An intelligent unmanned retail automatic interactive system
Through the intelligent unmanned retail automatic interaction system, transaction simulation and order matching model are used to solve the problems of unreasonable inventory management and route fixation in the unmanned retail vehicle system, and the linkage between unmanned retail vehicles is realized to avoid inventory accumulation and Improve user satisfaction.
Patent Information
- Application Number
- CN202410254223.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-06
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-03-06
AI Technical Summary
The lack of linkage between unmanned retail vehicles in the existing unmanned retail vehicle system has led to unreasonable inventory management, easy inventory accumulation, and fixed routes cannot be adapted to the inventory situation.
Design an intelligent unmanned retail automatic interaction system, by obtaining retail characteristic data of each route, conducting transaction simulation, predicting inventory volume, calculating retail evaluation coefficients, determining low-inventory and high-inventory unmanned retail vehicles, and adjusting route planning. And judge whether to generate a delivery instruction through the order pairing model to realize linkage between unmanned retail vehicles.
It effectively avoids inventory accumulation, breaks the fixed route limitations of unmanned retail vehicles, improves the rationality of inventory management and user satisfaction, and improves the shopping experience.
Smart Images

Figure CN118071459B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned retail vehicles, and more specifically, to an intelligent unmanned retail automatic interaction system. Background Art
[0002] With the advancement of science and technology, unmanned retail vehicles have gradually attracted attention as an innovative form of retail industry. The traditional retail model is limited to fixed space, while unmanned retail vehicles can realize commodity retail anytime and anywhere through automation technology and intelligent systems, providing consumers with a more flexible and convenient shopping experience. However, the current unmanned retail vehicle system is mainly deployed in tourist attractions, squares, subway stations and other areas with dense traffic, and circulates along fixed routes, causing some unmanned retail vehicles to have inventory backlogs, limiting the ability of unmanned retail vehicles to generate revenue. In order to solve these problems, a system is needed that can provide real-time user information during the operation of unmanned retail vehicles, improve intelligent interactive design and optimize inventory management. For example, the Chinese patent with publication number CN111147825A discloses an unmanned retail vehicle and a control method for waving and stopping. Although the method can replace manual operation to complete the action of waving and stopping by using an unmanned retail vehicle, it can bring pedestrians a good experience, can accurately identify whether pedestrians on the roadside are waving at the unmanned retail vehicle, and improve the service level of the unmanned retail vehicle, but the inventor has studied and applied the above method and the prior art and found that the above method and the prior art have at least the following defects:
[0003] (1) There is a lack of linkage between unmanned retail vehicles, and it is impossible to reasonably configure the inventory management of unmanned retail vehicles, which easily leads to inventory accumulation in unmanned retail vehicles. In particular, products with a short shelf life are more likely to deteriorate due to inventory accumulation;
[0004] (2) The routes of unmanned retail vehicles are fixed and cannot be adjusted according to the actual inventory situation.
[0005] To this end, the present invention provides an intelligent unmanned retail automatic interaction system. Summary of the invention
[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent unmanned retail automatic interaction system to solve the problems raised in the above-mentioned background technology.
[0007] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent unmanned retail automatic interaction system, comprising:
[0008] A retail evaluation module is used to obtain the retail feature data of the unmanned retail vehicle in each retail route, input the retail feature data into a pre-built transaction simulation model for simulation, so as to predict the retail inventory of each unmanned retail vehicle; and calculate the retail evaluation coefficient of each unmanned retail vehicle based on the retail feature data and the retail inventory;
[0009] An inventory analysis module, which determines low-inventory unmanned retail vehicles within a unit time based on the retail evaluation coefficient, and selects high-inventory unmanned retail vehicles within a unit time that have the same functions as the low-inventory unmanned retail vehicles;
[0010] Route planning module, which determines the best planned route based on high-inventory unmanned retail vehicles and low-inventory unmanned retail vehicles;
[0011] Interaction module, when the high-inventory unmanned retail vehicle arrives at the low-inventory unmanned retail location along the optimal planned route, the high-inventory unmanned retail vehicle collects commodity retail data, inputs the commodity retail data information into the pre-built order matching model to obtain the order matching coefficient, compares the order matching coefficient with the order matching coefficient threshold, and determines whether to generate a delivery instruction based on the comparison result.
[0012] Furthermore, the retail characteristic data includes the actual sales volume of each unmanned retail vehicle, the total operating time of each unmanned retail vehicle, the total inventory of goods of each unmanned retail vehicle, weather data and traffic volume; the retail inventory is the remaining inventory of goods of each unmanned retail vehicle;
[0013] The weather data is humidity or temperature; the weather data is obtained through a weather forecast platform; the actual sales volume of each unmanned retail vehicle is the number of commodities actually sold by the unmanned retail vehicle in unit time; the total operating time of each unmanned retail vehicle is all time between the start and end of the operation of the unmanned retail vehicle; the total inventory of commodities of each unmanned retail vehicle is the total number of all commodities carried by the unmanned retail vehicle; the pedestrian flow is the number of pedestrian flows per unit area of each unmanned retail vehicle; the actual sales volume of each unmanned retail vehicle, the total operating time of each unmanned retail vehicle, the total inventory of commodities of each unmanned retail vehicle, the weather data and the pedestrian flow are all pre-stored in the cloud server.
