Intelligent stereoscopic warehouse goods allocation control method and device and computer program product

Abstract

The invention discloses an intelligent stereoscopic warehouse goods allocation control method and device and a computer program product, and the method comprises the steps: collecting the multi-source data of goods and goods allocation, and extracting the feature vectors of the goods and goods allocation; the adaptation degree of the goods and the goods allocation is calculated, and the dispatching priority is determined through a preset model in combination with goods characteristics; on the basis of the priority and the adaptation degree, combining the dynamic volume demand of the goods, the carrying path parameters and the goods aggregation degree, performing global optimization by utilizing a preset space group effect function, and determining the optimal target goods allocation of each goods; and calculating the spatial deviation degree between the current position of the goods and the optimal target goods allocation, judging and adjusting urgency in combination with timeliness, and dynamically planning a path to realize self-adaptive allocation of the goods allocation. According to the invention, accurate adaptation of the goods and the goods locations is realized, the warehouse resource utilization efficiency and scheduling rationality are improved, the robustness of the system in a complex scene is enhanced, and the method is suitable for efficient operation of a high-density and automatic stereoscopic warehouse.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent warehousing and logistics automation, and particularly relates to a method and device for allocating and controlling a goods location in an intelligent warehouse, and a computer program product. BACKGROUND

[0002] With the rapid development of e-commerce, manufacturing industry and modern logistics industry, the warehousing system as a core link in the supply chain directly affects the collaborative performance of the entire industry chain. Under this background, the warehousing system is facing multiple challenges such as increasing variety of goods, continuously increasing frequency of warehouse entry and exit, and continuously improving management accuracy, which puts unprecedented high standards on the utilization rate of warehouse space, the efficiency of goods storage and retrieval, and the response speed of scheduling.

[0003] Specifically, in a high-density and automated warehouse environment, the traditional goods location management mode relying on manual experience or the static and fixed goods location allocation strategy has been difficult to adapt to the dynamic needs of modern warehousing. On the one hand, the static management mode cannot optimize the goods location layout according to dynamic data such as real-time warehouse entry and exit frequency and inventory turnover rate, resulting in low utilization rate of warehouse space. On the other hand, the response speed of manual decision is slow, which is difficult to meet the efficient scheduling needs in the high-frequency warehouse entry and exit scenario, thereby restricting the improvement of the overall operation efficiency of the intelligent warehouse.

[0004] To address the above problems, intelligent technology means have been gradually introduced to realize dynamic scheduling and intelligent allocation of goods locations. At the same time, the continuous maturity of software and hardware technologies such as automatic handling equipment (such as AGV), sensor networks, and warehouse management systems (WMS) provides a solid technical support for the refinement and automation of goods location management. However, the existing technologies still have deficiencies in the collaborative application of multi-source data fusion perception, real-time modeling of warehouse dynamic state, and intelligent decision optimization, which are difficult to realize the global optimal configuration of warehouse resources. SUMMARY

[0005] The technical problem to be solved by the embodiments of the present application is to provide a method and device for allocating and controlling a goods location in an intelligent warehouse, so as to improve the space utilization rate and storage and retrieval efficiency of the intelligent warehouse.

[0006] To solve the above technical problems, the present application provides a method for allocating and controlling a goods location in an intelligent warehouse, comprising:

[0007] Step S1, collecting multi-source data of goods and goods locations in the intelligent warehouse, and extracting feature vectors of the goods and feature vectors of the goods locations;

[0008] Step S2: Calculate the fit between the cargo and the storage location based on the feature vector of the cargo and the feature vector of the storage location, and calculate the scheduling priority of the cargo based on the fit and the feature vector of the cargo through a preset scheduling priority evaluation model.

[0009] Step S3: Based on the scheduling priority of the goods and the suitability between the goods and the storage location, and in conjunction with the dynamic volume requirements of the goods, the angle cost of the handling path, the handling distance, and the concentration of the goods, a global optimization calculation is performed through a preset spatial group effect function to determine the optimal target storage location for each goods.

[0010] Step S4: Calculate the spatial deviation between the current location of the goods and the optimal target location, determine the urgency of goods adjustment based on the timeliness and adaptability of goods scheduling, and dynamically plan the handling path to achieve adaptive allocation of the location.

