A remotely controlled intelligent vending machine vending system and method

Abstract

The present invention provides a vending system and method for a remotely controlled smart vending machine, wherein the system includes: an acquisition module, for acquiring the purchase demand information input by the target user into the smart vending machine; a verification module, for verifying the prepaid order corresponding to the purchase demand; a control module, for controlling the preset picking device of the smart vending machine to pick up the target commodity by the shortest path based on the corresponding purchase demand information if the verification is passed; a reminder module, for reminding the target user to pick up the target commodity after the placement is completed; a placement and arrangement module, for controlling the picking device to reasonably place and arrange the target commodity at the pickup position when placing the target commodity at the pickup position. The vending system and method for a remotely controlled smart vending machine of the present invention controls the picking device to reasonably place and arrange the target commodity at the pickup position, which is more suitable. At the same time, by picking up the target commodity through the picking device, the restrictions on the types of the target commodity are reduced, and the user experience is better.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a vending system and method for a remotely controlled intelligent vending machine. Background Art

[0002] At present, mainstream vending machine aisles (for example, spring spiral aisles, crawler aisles, and serpentine aisles) generally deliver goods by the goods' own gravity. After the goods fall freely, the goods are generally randomly arranged and stacked, which can easily cause damage to the goods when the packaging of the goods is fragile, and their suitability is poor. At the same time, traditional vending machine aisles have many restrictions on goods (for example, spring spiral aisles are not suitable for selling fragile goods, crawler aisles are not suitable for selling goods that are easy to fall over and have no fixed packaging, serpentine aisles are only suitable for selling beverages, etc.), there are fewer types of goods that can be sold, and the user experience is poor.

[0003] Therefore, a solution is urgently needed. Summary of the invention

[0004] One of the purposes of the present invention is to provide a vending system for a remotely controlled intelligent vending machine, which controls a picking device to reasonably place and arrange target goods at a picking position, rather than shipping the goods by their own gravity, thereby avoiding damage to the goods and being more suitable. At the same time, by picking up the target goods through the picking device, restrictions on the types of target goods are reduced, the types of goods sold are richer, and the user experience is improved.

[0005] An embodiment of the present invention provides a remotely controlled intelligent vending machine vending system, comprising:

[0006] An acquisition module, configured to acquire purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category;

[0007] A verification module, used to verify the prepaid order corresponding to the purchase demand;

[0008] A control module, for controlling a preset picking device of the smart vending machine to pick up the target commodity by the shortest path and placing the target commodity at a preset picking position based on the corresponding purchase demand information if the verification is passed;

[0009] The reminder module is used to remind the target user to pick up the target product after it is placed;

[0010] The placement and arrangement module is used to control the picking device to reasonably place and arrange the target goods at the pickup position when placing the target goods at the pickup position.

[0011] Preferably, the acquisition module acquires purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category, including:

[0012] Acquire the purchase demand information input by the target user corresponding to the purchase platform of the smart vending machine;

[0013] and / or,

[0014] Obtaining input instructions input by the target user into the vending machine;

[0015] Parse the input instructions and obtain the corresponding purchase demand information.

[0016] Preferably, the verification module verifies the prepaid order corresponding to the purchase demand, including:

[0017] Acquire inventory information of the vending machine, the inventory information including: inventory commodity types and inventory quantities corresponding to the inventory commodity types;

[0018] Traverse the first purchase commodity category in sequence, and use the first purchase commodity category currently being traversed as the second purchase commodity category;

[0019] During each traversal, it is determined whether there is an inventory commodity type that is consistent with the second purchased commodity type and whether the inventory quantity of the corresponding inventory commodity type is greater than or equal to the purchase commodity quantity corresponding to the second purchased commodity type;

[0020] If yes, obtain the payment status of the corresponding target user;

[0021] If the corresponding target user completes the payment, the prepaid order is verified.

[0022] Preferably, if the verification is passed, the control module controls the preset picking device of the smart vending machine to pick up the target commodity based on the corresponding purchase demand information, and places the target commodity at a preset picking position, including:

[0023] Analyze the corresponding purchase demand information and determine the target products and target quantities;

[0024] Obtaining the operation track diagram of the picking device;

[0025] Determine a first point position of the target product in the running track map, a second point position of the picking device in the running track map, a third point position of the picking location in the running track map, and a running track of the picking device;

[0026] Based on the path planning technology, according to the first point of the target product, the second point of the picking device, the third point of the pickup location and the running trajectory of the picking device, the shortest path for the picking device to pick up all the target products and deliver them to the pickup location is planned;

[0027] After the planning is completed, the picking device is controlled to pick up the target product based on the shortest path and send it to the pickup location.

