Tobacco distribution method and computer device applied to the front storage of tobacco drones

Abstract

Embodiments of the present application relate to the field of tobacco distribution applied to the front warehouse of tobacco drones, and specifically relate to a tobacco distribution method and a computer device applied to the front warehouse of tobacco drones. A specific implementation manner of the method includes: in response to receiving tobacco distribution request information corresponding to a target tobacco drone front warehouse, performing structured processing on the tobacco distribution request information and the current tobacco storage information of the target tobacco drone front warehouse to obtain structured tobacco information; determining tobacco distribution restriction information corresponding to the tobacco distribution request information; inputting the front warehouse structured information, the front warehouse tobacco structured information, and the front warehouse tobacco association information into a front warehouse tobacco distribution model to obtain front warehouse tobacco distribution information; and controlling the tobacco drones to be delivered to perform tobacco distribution according to the front warehouse tobacco distribution information. This implementation manner can improve the accuracy of the front warehouse tobacco distribution information, improve the utilization rate of the front warehouse, and reduce the waste of tobacco distribution resources.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of tobacco distribution applied to the front warehouse of tobacco drones, and particularly to a tobacco distribution method and a computer device applied to the front warehouse of tobacco drones. Background Art

[0002] As one of the important links, the transportation of tobacco products often requires transportation personnel to deliver tobacco products to corresponding delivery points (such as sales points) in advance within a fixed time period. For the distribution of tobacco products, the commonly adopted method is: through a tobacco distribution model, generate tobacco distribution information, transport the tobacco to a large warehouse uniformly, and then distribute it by the distribution vehicles in each district. However, when using the above method for tobacco distribution, the following technical problems often exist: the accuracy of the generated tobacco scheduling and distribution information is relatively low, there is a waste of tobacco distribution resources and an increase in distribution time, and it is difficult to effectively ensure the timeliness of tobacco product distribution. Summary of the Invention

[0003] This section of the present application is used to briefly introduce concepts, which will be described in detail in the following detailed implementation section. This section of the present application is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0004] Some embodiments of the present application propose a tobacco distribution method, a computer device, and a computer-readable storage medium applied to the front warehouse of tobacco drones to solve one or more of the technical problems mentioned in the above background art section.

[0005] In a first aspect, some embodiments of the present application provide a tobacco distribution method applied to a front warehouse of a tobacco drone. The method includes: in response to receiving tobacco distribution request information corresponding to a target tobacco drone front warehouse, performing structured processing on the tobacco distribution request information and the current tobacco storage information of the target tobacco drone front warehouse to obtain structured tobacco information, where the structured tobacco information includes: front warehouse structured information, front warehouse tobacco structured information, and front warehouse tobacco association information; determining tobacco distribution restriction information corresponding to the tobacco distribution request information, where the tobacco distribution restriction information is at least one restriction condition corresponding to a preset front warehouse tobacco distribution model; inputting the front warehouse structured information, the front warehouse tobacco structured information, and the front warehouse tobacco association information into the front warehouse tobacco distribution model to obtain front warehouse tobacco distribution information under the tobacco distribution restriction information; determining at least one to-be-distributed tobacco drone corresponding to the front warehouse tobacco distribution information; controlling an associated tobacco handling device to load tobacco onto the at least one to-be-distributed tobacco drone; and in response to determining that the tobacco loading is completed, controlling the at least one to-be-distributed tobacco drone to perform tobacco distribution.

[0006] In a second aspect, the present application further provides a computer device. The computer device includes a processor, a memory, and a computer program stored on the memory and executable by the processor. When the computer program is executed by the processor, the method described in any implementation manner of the first aspect is implemented.

[0007] In a third aspect, the present application further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the method described in any implementation manner of the first aspect is implemented.

[0008] The above-mentioned various embodiments of the present application have the following beneficial effects: Through the tobacco distribution method applied to the front-loading bin of tobacco drones in some embodiments of the present application, the accuracy of the tobacco distribution information of the front-loading bin can be improved, the utilization rate of the front-loading bin can be increased, and the waste of tobacco distribution resources can be reduced. First, in response to receiving the tobacco distribution request information corresponding to the target tobacco drone front-loading bin, the above-mentioned tobacco distribution request information and the current tobacco storage information of the above-mentioned target tobacco drone front-loading bin are structurally processed to obtain structured tobacco information, wherein the above-mentioned structured tobacco information includes: front-loading bin structured information, front-loading bin tobacco structured information, and front-loading bin tobacco association information. Secondly, determine the tobacco distribution restriction information corresponding to the above-mentioned tobacco distribution request information. Wherein, the above-mentioned tobacco distribution restriction information is at least one restriction condition corresponding to a preset front-loading bin tobacco distribution model. Thus, considering specific restriction information, the correlation between the front-loading bin tobacco distribution model and the tobacco distribution request information is improved, which is convenient for generating more accurate front-loading bin tobacco distribution information under the corresponding tobacco distribution request information. Then, input the above-mentioned front-loading bin structured information, the above-mentioned front-loading bin tobacco structured information, and the above-mentioned front-loading bin tobacco association information into the above-mentioned front-loading bin tobacco distribution model to obtain the front-loading bin tobacco distribution information under the above-mentioned tobacco distribution restriction information. Thus, the front-loading bin tobacco distribution information corresponding to the corresponding tobacco distribution request information is generated, and the accuracy of the front-loading bin tobacco distribution information and the utilization rate of the front-loading bin under the tobacco distribution request information are improved. Then, according to the above-mentioned front-loading bin tobacco distribution information, determine at least one to-be-distributed tobacco drone; control the associated tobacco handling equipment to load tobacco onto the above-mentioned at least one to-be-distributed tobacco drone. Finally, in response to determining that the tobacco loading is completed, control at least one to-be-distributed tobacco drone to perform tobacco distribution. Thus, the timeliness of the distribution of tobacco products can be effectively guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the various embodiments of the present application will become more apparent. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the elements and elements are not necessarily drawn to scale.

