on the preset shape to obtain the size information of the non-standard object.

[0066] Specifically, the preset shape includes a regular hexahedron such as a cuboid.

[0067] In an embodiment of the present application, the non-standard object is size mapped based on a preset shape to obtain size information of the non-standard object, that is, the non-standard object is abstracted into a minimum cuboid or minimum cube containing the non-standard object to obtain size information of the abstract minimum cuboid or minimum cube.

[0068] When loading non-standard objects that need to be assembled, it is necessary to pre-load them on the target carrier, specifically: call the linear programming model, calculate and determine the first stacking data of the non-standard objects that need to be assembled based on the first object information of the non-standard objects that need to be assembled and the carrier information; wherein the first stacking data refers to the stacking method of a single-layer non-standard objects that need to be assembled on the carrier, including but not limited to the placement direction of the single-layer non-standard objects that need to be assembled on the carrier, the total number of placements, the number of placements in the width direction of the carrier, and the number of placements in the length direction of the carrier.

[0069] In one embodiment, the objects to be supported include objects to be loaded whose size is smaller than a preset threshold, wherein the preset threshold may be 20 cm, that is, objects to be loaded are considered to be objects to be supported if their length, width and height are all smaller than 20 cm.

[0070] When loading objects to be assembled whose sizes are smaller than a preset threshold, it is necessary to pre-load them on the target carrier, specifically: call the linear programming model, and calculate and determine the first stacking data of the objects to be assembled whose sizes are smaller than the preset threshold based on the first object information of the objects to be assembled whose sizes are smaller than the preset threshold and the carrier information; wherein the first stacking data refers to the stacking method of the objects to be assembled whose single-layer sizes are smaller than the preset threshold on the carrier, including but not limited to the placement direction, the total number of placements, the number of placements in the width direction of the carrier, and the number of placements in the length direction of the carrier of the objects to be assembled whose single-layer sizes are smaller than the preset threshold on the carrier.

[0071] When there are objects to be stacked, a linear programming model is called to determine first stacking data of the objects to be stacked based on first object information and carrier information, thereby improving the efficiency of the spatial search algorithm.

[0072] S105. Calling a spatial search algorithm, based on the information of the object to be loaded, the first stacking data and the loading container information, performs a simulated stacking process of the object to be loaded on at least one subspace of the current remaining space of the target loading container, and obtains second stacking data of the object to be loaded in the target loading container; wherein at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container;

[0073] Figure 2 A flow chart of a subspace acquisition method provided in an embodiment of the present application, before performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information, the loading data processing method also includes a subspace acquisition method, please refer to Figure 2 , a subspace acquisition method provided in an embodiment of the present application includes:

[0074] S201. Obtaining stackable plane information in the current remaining space of the target loading container;

[0075] In the embodiment of the present application, loading containers are selected as target loading containers in sequence, and objects to be loaded are selected in sequence to perform simulated stacking processing in at least one subspace of the current remaining space of the target loading container. Before performing simulated stacking processing of the objects to be loaded on at least one subspace of the current remaining space of the target loading container, it is necessary to obtain stackable plane information in the current remaining space of the target loading container.

[0076] Among them, the current remaining space of the target loading container refers to the remaining storage space of the target loading container in the current state, including the storage space of the complete target loading container when the objects to be loaded are not stacked and the remaining storage space of the remaining target loading container when the objects to be loaded are partially stacked; the stackable plane refers to the plane on which the objects to be loaded can be stacked, and the stackable plane information includes but is not limited to the number of stackable planes and the size information of each stackable plane, wherein the number of stackable planes can be 1, 2...N.

[0077] Figure 3 This is a schematic diagram of the stackable plane of the current remaining space of the target loading container in an embodiment. Please refer to Figure 3 In one embodiment, after the first object to be loaded and the second object to be loaded are stacked in the target loading container, there are three stackable planes in the remaining accommodation space of the target loading container, namely, the first stacking plane, the second stacking plane and the third stacking plane.

[0078] S203. Based on the stackable plane information, perform space division for each stackable plane in the current remaining space of the target loading container to obtain at least one subspace.