[0028] Preferably, when the placement and arrangement module places the target commodity at the pickup position, the placement and arrangement module controls the picking device to reasonably place and arrange the target commodity at the pickup position, including:

[0029] Acquire first shape information of a storage table surface corresponding to a pickup location;

[0030] Obtain the second shape information of the largest surface on which the target product can be placed;

[0031] Based on the first shape information and the second shape information, determining whether all target products can be laid flat on the storage table;

[0032] If not, obtain a stacking arrangement plan;

[0033] Based on the stacking arrangement plan, the target products are placed and arranged.

[0034] Preferably, obtaining a stacking arrangement plan includes:

[0035] Acquire first space information of a placement space corresponding to a pickup location, and at the same time, acquire first commodity information of a target commodity, wherein the first commodity information includes: third shape information and packaging material information;

[0036] Get multiple goods stacking records;

[0037] According to the goods stacking records, train the goods stacking plan determination model;

[0038] Based on the product stacking plan determination model, a stacking arrangement plan is determined according to the first space information and the first product information.

[0039] Preferably, the target products are placed and arranged based on the stacking arrangement scheme, including:

[0040] The target product being placed is taken as the target placement product, and the fourth point position of the target placement product in the vending machine is determined;

[0041] Analyze the stacking arrangement plan and determine that the target placement product in the stacking arrangement plan corresponds to the fifth point in the vending machine;

[0042] Based on a preset position relationship feature extraction template, feature extraction is performed on the relative position relationship between the fourth point and the fifth point in real time to obtain a relative position relationship feature value;

[0043] Based on the relative position relationship feature value, construct a first position relationship description vector;

[0044] Matching the first position relationship description vector with a second position relationship description vector in a preset control strategy library, determining a control strategy corresponding to the second position relationship description vector that matches the first position relationship description vector, and using the control strategy as the target control strategy;

[0045] Based on the target control strategy, the picking device is controlled to place the corresponding target products.

[0046] Preferably, the vending system of the remotely controlled intelligent vending machine further includes:

[0047] The alarm module is used to obtain a first monitoring image taken by a camera pre-installed on the intelligent vending machine, and determine whether to issue an alarm based on the first monitoring image.

[0048] Preferably, the alarm module determines whether to issue an alarm based on the first monitoring image, including:

[0049] Based on face recognition technology, determining a second monitoring image containing a face in the first monitoring image;

[0050] Obtaining a shooting time of a second surveillance image;

[0051] Based on the shooting time, the second monitoring image is expanded on a preset time axis;

[0052] Traversing the second monitoring image on the time axis from beginning to end in sequence, and taking the second monitoring image currently being traversed as the third monitoring image;

[0053] Based on the behavior recognition technology, identifying the behavior of the person in the third monitoring image;

[0054] Obtain a regularly updated alarm behavior library;

[0055] Matching personnel behavior with the alarm behavior in the alarm behavior library;

[0056] If there is a match, obtain the alarm information of the corresponding alarm behavior and give a reminder based on the alarm information;

[0057] The obtaining of the regularly updated alarm behavior library includes:

[0058] Get the node types of multiple preset acquisition nodes;

[0059] Based on the node type, determining an availability verification information acquisition template corresponding to the acquisition node;

[0060] According to the availability verification information acquisition template, the availability verification information of the corresponding acquisition node is acquired;

[0061] Obtain type weighted values ​​of availability verification information corresponding to different information types;

[0062] Obtaining a preset availability verification template corresponding to the availability verification information;

[0063] Based on the availability verification template and the corresponding availability verification information, a verification value is determined;

[0064] Based on the verification value and the corresponding type weight value, a sum value is determined and associated with the corresponding acquisition node;

[0065] Accumulate and calculate the sum value associated with each acquisition node to obtain the verification target value;

[0066] If the verification target value is greater than or equal to the preset verification target value threshold, the supplementary alarm behavior provided by the corresponding acquisition node is obtained;

[0067] The supplementary alarm behavior is stored in the alarm behavior library at preset time intervals to complete the update.

[0068] An embodiment of the present invention provides a vending method for a remotely controlled intelligent vending machine, comprising:

[0069] Step 1: obtaining purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category;

[0070] Step 2: Verify the prepaid order corresponding to the purchase demand;

[0071] Step 3: If the verification is successful, based on the corresponding purchase demand information, the preset picking device of the smart vending machine is controlled to pick up the target product and place the target product at the preset picking position;

[0072] Step 4: When the target product is placed, remind the target user to pick it up;

[0073] Step 5: When placing the target product at the pickup location, control the picking device to reasonably place and arrange the target product at the pickup location.