[0014] Furthermore, the method for constructing the transaction simulation model includes:
[0015] Obtaining historical supervision data pre-stored in the cloud server, the historical supervision data including retail feature data of each unmanned retail vehicle and the corresponding actual retail inventory, dividing the historical supervision data into a data training set and a data test set, and building a training network;
[0016] The retail feature data of each unmanned retail vehicle in the historical supervision data is used as the input of the training network, and the actual retail inventory of each unmanned retail vehicle is used as the output of the training network. The training network is trained to obtain an initial machine learning model, and the initial machine learning model is verified using a data test set to output an initial machine learning model that meets a preset accuracy rate as a transaction simulation model; the transaction simulation model is one of a linear regression model, a logistic regression model or a neural network model.
[0017] Furthermore, the method for obtaining the retail evaluation coefficient of the unmanned retail vehicle includes:
[0018] A formulaic calculation is performed based on the retail characteristic data and the retail inventory to obtain the retail evaluation coefficient of each retail route; the calculation formula is:
[0019]
[0020] Where: Lsp is the retail assessment coefficient, Sx i is the actual sales volume of the i-th unmanned retail vehicle, Sc i is the total operating time of the i-th unmanned retail vehicle; Zk i is the total inventory of goods of the i-th unmanned retail vehicle, Yk i To predict the retail inventory of the i-th unmanned retail vehicle, Tq is the weather data, γ is the weight of the weather data; α 1 is the weight of the actual sales volume of the i-th unmanned retail vehicle; α 2 is the weight of the total operating time of the i-th unmanned retail vehicle; RL is the flow of people in the target supervision area during the T period; α 3 is the weight of the flow of people in the target supervision area during the T period;
[0021] Preset retail evaluation coefficient thresholds Y1 and Y2, Y1>Y2, and compare the retail evaluation coefficient with the retail evaluation coefficient threshold;
[0022] If the retail evaluation coefficient is greater than or equal to the retail evaluation coefficient threshold Y1, it is determined that the corresponding unmanned retail vehicle has a large transaction volume, and the first retail vehicle is marked;
[0023] If the retail evaluation coefficient is less than the retail evaluation coefficient threshold value Y1 and greater than the retail evaluation coefficient threshold value Y2, then the corresponding unmanned retail vehicle transaction volume is determined to be normal;
[0024] If the retail evaluation coefficient is less than or equal to the retail evaluation coefficient threshold Y2, the corresponding unmanned retail vehicle is judged to have a low transaction volume and is marked as a non-target retail vehicle.
[0025] Furthermore, the method for determining the low inventory of unmanned retail vehicles per unit time includes:
[0026] Get the inventory valuation coefficient of the first retail vehicle;
[0027] Calculate the difference between the inventory assessment coefficient of the first retail vehicle and the corresponding retail assessment coefficient, and mark it as the first assessment difference;
[0028] If the first evaluation difference is less than or equal to the preset first evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a low-inventory unmanned retail vehicle;
[0029] The calculation formula of the inventory valuation coefficient is:
[0030]
[0031] Among them, Kcp is the inventory assessment coefficient; Bz is the replenishment cycle of the i-th unmanned retail vehicle, β is the weight factor of the replenishment cycle of the i-th unmanned retail vehicle, and Lk i is the historical inventory of the ith unmanned retail vehicle; the historical inventory is the remaining inventory of goods of each unmanned retail vehicle in the past unit time, and the replenishment cycle and the historical inventory are pre-stored in the cloud server.
[0032] Furthermore, the method for screening out the high-inventory unmanned retail vehicle having the same function as the low-inventory unmanned retail vehicle within a unit time includes:
[0033] Determine whether there is a low-inventory unmanned retail vehicle among the non-target retail vehicles. If so, use the corresponding non-target retail vehicle as the first retail vehicle. If not, use the corresponding non-target retail vehicle as the second retail vehicle.
[0034] Extract and obtain the inventory valuation coefficient of the second retail vehicle;
[0035] Calculate the difference between the inventory valuation coefficient and the corresponding retail valuation coefficient, and mark the retail valuation coefficient as the second valuation difference;
[0036] If the second evaluation difference is less than or equal to the preset second evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a high-inventory unmanned retail vehicle, and Q high-inventory unmanned retail vehicles are obtained, where Q is an integer set greater than zero;
[0037] Sorting the second evaluation differences that are greater than or equal to a preset second evaluation difference threshold from large to small according to their values;
[0038] The unmanned retail vehicle corresponding to the second evaluation difference in the first ranking is regarded as a high-inventory unmanned retail vehicle.
[0039] Furthermore, the method for determining the best planned route based on the high-inventory unmanned retail vehicle and the low-inventory unmanned retail vehicle includes:
[0040] Obtain a layout map of the target supervision area, and based on the layout map of the target supervision area, use the location of the high-inventory unmanned retail vehicle as a moving point, and use the location of the low-inventory unmanned retail vehicle as a destination point;
[0041] Based on the layout diagram of the target supervision area, the low-inventory unmanned retail vehicle locations are connected with the high-inventory unmanned retail vehicle locations to obtain X retail routes, where X is an integer greater than zero;
[0042] Each planned route is simulated using a pre-built digital twin model to obtain route training data for each planned route, and a route evaluation coefficient for each planned route is calculated based on the route training data; the route training data includes the length of each planned route, the moving time of a high-inventory unmanned retail vehicle on each planned route, the parking time of an unmanned retail vehicle on each planned route, and the number of times an unmanned retail vehicle passes through each planned route; the route training data calculation formula is:
[0043]
[0044] In the formula, Xp is the line evaluation coefficient, Gh is the length of the planned route, and Wr G is the moving time of the unmanned retail vehicle with high inventory, Jz G is the parking time of the high-inventory unmanned retail vehicle, Lg is the number of times the unmanned retail vehicle passes by, θ 1 ,θ 2 and θ 3 is the correction factor, and θ 1 ,θ 2 and θ 3 are greater than zero;
[0045] Sort the route evaluation coefficient of each planned route from small to large;
[0046] The retail route corresponding to the minimum route evaluation coefficient is taken as the optimal planned route.