[0010] Figure 1 is a flowchart of some embodiments of the tobacco distribution method applied to the front-loading bin of tobacco drones according to the present application;

[0011] Figure 2 is a schematic structural diagram of a computer device suitable for implementing some embodiments of the present application;

[0012] Figure 3 is a schematic structural diagram of a to-be-distributed tobacco drone (new drone distribution terminal) in the tobacco distribution method applied to the front-loading bin of tobacco drones of the present application;

[0013] Figure 4 It is a front structural schematic diagram of a tobacco delivery unmanned aerial vehicle (new unmanned aerial vehicle delivery terminal) in the tobacco delivery method applied to the front warehouse of a tobacco unmanned aerial vehicle in this application. Specific embodiments

[0014] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not used to limit the protection scope of the present application.

[0015] In addition, it should be noted that for ease of description, only parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0016] It should be noted that the concepts such as "first" and "second" mentioned in the present application are only used to distinguish different devices, modules or units, and are not used to limit the order of functions executed by these devices, modules or units or their interdependent relationships.

[0017] It should be noted that the modifications of "one" and "multiple" mentioned in the present application are illustrative rather than restrictive. Those skilled in the art should understand that unless clearly stated otherwise in the context, it should be understood as "one or more".

[0018] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0019] The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0020] Figure 1 Flow 100 of some embodiments of the tobacco delivery method applied to the front warehouse of a tobacco unmanned aerial vehicle according to the present application is shown. The tobacco delivery method applied to the front warehouse of a tobacco unmanned aerial vehicle includes the following steps:

[0021] Step 101, in response to receiving the tobacco delivery request information corresponding to the target tobacco unmanned aerial vehicle front warehouse, perform structured processing on the tobacco delivery request information and the current tobacco storage information of the target tobacco unmanned aerial vehicle front warehouse to obtain structured tobacco information.

[0022] In some embodiments, the execution subject (e.g., a computing device) of the tobacco distribution method applied to the front-loading bin of a tobacco drone can, in response to receiving the tobacco distribution request information corresponding to the target front-loading bin of the tobacco drone, perform structured processing on the above-mentioned tobacco distribution request information and the current tobacco storage information of the target front-loading bin of the tobacco drone to obtain structured tobacco information. Among them, the above-mentioned structured tobacco information includes: front-loading bin structured information, front-loading bin tobacco structured information, and front-loading bin tobacco association information. The target front-loading bin of the tobacco drone can refer to the front-loading bin of tobacco loaded with a drone. This target front-loading bin of the tobacco drone can distribute tobacco by drone and also by vehicle. The tobacco distribution request information can be the request information for tobacco distribution. The above-mentioned tobacco distribution request information can be the tobacco demand information, front-loading bin tobacco information, and tobacco restriction information related to the target front-loading bin of the tobacco drone sent by the received front-end tobacco demand user terminal. The tobacco demand information can indicate which varieties of tobacco are needed and the quantity. The front-loading bin tobacco information can refer to the information of the front-loading bin storing the front-loading bin tobacco and the front-loading bin tobacco information. The tobacco restriction information can be the information restricting the tobacco distribution time. Among them, the above-mentioned front-loading bin tobacco association information can be the association information between the front-loading bin tobacco information, and the above-mentioned front-loading bin tobacco information can be the corresponding front-loading bin information and front-loading bin tobacco information. The above-mentioned front-loading bin structured information can be the structured information related to the front-loading bin that meets the input data requirements of the front-loading bin tobacco distribution model. The above-mentioned front-loading bin structured information includes: front-loading bin name, front-loading bin upper limit information, front-loading bin status information, and marked front-loading bin information, which is stored in the form of a data table. The above-mentioned front-loading bin upper limit information can include the information of the upper limit of tobacco storage in the front-loading bin warehouse and the volume of the front-loading bin. The above-mentioned front-loading bin status information can be the restriction information related to the front-loading bin in the above-mentioned tobacco distribution request information. For example, the above-mentioned front-loading bin status field can include: deletion of the information to be shipped out in the scenario of tobacco out-of-warehouse. The above-mentioned marked front-loading bin information can be the information of the front-loading bin with special requirements in the front-loading bin information and the tobacco distribution request information. For example, the above-mentioned marked front-loading bin information can include: automated front-loading bin information, semi-automated front-loading bin information. The above-mentioned front-loading bin tobacco structured information can be the structured information related to the front-loading bin tobacco that meets the input data requirements of the front-loading bin tobacco distribution model. The above-mentioned front-loading bin tobacco structured information can include the front-loading bin tobacco name and the front-loading bin tobacco storage information, and is stored in the form of a data table. The above-mentioned front-loading bin tobacco association structured information can be the structured information representing the restriction information related to the front-loading bin information and the tobacco information in the tobacco distribution request information in the form of a data table. The above-mentioned tobacco storage information can include the information of the front-loading bin storing the front-loading bin tobacco and the front-loading bin tobacco information. The front-loading bin can be composed of multiple sub-warehouses and is set up in the same area.