[0074] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0075] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0077] Figure 1 Schematic diagram of a vending system of a remotely controlled intelligent vending machine in an embodiment of the present invention;

[0078] Figure 2 Schematic diagram of a vending method of a remotely controlled intelligent vending machine in an embodiment of the present invention. DETAILED DESCRIPTION

[0079] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0080] The embodiment of the present invention provides a remotely controlled intelligent vending machine vending system, such as Figure 1 As shown, including:

[0081] The acquisition module 1 is used to acquire the purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category;

[0082] Verification module 2, used to verify the prepaid order corresponding to the purchase demand;

[0083] The control module 3 is used to control the preset picking device of the smart vending machine to pick up the target commodity by the shortest path and place the target commodity at the preset picking position based on the corresponding purchase demand information if the verification is passed;

[0084] Reminder module 4, used to remind the target user to pick up the target product after it is placed;

[0085] The placement and arrangement module 5 is used to control the picking device to reasonably place and arrange the target goods at the pickup position when the target goods are placed at the pickup position.

[0086] The working principle and beneficial effects of the above technical solution are:

[0087] The target users are: users who use smart vending machines to shop. The purchase demand information includes: the first purchase commodity category and the quantity of the purchase commodity corresponding to the first purchase commodity category. The first purchase commodity category is, for example: xx beverage, and the quantity of the purchase commodity is, for example: 2 bottles. The prepaid order corresponding to the purchase demand is verified from two aspects: first, verify whether the inventory of the vending machine meets the purchase demand; second, verify whether the target user has completed the payment of the corresponding purchase commodity. If the verification is passed, according to the purchase information, the pre-set picking device (for example: a manipulator) of the smart vending machine is controlled to pick up the corresponding number of target commodities in the shortest path (based on the path planning technology, which belongs to the prior art and will not be described in detail), and the target commodities are placed in the preset pickup position (for example: a storage table manually set in advance in the vending machine). After all the target commodities are placed, the target user is reminded to pick them up (for example: through the smart display screen of the smart vending machine, it is displayed: "All the commodities you purchased have been placed on the storage table under the vending machine, please pick them up in time");

[0088] Generally, when users use traditional vending machines to buy goods, they can only buy them one by one. At the same time, there are situations where the vending machine allows the goods to freely fall to the storage area of ​​the vending machine. If the mass of the target product is too large and the outer packaging has a small bearing pressure, it is easy to cause the outer packaging of the target product to rupture, causing losses to the user. Therefore, the control device is used to reasonably place and arrange the target products at the pickup position (for example: when all the products cannot be laid out flat, the products that are not easily compressed are stacked on the upper layer of the pile).

[0089] The present application controls the picking device to reasonably arrange the target goods at the picking position, rather than shipping the goods by their own gravity, thereby avoiding damage to the goods and being more suitable. At the same time, by picking up the target goods through the picking device, the restrictions on the types of target goods are reduced, the types of goods sold are richer, and the user experience is improved.

[0090] In one embodiment, the acquisition module acquires purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category, including:

[0091] Acquire the purchase demand information input by the target user corresponding to the purchase platform of the smart vending machine;

[0092] and / or,

[0093] Obtaining input instructions input by the target user into the vending machine;

[0094] Parse the input instructions and obtain the corresponding purchase demand information.

[0095] The working principle and beneficial effects of the above technical solution are:

[0096] There are two methods to obtain the target user's purchase demand information. The first method is to obtain the purchase demand information input by the target user on the purchase platform corresponding to the vending machine (for example: xxx vending machine online mall). When inputting, the target user's device end and the network node corresponding to the vending machine communicate and connect. The second method is to configure the vending machine with an operation input terminal, and parse the input instructions entered by the target user on the operation input terminal to obtain the corresponding purchase demand information.

[0097] This application introduces two methods to obtain the purchase demand information of target users, thereby improving the comprehensiveness of the purchase demand information acquisition.

[0098] In one embodiment, the verification module verifies the prepaid order corresponding to the purchase demand;

[0099] Acquire inventory information of the vending machine, the inventory information including: inventory commodity types and inventory quantities corresponding to the inventory commodity types;

[0100] Traverse the first purchase commodity category in sequence, and use the first purchase commodity category currently being traversed as the second purchase commodity category;

[0101] During each traversal, it is determined whether there is an inventory commodity type that is consistent with the second purchased commodity type and whether the inventory quantity of the corresponding inventory commodity type is greater than or equal to the purchase commodity quantity corresponding to the second purchased commodity type;

[0102] If yes, obtain the payment status of the corresponding target user;

[0103] If the corresponding target user completes the payment, the prepaid order is verified.