[0047] Furthermore, the method of obtaining the order matching coefficient includes:
[0048] It should be noted that the commodity retail data includes the quantity of commodities, the amount of commodities and the payment voucher matching coefficient; the commodity retail data is obtained by the client device sending an order to the cloud server.
[0049] The method for constructing the order matching model includes:
[0050] The order matching model adopts a linear regression model, with the product quantity Ss, product amount Je and payment voucher matching coefficient Px as input and the order matching coefficient Dp as output;
[0051] Build a linear regression model:
[0052] Dp=β 1 ×Ss+β 2 ×Je+β 3 ×Px;
[0053] In the formula, β 1 , β 2 With β 3 They are all preset proportional coefficients, and their values are all greater than zero; the payment credential matching coefficient refers to the security verification data sent by the user to the cloud server collected by the high-inventory unmanned retail vehicle, and obtained by matching analysis of the security verification data.
[0054] Furthermore, the method for determining whether to generate a shipping instruction according to the comparison result includes:
[0055] Preset an order matching coefficient threshold, and compare the order matching coefficient with the order matching coefficient threshold;
[0056] When the order matching coefficient is less than or equal to the order matching coefficient threshold, a shipping instruction is generated;
[0057] When the order matching coefficient is greater than the order matching coefficient threshold, no shipping instruction is generated;
[0058] In a second aspect, the present invention provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;
[0059] The processor executes the above-mentioned intelligent unmanned retail automatic interaction system by calling the computer program stored in the memory.
[0060] Technical effects and advantages of the present invention:
[0061] 1. The present invention obtains retail feature data of unmanned retail vehicles in each retail route, inputs the retail feature data into a pre-built transaction simulation model for simulation, so as to predict the retail inventory of each unmanned retail vehicle; and calculates the retail evaluation coefficient of each unmanned retail vehicle based on the retail feature data and the retail inventory; determines the low-inventory unmanned retail vehicle in a unit time based on the retail evaluation coefficient, and screens out the high-inventory unmanned retail vehicle with the same function as the low-inventory unmanned retail vehicle in a unit time; determines the best planned route based on the high-inventory unmanned retail vehicle and the low-inventory unmanned retail vehicle; when the high-inventory unmanned retail vehicle arrives at the retail location of the low-inventory unmanned retail vehicle along the best planned route, the high-inventory unmanned retail vehicle collects commodity retail data, inputs the commodity retail data information into a pre-built order matching model to obtain an order matching coefficient, compares the order matching coefficient with the order matching coefficient threshold, and determines whether to generate a shipping instruction according to the comparison result; the present invention increases linkage between unmanned retail vehicles, thereby facilitating reasonable configuration of inventory management of unmanned retail vehicles and facilitating avoiding inventory accumulation of unmanned retail vehicles;
[0062] 2. The present invention simulates the inventory of unmanned retail vehicles to determine the locations of low-inventory unmanned retail vehicles and high-inventory unmanned retail vehicles, and obtains a layout map of the target supervision area. Based on the layout map of the target supervision area, the location of the high-inventory unmanned retail vehicle is used as a moving point, and the location of the low-inventory unmanned retail vehicle is used as a destination point; the location of the low-inventory unmanned retail vehicle is connected with the location of the high-inventory unmanned retail vehicle by line, and the line is analyzed to obtain the best planned route, which is conducive to breaking the limitations of the fixed operating route of the unmanned retail vehicle and responding at any time, thereby helping to improve user satisfaction and shopping experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] Figure 1 is a system schematic diagram of Example 1;
[0064] Figure 2 A local layout diagram of the target regulatory area in Example 1;
[0065] Figure 3 This is a schematic diagram of an electronic device according to Embodiment 2;
[0066] Figure 4 A schematic diagram of a computer-readable storage medium of Example 3. DETAILED DESCRIPTION
[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0068] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.
[0069] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0070] Example 1
[0071] See also Figure 1 As shown, this embodiment discloses a smart unmanned retail automatic interaction system, including: a retail assessment module, an inventory analysis module, a route planning module and an interaction module. The modules are connected by wired and / or wireless means to achieve data transmission.
