Smart Go guide type smart shopping cart recognition system

Abstract

The invention relates to the technical field of shopping carts, and particularly discloses a SmartGo guide type smart shopping cart recognition system, which comprises a cart body, and a motion control module, an image recognition processing and display module, a positioning and navigation module and a cloud platform module which are arranged in the cart body, the intelligent shopping cart integrates various advanced technologies such as computer vision, artificial intelligence and machine learning, can accurately identify commodities and calculate prices, greatly improves the shopping efficiency, has various convenient functions such as automatic list checking, automatic navigation and automatic checkout, greatly shortens the shopping time of consumers, and improves the shopping experience of the consumers. And great convenience is provided for consumers. In addition, the intelligent shopping cart can also be connected with mobile phones, intelligent homes and the like of consumers, commodities are recommended in a personalized mode, professional shopping suggestions are provided for the consumers, and therefore better shopping experience is provided.

Description

Technical Field

[0001] The present invention belongs to the technical field of shopping carts, and in particular relates to a SmartGo guided intelligent shopping cart identification system. Background Art

[0002] , "contactless shopping" and "digital transformation of offline retailers" are increasingly becoming hot topics in the retail industry. The entire industry is seeking ways to achieve contactless shopping for consumers and digital transformation of offline retail. The emergence of smart shopping carts is undoubtedly one of the best solutions to achieve "contactless shopping" and "digital transformation of offline retailers". As the economy gradually recovers, residents' shopping needs and abilities gradually recover, and offline retailers' investment budgets in digital transformation will increase, smart shopping carts are expected to become one of the main means of implementing supermarket digital transformation.

[0003] Most of the current smart shopping carts are developed by foreign companies. For example: On July 14, 2020, Amazon announced the launch of the smart shopping cart "Dash Carts", but the product is equipped with multiple cameras, which is costly, cumbersome to operate, and provides a poor shopping experience for customers. Most supermarkets in China are still using traditional shopping carts. In recent years, some domestic units have also designed and developed smart shopping carts. For example, Pingdingshan University has designed a smart shopping cart based on STM32, which has functions such as automatic following and barcode recognition, but the automatic following function is difficult to implement in a complex supermarket shopping environment, and its feasibility is poor; Haikou University of Economics has developed a SmartBuy smart shopping cart based on a single-chip microcomputer design, which has functions such as product search and navigation, but the precise positioning of a single product still faces great challenges.

[0004] Currently, smart shopping carts have appeared both at home and abroad, but this product uses the K210 development board for dual-system technology that combines image recognition and GUI display. This technology has not yet appeared in China. This technology can improve system operating efficiency, is low-cost, advanced, and feasible, so it is highly feasible and can be successfully implemented. It has successfully filled the gap in related domestic technologies and achieved a leap in the intelligence of domestic smart shopping carts.

[0005] In this regard, the inventor proposes a SmartGo guided intelligent shopping cart recognition system to solve the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a SmartGo guided intelligent shopping cart identification system to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A SmartGo guided intelligent shopping cart recognition system, comprising:

[0009] The vehicle body, as well as the motion control module, image recognition processing and display module, positioning and navigation module and cloud platform module built into the vehicle body.

[0010] Preferably, the vehicle body further comprises a basket: used for placing shopping items, with an ergonomic design, reasonable size estimation, and a reasonable and stable center of gravity position;

[0011] Trolley armrest: The pull-rod armrest can be adjusted in height to improve the convenience of use;

[0012] Casters: High-performance casters ensure smooth and stable movement;

[0013] Screen: integrated touch panel to enable interaction with users;

[0014] Vehicle chassis: The design reserves space for placing large objects, enhancing the humanized design.

[0015] Preferably, the motion control module uses a dual-axis XY rocker for power control, and the control unit uses an STM32F103C8T6 chip for ADC conversion, outputs a PWM value to control a high-torque motor, and realizes intelligent control;

[0016] The obstacle avoidance unit uses a diffuse reflection photoelectric switch with a detection distance of 0.2 to 0.3m. When an obstacle is detected, the vehicle will immediately slow down and brake to ensure driving safety.

[0017] Preferably, the image recognition processing and display module includes a K210 visual recognition unit, a Scanner v3.0 barcode recognition unit and a GUI graphical user interface unit;

[0018] The K210 visual recognition unit includes a main control board, which uses K210, has visual and auditory processing capabilities, low power consumption and high performance, and is suitable for the fields of Internet of Things and artificial intelligence;

[0019] Neural network processor KPU, which implements convolutional neural network calculations and obtains product information in real time;

[0020] Model training: Use the Maixhub platform to train product image datasets, and load the model to K210 for product recognition and classification;

[0021] The Scanner v3.0 barcode recognition unit includes a barcode recognition module, which uses a professional image processing chip, has high-performance code reading in complex environments, and can quickly read various barcodes and QR codes;

[0022] Serial communication, by configuring the baud rate and recognition mode, realizes communication with the microcontroller to achieve efficient barcode recognition effect;

[0023] The GUI graphical user interface unit includes a main control board, which adopts TKM32F499 to realize GUI design on an RGB touch screen, display product information, shopping lists, and multi-mode selection, thereby realizing simple and comfortable human-computer interaction.