[0104] The working principle and beneficial effects of the above technical solution are:

[0105] The inventory information of the vending machine includes: the types of inventory goods and the corresponding inventory quantities of the inventory goods types. The types of inventory goods are: the types of all goods stored in the vending machine, and the inventory quantities are the number of goods corresponding to the inventory goods types. Traverse the first purchase goods types in sequence, and take the first purchase goods type currently being traversed as the second purchase goods type. If the purchase quantity of the second purchase goods type is greater than or equal to the inventory quantity of the corresponding inventory goods type, it means that the stock is sufficient and supply can be realized. If all the first purchase goods types that are traversed meet the above conditions, it means that all the goods that the target user wants to purchase are in sufficient stock, then the next step is to verify the user's payment status. Get the payment status of the corresponding target user, and the payment status includes: payment completed and unpaid. If the target user completes the payment, the prepaid order verification is passed.

[0106] This application introduces inventory information and payment status to verify prepaid orders, which improves the rationality and comprehensiveness of the verification.

[0107] In one embodiment, if the verification is passed, the control module controls the preset picking device of the smart vending machine to pick up the target product based on the corresponding purchase demand information, and places the target product at a preset picking position, including:

[0108] Analyze the corresponding purchase demand information and determine the target products and target quantities;

[0109] Obtaining the operation track diagram of the picking device;

[0110] Determine a first point position of the target product in the running track map, a second point position of the picking device in the running track map, a third point position of the picking location in the running track map, and a running track of the picking device;

[0111] Based on the path planning technology, according to the first point of the target product, the second point of the picking device, the third point of the pickup location and the running trajectory of the picking device, the shortest path for the picking device to pick up all the target products and deliver them to the pickup location is planned;

[0112] After the planning is completed, the picking device is controlled to pick up the target product based on the shortest path and send it to the pickup location.

[0113] The working principle and beneficial effects of the above technical solution are:

[0114] Analyze the verified purchase needs and determine the target products and target quantities (for example: 2 bottles of mineral water, 1 bag of potato chips). The operation track diagram of the picking device is: the annotation diagram obtained by marking the movable track of the picking device in the internal distribution map of the vending machine. Determine the first point of the target product in the operation track diagram, the second point of the picking device, the third point of the user's pickup position, and the operation track of the picking device. Based on the path planning technology, plan the shortest path for the picking device to pick up all target products and deliver all target products to the pickup location (path planning technology belongs to the prior art and will not be described in detail). After the planning is completed, the picking device is controlled to pick up the target product based on the shortest path and send it to the pickup location. When the picking device picks up the product, the target image of the target product to be picked up is obtained (which can be obtained by a camera device pre-set on the picking device), and the target image is mapped in a preset plane rectangular coordinate system. The target area corresponding to the target image is determined, and the centroid coordinates of the target area σ are calculated. in:

[0115]

[0116]

[0117] in, The abscissa of the centroid, is the centroid ordinate, x and y are the integrands, ∫∫[…]dσ is the double integral of […] in the target area σ, and the centroid coordinates are obtained corresponding to the grasping position of the target product to be taken. Based on the grasping position, the grasping device is controlled to take the corresponding target product.

[0118] This application introduces a picking device operation track map, and based on the path planning technology, plans the shortest path for picking up all target commodities in the picking device operation track map, thereby improving the efficiency of the picking device in picking up target commodities and reducing the energy consumption of the system.

[0119] In one embodiment, when the placement and arrangement module places the target commodity at the pickup position, the placement and arrangement module controls the picking device to reasonably place and arrange the target commodity at the pickup position, including:

[0120] Acquire first shape information of a storage table surface corresponding to a pickup location;

[0121] Obtain the second shape information of the largest surface on which the target product can be placed;

[0122] Based on the first shape information and the second shape information, determining whether all target products can be laid flat on the storage table;

[0123] If not, obtain a stacking arrangement plan;

[0124] Based on the stacking arrangement plan, the target products are placed and arranged.