[0079] Based on the stackable plane information, for each stackable plane, the current remaining space of the target loading container is divided into at least one subspace; and the at least one subspace is a part of the current remaining space of the target loading container.

[0080] The number of subspaces may be 1, 2, ...N.

[0081] In one embodiment, each subspace is a cuboid or a cube.

[0082] Figure 4 For a flow chart of a method for simulating stacking of objects to be loaded in a subspace provided in an embodiment of the present application, please refer to Figure 4 , a method for simulating stacking of objects to be loaded in a subspace provided in an embodiment of the present application includes:

[0083] S401. Obtain subspace information of at least one subspace;

[0084] In an embodiment of the present application, the current remaining space of the target loading container is divided into at least one subspace based on the stackable plane information, and subspace information of the at least one subspace is obtained, and the subspace information includes but is not limited to the length, width and height information of the subspace.

[0085] S403. For each subspace of the target loading container, calling a spatial search algorithm based on the subspace information, the information of the object to be loaded and the first stacking data, determining at least one target loading object and the stacking parameters of the target loading object corresponding to each subspace from the currently remaining objects to be loaded;

[0086] In the embodiment of the present application, the current remaining space of the target loading container is divided into at least one subspace. For each subspace of the target loading container, a spatial search algorithm is called to determine at least one target loading object and the stacking parameters of the target loading object corresponding to each subspace from the currently remaining objects to be loaded based on the information of each subspace, the information of the object to be loaded and the first stacking data.

[0087] Among them, the target loading object refers to the object to be loaded that can be simulated and stacked in the subspace of the target loading container. If the object to be loaded cannot be simulated and stacked in the subspace of the current remaining space of any target loading container due to overweight or oversize, the next target loading container will be selected in turn for simulated stacking; the stacking parameters of the target loading object refer to the stacking method of the target loading object in the subspace of the target loading container, including the stacking direction information of the target loading object.

[0088] The target loading objects in each subspace of the target loading container are not repeated.

[0089] The determination of the target loading objects and the stacking parameters of the target loading objects in each subspace of the target loading container is performed simultaneously, which can improve the efficiency of the space search algorithm.

[0090] In one embodiment, the objects to be loaded include non-assembly objects, and a spatial search algorithm is called to determine at least one target non-assembly object and the stacking parameters of the target non-assembly object from the currently remaining non-assembly objects based on subspace information and second object information of the non-assembly object; wherein the target non-assembly objects in each subspace of the target loading container are not repeated; wherein the stacking parameters of the target non-assembly object include the first stacking direction information of the target non-assembly object.

[0091] In another embodiment, the objects to be loaded include objects to be assembled, and a spatial search algorithm is called to determine at least one target object to be assembled, a corresponding target carrier, and stacking parameters of the target object to be assembled from the currently remaining objects to be assembled based on subspace information, first object information of the object to be assembled, and first stacking data; wherein the objects to be assembled in each subspace of the target loading container are not repeated; wherein the stacking parameters of the target object to be assembled include the number of stacking layers of the target object to be assembled on the target carrier, and the size information of the whole composed of the target carrier and the target object to be assembled when the simulated stacking is completed, and the second stacking direction information.

[0092] Among them, the stacking number of target objects to be assembled on the target carrier refers to the total number of layers of target objects to be assembled on the target carrier calculated and determined based on the target subspace information and the first object information of the target objects to be assembled. The total number of layers of target objects to be assembled on the target carrier can be 1, 2...N.

[0093] S405. Based on at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object, update the stackable plane information of the target loading container;

[0094] In an embodiment of the present application, based on at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object, the remaining space of the target loading container is calculated, the current remaining space of the target loading container is updated, and the stackable plane information of the current remaining space of the target loading container is updated.

[0095] S407. Repeat the steps of dividing the space for each stackable plane, obtaining subspace information of at least one subspace, and determining at least one target loading object corresponding to each subspace from the currently remaining objects to be loaded, until all objects to be loaded have completed simulated stacking or there is no valid remaining space in the target loading container;

[0096] In one embodiment, the following steps are repeated:

[0097] S203: Based on the stackable plane information, perform space division for each stackable plane in the current remaining space of the target loading container to obtain at least one subspace;

[0098] S401: Acquire subspace information of at least one subspace;

[0099] S403: for each subspace of the target loading container, calling a spatial search algorithm based on the subspace information, the information of the object to be loaded and the first stacking data, to determine at least one target loading object and the stacking parameters of the target loading object corresponding to each subspace from the currently remaining objects to be loaded;

[0100] S405: Based on at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object, update the stackable plane information of the target loading container.