[0023] For example, the above-mentioned execution entity may convert the above-mentioned tobacco distribution request information and the above-mentioned tobacco storage information into tabular structured data required by the above-mentioned front-warehouse tobacco distribution model to obtain structured tobacco information.

[0024] Step 102, determine the tobacco distribution restriction information corresponding to the above-mentioned tobacco distribution request information.

[0025] In some embodiments, the above-mentioned execution entity may determine the tobacco distribution restriction information corresponding to the above-mentioned tobacco distribution request information. Among them, the above-mentioned tobacco distribution restriction information is at least one restriction condition corresponding to a preset front-warehouse tobacco distribution model. Among them, the above-mentioned front-warehouse tobacco distribution model includes: a front-warehouse tobacco restriction condition group. The above-mentioned front-warehouse tobacco distribution model includes: a front-warehouse tobacco calculation function group. The above-mentioned front-warehouse tobacco distribution model may be a model for performing front-warehouse tobacco scheduling planning on the input tobacco distribution request information and front-warehouse information to generate front-warehouse tobacco scheduling information between multiple warehouses. For example, the above-mentioned front-warehouse tobacco distribution model may include: a reinforcement learning model with an objective function of minimizing distribution cost, an objective function of minimizing distribution distance, a distribution time limit, and a front-warehouse inventory capacity limit. The state space of the front-warehouse tobacco distribution model may include: encoded tobacco distribution request information and front-warehouse information. The action space of the front-warehouse tobacco distribution model may include: the quantity of front-warehouse tobacco to be distributed after encoding and the distribution path. The incentive function of the front-warehouse tobacco distribution model may be: the weighted sum of the objective function of minimizing distribution cost and the objective function of minimizing distribution distance, minus a penalty value to obtain the incentive function. The above-mentioned penalty value may be a value that needs to be subtracted after the action performed by the drone violates the actions specified by the distribution time limit and the front-warehouse inventory capacity limit conditions.

[0026] For example, first, match the above-mentioned tobacco distribution request information with the set of model restriction conditions included in the preset model restriction condition pool to obtain a model restriction matching result set. Among them, the above-mentioned preset model restriction condition pool may be a condition pool storing all the restriction conditions corresponding to multiple preset scenarios. Then, determine at least one model restriction condition corresponding to at least one model restriction matching result indicating successful matching in the above-mentioned model restriction matching result set as the tobacco distribution restriction information.

[0027] In a practical application scenario, the above-mentioned execution entity can determine each pre-warehouse tobacco restriction condition corresponding to the above-mentioned tobacco distribution request information in the above-mentioned pre-warehouse tobacco restriction condition group as tobacco distribution restriction information. Among them, the above-mentioned tobacco distribution restriction information includes: a basic pre-warehouse tobacco restriction condition group, a tobacco distribution restriction condition group, and a pre-warehouse tobacco adjustment restriction condition group. Among them, the pre-warehouse tobacco restriction conditions in the above-mentioned pre-warehouse tobacco restriction condition group can be all the restriction conditions related to pre-warehouse information and pre-warehouse tobacco information included in the pre-warehouse tobacco distribution model in multiple scenarios. Among them, the basic pre-warehouse tobacco restriction conditions in the above-mentioned basic pre-warehouse tobacco restriction condition group can be restriction information related to pre-warehouse information and pre-warehouse tobacco information that is common in multiple scenario information. The above-mentioned basic pre-warehouse tobacco restriction condition group can include: pre-warehouse storage restriction information, pre-warehouse capacity upper limit restriction information, pre-warehouse capacity lower limit restriction information, pre-warehouse tobacco association restriction information, and pre-warehouse usage restriction information. The above-mentioned pre-warehouse storage restriction information can represent the restriction information that a certain category of pre-warehouse tobacco can only be stored in one warehouse. The above-mentioned pre-warehouse storage restriction information can be expressed as: the sum of the pre-warehouse tobacco storage variables within the value range of the pre-warehouse tobacco information set and the pre-warehouse information is 1. Among them, the above-mentioned pre-warehouse tobacco information set can be the information of all pre-warehouse tobacco that can be stored in the pre-warehouse cluster. The above-mentioned pre-warehouse tobacco storage variable can represent a variable indicating whether the pre-warehouse tobacco is stored in the pre-warehouse. If the pre-warehouse tobacco is stored in the warehouse, the pre-warehouse tobacco storage variable is 1. Otherwise, the value of the pre-warehouse tobacco storage variable is 0. The above-mentioned pre-warehouse capacity upper limit restriction information can represent that after the pre-warehouse tobacco is distributed, the quantity of the pre-warehouse tobacco stored in the adjusted pre-warehouse cannot exceed the upper limit of the pre-warehouse capacity in each month. The above-mentioned pre-warehouse capacity upper limit restriction information can represent that after the pre-warehouse tobacco is distributed, the quantity of the pre-warehouse tobacco stored in the adjusted pre-warehouse cannot be lower than the lower limit of the pre-warehouse capacity in each month. The above-mentioned pre-warehouse tobacco association restriction information can represent whether any two warehouse tobaccos are stored in the same pre-warehouse. The above-mentioned pre-warehouse usage restriction information can represent that for any one warehouse, when the warehouse items are stored in the warehouse, the warehouse usage variable is 1, that is, when any warehouse tobacco is stored in the warehouse, the warehouse starts to be used. The pre-warehouse tobacco adjustment restriction conditions in the above-mentioned pre-warehouse tobacco adjustment restriction condition group can be restriction conditions representing the change in the proportion of the stored quantity of warehouse tobacco before and after tobacco scheduling and distribution. The above-mentioned pre-warehouse tobacco adjustment restriction condition group can include: target pre-warehouse tobacco proportion restriction information. The above-mentioned target pre-warehouse tobacco proportion restriction information can be restriction information representing that considering the type of tobacco rack in the warehouse, the proportion of the target pre-warehouse tobacco in the warehouse after pre-warehouse tobacco scheduling is greater than or equal to the proportion before pre-warehouse tobacco scheduling. The above-mentioned target pre-warehouse tobacco can be the tobacco with the largest stored quantity of tobacco.The tobacco distribution restriction condition group can represent the restriction conditions of distribution time and distribution route.