[0125] The working principle and beneficial effects of the above technical solution are:

[0126] The first shape information is: the shape and size of the storage table surface at the pickup location, for example: a rectangular area with a length of 80 cm and a width of 60 cm. The second shape information is: the shape and size of the largest surface on which the target product can be placed, for example: a can of Coke, the largest surface that can be placed is a circular area with a diameter of 8 cm at the bottom of the can. Based on the area coverage technology, according to the first shape information and the second shape information, determine whether all target products can be laid out on the storage table. The area coverage technology belongs to the prior art and will not be elaborated on. If not, obtain the stacking arrangement plan. The stacking arrangement plan is: how to stack and place the target products. Based on the stacking arrangement plan, arrange the target products.

[0127] The present application introduces regional planning technology, which determines whether the target goods need to be stacked and arranged according to the first shape information and the second shape information, and is more applicable.

[0128] In one embodiment, obtaining a stacking arrangement scheme includes:

[0129] Acquire first space information of a placement space corresponding to a pickup location, and at the same time, acquire first commodity information of a target commodity, wherein the first commodity information includes: third shape information and packaging material information;

[0130] Get multiple goods stacking records;

[0131] According to the goods stacking records, train the goods stacking plan determination model;

[0132] Based on the product stacking plan determination model, a stacking arrangement plan is determined according to the first space information and the first product information.

[0133] The working principle and beneficial effects of the above technical solution are:

[0134] The first spatial information is: the spatial distribution information at the pickup location, and the first product information includes: the third shape information and packaging material information of the target product, for example: what shape the target product is, how big its volume is, what material and process the packaging material is, etc. The goods stacking record is the goods stacking record obtained through the big data platform, for example: what kind of goods information is stacked where. According to the goods stacking record, the goods stacking plan determination model is trained. The model training technology belongs to the prior art and will not be elaborated on. Based on the goods stacking plan determination model, the stacking arrangement plan (for example: where and how to stack the target goods) is determined according to the first spatial information and the first product information.

[0135] The present application introduces a product stacking plan determination model, which determines a suitable stacking arrangement plan based on the first space information and the first product information, which is more reasonable.

[0136] In one embodiment, the method of placing and arranging the target products based on the stacking arrangement scheme includes:

[0137] The target product being placed is taken as the target placement product, and the fourth point position of the target placement product in the vending machine is determined;

[0138] Analyze the stacking arrangement plan and determine that the target placement product in the stacking arrangement plan corresponds to the fifth point in the vending machine;

[0139] Based on a preset position relationship feature extraction template, feature extraction is performed on the relative position relationship between the fourth point and the fifth point in real time to obtain a relative position relationship feature value;

[0140] Based on the relative position relationship feature value, construct a first position relationship description vector;

[0141] Matching the first position relationship description vector with a second position relationship description vector in a preset control strategy library, determining a control strategy corresponding to the second position relationship description vector that matches the first position relationship description vector, and using the control strategy as the target control strategy;

[0142] Based on the target control strategy, the picking device is controlled to place the corresponding target products.

[0143] The working principle and beneficial effects of the above technical solution are:

[0144] When goods are stacked, they are generally stacked in sequence. Therefore, the fourth point of the target placed goods currently being placed (the position of the target placed goods in the vending machine) is determined. The fifth point is: the position where the target placed goods should be placed. The preset position relationship feature extraction template is: a template pre-set for extracting position relationship feature values. Based on the position relationship feature extraction template, the relative position relationship between the fourth point and the fifth point is feature extracted to obtain the relative position relationship feature value (for example: in which direction is the fifth point in the fourth point, how much is the distance between the two, etc.). Based on the relative position relationship feature value, a first position relationship description vector is constructed (constructing vectors based on feature values ​​belongs to the prior art and will not be repeated). The preset control strategy library includes: multiple one-to-one corresponding second position relationship description vectors and control strategies, matching the first position relationship description vector and the second position relationship description vector, determining the control strategy corresponding to the second position relationship description vector that matches the target control strategy (for example: controlling where to move the picking device, how much to move, etc.), and controlling the picking device to place the corresponding target placed goods based on the target control strategy.

[0145] The present application introduces a position relationship feature extraction template to extract the relative position relationship feature value between the fourth point and the fifth point, thereby improving the accuracy of obtaining the relative position relationship feature value; introduces a control strategy library to determine the target control strategy based on the first position relationship description vector and the second position relationship description vector, thereby improving the suitability of the control of the picking device.

[0146] In one embodiment, the vending system of the remotely controlled smart vending machine further includes:

[0147] The alarm module is used to obtain a first monitoring image taken by a camera pre-installed on the intelligent vending machine, and determine whether to issue an alarm based on the first monitoring image.