[0101] Until all objects to be loaded are stacked in a target loading container.

[0102] In another embodiment, the following steps are repeated:

[0103] S203: Based on the stackable plane information, perform space division for each stackable plane in the current remaining space of the target loading container to obtain at least one subspace;

[0104] S401: Acquire subspace information of at least one subspace;

[0105] S403: for each subspace of the target loading container, calling a spatial search algorithm based on the subspace information, the information of the object to be loaded and the first stacking data, to determine at least one target loading object and the stacking parameters of the target loading object corresponding to each subspace from the currently remaining objects to be loaded;

[0106] S405: Based on at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object, update the stackable plane information of the target loading container.

[0107] When there is no effective remaining space in the target loading container, but there are remaining objects to be loaded, the next target loading container is selected in turn and the above steps are repeated to perform simulated stacking of the remaining objects to be loaded until all the objects to be loaded have completed simulated stacking.

[0108] S409. Generate second stacking data based on the stacking parameters of the target loading objects in the target loading container when all the objects to be loaded have completed simulated stacking or there is no valid remaining space.

[0109] In one embodiment, when all objects to be loaded have completed simulated stacking in a target loading container, second stacking data is generated based on the stacking parameters of the target loading objects in the target loading container at this time; wherein the second stacking data refers to the stacking method of the objects to be loaded in the target loading container.

[0110] In another embodiment, when there is no effective remaining space in the target loading container, but there are remaining objects to be loaded, the next target loading container is selected in turn and the above steps are repeated to perform simulated stacking of the currently remaining objects to be loaded, until all objects to be loaded have completed simulated stacking, and second stacking data is generated based on the stacking parameters of the target loading objects in the target loading container at this time; wherein the second stacking data refers to the stacking method of the objects to be loaded in the target loading container.

[0111] S107. Calling a target local search algorithm and a target reinforcement learning algorithm to optimize the second stacking data, and obtaining target stacking data corresponding to the second stacking data, the target stacking data indicating a target stacking mode of the objects to be loaded in the target loading container.

[0112] In an embodiment of the present application, a target local search algorithm is called to optimize the second stacking data based on the current operation operator until an iteration termination condition is met, and the iteration is stopped to obtain the target stacking data; wherein the target stacking data refers to the target stacking method of the object to be loaded in the target loading container.

[0113] In an embodiment of the present application, during the optimization calculation process, the intermediate parameters of the optimization calculation in the target local search algorithm are extracted; and the target reinforcement learning algorithm is called to update the current operation operator of the target local search algorithm based on the intermediate parameters of the optimization calculation.

[0114] In the embodiment of the present application, the target local search algorithm is a variable neighborhood search algorithm, and the target reinforcement learning algorithm is a Q-learning algorithm.

[0115] The operator includes a first operator and a second operator.

[0116] The first operation operator is defined as: based on the second stacking data, calculating and determining the stacking layer with the largest size difference of the objects to be loaded in the target loading container, and sorting the objects to be loaded in the stacking layer with the largest size difference from large to small according to the size difference value, determining the objects to be loaded with larger size difference and rotating their stacking method.

[0117] In an embodiment of the present application, the number of objects to be loaded with a large size difference is determined based on the percentage of the number of objects to be loaded in the stacking layer with the largest size difference in the target loading container, which may be specifically 10%; that is, the number of objects to be loaded that need to be rotated is determined based on 10% of the number of objects to be loaded in the stacking layer with the largest size difference, and the objects to be loaded that need to be rotated are determined based on the sorting of the objects to be loaded in the stacking layer with the largest size difference from large to small.

[0118] The second operation operator is defined as: based on the loading rates of the loading containers, determining the loading container with the lowest loading rate, and randomly exchanging the stacking order of at least two objects to be loaded in the loading container with the lowest loading rate.