[0024] Preferably, the positioning and navigation module includes a commodity positioning and shopping cart positioning unit and a path planning and navigation unit;

[0025] The product positioning and shopping cart positioning unit includes positioning technology, adopts Beacon Bluetooth positioning, and realizes the positioning of products and shopping carts through RSSI field strength information;

[0026] The path planning and navigation unit adopts the A* algorithm to realize global path planning and navigation, combines the heuristic method with the conventional method, has high efficiency and the shortest path.

[0027] Preferably, the cloud platform module includes a visual management platform and mini-program development and management based on the Internet of Things cloud platform;

[0028] The visualization management platform based on the IoT cloud platform includes big data visualization, which displays complex data through dimensions such as geographic space, time series, and logical relationships, thereby improving user decision-making capabilities and brand display effects;

[0029] OneNET platform realizes visual management and information transmission of shopping carts. It uploads data to Huawei Cloud Platform through NB-IOT transmission protocol to realize real-time monitoring of shopping cart location, power consumption and faults.

[0030] The mini program is developed and managed using WeChat mini program, the user end is developed using WeChat mini program, the cloud platform server processes data, and transmits data to the WeChat mini program server through the MQTT protocol to realize the front-end display of data.

[0031] Preferably, the center of gravity position calculation is performed by calculating the mass and position of the cart basket, cart handrails, casters and screen to determine the center of gravity position of the entire shopping cart, specifically:

[0032]

[0033] Among them, mi is the mass of each part, xi, yi, zi are the coordinates of each part.

[0034] Preferably, the convolutional neural network calculation uses a neural network processor KPU to perform convolution operations to identify commodity information, and the convolution operation is expressed as:

[0035]

[0036] Among them, I is the input image and K is the convolution kernel.

[0037] Preferably, RSSI field strength ranging positioning is divided into four stages: data collection, data correction, distance fax, and position determination, specifically:

[0038] a. Initialize the anchor node and send Beacon broadcast packets regularly;

[0039] b. At fixed intervals from the anchor node, the onboard controller collects RSSI data and sends it to the host position together with the distance;

[0040] c. After receiving the data, the upper station performs KGMM hybrid filtering on the RSSI data of the fixed point;

[0041] d. Use the filtered RSSI value and the corresponding distance d data to build a ranging model;

[0042] The distance can be calculated using the following formula:

[0043]

[0044] Where A is the signal strength when the transmitter is 1 meter away, n is the environmental attenuation factor, and RSSI is the received signal strength.

[0045] Preferably, the A* algorithm realizes global path planning and navigation by combining heuristic methods with conventional methods, and its evaluation function formula is expressed as:

[0046] f(n)=g(n)+h(n)

[0047] Among them, f(n) is the comprehensive priority of node n, g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the end point.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] (1) The smart shopping cart of the present invention integrates a variety of advanced technologies, such as computer vision, artificial intelligence, machine learning, etc., which can accurately identify products and calculate prices, greatly improving shopping efficiency. The smart shopping cart can have a variety of convenient functions such as automatic list, automatic navigation, automatic checkout, etc., greatly shortening consumers' shopping time and providing great convenience for consumers. The smart shopping cart can also be connected with consumers' mobile phones, smart homes, etc., to recommend personalized products and provide consumers with professional shopping suggestions, thereby providing a better shopping experience.

[0050] (2) The present invention adopts intelligent program settings to largely avoid accidental damage such as collision and scratching of shopping carts, realizes autonomous damage prevention of shopping carts during shopping, increases the service life of shopping carts, and reduces the loss cost of shopping carts in supermarkets. It adopts indoor positioning technology to solve the problem that ordinary shopping carts cannot be accurately captured and located, and are difficult to recycle. This technology accurately locates the position of shopping carts, making the recycling management of shopping carts more convenient, helping supermarkets to achieve accurate and efficient recycling of shopping carts, and facilitating management and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A block diagram of a SmartGo guided intelligent shopping cart identification system of the present invention;

[0052] Figure 2 It is a block diagram of the vehicle body structure of the present invention;

[0053] Figure 3 It is a block diagram of the motion control module of the present invention;

[0054] Figure 4 It is a block diagram of the image recognition module of the present invention;

[0055] Figure 5 It is a block diagram of the positioning and navigation composition of the present invention;

[0056] Figure 6 This is a block diagram of the cloud platform module composition of the present invention. DETAILED DESCRIPTION

[0057] 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.