[0148] The working principle and beneficial effects of the above technical solution are:

[0149] Generally, vending machines are generally equipped with alarm devices to prevent theft of goods. Therefore, a first surveillance image is obtained; the first surveillance image is an image taken by a shooting device (e.g., a panoramic camera) installed on the smart vending machine. Based on the first surveillance image, it is determined whether to issue an alarm (e.g., when a person nearby steals, an alarm is required). This application introduces the first surveillance image for alarm determination, which improves the rationality of the alarm.

[0150] In one embodiment, the alarm module determines whether to issue an alarm based on the first monitoring image, including:

[0151] Based on face recognition technology, determining a second monitoring image containing a face in the first monitoring image;

[0152] Obtaining a shooting time of a second surveillance image;

[0153] Based on the shooting time, the second monitoring image is expanded on a preset time axis;

[0154] Traversing the second monitoring image on the time axis from beginning to end in sequence, and taking the second monitoring image currently being traversed as the third monitoring image;

[0155] Based on the behavior recognition technology, identifying the behavior of the person in the third monitoring image;

[0156] Obtain a regularly updated alarm behavior library;

[0157] Matching personnel behavior with the alarm behavior in the alarm behavior library;

[0158] If there is a match, obtain the alarm information of the corresponding alarm behavior and give a reminder based on the alarm information;

[0159] The obtaining of the regularly updated alarm behavior library includes:

[0160] Get the node types of multiple preset acquisition nodes;

[0161] Based on the node type, determining an availability verification information acquisition template corresponding to the acquisition node;

[0162] According to the availability verification information acquisition template, the availability verification information of the corresponding acquisition node is acquired;

[0163] Obtain type weighted values ​​of availability verification information corresponding to different information types;

[0164] Obtaining a preset availability verification template corresponding to the availability verification information;

[0165] Based on the availability verification template and the corresponding availability verification information, a verification value is determined;

[0166] Based on the verification value and the corresponding type weighted value, a sum value is determined and associated with the corresponding acquisition node;

[0167] Accumulate and calculate the sum value associated with each acquisition node to obtain the verification target value;

[0168] If the verification target value is greater than or equal to the preset verification target value threshold, the supplementary alarm behavior provided by the corresponding acquisition node is obtained;

[0169] The supplementary alarm behavior is stored in the alarm behavior library at preset time intervals to complete the update.

[0170] The working principle and beneficial effects of the above technical solution are:

[0171] Generally, when a thief steals goods from a vending machine, he needs to get close to the vending machine. It is more efficient to make an alarm judgment when there is a person in the first monitoring image. Therefore, based on face recognition technology, a second monitoring image containing a face in the first monitoring image is determined. Face recognition technology belongs to the prior art and is not described in detail. The shooting time of the second monitoring image is obtained (the shooting record stored locally by the shooting device can be retrieved in real time and obtained by parsing the shooting record).

[0172] Expand the second surveillance image on the preset timeline according to the shooting time. Traverse the second surveillance image on the timeline from beginning to end, and use the second surveillance image currently being traversed as the third surveillance image. Based on behavior recognition technology, identify the behavior of people in the third surveillance image (for example: picking locks, kicking cabinet doors, etc.). Behavior recognition technology belongs to the prior art and will not be described in detail. The alarm behavior library is a database, including multiple alarm behaviors. The specific process of obtaining a regularly updated alarm behavior library is as follows:

[0173] Obtain node types of multiple preset acquisition nodes; the acquisition node is: an information collection node for alarm behavior, and the node type is, for example: an associated acquisition node and an unassociated acquisition node. The associated acquisition node is: a network node corresponding to other vending machines on the same platform, and the unassociated acquisition node is: a network node corresponding to other vending machines obtained through big data.

[0174] Based on the node type, determine the availability verification information acquisition template corresponding to the acquisition node; the availability verification information acquisition template is: constraining only the acquisition of availability verification information; based on the availability verification information acquisition template, obtain the availability verification information of the corresponding acquisition node, such as: historical alarm behavior supplementary records, historical evaluation information, etc. Obtain the type weighted values ​​corresponding to different information types of the availability verification information. The higher the verification strength of the corresponding information type, the greater the type weighted value. According to the availability verification information acquisition template, obtain the availability verification information of the corresponding acquisition node; the preset availability verification template is: a template that constrains the verification behavior to only perform availability verification. Based on the availability verification template, determine the verification value according to the corresponding availability verification information. Based on the verification value and the corresponding type weighted value, determine the sum value to sum up and obtain the verification target value. The larger the verification target value, the more available the alarm behavior of the corresponding acquisition node. If the verification target value is greater than or equal to the preset verification target value threshold (the verification target value threshold is manually preset), it means that the supplementary alarm behavior provided by the corresponding acquisition node is available, and it is stored in the alarm behavior library at a preset time interval (for example: 2 days).