[0119] Among them, the loading rate of each loading container is calculated by the following formula:

[0120]

[0121] Where r is the loading rate of the target loading container; l i 、w i and h i are the length, width and height of the object i to be loaded in the target loading container; L, W and H are the length, width and height of the target loading container respectively.

[0122] In the embodiment of the present application, the stacking order of two objects to be loaded in the loading container with the lowest loading rate is randomly exchanged.

[0123] In one embodiment, all objects to be loaded are stacked in a target loading container in a simulated manner, a target local search algorithm is called, and the second stacking data is optimized and calculated based on the first operation operator.

[0124] In another embodiment, all objects to be loaded are stacked in simulated form in at least two target loading containers, a target local search algorithm is called, and the second stacking data is optimized and calculated based on the first operation operator and the second operation operator.

[0125] Figure 5 For a flow chart of a method for obtaining target stacking data provided in an embodiment of the present application, please refer to Figure 5 , a method for obtaining target stacking data provided by an embodiment of the present application includes:

[0126] S501. Determine the current initial solution of the target local search algorithm based on the second stacked data;

[0127] In the embodiment of the present application, a current initial solution of the target local search algorithm is determined based on the second stacked data.

[0128] S503. Call the target local search algorithm, search and calculate the current initial solution based on the current operation operator, and obtain the current local optimal solution;

[0129] Figure 6 For a flow chart of a method for calling a target local search algorithm to obtain the current local optimal solution provided in an embodiment of the present application, please refer to Figure 6 , a method for calling a target local search algorithm to obtain a current local optimal solution provided by an embodiment of the present application includes:

[0130] S601. Perturb the current initial solution and obtain the neighborhood solution of the current disturbance;

[0131] In the embodiment of the present application, a target local search algorithm is called to perturb the current initial solution and obtain a neighborhood solution of the current perturbation.

[0132] S603. Based on the current operation operator, search and calculate the neighborhood solution of the current disturbance to obtain the neighborhood solution of the current local search;

[0133] In an embodiment of the present application, a target local search algorithm is called to search and calculate a neighborhood solution of the current disturbance to obtain a neighborhood solution of the current local search.

[0134] S605. Calculate the objective function value of the neighborhood solution of the current local search;

[0135] S607. If the objective function value of the neighborhood solution of the current local search is better, the neighborhood solution of the current local search is updated to the current local optimal solution;

[0136] In an embodiment of the present application, if the objective function value of the neighborhood solution of the current local search is better, the neighborhood solution of the current local search is updated to the current local optimal solution; otherwise, the current local optimal solution is not updated.

[0137] S505. If the iteration termination condition is not met, the current local optimal solution is updated to the current initial solution;

[0138] S507. Call the target reinforcement learning algorithm to update the current operation operator;

[0139] Figure 7 For a flow chart of a method for calling a target reinforcement learning algorithm to update the current operation operator provided in an embodiment of the present application, please refer to Figure 7 , a method for calling a target reinforcement learning algorithm to update a current operation operator provided by an embodiment of the present application includes:

[0140] S701. Obtaining initial parameters of the target reinforcement learning algorithm;

[0141] In the embodiment of the present application, the parameters of the target reinforcement learning algorithm include but are not limited to state, action and Q table. The parameters of the target reinforcement learning algorithm are obtained and initialized to obtain the initial parameters of the target reinforcement learning algorithm.

[0142] S703. Obtaining intermediate parameters of the target local search algorithm based on the actions, states and feedback values ​​of steps S601, S603 and S605;

[0143] S705. Perform random exploration actions based on the intermediate parameters of the target local search algorithm and the initial parameters of the target reinforcement learning algorithm to obtain feedback and rewards of the target reinforcement learning algorithm;

[0144] In an embodiment of the present application, a greedy strategy random exploration action is performed based on the intermediate parameters of the target local search algorithm and the initial parameters of the target reinforcement learning algorithm to obtain feedback and rewards for the target reinforcement learning algorithm.

[0145] S707. Update the Q table of the target reinforcement learning algorithm based on the feedback and reward of the target reinforcement learning algorithm;

[0146] S709. Based on the updated Q-table, adjust the current operation operator of the target local search algorithm.