[0072] The retail evaluation module is used to obtain the retail feature data of the unmanned retail vehicle in each retail route, input the retail feature data into the pre-built transaction simulation model for simulation, so as to predict the retail inventory of each unmanned retail vehicle; and calculate the retail evaluation coefficient of each unmanned retail vehicle based on the retail feature data and the retail inventory;
[0073] It should be noted that: the present invention is applied to a target supervision area with i retail routes, the target management area includes but is not limited to campuses, parks or tourist attractions, etc.; i is an integer greater than zero, and each retail route includes an unmanned retail vehicle; and monitoring equipment and sensors are arranged in the target supervision area, the sensors include but are not limited to GPS positioning sensors, inductive sensors, laser radars and people counters, etc., the monitoring equipment and sensors are remotely connected to the cloud server by electrical and / or wireless means, and the cloud server is also wirelessly connected to each client device; the monitoring equipment is used to monitor the retail situation and operating status of each retail route and the unmanned retail vehicle in each retail route, etc.;
[0074] It should be noted that the retail characteristic data includes but is not limited to the actual sales volume of each unmanned retail vehicle, the total operating time of each unmanned retail vehicle, the total inventory of goods of each unmanned retail vehicle, weather data and traffic volume, etc.; the retail inventory is the remaining inventory of goods of each unmanned retail vehicle;
[0075] It should be noted that the weather data includes but is not limited to rainfall, humidity, temperature, etc.; the weather data is provided by the weather forecast platform; the actual sales volume of each unmanned retail vehicle is the number of commodities actually sold by the unmanned retail vehicle in a unit time; the total operation time of each unmanned retail vehicle is all the time between the start and end of the operation of the unmanned retail vehicle; the total inventory of commodities of each unmanned retail vehicle is the total number of all commodities carried by the unmanned retail vehicle; the pedestrian flow is the number of pedestrians per unit area of each unmanned retail vehicle, and the area of the ground space where the unmanned retail vehicle per unit area is located, and the specific area is obtained by artificial pre-division and is not limited too much here; the actual sales volume of each unmanned retail vehicle, the total operation time of each unmanned retail vehicle, the total inventory of commodities of each unmanned retail vehicle, weather data, pedestrian flow and other data are all pre-stored in the cloud server;
[0076] It should be noted that: the digital twin model is a virtual simulation model of a target regulatory area with M retail routes, and the digital twin model includes a physical model, a logical model and a data model; the construction process of the digital twin model includes: first, collecting basic information in the target regulatory area, including but not limited to structural parameters, geometric parameters, material parameters, state parameters and boundary conditions of retail routes, unmanned retail vehicles and pedestrian flow; using the acquired basic information, using 3D modeling tools (such as Creoparametric, UGNX or 3DMax, etc.) to build a physical model to simulate the actual situation of the target regulatory area; secondly, on the basis of the physical model, a logical model is established to describe the components, organizational structure and operation mechanism of the model, so as to optimize and adjust the physical model; based on the physical model, the logical model and other relevant data, a computable data model is constructed; finally, the physical model, the logical model, the data model and the transaction simulation model are integrated together to construct a complete digital twin model, so that the digital twin model can be accurately simulated and analyzed. The digital twin model is a prior art and will not be described in detail.
[0077] It should also be noted that: the transaction simulation model is used to simulate and predict the retail inventory of each unmanned retail vehicle; further, the transaction simulation model is specifically one of a linear regression model, a logistic regression model or a neural network model;
[0078] The method for constructing the transaction simulation model includes:
[0079] Obtain a large amount of historical supervision data pre-stored in the cloud server, wherein the large amount of historical supervision data includes retail feature data of each unmanned retail vehicle and the corresponding actual retail inventory, divide the large amount of historical supervision data into a data training set and a data test set, and construct a training network;
[0080] The retail feature data of each unmanned retail vehicle in a large amount of historical regulatory data is used as the input of the training network, and the actual retail inventory of each unmanned retail vehicle is used as the output of the training network. The training network is trained to obtain an initial machine learning model, and the initial machine learning model is verified using a data test set to output an initial machine learning model that meets the preset accuracy rate as a transaction simulation model;
[0081] Specifically, the method for obtaining the retail evaluation coefficient of the unmanned retail vehicle includes:
[0082] A formulaic calculation is performed based on the retail characteristic data and the retail inventory to obtain the retail evaluation coefficient of each retail route; the calculation formula is:
[0083]
[0084] Where: Lsp is the retail assessment coefficient, Sx i is the actual sales volume of the i-th unmanned retail vehicle, Sc i is the total operating time of the i-th unmanned retail vehicle; Zk i is the total inventory of goods of the i-th unmanned retail vehicle, Yk i To predict the retail inventory of the i-th unmanned retail vehicle, Tq is the weather data, γ is the weight of the weather data; α 1 is the weight of the actual sales volume of the i-th unmanned retail vehicle; α 2 is the weight of the total operating time of the i-th unmanned retail vehicle; RL is the flow of people in the target supervision area during the T period; α 3 is the weight of the flow of people in the target supervision area during the T period;
[0085] Preset retail evaluation coefficient thresholds Y1 and Y2, Y1>Y2, and compare the retail evaluation coefficient with the retail evaluation coefficient threshold;
[0086] If the retail evaluation coefficient is greater than or equal to the retail evaluation coefficient threshold Y1, it is determined that the corresponding unmanned retail vehicle has a large transaction volume, and the first retail vehicle is marked;
[0087] If the retail evaluation coefficient is less than the retail evaluation coefficient threshold value Y1 and greater than the retail evaluation coefficient threshold value Y2, then the corresponding unmanned retail vehicle transaction volume is determined to be normal;
[0088] If the retail evaluation coefficient is less than or equal to the retail evaluation coefficient threshold Y2, the corresponding unmanned retail vehicle is judged to have a low transaction volume and is marked as a non-target retail vehicle;
[0089] It should be noted that: judging that the transaction volume of the unmanned retail vehicle is high means that the transaction volume of the unmanned retail vehicle is large, and there may be insufficient inventory of goods remaining; judging that the transaction volume of the unmanned retail vehicle is normal means that there is no insufficient inventory of goods in the unmanned retail vehicle; judging that the transaction volume of the unmanned retail vehicle is low means that there is no insufficient inventory of goods in the unmanned retail vehicle, but because the user has purchased goods in the first retail vehicle and has no purchase demand, the unmanned retail vehicle has a large amount of remaining inventory;
[0090] It should also be noted that: if the retail evaluation coefficient is less than or equal to the retail evaluation coefficient threshold value Y2, it is determined that the corresponding unmanned retail vehicle has a low transaction volume. This situation may be caused by the first retail vehicle. Further explanation is that if the user has purchased goods in the first retail vehicle and has no purchase demand, it will affect the remaining inventory of subsequent unmanned retail vehicles, thereby causing the unmanned retail vehicle to have a low transaction volume;
[0091] It should be further explained that a large transaction volume of unmanned retail vehicles does not mean that the unmanned retail vehicles have a large inventory surplus, and further analysis is needed for this;
[0092] An inventory analysis module, which determines low-inventory unmanned retail vehicles within a unit time based on the retail evaluation coefficient, and selects high-inventory unmanned retail vehicles within a unit time that have the same functions as the low-inventory unmanned retail vehicles;
[0093] It should be noted that the unit time is obtained by artificial division in advance. For example, assuming that the operating time of each unmanned retail vehicle in a day is 300 minutes, if 300 minutes are divided into 5 time periods, the span of each time period is 60 minutes; if 300 minutes are divided into 10 time periods, the span of each time period is 30 minutes. The specific division rules are determined according to the actual situation of each unmanned retail vehicle, and the present invention does not impose too many restrictions on this.