[0011] Preferably, in step S1, the state feature vector of the goods is defined as G. i =[F i V i W i ,T i ,S i ], where F i V represents the frequency of goods i leaving the warehouse per unit time. i Let W be the volume of cargo i. i Let T be the weight of cargo i. i S represents the current storage time of goods i. i Let P be the shelf life of goods i; each storage location j in the warehouse is modeled as a feature vector P. j =[A j B j C j ], where A j B represents the available space area of ​​storage location j. j C is the maximum load-bearing capacity of storage location j. j The maximum storage time for recommended storage location j.

[0012] Preferably, the fit between goods and storage location in step S2 is calculated using the following formula:

[0013]

[0014] Where, α ij It is the fit between cargo i and storage location j; F j This represents the historical average shipping frequency of the storage location; F max It is the maximum outbound frequency for all goods and storage locations; A max This is the maximum volumetric capacity of all storage locations; B max This is the maximum load-bearing capacity of all storage locations; C maxis the maximum value of the recommended storage time among all storage locations.

[0015] Preferably, the dispatch priority of the goods in the step S2 is determined by the following formula:

[0016]

[0017] wherein P i is the allocation priority weight of the goods i; w1, w2, w3 are preset weight coefficients for balancing the influences among the out-of-warehouse frequency, the quality guarantee urgency and the storage history of the goods; is the quality guarantee urgency factor; is the penalty factor for the stagnant inventory; is the matching penalty term.

[0018] Preferably, the space colony effect function in the step S3 is defined as:

[0019]

[0020] wherein, is the space colony effect function; N is the total number of goods; M is the total number of storage locations; D i = V i ·(F i +1) represents the dynamic volume demand of the goods; R j is the current remaining space capacity of the storage location j; ∈ is a constant for preventing zero; θ ij is the angle between the current position of the goods i and the operation path of the storage location j; Δ ij is the Euclidean space distance from the goods i to the storage location j; β ij is the aggregation index of the goods i in the vicinity of the goods of the same type in the storage location j.

[0021] Preferably, the space colony effect function is maximized in the step S3 by means of simulated annealing search, segmented climbing or nonlinear constraint processing to obtain the optimal target storage location coordinates of each goods.

[0022] Preferably, the space deviation degree of the current position of the goods from the optimal target storage location in the step S4 is measured by a sparsity index, and the calculation formula of the sparsity index is:

[0023]

[0024] wherein ζ i is the sparsity index of the goods i; (x i , y i ) is the current actual coordinates of the goods i; is the optimal target storage location coordinates of the goods i; X max , Y max are the physical boundary sizes of the warehouse. represents the dynamic aging coefficient of the goods i; ∈ is a prevention zero constant.

[0025] Preferably, the dynamic planning of the carrying path in the step S4 comprises: constructing a sparse field atlas according to the sparsity indexes of all goods, setting a priority moving task for the goods with the largest sparsity index, and constraining the path resources of adjacent areas to prevent path conflicts, and gradually approaching the optimal state of space and efficiency through multi-period iteration.

[0026] The application further provides an intelligent stereoscopic warehouse goods location allocation control device, comprising:

[0027] one or more processors;

[0028] a memory;

[0029] one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to execute the intelligent stereoscopic warehouse goods location allocation control method.

[0030] The application further provides a computer program product comprising computer instructions instructing a computer device to perform operations corresponding to the method.

[0031] The application has the following beneficial effects: the application significantly improves the overall operation efficiency of the intelligent stereoscopic warehouse through deep fusion of multi-dimensional feature modeling and intelligent optimization decision. Specifically, the beneficial effects are as follows: first, by constructing a goods feature vector containing elements such as out-of-warehouse frequency, volume, weight, shelf life, storage duration, and comparing and analyzing the multi-dimensional features of the goods feature vector and the space carrying capacity of the goods location and the use cycle, the accuracy of the goods location adaptation degree is realized, the space waste and operation conflicts caused by extensive matching are effectively avoided, and the collaborative utilization efficiency of warehouse resources is greatly improved; second, on the basis of considering the static properties of goods, the dynamic relationship between the out-of-warehouse activity level, the storage period and the shelf life is fully integrated, a scheduling priority weight model is constructed and corrected in combination with the real-time adaptation degree, dynamic, reasonable and orderly task allocation is realized, and the scheduling conflicts in the high-frequency in-and-out-of-warehouse scenario are effectively alleviated; third, by establishing a space group effect function integrating the dynamic volume demand of goods, path distance, carrying angle and surrounding goods aggregation degree, local operation factors are included in the global optimization target, the limitations of traditional local optimization or operation bottlenecks are broken through, and the goods location use density and logistics throughput capacity are significantly improved; finally, by calculating the spatial deviation between the current position of the goods and the optimal target goods location, and dynamically adjusting the scheduling priority in combination with the timeliness, a closed-loop mechanism of autonomous adjustment and rolling update is formed, the robustness and rapid response capability of the system to sudden business changes are enhanced, and strong support is provided for efficient and stable operation of the intelligent stereoscopic warehouse. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating an intelligent automated warehouse location allocation and control method according to an embodiment of the present invention. Detailed Implementation