[0147] At this point, the updated current operation operator is obtained.

[0148] S509. Repeatedly call the target local search algorithm, search and calculate the current initial solution based on the current operation operator, and obtain the current local optimal solution. When the iteration termination condition is met, the current local optimal solution that meets the iteration termination condition is determined as the target stacking data.

[0149] In the embodiment of the present application, steps S601, S603, S605 and S607 are repeated until the iteration termination condition is met, then the iteration is stopped and the current local optimal solution that meets the iteration termination condition is determined as the target stacking data.

[0150] On the one hand, existing loading data processing methods use approximate algorithms, heuristic algorithms, etc. to solve such problems. However, the usual heuristic algorithms often have strong problem dependence. On the other hand, they use exact algorithms to establish a mixed integer programming model to solve the optimal solution. However, when the amount of data increases, the exact algorithm is often difficult to solve within a reasonable time frame.

[0151] In the embodiment of the present application, through the combined calculation of the local search algorithm and the reinforcement learning algorithm, the search range of the algorithm in the solution space is improved, and the algorithm performance is also improved, thereby significantly improving the loading data processing efficiency.

[0152] It can be seen from the loading data processing method provided by the above-mentioned embodiment of the present application that the embodiment of the present application obtains information on objects to be loaded, information on carriers and information on loading containers, wherein the information on objects to be loaded includes first object information on at least one object to be assembled or second object information on at least one non-assembled object; in the case where there are objects to be assembled, a linear programming model is called to determine first stacking data of the objects to be assembled based on the first object information and the carrier information; the first stacking data indicates the stacking method of a single layer of objects to be assembled on the carrier; a spatial search algorithm is called to perform simulated stacking processing of the objects to be loaded on at least one subspace of the current remaining space of the target loading container based on the information on objects to be loaded, the first stacking data and the loading container information, to obtain second stacking data of the objects to be loaded in the target loading container; wherein at least one subspace is based on The stackable plane information in the current remaining space of the target loading container is divided; the target local search algorithm and the target reinforcement learning algorithm are called to optimize the second stacking data to obtain the target stacking data corresponding to the second stacking data, and the target stacking data indicates the target stacking method of the object to be loaded in the target loading container; the loading data processing method provided in the embodiment of the present application uses the preset first stacking data as the input item of the spatial search algorithm, and simulates the stacking of multiple subspaces of the remaining space of the target loading container at the same time, so as to improve the efficiency of the spatial search algorithm; through the combined calculation of the local search algorithm and the reinforcement learning algorithm, the search range of the algorithm in the solution space is improved, and the algorithm performance is also improved; the loading data processing method provided in the embodiment of the present application can quickly and accurately obtain the optimized stacking method of the object to be loaded.

[0153] The present application also provides a loading data processing device, please refer to Figure 8, a loading data processing device provided in an embodiment of the present application includes:

[0154] The first acquisition module 810 is used to acquire information about the object to be loaded, information about the carrier, and information about the loading container, wherein the information about the object to be loaded includes first object information of at least one object to be assembled or second object information of at least one object not to be assembled;

[0155] The first determination module 820 is used to call the linear programming model to determine the first stacking data of the object to be assembled based on the first object information and the carrier information when there is an object to be assembled. The first stacking data indicates the stacking mode of a single layer of the object to be assembled on the carrier.

[0156] The first processing module 830 is used to call a space search algorithm, and based on the information of the object to be loaded, the first stacking data and the loading container information, perform simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container to obtain second stacking data of the object to be loaded in the target loading container; wherein at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container;

[0157] The first optimization module 840 is used to call the target local search algorithm and the target reinforcement learning algorithm to optimize the second stacking data to obtain the target stacking data corresponding to the second stacking data, and the target stacking data indicates the stacking method of the objects to be loaded in the target loading container.

[0158] In the embodiment of the present application, it also includes:

[0159] The second acquisition module is used to acquire the stackable plane information in the current remaining space of the target loading container;

[0160] The third acquisition module: based on the stackable plane information, performs space division for each stackable plane in the current remaining space of the target loading container to obtain at least one subspace.