[0028] Step 103: Input the above-mentioned pre-warehouse structured information, the above-mentioned pre-warehouse tobacco structured information, and the above-mentioned pre-warehouse tobacco association information into the above-mentioned pre-warehouse tobacco distribution model to obtain the pre-warehouse tobacco distribution information under the above-mentioned tobacco distribution restriction information.

[0029] In some embodiments, the above-mentioned execution entity can input the above-mentioned pre-warehouse structured information, the above-mentioned pre-warehouse tobacco structured information, and the above-mentioned pre-warehouse tobacco association information into the above-mentioned pre-warehouse tobacco distribution model to obtain the pre-warehouse tobacco distribution information under the above-mentioned tobacco distribution restriction information. Among them, the above-mentioned pre-warehouse tobacco distribution information can be the information for adjusting and distributing pre-warehouse tobacco among multiple warehouses. For example, the pre-warehouse tobacco distribution information can be: scheduling the "001" tobacco in Warehouse A and Warehouse B to the C pre-warehouse and delivering it by drone, and the delivery time is XX year XX month XX day.

[0030] In an actual application scenario, the above-mentioned execution entity can input the above-mentioned pre-warehouse structured information, the above-mentioned pre-warehouse tobacco structured information, and the above-mentioned pre-warehouse tobacco association information into the above-mentioned pre-warehouse tobacco distribution model through the following steps:

[0031] First step, determine each pre-position warehouse tobacco calculation function corresponding to the above tobacco distribution request information in the above pre-position warehouse tobacco calculation function group as the target pre-position warehouse tobacco calculation function group. Among them, the pre-position warehouse tobacco calculation functions in the above pre-position warehouse tobacco calculation function group can be full-scale optimization functions related to pre-position warehouse information and pre-position warehouse tobacco information included in the above pre-position warehouse tobacco distribution model in multiple scenarios. The above target pre-position warehouse tobacco calculation function group may include: a target function for minimizing the usage quantity of pre-position warehouses and a target function for maximizing the correlation degree of pre-position warehouse tobacco. The above target function for minimizing the usage quantity of pre-position warehouses can be a target function that reduces the usage quantity of pre-position warehouses by optimizing the utilization rate of pre-position warehouse energy and production capacity. The above target function for minimizing the usage quantity of pre-position warehouses can be expressed as: minimizing the cumulative sum of the values of the pre-position warehouse usage variable within the value range of the warehouse information set to obtain the target function for minimizing the usage quantity of pre-position warehouses. The above target function for maximizing the correlation degree of pre-position warehouse tobacco can be a target function that combines pre-position warehouse tobacco based on the correlation degree of pre-position warehouse tobacco to reduce the order splitting rate of UAV distribution. Among them, the above target function for maximizing the correlation degree of pre-position warehouse tobacco can be expressed as: maximizing the cumulative sum of the product of the pre-position warehouse tobacco correlation degree and the pre-position warehouse tobacco correlation degree variable within the value range of the pre-position warehouse tobacco information set to obtain the target function for maximizing the correlation degree of pre-position warehouse tobacco. The above pre-position warehouse tobacco correlation degree can be the correlation degree value between any two warehouse tobaccos. The above pre-position warehouse tobacco correlation degree can be a correlation degree value obtained by statistically analyzing the proportion of the quantity of warehouse tobacco included in each historical distribution information in the historical distribution information set.

[0032] Second step, according to the above basic pre-position warehouse tobacco restriction condition group, tobacco distribution restriction condition group, and pre-position warehouse tobacco adjustment restriction condition group, input the above pre-position warehouse structured information, the above pre-position warehouse tobacco structured information, and the above pre-position warehouse tobacco correlation information into the above target pre-position warehouse tobacco calculation function group to obtain pre-position warehouse tobacco distribution information.