[0034] The following description of the embodiments is taken with reference to the accompanying drawings, which illustrate specific embodiments in which the invention can be implemented.

[0035] Please refer to Figure 1 As shown, Embodiment 1 of the present invention provides a method for intelligent automated warehouse location allocation and control, comprising:

[0036] Step S1: Collect multi-source data on goods and storage locations in the intelligent automated warehouse, and extract feature vectors for the goods and storage locations;

[0037] Step S2: Calculate the fit between the cargo and the storage location based on the feature vector of the cargo and the feature vector of the storage location, and calculate the scheduling priority of the cargo based on the fit and the feature vector of the cargo through a preset scheduling priority evaluation model.

[0038] Step S3: Based on the scheduling priority of the goods and the suitability between the goods and the storage location, and in conjunction with the dynamic volume requirements of the goods, the angle cost of the handling path, the handling distance, and the concentration of the goods, a global optimization calculation is performed through a preset spatial group effect function to determine the optimal target storage location for each goods.

[0039] Step S4: Calculate the spatial deviation between the current location of the goods and the optimal target location, determine the urgency of goods adjustment based on the timeliness and adaptability of goods scheduling, and dynamically plan the handling path to achieve adaptive allocation of the location.

[0040] Specifically, in step S1, multi-source data on goods and storage locations in the intelligent automated warehouse are collected through multi-source sensing devices deployed in the warehousing environment. These devices mainly include RFID tags bound to goods upon entry, laser rangefinders installed on shelves and aisles, weight sensors, and volume recognition cameras. The volume, weight, entry time, current storage location, and outbound frequency calculated based on historical order information are acquired in real time and synchronized to the warehouse management system (WMS) to form a structured data stream. Feature vectors for goods and storage locations are extracted from the collected data.

[0041] For each piece of goods entering the stereoscopic warehouse, a multi-dimensional state model is established, and the basic characteristic parameters for allocation control are extracted. The state characteristic vector of the goods is defined as G i = [F i , V i , W i , T i , S i ], wherein F i represents the frequency of goods i (times / day) in unit time, V i is the volume of goods i (cubic meters), W i is the weight of goods i (kilograms), T i is the current storage time of goods i (days), and S i is the shelf life of goods i (days). At the same time, each storage location j in the warehouse is modeled as a characteristic vector P j = [A j , B j , C j ], wherein A j represents the available space area of storage location j (square meters), B j is the maximum load capacity of storage location j (kilograms), and C j is the maximum storage time of recommended storage location j (days).

[0042] In step S2, in order to measure the matching degree between each piece of goods and each storage location in multiple physical and operational dimensions, an adaptation degree function is constructed to measure the comprehensive deviation degree of goods i and storage location j in terms of delivery frequency, space volume, load capacity and storage time.

[0043] Specifically, the adaptation degree function calculates the difference between the delivery frequency, volume, weight of the goods and the corresponding dimensional parameters of the storage location item by item, and scales all features to the same numerical dimension range through a normalization factor (such as the global maximum frequency, maximum area, maximum load capacity, and maximum recommended storage time). Then, the normalized difference is added up to form a scalar value reflecting the degree of multi-dimensional mismatch. The adaptation degree function formula is as follows:

[0044]

[0045] Wherein, α ij is the adaptation degree between goods i and storage location j, which is a dimensionless matching error measure, and the smaller the value is, the higher the adaptation degree is; F j is the historical average delivery frequency of the storage location (used for position sensitivity normalization); F max is the maximum value of the delivery frequency of all goods and storage locations, used for normalizing the frequency difference; A maxis the maximum volume capacity of all the storage spaces for normalization; B max is the maximum load capacity of all the storage spaces; C max is the maximum value of the recommended storage time in all the storage spaces. Each ratio reflects the relative deviation of a one-dimensional feature, and the smaller the overall fitness is, the more reasonable and coordinated the deployment of the goods on the corresponding storage space is.