[0161] In the embodiment of the present application, the first processing module 830 includes:

[0162] A first acquisition unit: used to acquire subspace information of at least one subspace;

[0163] A first determining unit is used for determining, for each subspace of the target loading container, at least one target loading object and stacking parameters of the target loading object corresponding to each subspace from the currently remaining objects to be loaded by calling a spatial search algorithm based on the subspace information, the information of the object to be loaded and the first stacking data;

[0164] A first updating unit: used for updating the stackable plane information of the target loading container based on at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object;

[0165] A second determining unit is used to repeat the steps of dividing each stackable plane into spaces, acquiring subspace information of at least one subspace, and determining at least one target loading object corresponding to each subspace from the currently remaining objects to be loaded, until all the objects to be loaded have completed simulated stacking or there is no valid remaining space in the target loading container;

[0166] The second acquisition unit is used to generate second stacking data based on the stacking parameters of the target loading objects in the target loading container when all the objects to be loaded have completed simulated stacking or there is no valid remaining space.

[0167] In the embodiment of the present application, the first optimization module 840 includes:

[0168] A third determining unit: used for determining a current initial solution of the target local search algorithm based on the second stacking data;

[0169] The first calculation unit is used to call the target local search algorithm, search and calculate the current initial solution based on the current operation operator, and obtain the current local optimal solution;

[0170] The first updating unit is used to update the current local optimal solution to the current initial solution if the iteration termination condition is not met;

[0171] The second updating unit is used to call the target reinforcement learning algorithm to update the current operation operator;

[0172] The fourth determination unit is used to repeatedly call the target local search algorithm, search and calculate the current initial solution based on the current operation operator, obtain the current local optimal solution step, and when the iteration termination condition is met, determine the current local optimal solution that meets the iteration termination condition as the target stacking data.

[0173] In the embodiment of the present application, the first calculation unit includes:

[0174] The first acquisition subunit is used to perturb the current initial solution and obtain the neighborhood solution of the current perturbation;

[0175] The second acquisition subunit is used to search and calculate the neighborhood solution of the current disturbance based on the current operation operator to obtain the neighborhood solution of the current local search;

[0176] The first calculation subunit is used to calculate the objective function value of the neighborhood solution of the current local search;

[0177] The first updating subunit is used to update the neighborhood solution of the current local search to the current local optimal solution if the objective function value of the neighborhood solution of the current local search is better.

[0178] In the embodiment of the present application, the first updating unit includes:

[0179] The third acquisition subunit is used to obtain the initial parameters of the target reinforcement learning algorithm;

[0180] The fourth acquisition subunit is used to obtain the neighborhood solution of the current disturbance based on the current initial solution of the disturbance, search and calculate the neighborhood solution of the current disturbance based on the current operation operator, obtain the neighborhood solution of the current local search, and calculate the action, state and feedback value of the objective function value step of the neighborhood solution of the current local search to obtain the intermediate parameters;

[0181] The fifth acquisition subunit is used to perform random exploration actions based on the intermediate parameters and the initial parameters of the target reinforcement learning algorithm to obtain feedback and rewards of the target reinforcement learning algorithm;

[0182] The second updating subunit is used to update the Q table of the target reinforcement learning algorithm based on the feedback and reward of the target reinforcement learning algorithm;

[0183] The third updating subunit is used to adjust the current operation operator of the target local search algorithm based on the updated Q table.

[0184] The device and method embodiments in the described device embodiments are based on the same application concept.

[0185] Please refer to Fig. 9 An embodiment of the present application provides an electronic device for implementing the above-mentioned loading data processing method, the electronic device includes a processor and a memory, the memory stores at least one instruction or at least one program, the at least one instruction or the at least one program is loaded and executed by the processor to implement the loading data processing method provided in the above-mentioned method embodiment.

[0186] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0187] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal, a server or a similar computing device, that is, the above-mentioned electronic device may include a mobile terminal, a computer terminal, a server or a similar computing device. Among them, the above-mentioned server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited to this.