[0094] Specifically, the method for determining the low inventory of unmanned retail vehicles per unit time includes:
[0095] Get the inventory valuation coefficient of the first retail vehicle;
[0096] It should be noted that the calculation formula of the inventory assessment coefficient is:
[0097]
[0098] Among them, Kcp is the inventory assessment coefficient; Bz is the replenishment cycle of the i-th unmanned retail vehicle, β is the weight factor of the replenishment cycle of the i-th unmanned retail vehicle, and Lk i is the historical inventory of the ith unmanned retail vehicle; the historical inventory is the remaining inventory of goods of each unmanned retail vehicle in the past unit time, the replenishment cycle and the historical inventory are pre-stored in the cloud server, based on the above, the predicted retail inventory of each unmanned retail vehicle is obtained based on the simulation of the digital twin model;
[0099] Calculate the difference between the inventory assessment coefficient of the first retail vehicle and the corresponding retail assessment coefficient, and mark it as the first assessment difference;
[0100] If the first evaluation difference is less than or equal to the preset first evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a low-inventory unmanned retail vehicle;
[0101] Specifically, the method for screening out the high-inventory unmanned retail vehicle having the same function as the low-inventory unmanned retail vehicle within a unit time includes:
[0102] Determine whether there is a low-inventory unmanned retail vehicle among the non-target retail vehicles. If so, use the corresponding non-target retail vehicle as the first retail vehicle. If not, use the corresponding non-target retail vehicle as the second retail vehicle.
[0103] It should be noted that: based on the above, if the user has not purchased goods in the first retail vehicle, it will affect the remaining inventory of the subsequent unmanned retail vehicle, thereby causing the unmanned retail vehicle to have a large transaction volume. Therefore, this type of unmanned retail vehicle still belongs to the first retail vehicle; on the contrary, if there is no low-inventory unmanned retail vehicle in the non-target retail vehicle, it will be used as the second retail vehicle. Through the above settings, it is possible to distinguish whether there is a low-inventory unmanned retail vehicle in each retail route, which is conducive to establishing a foundation for the subsequent determination of high-inventory unmanned retail vehicles, and at the same time, it is conducive to promoting the linkage between subsequent unmanned retail vehicles;
[0104] Extract and obtain the inventory valuation coefficient of the second retail vehicle;
[0105] Calculate the difference between the inventory valuation coefficient and the corresponding retail valuation coefficient, and mark the retail valuation coefficient as the second valuation difference;
[0106] If the second evaluation difference is less than or equal to the preset second evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a high-inventory unmanned retail vehicle, and Q high-inventory unmanned retail vehicles are obtained, where Q is an integer set greater than zero;
[0107] Sorting the second evaluation differences that are greater than or equal to a preset second evaluation difference threshold from large to small according to their values;
[0108] It should be understood that: according to the value from small to large, the smaller the second evaluation difference is, the smaller the sales volume of the corresponding high-inventory unmanned retail vehicle is. Further explanation is that the smaller the second evaluation difference is, the corresponding high-inventory unmanned retail vehicle is in a state of low sales volume, or may be infinitely close to or equal to a state of no sales volume;
[0109] The unmanned retail vehicle corresponding to the second evaluation difference in the first ranking is regarded as a high-inventory unmanned retail vehicle;
[0110] Route planning module, which determines the best planned route based on high-inventory unmanned retail vehicles and low-inventory unmanned retail vehicles;
[0111] It should be noted that the methods for determining the best planning route include:
[0112] Obtain a layout map of the target supervision area, and based on the layout map of the target supervision area, use the location of the high-inventory unmanned retail vehicle as a moving point, and use the location of the low-inventory unmanned retail vehicle as a destination point;
[0113] Based on the layout diagram of the target supervision area, the low-inventory unmanned retail vehicle locations are connected with the high-inventory unmanned retail vehicle locations to obtain X retail routes, where X is an integer greater than zero;
[0114] It should be noted that: the layout diagram of the target supervision area is pre-stored in the cloud server, and the layout diagram of the target supervision area at least includes the length data of each route segment, the number of unmanned retail vehicles, and the position of each unmanned retail vehicle;
[0115] For example, see Figure 2 As shown in the local layout diagram of the target supervision area, it is assumed that A is the location of the low-inventory unmanned retail vehicle, that is, A is the destination point; B is the location of the high-inventory unmanned retail vehicle, that is, B is the moving point; it is assumed that the destination point is connected to the moving point by line, and three planned routes X1, X2 and X3 are obtained, so at this time X=3;
[0116] Each planned route is simulated using a pre-built digital twin model to obtain route training data for each planned route, and a route evaluation coefficient for each planned route is calculated based on the route training data; the route training data includes the length of each planned route, the moving time of a high-inventory unmanned retail vehicle on each planned route, the parking time of an unmanned retail vehicle on each planned route, and the number of times an unmanned retail vehicle passes through each planned route; the route training data calculation formula is:
[0117]
[0118] In the formula, Xp is the line evaluation coefficient, Gh is the length of the planned route, and Wr G is the moving time of the unmanned retail vehicle with high inventory, Jz G is the parking time of the high-inventory unmanned retail vehicle, Lg is the number of times the unmanned retail vehicle passes by, θ 1 ,θ 2 and θ 3 is the correction factor, and θ 1 ,θ 2 and θ 3 are greater than zero;
[0119] It should be understood that the parking time of the high-inventory unmanned retail vehicle refers to the stop time generated to avoid other unmanned retail vehicles; the larger the route evaluation coefficient, the less moving time the high-inventory unmanned retail vehicle has, the shorter the parking time of the high-inventory unmanned retail vehicle, and the shorter the length of the planned route;
[0120] Sort the route evaluation coefficient of each planned route from small to large;
[0121] The retail route corresponding to the minimum route evaluation coefficient is taken as the best planned route;
[0122] Interaction module, when the high-inventory unmanned retail vehicle arrives at the low-inventory unmanned retail location along the optimal planned route, the high-inventory unmanned retail vehicle collects commodity retail data, inputs the commodity retail data information into the pre-built order matching model to obtain the order matching coefficient, compares the order matching coefficient with the order matching coefficient threshold, and determines whether to generate a delivery instruction based on the comparison result.
[0123] Methods for obtaining order matching coefficients include:
[0124] It should be noted that the commodity retail data includes the quantity of commodities, the amount of commodities and the payment voucher matching coefficient; the commodity retail data is obtained by the client device sending an order to the cloud server.
[0125] The method for constructing the order matching model includes:
[0126] The order matching model adopts a linear regression model, with the product quantity Ss, product amount Je and payment voucher matching coefficient Px as input and the order matching coefficient Dp as output;
[0127] Build a linear regression model:
[0128] Dp=β 1 ×Ss+β 2 ×Je+β 3 ×Px;
[0129] In the formula, β 1 , β 2 With β 3 They are all preset proportional coefficients, and their values are all greater than zero.
[0130] It should be noted that the payment credential matching coefficient refers to the security verification data sent by the user to the cloud server collected by the high-inventory unmanned retail vehicle, and the security verification data is obtained by matching analysis. The security verification data includes but is not limited to verification codes, passwords, etc. The security verification data is used to ensure effective identity authentication during order generation. When the payment credential matching coefficient is higher, it means that the security verification data is more closely matched with the user's identity, and the order matching coefficient will also increase accordingly, thereby increasing the credibility of the order. Therefore, the larger the payment credential matching coefficient, the higher the matching and credibility of the order can be, ensuring the safety and effectiveness of the transaction process. Matching analysis of security verification data is a prior art, and no further elaboration will be given.
[0131] Methods for determining whether to generate a shipping instruction based on the comparison results include:
[0132] Preset an order matching coefficient threshold, and compare the order matching coefficient with the order matching coefficient threshold;
[0133] When the order matching coefficient is less than or equal to the order matching coefficient threshold, a shipping instruction is generated;
[0134] When the order matching coefficient is greater than the order matching coefficient threshold, no shipping instruction is generated;
[0135] The present invention obtains retail feature data of unmanned retail vehicles in each retail route, inputs the retail feature data into a pre-built transaction simulation model for simulation, so as to predict the retail inventory of each unmanned retail vehicle; and calculates the retail evaluation coefficient of each unmanned retail vehicle based on the retail feature data and the retail inventory; determines the low-inventory unmanned retail vehicle in a unit time based on the retail evaluation coefficient, and screens out the high-inventory unmanned retail vehicle with the same function as the low-inventory unmanned retail vehicle in a unit time; determines the best planned route based on the high-inventory unmanned retail vehicle and the low-inventory unmanned retail vehicle; when the high-inventory unmanned retail vehicle arrives at the retail location of the low-inventory unmanned retail vehicle along the best planned route, the high-inventory unmanned retail vehicle collects commodity retail data, inputs the commodity retail data information into a pre-built order matching model, so as to obtain an order matching coefficient, compare the order matching coefficient with the order matching coefficient threshold, and judge whether to generate a shipping instruction according to the comparison result; the present invention increases linkage between unmanned retail vehicles, thereby facilitating reasonable configuration of inventory management of unmanned retail vehicles and facilitating avoiding the situation of inventory accumulation of unmanned retail vehicles;
[0136] The present invention simulates the inventory of unmanned retail vehicles, determines the locations of low-inventory unmanned retail vehicles and high-inventory unmanned retail vehicles, and obtains the layout map of the target supervision area.