[0046] In the dynamic calculation phase of the priority and allocation weight of the goods, a scheduling priority evaluation model of each piece of goods in the current time window is constructed to realize the reasonable allocation of the storage space resources and the orderly advancement of the space scheduling. The priority depends not only on the static attributes of the goods (such as the shelf life and the current storage time length), but also on the dynamic characteristics, especially the frequency of the goods out of the warehouse, that is, the frequency of the actual calling of the current goods in the warehouse.

[0047] Specifically, the frequency of the goods out of the warehouse directly affects whether the goods should be allocated to the efficient storage space near the entrance. The time factor is constructed by combining the shelf life and the current storage time length to reflect whether the goods have the risk of expiration or invalidation, and a non-linear descending trend is constructed by a score type expression to improve the discrimination sensitivity. Finally, the time factor and the result of the frequency of the goods out of the warehouse are weighted and summed to form the basic priority value, and then multiplied by an adjustment factor reflecting the adaptation degree of the goods and the target storage space to ensure that the priority does not deviate from the actual conditions of the storage space, and to avoid that the theoretically high-priority goods fall into the low-adaptation storage space, so as to constitute the following priority weight formula:

[0048]

[0049] wherein, P i is the allocation priority weight of the goods i, and the larger the value is, the more priority the goods i has in the current scheduling period; w1, w2, and w3 are preset weight coefficients for balancing the influence between the frequency of the goods out of the warehouse, the shelf life urgency, and the storage history of the goods; is the shelf life urgency factor, which measures the time window width of the goods from expiration; is the penalty factor for the stagnant inventory, which is used to punish the goods with long storage time but poor liquidity; is the matching penalty term, which is used to adjust the executability of the results of the above three scheduling factors. If the physical, operational, or time length attributes of a certain goods i and a certain storage space j are poorly matched, even if it is scheduled urgently, it is not suitable to be allocated to the storage space, and its priority should be reduced. Otherwise, the original priority is retained.

[0050] In step S3, the obtained priority weight of each goods is combined with the fitness of the set of feasible locations in the warehouse space to model a comprehensive evaluation model, and the allocation relationship between all goods and all candidate locations is globally optimized. Not only the fitness between goods and locations is considered, but also the dynamic volume demand of goods, i.e., the space usage trend in the future period of time, is further introduced; and the angle cost of the current coordinates of each goods and the target location, as well as the space length of the handling path, are introduced to describe the operation complexity and operation delay, respectively; the idea of particle swarm optimization in the constraint space is used to construct a space group effect function, which is aggregated by the space group effect function and maximized by searching to obtain the optimal target location coordinates of all goods in the current period. The space group effect function is defined as follows:

[0051]

[0052] wherein, is the space group effect function, which is a comprehensive index function for evaluating the global optimization degree of the location allocation strategy of the whole three-dimensional warehouse, and the goal is to achieve optimal location allocation under the constraint of multiple factors; N is the total number of goods; M is the total number of locations; D i = V i ·(F i +1) represents the dynamic volume demand of the goods; R j is the current remaining space capacity of the location j; ∈ is a constant to prevent zero; θ ij is the operation path angle between the current position of the goods i and the location j (indicating the angle cost of the mechanical arm); Δ ij is the Euclidean space distance from the goods i to the location j; β ij is the aggregation degree index of the goods i in the vicinity of the location j, which reflects the local congestion degree of the goods layout.

[0053] The space group effect function integrates the priority, fitness, path cost, space remaining degree, and goods type aggregation degree, and is jointly optimized by means of simulated annealing search, segmented climbing, and nonlinear constraint processing. After the optimal value is obtained, the optimal target location j * for each goods i is determined, and the two-dimensional coordinates are marked as