[0188] Fig. 9 1 is a hardware structure block diagram of an electronic device for implementing the above-mentioned loading data processing method provided by an embodiment of the present application. Fig. 9 As shown, the electronic device 900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 910 (the processor 910 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 930 for storing data, and one or more storage media 920 (such as one or more mass storage devices) for storing application programs 923 or data 922. Among them, the memory 930 and the storage medium 920 can be short-term storage or permanent storage. The program stored in the storage medium 920 may include one or more modules, each of which may include a series of instruction operations in the electronic device. Furthermore, the central processing unit 910 can be configured to communicate with the storage medium 920 and execute a series of instruction operations in the storage medium 920 on the electronic device 900. The electronic device 900 may also include one or more power supplies 960, one or more wired or wireless network interfaces 950, one or more input and output interfaces 940, and / or, one or more operating systems 921, such as Windows Server TM , Mac OS X TM , Unix TM ,Linux TM , FreeBSD TM etc.

[0189] Processor 910 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0190] The input / output interface 940 may be used to receive or send data via a network. The specific example of the network may include a wireless network provided by a communication provider of the electronic device 900. In one example, the input / output interface 940 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 940 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0191] The operating system 921 may include system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic businesses and processing hardware-based tasks.

[0192] It can be understood by those skilled in the art that Fig. 9 The structure shown is only for illustration and does not limit the structure of the above electronic device. Fig. 9 More or fewer components as shown, or with Fig. 9 Different configurations are shown.

[0193] An embodiment of the present application also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to a loading data processing method in a method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the loading data processing method provided in the above method embodiment.

[0194] Optionally, in this embodiment, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0195] The embodiments of the present application also provide a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above-mentioned various optional implementations.

[0196] It should be noted that the above-mentioned sequence of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0197] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, system and server embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0198] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0199] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A loading data processing method, characterized in that: include: Acquire information of objects to be loaded, information of carriers and information of loading containers, wherein the information of objects to be loaded includes first object information of at least one object to be assembled or second object information of at least one object not to be assembled; In the case where the object to be assembled exists, a linear programming model is called to determine first stacking data of the object to be assembled based on the first object information and the carrier information; the first stacking data indicates a stacking mode of a single layer of the object to be assembled on the carrier; Calling a spatial search algorithm, based on the information of the object to be loaded, the first stacking data and the loading container information, performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container, to obtain second stacking data of the object to be loaded in the target loading container; wherein the at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container; Calling a target local search algorithm and a target reinforcement learning algorithm to perform data optimization on the second stacking data to obtain target stacking data corresponding to the second stacking data, wherein the target stacking data indicates a target stacking mode of the object to be loaded in the target loading container; The method of performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information to obtain second stacking data of the object to be loaded in the target loading container includes: Acquiring subspace information of the at least one subspace; For each subspace of the target loading container, calling the space search algorithm to determine at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object from the currently remaining objects to be loaded based on the subspace information, the information of the object to be loaded and the first stacking data; Based on at least one target loading object corresponding to each subspace and stacking parameters of the target loading object, updating the stackable plane information of the target loading container; Repeat the steps of dividing the space for each stackable plane, acquiring the subspace information of the at least one subspace, and determining at least one target loading object corresponding to each subspace from the currently remaining objects to be loaded, until all the objects to be loaded have completed simulated stacking or there is no valid remaining space in the target loading container; The second stacking data is generated based on the stacking parameters of the target loading objects in the target loading container when all the objects to be loaded have completed simulated stacking or there is no effective remaining space.

2. The loading data processing method according to claim 1, characterized in that: The first object information includes size information of the object to be assembled and supported, and the object to be assembled and supported includes a non-standard object. Before calling the linear programming model to determine the first stacking data of the object to be assembled and supported based on the first object information and the carrier information, the method further includes: The non-standard object is dimensionally mapped based on a preset shape to obtain dimension information of the non-standard object.

3. The loading data processing method according to claim 1, characterized in that: Before performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information, the loading data processing method further includes: Obtaining stackable plane information in the current remaining space of the target loading container; Based on the stackable plane information, space division is performed for each stackable plane in the current remaining space of the target loading container to obtain the at least one subspace.