[0033] Step 104, determine at least one to-be-distributed tobacco UAV corresponding to the above pre-position warehouse tobacco distribution information.

[0034] In some embodiments, the above execution entity may determine at least one to-be-distributed tobacco UAV corresponding to the above pre-position warehouse tobacco distribution information. For example, at least one to-be-distributed tobacco UAV can be determined according to the quantity of tobacco, the volume and weight of the tobacco included in the pre-position warehouse tobacco distribution information. For example, the maximum distribution volume and weight of each tobacco UAV can be determined, and thus, it can be determined according to the quantity of tobacco, the volume and weight of the tobacco.

[0035] Step 105, control the associated tobacco handling equipment to load tobacco onto the above at least one to-be-distributed tobacco UAV.

[0036] In some embodiments, the above-mentioned execution entity may control the associated tobacco handling device to load tobacco into at least one tobacco delivery drone to be delivered. The tobacco handling device may be a robotic arm for handling tobacco, which can grab the tobacco into the cabin of the drone.

[0037] Step 106, in response to determining that the tobacco loading is completed, control at least one tobacco delivery drone to perform tobacco delivery.

[0038] In some embodiments, the above-mentioned execution entity may, in response to determining that the tobacco loading is completed, control at least one tobacco delivery drone to perform tobacco delivery. For example, it may control at least one tobacco delivery drone to perform tobacco delivery according to the delivery route included in the pre-position warehouse tobacco delivery information. As Figure 3 shown in the example, a schematic structural diagram of a tobacco delivery drone (new drone delivery terminal) to be delivered is shown, including the drone and the drone cabin. It should be noted that Figure 3 the size of the drone and the size of the drone cabin in are only schematic renderings and not actual size drawings.

[0039] Again, as Figure 4 shown in the example, a front structural diagram of a tobacco delivery drone (new drone delivery terminal) to be delivered is shown, including the drone and the drone cabin, and the example shows each sub-compartment (such as compartments ABCDEFGHIJK, etc.) included in the drone cabin, and each sub-compartment is used to fixedly store tobacco products.

[0040] Further, in response to receiving a prediction request for pre-position warehouse resource conversion of the above-mentioned target tobacco drone pre-position warehouse, the pre-position warehouse resource conversion efficiency prediction model trained in advance is used to predict the pre-position warehouse resource conversion efficiency of the above-mentioned target tobacco drone pre-position warehouse at the target time.

[0041] In some embodiments, the above-mentioned execution entity may, in response to receiving a prediction request for the conversion of the resources in the front-end warehouse of the target tobacco drone, predict the conversion efficiency of the resources in the front-end warehouse of the target tobacco drone at the target time through a pre-trained prediction model for the conversion efficiency of the resources in the front-end warehouse. The prediction request for the conversion of the resources in the front-end warehouse may refer to a detection and prediction request for the resource utilization rate of the front-end warehouse of the target tobacco drone. The resources in the front-end warehouse of the target tobacco drone may include warehousing resources and distribution resources. The warehousing resources may represent the storage resources of each warehouse. The distribution resources may represent the distribution resources of the distribution drones and distribution vehicles. The prediction model for the conversion efficiency of the resources in the front-end warehouse may be a neural network model pre-trained with the resource usage information of the front-end warehouse of the tobacco drone as the input and the prediction result of the conversion efficiency of the resources in the front-end warehouse as the output. The prediction result of the conversion efficiency of the resources in the front-end warehouse may represent the resource utilization rate of the front-end warehouse of the target tobacco drone.

[0042] Further, according to the generated prediction result of the conversion efficiency of the resources in the front-end warehouse, adjust the drone distribution resources and tobacco storage resources of the front-end warehouse of the target tobacco drone.

[0043] In some embodiments, the above-mentioned execution entity may adjust the drone distribution resources and tobacco storage resources of the front-end warehouse of the target tobacco drone according to the generated prediction result of the conversion efficiency of the resources in the front-end warehouse. For example, if the prediction result of the conversion efficiency of the resources in the front-end warehouse is less than 85%, the redundant distribution resources and warehouse resources may be scheduled to other front-end warehouses.

[0044] It should be noted that the prediction model for the conversion efficiency of the resources in the front-end warehouse may be trained through the following steps:

[0045] First step, obtain a historical sequence set of the conversion information of the resources in the front-end warehouses of each tobacco drone in the target area. Among them, one front-end warehouse of a tobacco drone corresponds to a historical sequence of the conversion information of the resources in the front-end warehouse. The historical sequence of the conversion information of the resources in the front-end warehouse is usually a set of data sorted by time. The historical conversion information of the resources in the front-end warehouse can be used to represent the data of the resource conversion situation after allocating relevant resources to the front-end warehouse. Among them, the historical conversion information of the resources in the front-end warehouse may include the conversion efficiency of the resources, the resource allocation location, and the time. Among them, the conversion efficiency of the resources may be index data representing the resource conversion situation. The resource allocation location may represent the location information of the allocated resources. The time may represent the time information for data collection and acquisition.