[0175] Match the personnel behavior with the alarm behavior in the alarm behavior library. If the match is confirmed, it means that the alarm behavior is identified and a reminder is needed (for example: inform the corresponding personnel that their behavior is being monitored and deter them), obtain the corresponding reminder information (for example: do not kick the cabinet door, your behavior is being recorded in real time, please stop immediately), and remind the corresponding personnel based on the alarm information, for example: play a reminder voice through the microphone of the vending machine.

[0176] This application introduces face recognition technology to determine the second monitoring image for alarm behavior detection, thereby improving the recognition efficiency of alarm behaviors; introduces behavior recognition technology and an alarm behavior library, thereby improving the rationality of alarm behavior recognition; and performs availability verification on the acquisition nodes for supplementary alarm behaviors, thereby ensuring the comprehensiveness of the alarm behavior library update and improving the rationality of the update.

[0177] The embodiment of the present invention provides a vending method of a remotely controlled intelligent vending machine, such as Figure 2 As shown, including:

[0178] Step 1: obtaining purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category;

[0179] Step 2: Verify the prepaid order corresponding to the purchase demand;

[0180] Step 3: If the verification is successful, based on the corresponding purchase demand information, the preset picking device of the smart vending machine is controlled to pick up the target product and place the target product at the preset picking position;

[0181] Step 4: When the target product is placed, remind the target user to pick it up;

[0182] Step 5: When placing the target product at the pickup location, control the picking device to reasonably place and arrange the target product at the pickup location.

[0183] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A remotely controlled intelligent vending machine vending system, characterized in that: include: An acquisition module, configured to acquire purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category; A verification module is used to verify the prepaid order corresponding to the purchase demand; A control module, for controlling a preset picking device of the smart vending machine to pick up the target commodity by the shortest path and placing the target commodity at a preset picking position based on the corresponding purchase demand information if the verification is passed; The reminder module is used to remind the target user to pick up the target product after it is placed; The placement and arrangement module is used to control the picking device to reasonably place and arrange the target goods at the pickup position when placing the target goods at the pickup position; When the placement and arrangement module places the target commodity at the pickup position, the placement and arrangement module controls the picking device to reasonably place and arrange the target commodity at the pickup position, including: Acquire first shape information of a storage table surface corresponding to a pickup location; Obtain the second shape information of the largest surface on which the target product can be placed; Based on the first shape information and the second shape information, determining whether all target products can be laid flat on the storage table; If not, obtain a stacking arrangement plan; Arrange the target products based on the stacking arrangement plan; The obtaining of the stacking arrangement scheme includes: Acquire first space information of a placement space corresponding to a pickup location, and at the same time, acquire first commodity information of a target commodity, wherein the first commodity information includes: third shape information and packaging material information; Get multiple goods stacking records; According to the goods stacking records, train the goods stacking plan determination model; Based on the product stacking plan determination model, determine the stacking arrangement plan according to the first space information and the first product information; The target products are placed and arranged based on the stacking arrangement scheme, including: The target product being placed is taken as the target placement product, and the fourth point position of the target placement product in the vending machine is determined; Analyze the stacking arrangement plan and determine that the target placement product in the stacking arrangement plan corresponds to the fifth point in the vending machine; Based on a preset position relationship feature extraction template, feature extraction is performed on the relative position relationship between the fourth point and the fifth point in real time to obtain a relative position relationship feature value; Based on the relative position relationship feature value, construct a first position relationship description vector; Matching the first position relationship description vector with the second position relationship description vector in the preset control strategy library, determining the control strategy corresponding to the second position relationship description vector that matches the first position relationship description vector, and using it as the target control strategy; Based on the target control strategy, the picking device is controlled to place the corresponding target products.

2. The remotely controlled smart vending machine vending system according to claim 1, characterized in that: The acquisition module acquires purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category, including: Acquire the purchase demand information input by the target user corresponding to the purchase platform of the smart vending machine; and / or, Obtaining input instructions input by the target user into the vending machine; Parse the input instructions and obtain the corresponding purchase demand information.