4. The loading data processing method according to claim 1, characterized in that: The calling of the spatial search algorithm to determine at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object from the currently remaining objects to be loaded based on the subspace information, the information of the objects to be loaded and the first stacking data includes: In the case where the non-supported objects exist, calling the spatial search algorithm to determine at least one target non-supported object and stacking parameters of the target non-supported object from the currently remaining non-supported objects based on the subspace information and the second object information; When the objects to be assembled exist, the spatial search algorithm is called to determine at least one target object to be assembled, the corresponding target carrier and the stacking parameters of the target object to be assembled from the currently remaining objects to be assembled based on the subspace information, the first object information and the first stacking data.

5. The loading data processing method according to claim 1, characterized in that: The calling of the target local search algorithm and the target reinforcement learning algorithm to optimize the second stacking data to obtain the target stacking data corresponding to the second stacking data includes: Calling the target local search algorithm, optimizing and calculating the second stacking data based on the current operation operator until an iteration termination condition is satisfied, and obtaining the target stacking data; Among them, during the optimization calculation process, the intermediate parameters of the optimization calculation of the target local search algorithm are extracted; and the target reinforcement learning algorithm is called to update the current operation operator of the target local search algorithm based on the intermediate parameters.

6. The loading data processing method according to claim 5, characterized in that: The calling of the target local search algorithm, optimizing the second stacking data based on the current operation operator until an iteration termination condition is satisfied to obtain the target stacking data, includes: determining a current initial solution of the target local search algorithm based on the second stacked data; Calling the target local search algorithm, searching and calculating the current initial solution based on the current operation operator, and obtaining the current local optimal solution; If the iteration termination condition is not met, updating the current local optimal solution to the current initial solution; Calling the target reinforcement learning algorithm to update the current operation operator; Repeat the step of calling the target local search algorithm, searching and calculating the current initial solution based on the current operation operator to obtain the current local optimal solution, and when the iteration termination condition is met, determine the current local optimal solution that meets the iteration termination condition as the target stacking data.

7. The loading data processing method according to claim 1, characterized in that: The target local search algorithm is a variable neighborhood search algorithm, and the target reinforcement learning algorithm is a Q-learning algorithm.

8. A loading data processing device, characterized in that: The device comprises: A first acquisition module: used to acquire information about the object to be loaded, information about the carrier, and information about the loading container, wherein the information about the object to be loaded includes first object information of at least one object to be assembled or second object information of at least one object not to be assembled; A first determining module is used for, when the object to be assembled exists, calling a linear programming model to determine first stacking data of the object to be assembled based on the first object information and the carrier information; the first stacking data indicates a stacking mode of a single layer of the object to be assembled on the carrier; The first processing module is used to call a spatial search algorithm to perform simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information, so as to obtain second stacking data of the object to be loaded in the target loading container; wherein the at least one subspace is divided based on the stackable plane information in the current remaining space of the target loading container; A first optimization module: used for calling a target local search algorithm and a target reinforcement learning algorithm to perform data optimization on the second stacking data to obtain target stacking data corresponding to the second stacking data, wherein the target stacking data indicates a stacking method of the object to be loaded in the target loading container; The method of performing simulated stacking processing of the object to be loaded on at least one subspace of the current remaining space of the target loading container based on the information of the object to be loaded, the first stacking data and the loading container information to obtain second stacking data of the object to be loaded in the target loading container includes: Acquiring subspace information of the at least one subspace; For each subspace of the target loading container, calling the space search algorithm to determine at least one target loading object corresponding to each subspace and the stacking parameters of the target loading object from the currently remaining objects to be loaded based on the subspace information, the information of the object to be loaded and the first stacking data; Based on at least one target loading object corresponding to each subspace and stacking parameters of the target loading object, updating the stackable plane information of the target loading container; Repeat the steps of dividing the space for each stackable plane, acquiring the subspace information of the at least one subspace, and determining at least one target loading object corresponding to each subspace from the currently remaining objects to be loaded, until all the objects to be loaded have completed simulated stacking or there is no valid remaining space in the target loading container; The second stacking data is generated based on the stacking parameters of the target loading objects in the target loading container when all the objects to be loaded have completed simulated stacking or there is no effective remaining space.

9. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor according to any one of claims 1 to 7.

Images