[0046] Second step, select a historical sequence of the conversion information of the resources in the front-end warehouse from the above-mentioned historical sequence set of the conversion information of the resources in the front-end warehouse as the target historical sequence of the conversion information of the resources in the front-end warehouse. Among them, the target historical conversion information of the resources in the front-end warehouse includes: the conversion efficiency of the resources in the front-end warehouse.

[0047] Step 3: Perform masking processing on the pre-warehouse resource conversion efficiency included in any number of target historical pre-warehouse resource conversion information in the above-mentioned target historical pre-warehouse resource conversion information sequence to obtain a first pre-warehouse resource conversion information sample. Here, any number is less than the number of target historical pre-warehouse resource conversion information included in the target historical pre-warehouse resource conversion information sequence. For example, the execution subject can randomly select some target historical pre-warehouse resource conversion information from the target historical pre-warehouse resource conversion information sequence, and then perform masking processing on the pre-warehouse resource conversion efficiency in the selected target historical pre-warehouse resource conversion information. The method of masking processing is not limited, such as occlusion or marking. Then, the target historical pre-warehouse resource conversion information sequence after masking processing can be intercepted to obtain a historical pre-warehouse resource conversion information subsequence group. Each intercepted historical pre-warehouse resource conversion information subsequence contains at least one data that has been masked. After that, the masked historical pre-warehouse resource conversion information subsequence group after masking processing can be determined as the first pre-warehouse resource conversion information sample.

[0048] Among them, the above Step 3 can be implemented by the following steps:

[0049] Step 1: Intercept the above-mentioned target historical pre-warehouse resource conversion information sequence in the way of a sliding window to obtain a historical pre-warehouse resource conversion information subsequence group.

[0050] Step 2: For each historical pre-warehouse resource conversion information subsequence in the above-mentioned historical pre-warehouse resource conversion information subsequence group, select a preset number of historical pre-warehouse resource conversion information from this historical pre-warehouse resource conversion information subsequence, and perform masking processing on the pre-warehouse resource conversion efficiency included in each selected historical pre-warehouse resource conversion information through masking marks to obtain each masked historical pre-warehouse resource conversion information.

[0051] Step 3: Determine the masked historical pre-warehouse resource conversion information subsequence group after masking processing as the first pre-warehouse resource conversion information sample.

[0052] In the fourth step, the initial pre-warehouse resource conversion efficiency prediction model is pre-trained using the above-mentioned first pre-warehouse resource conversion information sample to obtain a pre-trained initial pre-warehouse resource conversion efficiency prediction model. The initial pre-warehouse resource conversion efficiency prediction model includes: an information input layer, an encoder network, and a prediction output layer. Among them, the information input layer is used to vectorize the input first pre-warehouse resource conversion information sample; the encoder network is used to learn the association relationship between the pre-warehouse resource conversion efficiency, the pre-warehouse resource configuration location, and the information collection time in the vector data of the first pre-warehouse resource conversion information sample, and to learn the association relationship between the pre-warehouse resource conversion efficiencies corresponding to different times at the same pre-warehouse resource configuration location. Each encoder in the encoder network includes a multi-head self-attention mechanism network layer and a feedforward neural network layer; the prediction output layer is used to output the pre-warehouse resource conversion efficiency prediction information at the masked position of the first pre-warehouse resource conversion information sample. The initial pre-warehouse resource conversion efficiency prediction model can be used to predict the resource conversion efficiency after masking. For example, the execution subject can input the first pre-warehouse resource conversion information sample into the initial pre-warehouse resource conversion efficiency prediction model to output the resource conversion efficiency prediction value at the masked position. Then, according to the resource conversion efficiency prediction value and the actual resource conversion efficiency value in the corresponding target historical pre-warehouse resource conversion information sequence, the model parameters of the initial pre-warehouse resource conversion efficiency prediction model are adjusted. The information input layer can be used to vectorize the input first pre-warehouse resource conversion information sample. That is, the resource conversion efficiency, the resource configuration location, and the time in the first pre-warehouse resource conversion information sample are vectorized. And these vector data can be output to the encoder network. The encoder network can be used to learn the relationship between the vector data output by the information input layer. The prediction output layer can be used to output the resource conversion efficiency prediction value at the masked position of the first pre-warehouse resource conversion information sample. The encoder network can adopt a multi-layer Transformer encoder. Each encoder includes a multi-head self-attention mechanism network layer (Multi-head Self-Attention, MSA) and a feedforward neural network layer (Feedforward Neural Network, FNN). The multi-layer stacking of the encoder network enables the model to learn different levels of abstract representations, from the underlying efficiency level to higher-level embedding representations. In addition, the prediction output layer can include a fully connected layer. The activation function (ReLU) in the fully connected layer can be used to constrain the output result to be positive. That is, to ensure that the prediction result is positive.

[0053] Among them, the above-mentioned fourth step can be implemented by the following steps:

[0054] Step 1: Input the above first pre-warehouse resource conversion information sample into the initial pre-warehouse resource conversion efficiency prediction model to obtain the pre-warehouse resource conversion efficiency prediction information at the masked positions.

[0055] Step 2: Adjust the network parameters of the above initial pre-warehouse resource conversion efficiency prediction model according to the above pre-warehouse resource conversion efficiency prediction information and the actual pre-warehouse resource conversion efficiency information corresponding to the target historical pre-warehouse resource conversion information sequence.