3. The remote controlled smart vending machine vending system according to claim 1, characterized in that: The verification module verifies the prepaid order corresponding to the purchase demand, including: Acquire inventory information of the vending machine, the inventory information including: inventory commodity types and inventory quantities corresponding to the inventory commodity types; Traverse the first purchase commodity category in sequence, and use the first purchase commodity category currently being traversed as the second purchase commodity category; During each traversal, it is determined whether there is an inventory commodity type that is consistent with the second purchased commodity type and whether the inventory quantity of the corresponding inventory commodity type is greater than or equal to the purchase commodity quantity corresponding to the second purchased commodity type; If yes, obtain the payment status of the corresponding target user; If the corresponding target user completes the payment, the prepaid order is verified.

4. The remotely controlled smart vending machine vending system according to claim 1, characterized in that: If the verification is successful, the control module controls the preset picking device of the smart vending machine to pick up the target product based on the corresponding purchase demand information, and places the target product at a preset picking position, including: Analyze the corresponding purchase demand information and determine the target products and target quantities; Obtaining the operation track diagram of the picking device; Determine a first point position of the target product in the running track map, a second point position of the picking device in the running track map, a third point position of the picking location in the running track map, and a running track of the picking device; Based on the path planning technology, according to the first point of the target product, the second point of the picking device, the third point of the pickup location and the running trajectory of the picking device, the shortest path for the picking device to pick up all the target products and deliver them to the pickup location is planned; After the planning is completed, the picking device is controlled to pick up the target product based on the shortest path and send it to the pickup location.

5. The remotely controlled smart vending machine vending system according to claim 1, characterized in that: Also includes: The alarm module is used to obtain a first monitoring image taken by a camera pre-installed on the intelligent vending machine, and determine whether to issue an alarm based on the first monitoring image.

6. The remotely controlled intelligent vending machine vending system according to claim 5, characterized in that: The alarm module determines whether to issue an alarm based on the first monitoring image, including: Based on face recognition technology, determining a second monitoring image containing a face in the first monitoring image; Obtaining a shooting time of a second surveillance image; Based on the shooting time, the second monitoring image is expanded on a preset time axis; Traversing the second monitoring image on the time axis from beginning to end in sequence, and taking the second monitoring image currently being traversed as the third monitoring image; Based on the behavior recognition technology, identifying the behavior of the person in the third monitoring image; Obtain a regularly updated alarm behavior library; Matching personnel behavior with the alarm behavior in the alarm behavior library; If there is a match, obtain the alarm information of the corresponding alarm behavior and give a reminder based on the alarm information.

7. A vending method for a remotely controlled intelligent vending machine, characterized in that: include: Step 1: obtaining purchase demand information input by at least one target user into the intelligent vending machine, wherein the purchase demand includes: a first purchase commodity category and a purchase commodity quantity corresponding to the first purchase commodity category; Step 2: Verify the prepaid order corresponding to the purchase demand; Step 3: If the verification is successful, based on the corresponding purchase demand information, the preset picking device of the smart vending machine is controlled to pick up the target product and place the target product at the preset picking position; Step 4: When the target product is placed, remind the target user to pick it up; Step 5: When placing the target product at the pickup location, control the picking device to reasonably place and arrange the target product at the pickup location; The step 5: when placing the target commodity at the pickup position, controlling the picking device to reasonably place and arrange the target commodity at the pickup position includes: Acquire first shape information of a storage table surface corresponding to a pickup location; Obtain the second shape information of the largest surface on which the target product can be placed; Based on the first shape information and the second shape information, determining whether all target products can be laid flat on the storage table; If not, obtain a stacking arrangement plan; Arrange the target products based on the stacking arrangement plan; The obtaining of the stacking arrangement scheme includes: Acquire first space information of a placement space corresponding to a pickup location, and at the same time, acquire first commodity information of a target commodity, wherein the first commodity information includes: third shape information and packaging material information; Get multiple goods stacking records; According to the goods stacking records, train the goods stacking plan determination model; Based on the product stacking plan determination model, determine the stacking arrangement plan according to the first space information and the first product information; The target products are placed and arranged based on the stacking arrangement scheme, including: The target product being placed is taken as the target placement product, and the fourth point position of the target placement product in the vending machine is determined; Analyze the stacking arrangement plan and determine that the target placement product in the stacking arrangement plan corresponds to the fifth point in the vending machine; Based on a preset position relationship feature extraction template, feature extraction is performed on the relative position relationship between the fourth point and the fifth point in real time to obtain a relative position relationship feature value; Based on the relative position relationship feature value, construct a first position relationship description vector; Matching the first position relationship description vector with the second position relationship description vector in the preset control strategy library, determining the control strategy corresponding to the second position relationship description vector that matches the first position relationship description vector, and using it as the target control strategy; Based on the target control strategy, the picking device is controlled to place the corresponding target products.

Images