[0137] The location of the unmanned retail vehicle with high inventory is used as the moving point, and the location of the unmanned retail vehicle with low inventory is used as the destination point; the location of the unmanned retail vehicle with low inventory is connected with the location of the unmanned retail vehicle with high inventory, and the route is analyzed to obtain the best planned route, which is conducive to breaking the limitations of the fixed operating route of the unmanned retail vehicle and responding at any time, which is conducive to improving user satisfaction and shopping experience.
[0138] The formulas involved in the above are all calculated by removing dimensions and taking their numerical values. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The weight factors in the formula and the preset thresholds in the analysis process are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data; the size of the weight factor is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the weight factor depends on the amount of sample data and the corresponding processing coefficient initially set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.
[0139] Example 2
[0140] See also Figure 3 As shown, this embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;
[0141] The processor executes an intelligent unmanned retail automatic interaction system of Example 1 by calling the computer program stored in the memory.
[0142] Example 3
[0143] See also Figure 4 As shown, this embodiment provides a computer-readable storage medium storing instructions. When the instructions are executed on a computer, the computer executes an intelligent unmanned retail automatic interaction system of embodiment 1.
[0144] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center via a wired network. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0145] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0146] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0147] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only one. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0148] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0149] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0150] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0151] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0152] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. An intelligent unmanned retail automatic interactive system, It is characterized in that include: A retail evaluation module is used to obtain the retail feature data of the unmanned retail vehicle in each retail route, and input the retail feature data into a pre-built transaction simulation model for simulation to predict the retail inventory of each unmanned retail vehicle; The retail evaluation coefficient of each unmanned retail vehicle is calculated based on the retail characteristic data and retail inventory; An inventory analysis module, which determines low-inventory unmanned retail vehicles within a unit time based on the retail evaluation coefficient, and selects high-inventory unmanned retail vehicles within a unit time that have the same functions as the low-inventory unmanned retail vehicles; Route planning module, which determines the best planned route based on high-inventory unmanned retail vehicles and low-inventory unmanned retail vehicles; Methods for determining the best planning routes based on high-inventory unmanned retail vehicles and low-inventory unmanned retail vehicles include: Obtain a layout map of the target supervision area, and based on the layout map of the target supervision area, use the location of the high-inventory unmanned retail vehicle as a moving point, and use the location of the low-inventory unmanned retail vehicle as a destination point; Based on the layout diagram of the target supervision area, the low-inventory unmanned retail vehicle locations are connected with the high-inventory unmanned retail vehicle locations to obtain X retail routes, where X is an integer greater than zero; Each planned route is simulated using a pre-built digital twin model to obtain route training data for each planned route, and a route evaluation coefficient for each planned route is calculated based on the route training data; the route training data includes the length of each planned route, the moving time of a high-inventory unmanned retail vehicle on each planned route, the parking time of an unmanned retail vehicle on each planned route, and the number of times an unmanned retail vehicle passes through each planned route; the route training data calculation formula is: In the formula, Xp is the line evaluation coefficient, Gh is the length of the planned route, and Wr G is the moving time of the unmanned retail vehicle with high inventory, Jz G is the parking time of the high-inventory unmanned retail vehicle, Lg is the number of times the unmanned retail vehicle passes by, θ 1 ,θ 2 and θ 3 is the correction factor, and θ 1 ,θ 2 and θ 3 are greater than zero; Sort the route evaluation coefficient of each planned route from small to large; The retail route corresponding to the minimum route evaluation coefficient is taken as the best planned route; Interaction module, when the high-inventory unmanned retail vehicle arrives at the low-inventory unmanned retail location along the optimal planned route, the high-inventory unmanned retail vehicle collects commodity retail data, inputs the commodity retail data information into the pre-built order matching model to obtain the order matching coefficient, compares the order matching coefficient with the order matching coefficient threshold, and determines whether to generate a delivery instruction based on the comparison result.
2. According to claim 1, an intelligent unmanned retail automatic interactive system, It is characterized in that The retail characteristic data includes the actual sales volume of each unmanned retail vehicle, the total operating time of each unmanned retail vehicle, the total inventory of goods of each unmanned retail vehicle, weather data and traffic volume; the retail inventory is the remaining inventory of goods of each unmanned retail vehicle; The weather data is humidity or temperature; the weather data is obtained through a weather forecast platform; the actual sales volume of each unmanned retail vehicle is the number of commodities actually sold by the unmanned retail vehicle in unit time; the total operating time of each unmanned retail vehicle is all time between the start and end of the operation of the unmanned retail vehicle; the total inventory of commodities of each unmanned retail vehicle is the total number of all commodities carried by the unmanned retail vehicle; the pedestrian flow is the number of pedestrian flows per unit area of each unmanned retail vehicle; the actual sales volume of each unmanned retail vehicle, the total operating time of each unmanned retail vehicle, the total inventory of commodities of each unmanned retail vehicle, the weather data and the pedestrian flow are all pre-stored in the cloud server.
3. According to claim 2, an intelligent unmanned retail automatic interactive system, It is characterized in that The method for constructing the transaction simulation model includes: Obtaining historical supervision data pre-stored in the cloud server, the historical supervision data including retail feature data of each unmanned retail vehicle and the corresponding actual retail inventory, dividing the historical supervision data into a data training set and a data test set, and building a training network; The retail feature data of each unmanned retail vehicle in the historical supervision data is used as the input of the training network, and the actual retail inventory of each unmanned retail vehicle is used as the output of the training network. The training network is trained to obtain an initial machine learning model, and the initial machine learning model is verified using a data test set to output an initial machine learning model that meets a preset accuracy rate as a transaction simulation model; the transaction simulation model is one of a linear regression model, a logistic regression model or a neural network model.