[0056] Fifth step: Mask the pre-warehouse resource conversion efficiency included in at least two target historical pre-warehouse resource conversion information at the end of the above target historical pre-warehouse resource conversion information sequence to obtain a second pre-warehouse resource conversion information sample. For example, in order to improve the learning effect of the model, the execution entity can also re-intercept the data in the target historical pre-warehouse resource conversion information sequence to obtain a historical pre-warehouse resource conversion information subsequence group. For each historical pre-warehouse resource conversion information subsequence, the resource conversion efficiency in several consecutive data at the end of the historical pre-warehouse resource conversion information subsequence can be masked, such as the first five data counted from the end of the sequence. Then, each masked historical pre-warehouse resource conversion information subsequence is used as sample data to obtain a second pre-warehouse resource conversion information sample.

[0057] Sixth step: Based on the above second pre-warehouse resource conversion information sample, fine-tune the pre-trained initial pre-warehouse resource conversion efficiency prediction model to obtain a fine-tuned pre-warehouse resource conversion efficiency prediction model. For example, the second pre-warehouse resource conversion information sample can be input into the pre-trained initial pre-warehouse resource conversion efficiency prediction model, and the resource conversion efficiency prediction value in the second pre-warehouse resource conversion information sample is output. Then, according to the resource conversion efficiency prediction value and the corresponding actual resource conversion efficiency value, fine-tune the model parameters of the pre-trained initial pre-warehouse resource conversion efficiency prediction model until the fine-tuning training is completed.

[0058] Thus, the resource utilization rate of the tobacco pre-warehouse can be predicted, and thus, it is convenient to adjust and optimize the resources of the tobacco pre-warehouse to avoid the vacancy and waste of resources.

[0059] Figure 2 The following is a schematic block diagram of the structure of a computer device provided by an embodiment of the present disclosure. This computer device can be a terminal.

[0060] As Figure 2 shown, the computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the memory can include a non-volatile storage medium and an internal memory.

[0061] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions that, when executed, enable the processor to execute any one of the tobacco distribution methods applied to the front compartment of a tobacco drone.

[0062] The processor is used to provide computing and control capabilities to support the operation of the entire computer device.

[0063] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, it enables the processor to execute any one of the tobacco distribution methods applied to the front compartment of a tobacco drone.

[0064] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that Figure 2 the structure shown in [diagram] is only a block diagram of some of the structures related to the solution of the present disclosure, and does not constitute a limitation on the computer device to which the solution of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout.

[0065] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0066] Among them, in one embodiment, the above-mentioned processor is used to run a computer program stored in the memory to implement the following steps: in response to receiving at least one tobacco change data sent by an external tobacco data monitoring terminal, perform the following update steps on the tobacco information in the associated target tobacco storage area: in response to receiving the tobacco delivery request information corresponding to the target tobacco drone front warehouse, perform structured processing on the above-mentioned tobacco delivery request information and the current tobacco storage information of the above-mentioned target tobacco drone front warehouse to obtain structured tobacco information, where the above-mentioned structured tobacco information includes: front warehouse structured information, front warehouse tobacco structured information, and front warehouse tobacco association information; determine the tobacco delivery restriction information corresponding to the above-mentioned tobacco delivery request information, where the above-mentioned tobacco delivery restriction information is at least one restriction condition corresponding to a preset front warehouse tobacco delivery model; input the above-mentioned front warehouse structured information, the above-mentioned front warehouse tobacco structured information, and the above-mentioned front warehouse tobacco association information into the above-mentioned front warehouse tobacco delivery model to obtain the front warehouse tobacco delivery information under the above-mentioned tobacco delivery restriction information; determine at least one to-be-delivered tobacco drone corresponding to the above-mentioned front warehouse tobacco delivery information; control the associated tobacco handling equipment to load tobacco onto the above-mentioned at least one to-be-delivered tobacco drone; in response to determining that the tobacco loading is completed, control at least one to-be-delivered tobacco drone to perform tobacco delivery.

[0067] The embodiments of the present disclosure further provide a computer-readable storage medium, on which a computer program is stored, and the computer program includes program instructions. The method implemented when the program instructions are executed can refer to the various embodiments of the tobacco delivery method applied to the tobacco drone front warehouse in the present disclosure.

[0068] Among them, the above-mentioned computer-readable storage medium may be an internal storage unit of the computer device in the foregoing embodiment, such as the hard disk or memory of the computer device. The above-mentioned computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device.

[0069] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or system. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or system including the element.