[0054] The optimal location allocation coordinates determined by the space group effect function Although it is a globally optimal solution in theory, in actual warehouse operation, due to factors such as channel blockage, mechanical arm scheduling conflict, temporary change of location occupation, and the like, the goods may not be able to move to its theoretically optimal position immediately. Therefore, a quantitative index is constructed in step S4 to describe the current actual coordinates (x i , y i) and its optimal target coordinate The degree of spatial deviation between the optimal target coordinate of the goods and the current coordinate of the goods is combined with the timeliness of the goods scheduling and the current adaptability to comprehensively judge whether the goods need to be immediately re-allocated or delayed, and the index is "sparsity", and the greater the value is, the greater the deviation between the current state of the goods and the optimal storage state of the goods is, the higher the scheduling urgency is, and the carrying path and the release of the target position should be processed in priority. The sparsity function is expressed as:

[0055]

[0056] Wherein, ζ i is the sparsity index of the goods i, indicating the deviation degree between the current state of the goods i and the optimal storage state of the goods i, and the greater the value is, the farther the goods i is from the optimal storage position, the more unreasonable the current state is, and the dynamic scheduling or position adjustment should be performed in priority; (x i ,y i ) is the current actual coordinate of the goods i; X max , Y max are the physical boundary sizes of the warehouse; represents the dynamic timeliness coefficient of the goods i, and is used for amplifying the scheduling urgency with the passage of time.

[0057] The sparsity field map is constructed according to the sparsity ζ i values of all the goods, the goods with the greatest deviation are set to have a priority moving task, and the path resources of the adjacent areas are constrained to prevent path conflicts; the search process is iterated in multiple periods, so that the warehouse system gradually approaches the optimal state of space and efficiency under the premise of not completely re-allocating, and realizes the system-level self-repairing and responding.

[0058] Corresponding to the intelligent three-dimensional warehouse storage allocation control method in the foregoing embodiment one, the embodiment two further provides an intelligent three-dimensional warehouse storage allocation control device, which comprises:

[0059] one or more processors;

[0060] a memory;

[0061] one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to execute the intelligent three-dimensional warehouse storage allocation control method in the foregoing embodiment one.

[0062] Corresponding to the intelligent three-dimensional warehouse storage allocation control method in the foregoing embodiment one, the embodiment three further provides a computer program product comprising computer instructions, which instruct a computer device to execute the operations corresponding to the intelligent three-dimensional warehouse storage allocation control method in the foregoing embodiment one.

[0063] Preferably, the processor can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor. The processor is a control center of the device, and connects various parts of the device through various interfaces and lines.

[0064] The memory mainly includes a program storage area and a data storage area. The program storage area can store an operating system, at least one application required by a function, etc., and the data storage area can store relevant data, etc. In addition, the memory can be a high-speed random access memory, and can also be a non-volatile memory such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., or the memory can also be other volatile solid-state storage devices.

[0065] It should be noted that the above device can include but is not limited to the processor and the memory, which can be understood by those skilled in the art.

[0066] From the above description, compared with the prior art, the beneficial effects of the present application are that the present application significantly improves the overall operation efficiency of the intelligent stereoscopic warehouse through the deep integration of multi-dimensional feature modeling and intelligent optimization decision. Specifically, the beneficial effects are as follows: first, by constructing a cargo feature vector containing factors such as delivery frequency, volume, weight, shelf life, storage duration, etc., and comparing and analyzing the multi-dimensional characteristics of the space carrying capacity and service life of the storage space, the precision of the cargo and storage space adaptation degree is realized, effectively avoiding the space waste and operation conflict caused by extensive matching, and greatly improving the collaborative utilization efficiency of warehouse resources; second, based on the consideration of the static properties of the goods, the dynamic relationship between the delivery activity level, storage period and shelf life is fully integrated, a scheduling priority weight model is constructed and corrected combined with real-time adaptation, realizing dynamic, reasonable and orderly task allocation, effectively alleviating the scheduling conflict in the high-frequency warehouse-in and warehouse-out scenario; third, by establishing a spatial group effect function integrating dynamic volume demand, path distance, handling angle and surrounding goods aggregation, local operation factors are included in the global optimization target, breaking through the limitations of traditional local optimization or operation bottlenecks, and significantly improving the storage space utilization density and logistics throughput capacity; finally, by calculating the spatial deviation between the current position of the goods and the optimal target storage space, and dynamically adjusting the scheduling priority combined with the timeliness, a closed-loop mechanism of autonomous adjustment and rolling update is formed, enhancing the robustness and rapid response capability of the system to sudden business changes, and providing strong support for the efficient and stable operation of the intelligent stereoscopic warehouse.

[0067] The above disclosure is only the preferred embodiment of the present application, and of course cannot limit the scope of the right of the present application, therefore the equivalent changes made according to the claims of the present application still belong to the scope covered by the present application.