4. The intelligent unmanned retail automatic interactive system according to claim 3, It is characterized in that Methods for obtaining the retail evaluation coefficient of unmanned retail vehicles include: A formulaic calculation is performed based on the retail characteristic data and the retail inventory to obtain the retail evaluation coefficient of each retail route; the calculation formula is: Where: Lsp is the retail assessment coefficient, Sx i is the actual sales volume of the i-th unmanned retail vehicle, Sc i is the total operating time of the i-th unmanned retail vehicle; Zk i is the total inventory of goods of the i-th unmanned retail vehicle, Yk i To predict the retail inventory of the i-th unmanned retail vehicle, Tq is the weather data, γ is the weight of the weather data; α 1 is the weight of the actual sales volume of the i-th unmanned retail vehicle; α 2 is the weight of the total operating time of the i-th unmanned retail vehicle; RL is the flow of people in the target supervision area during the T period; α 3 is the weight of the flow of people in the target supervision area during the T period; Preset retail evaluation coefficient thresholds Y1 and Y2, Y1>Y2, and compare the retail evaluation coefficient with the retail evaluation coefficient threshold; If the retail evaluation coefficient is greater than or equal to the retail evaluation coefficient threshold Y1, it is determined that the corresponding unmanned retail vehicle has a large transaction volume, and the first retail vehicle is marked; If the retail evaluation coefficient is less than the retail evaluation coefficient threshold value Y1 and greater than the retail evaluation coefficient threshold value Y2, then the corresponding unmanned retail vehicle transaction volume is determined to be normal; If the retail evaluation coefficient is less than or equal to the retail evaluation coefficient threshold Y2, the corresponding unmanned retail vehicle is judged to have a low transaction volume and is marked as a non-target retail vehicle.
5. The intelligent unmanned retail automatic interactive system according to claim 4, It is characterized in that Methods for determining low inventory unmanned retail vehicles per unit time include: Get the inventory valuation coefficient of the first retail vehicle; Calculate the difference between the inventory assessment coefficient of the first retail vehicle and the corresponding retail assessment coefficient, and mark it as the first assessment difference; If the first evaluation difference is less than or equal to the preset first evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a low-inventory unmanned retail vehicle; The calculation formula of the inventory valuation coefficient is: Among them, Kcp is the inventory assessment coefficient; Bz is the replenishment cycle of the i-th unmanned retail vehicle, β is the weight factor of the replenishment cycle of the i-th unmanned retail vehicle, and Lk i is the historical inventory of the ith unmanned retail vehicle; the historical inventory is the remaining inventory of goods of each unmanned retail vehicle in the past unit time, and the replenishment cycle and the historical inventory are pre-stored in the cloud server.
6. The intelligent unmanned retail automatic interactive system according to claim 5, It is characterized in that The method for selecting a high-inventory unmanned retail vehicle having the same function as the low-inventory unmanned retail vehicle within a unit time includes: Determine whether there is a low-inventory unmanned retail vehicle among the non-target retail vehicles. If so, use the corresponding non-target retail vehicle as the first retail vehicle. If not, use the corresponding non-target retail vehicle as the second retail vehicle. Extract and obtain the inventory valuation coefficient of the second retail vehicle; Calculate the difference between the inventory valuation coefficient and the corresponding retail valuation coefficient, and mark the retail valuation coefficient as the second valuation difference; If the second evaluation difference is less than or equal to the preset second evaluation difference threshold, the corresponding unmanned retail vehicle is marked as a high-inventory unmanned retail vehicle, and Q high-inventory unmanned retail vehicles are obtained, where Q is an integer set greater than zero; Sort the second evaluation differences that are greater than or equal to a preset second evaluation difference threshold value from large to small according to their values; The unmanned retail vehicle corresponding to the second evaluation difference in the first ranking is regarded as a high-inventory unmanned retail vehicle.
7. The intelligent unmanned retail automatic interactive system according to claim 6, It is characterized in that Methods for obtaining order matching coefficients include: It should be noted that the commodity retail data includes the quantity of commodities, the commodity amount and the payment voucher matching coefficient; the commodity retail data is obtained by the client device sending an order to the cloud server; The method for constructing the order matching model includes: The order matching model adopts a linear regression model, with the product quantity Ss, product amount Je and payment voucher matching coefficient Px as input and the order matching coefficient Dp as output; Build a linear regression model: Dp=β 1 ×Ss+β 2 ×Je+β 3 ×Px; In the formula, β 1 , β 2 With β 3 They are all preset proportional coefficients, and their values are all greater than zero; the payment credential matching coefficient refers to the security verification data sent by the user to the cloud server collected by the high-inventory unmanned retail vehicle, and obtained by matching analysis of the security verification data.
8. The intelligent unmanned retail automatic interactive system according to claim 7, It is characterized in that Methods for determining whether to generate a shipping instruction based on the comparison results include: Preset an order matching coefficient threshold, and compare the order matching coefficient with the order matching coefficient threshold; When the order matching coefficient is less than or equal to the order matching coefficient threshold, a shipping instruction is generated; When the order matching coefficient is greater than the order matching coefficient threshold, no shipping instruction is generated.
9. An electronic device, It is characterized in that include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes an intelligent unmanned retail automatic interaction system according to any one of claims 1 to 8 by calling a computer program stored in the memory.