[0070] The serial numbers of the above embodiments of the present disclosure are only for description and do not represent the advantages or disadvantages of the embodiments. As described above, the above is only the specific implementation manner of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present disclosure can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A tobacco distribution method applied to a tobacco drone forward warehouse, comprising: In response to receiving tobacco delivery request information corresponding to a target tobacco drone forward warehouse, the tobacco delivery request information and the current tobacco storage information of the target tobacco drone forward warehouse are structured to obtain structured tobacco information, wherein the structured tobacco information includes: forward warehouse structured information, forward warehouse tobacco structured information, and forward warehouse tobacco associated information; Determining tobacco delivery restriction information corresponding to the tobacco delivery request information, wherein the tobacco delivery restriction information is at least one restriction condition corresponding to a preset forward warehouse tobacco delivery model; Inputting the forward warehouse structured information, the forward warehouse tobacco structured information, and the forward warehouse tobacco associated information into the forward warehouse tobacco distribution model to obtain the forward warehouse tobacco distribution information under the tobacco distribution restriction information; Determine at least one corresponding tobacco drone to be delivered according to the tobacco delivery information of the forward warehouse; Controlling the associated tobacco handling equipment to load tobacco onto the at least one tobacco delivery drone; In response to determining that tobacco loading is complete, controlling at least one tobacco delivery drone to deliver the tobacco; In response to receiving a forward warehouse resource conversion prediction request for the target tobacco drone forward warehouse, predicting the forward warehouse resource conversion efficiency of the target tobacco drone forward warehouse at a target time through a pre-trained forward warehouse resource conversion efficiency prediction model; According to the generated forward warehouse resource conversion efficiency prediction results, the drone delivery resources and tobacco storage resources of the target tobacco drone forward warehouse are adjusted; Among them, the forward warehouse resource conversion efficiency prediction model is trained through the following steps: Obtain a historical forward warehouse resource conversion information sequence set of each tobacco drone forward warehouse in the target area, wherein one tobacco drone forward warehouse corresponds to one historical forward warehouse resource conversion information sequence; Selecting a historical forward warehouse resource conversion information sequence from the historical forward warehouse resource conversion information sequence set as a target historical forward warehouse resource conversion information sequence, wherein the target historical forward warehouse resource conversion information includes: forward warehouse resource conversion efficiency; Masking the forward warehouse resource conversion efficiencies included in any number of target historical forward warehouse resource conversion information in the target historical forward warehouse resource conversion information sequence to obtain a first forward warehouse resource conversion information sample, wherein the any number is less than the number of target historical forward warehouse resource conversion information included in the target historical forward warehouse resource conversion information sequence; Pre-training an initial forward warehouse resource conversion efficiency prediction model through the first forward warehouse resource conversion information sample to obtain a pre-trained initial forward warehouse resource conversion efficiency prediction model; Performing mask processing on forward warehouse resource conversion efficiencies included in at least two target historical forward warehouse resource conversion information at the end of the target historical forward warehouse resource conversion information sequence to obtain a second forward warehouse resource conversion information sample; Based on the second forward warehouse resource conversion information sample, fine-tune the pre-trained initial forward warehouse resource conversion efficiency prediction model to obtain a fine-tuned forward warehouse resource conversion efficiency prediction model; The step of performing mask processing on the forward warehouse resource conversion efficiencies included in any number of target historical forward warehouse resource conversion information in the target historical forward warehouse resource conversion information sequence to obtain a first forward warehouse resource conversion information sample includes: By means of a sliding window, the target historical forward warehouse resource conversion information sequence is intercepted to obtain a historical forward warehouse resource conversion information subsequence group; For each historical forward warehouse resource conversion information subsequence in the historical forward warehouse resource conversion information subsequence group, a preset number of historical forward warehouse resource conversion information are selected from the historical forward warehouse resource conversion information subsequence, and the forward warehouse resource conversion efficiency included in each selected historical forward warehouse resource conversion information is masked through mask marking to obtain each masked historical forward warehouse resource conversion information; The masked historical forward warehouse resource conversion information subsequence group after mask processing is determined as the first forward warehouse resource conversion information sample.

2. The tobacco distribution method applied to the tobacco drone forward warehouse according to claim 1 is characterized in that: The forward warehouse tobacco distribution model includes: a forward warehouse tobacco restriction condition group; and the tobacco distribution restriction information corresponding to the tobacco distribution request information is determined, including: Each forward warehouse tobacco restriction condition in the forward warehouse tobacco restriction condition group corresponding to the tobacco delivery request information is determined as tobacco delivery restriction information.

3. The tobacco distribution method applied to the tobacco drone forward warehouse according to claim 2 is characterized in that: The tobacco distribution restriction information includes: a basic forward warehouse tobacco restriction condition group, a tobacco distribution restriction condition group and a forward warehouse tobacco adjustment restriction condition group; the forward warehouse tobacco distribution model includes: a forward warehouse tobacco calculation function group; and the forward warehouse structured information, the forward warehouse tobacco structured information, and the forward warehouse tobacco associated information are input into the forward warehouse tobacco distribution model to obtain the forward warehouse tobacco distribution information under the tobacco distribution restriction information, including: Determine each forward warehouse tobacco calculation function corresponding to the tobacco delivery request information in the forward warehouse tobacco calculation function group as a target forward warehouse tobacco calculation function group; According to the basic forward warehouse tobacco restriction condition group, the tobacco delivery restriction condition group and the forward warehouse tobacco adjustment restriction condition group, the forward warehouse structured information, the forward warehouse tobacco structured information and the forward warehouse tobacco associated information are input into the target forward warehouse tobacco calculation function group to obtain the forward warehouse tobacco delivery information.

4. A computer device, wherein: The computer device comprises a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 3 are implemented.

5. A computer-readable storage medium, wherein: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images