Claims

1. An intelligent three-dimensional warehouse storage location allocation control method, characterized by, The method comprises the following steps: Step S1, collecting multi-source data of goods and storage locations in an intelligent stereoscopic warehouse, and extracting feature vectors of the goods and the storage locations; Step S2, calculating the adaptability of the goods and the storage locations based on the feature vectors of the goods and the storage locations, and calculating the scheduling priority of the goods based on the adaptability and the feature vectors of the goods through a preset scheduling priority evaluation model; Step S3, based on the scheduling priority of the goods and the adaptability of the goods and the storage locations, combining the dynamic volume demand of the goods, the path angle cost, the carrying distance and the goods aggregation degree, and performing global optimization calculation through a preset space group effect function to determine the optimal target storage location of each good; Step S4, calculating the spatial deviation degree of the current position of the goods and the optimal target storage location, combining the scheduling timeliness of the goods and the adaptability to determine the adjustment urgency of the goods, and dynamically planning the carrying path to realize adaptive allocation of the storage locations.

2. The method of claim 1, wherein, The state feature vector of the goods in the step S1 is defined as G i = [F i , V i , W i , T i , S i ], wherein F i represents the outbound frequency of the goods i per unit time, V i is the volume of the goods i, W i is the weight of the goods i, T i is the current storage time of the goods i, and S i is the shelf life of the goods i; each goods location j in the warehouse is modeled as a feature vector P j = [A j , B j , C j ], wherein A j represents the available space area of the goods location j, B j is the maximum load capacity of the goods location j, and C j is the maximum storage time of the recommended goods location j.

3. The method of claim 2, wherein, The adaptability of the goods and the storage locations in step S2 is calculated by the following formula: where, a ij is the fit between the goods i and the location j; F j is the historical average delivery frequency of the location; F max is the maximum value of the delivery frequency of all goods and locations; A max is the maximum volume capacity of all locations; B max is the maximum load-bearing capacity of all locations; C max is the maximum recommended storage time in all locations.

4. The method of claim 3, wherein, The scheduling priority of the goods in step S2 is determined by the following formula: wherein P i is the distribution priority weight of the goods i; w1, w2, w3 are preset weight coefficients for balancing the influence between the out-of-warehouse frequency, the quality guarantee urgency and the goods storage history; is the quality guarantee urgency factor; is the stay-in-stock penalty factor; is the matching penalty term.

5. The method of claim 1, wherein, The space group effect function in step S3 is defined as: where, is the spatial group effect function; N is the total number of goods; M is the total number of storage spaces; D i = V i · (F i + 1) represents the dynamic volume demand of the goods; R j is the current remaining space capacity of the storage space j; ∈ is a zero-prevention constant; θ ij is the angle between the current position of the goods i and the operating path of the storage space j; Δ ij is the Euclidean spatial distance from the goods i to the storage space j; β ij is the aggregation index of the goods i in the vicinity of the goods of the same kind in the storage space j.

6. The method of claim 1, wherein, In step S3, the space group effect function is maximized through simulated annealing search, segmented climbing or nonlinear constraint processing to obtain the optimal target storage location coordinates of each good.

7. The method of claim 1, wherein, In step S4, the spatial deviation degree of the current position of the goods and the optimal target storage location is measured by the sparsity index, and the calculation formula of the sparsity index is: wherein ζ i is the sparsity index of the goods i; (x i , y i ) is the current actual coordinate of the goods i; is the optimal target storage location coordinate of the goods i; X max , Y max are the physical boundary dimensions of the warehouse; denotes the dynamic time coefficient of the goods i; and ∈ is a prevention zero constant.

8. The method of claim 1, wherein, The dynamic planning of the carrying path in step S4 comprises: constructing a sparse field atlas according to the sparsity index of all goods, setting a priority moving task for the goods with the largest sparsity index, and constraining the path resources in the adjacent area to prevent path conflicts, and gradually approaching the optimal state of space and efficiency through multi-cycle iteration.

9. An intelligent three-dimensional warehouse storage location allocation control device characterized by comprising: One or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to execute the intelligent stereoscopic warehouse storage location allocation control method according to any one of claims 1-8. The computer instructions instruct the computer device to perform operations corresponding to the method according to any one of claims 1-8.

10. A computer program product, characterised in that, ​

Images