[0058] Embodiment 1:

[0059] See also Figures 1 to 6 As shown, a SmartGo guided intelligent shopping cart recognition system includes:

[0060] The vehicle body, as well as the motion control module, image recognition processing and display module, positioning and navigation module and cloud platform module built into the vehicle body.

[0061] Specifically, the vehicle body also includes a basket: used to place shopping items, the design is ergonomic, the size is reasonably estimated, and the center of gravity is reasonably and stably located;

[0062] Trolley armrest: The pull-rod armrest can be adjusted in height to improve the convenience of use;

[0063] Casters: High-performance casters ensure smooth and stable movement;

[0064] Screen: integrated touch panel to enable interaction with users;

[0065] Vehicle chassis: The design reserves space for placing large objects, enhancing the humanized design.

[0066] Specifically, the motion control module uses a dual-axis XY rocker for power control, and the control unit uses an STM32F103C8T6 chip for ADC conversion, outputs PWM values ​​to control a high-torque motor, and realizes intelligent control;

[0067] The obstacle avoidance unit uses a diffuse reflection photoelectric switch with a detection distance of 0.2 to 0.3m. When an obstacle is detected, the vehicle will immediately slow down and brake to ensure driving safety.

[0068] Specifically, the image recognition processing and display module includes a K210 visual recognition unit, a Scanner v3.0 barcode recognition unit and a GUI graphical user interface unit;

[0069] The K210 visual recognition unit includes a main control board, which uses K210, has visual and auditory processing capabilities, low power consumption and high performance, and is suitable for the fields of Internet of Things and artificial intelligence;

[0070] Neural network processor KPU, which implements convolutional neural network calculations and obtains product information in real time;

[0071] Model training: Use the Maixhub platform to train product image datasets, and load the model to K210 for product recognition and classification;

[0072] The Scanner v3.0 barcode recognition unit includes a barcode recognition module, which uses a professional image processing chip, has high-performance code reading in complex environments, and can quickly read various barcodes and QR codes;

[0073] Serial communication, by configuring the baud rate and recognition mode, realizes communication with the microcontroller to achieve efficient barcode recognition effect;

[0074] The GUI graphical user interface unit includes a main control board, which adopts TKM32F499 to realize GUI design on an RGB touch screen, display product information, shopping lists, and multi-mode selection, and realize simple and comfortable human-computer interaction.

[0075] Specifically, the positioning and navigation module includes a product positioning and shopping cart positioning unit and a path planning and navigation unit;

[0076] The product positioning and shopping cart positioning unit includes positioning technology, adopts Beacon Bluetooth positioning, and realizes the positioning of products and shopping carts through RSSI field strength information;

[0077] The path planning and navigation unit adopts the A* algorithm to realize global path planning and navigation, combines the heuristic method with the conventional method, has high efficiency and the shortest path.

[0078] Specifically, the cloud platform module includes a visual management platform based on the IoT cloud platform and small program development and management;

[0079] The visualization management platform based on the IoT cloud platform includes big data visualization, which displays complex data through dimensions such as geographic space, time series, and logical relationships, thereby improving user decision-making capabilities and brand display effects;

[0080] OneNET platform realizes visual management and information transmission of shopping carts. It uploads data to Huawei Cloud Platform through NB-IOT transmission protocol to realize real-time monitoring of shopping cart location, power consumption and faults.

[0081] The mini program is developed and managed using WeChat mini program, the user end is developed using WeChat mini program, the cloud platform server processes data, and transmits data to the WeChat mini program server through the MQTT protocol to realize the front-end display of data.

[0082] Specifically, the center of gravity position calculation is performed by calculating the mass and position of the cart basket, cart handrails, casters and screen to determine the center of gravity position of the entire shopping cart, specifically:

[0083]

[0084] Among them, mi is the mass of each part, xi, yi, zi are the coordinates of each part.

[0085] Specifically, the convolutional neural network calculation uses a neural network processor KPU to perform convolution operations to identify product information. The convolution operation is expressed as:

[0086]

[0087] Among them, I is the input image and K is the convolution kernel.

[0088] Specifically, RSSI field strength ranging positioning is divided into four stages: data collection, data correction, distance fax, and position determination, specifically:

[0089] a. Initialize the anchor node and send Beacon broadcast packets regularly;

[0090] b. At fixed intervals from the anchor node, the onboard controller collects RSSI data and sends it to the host position together with the distance;

[0091] c. After receiving the data, the upper station performs KGMM hybrid filtering on the RSSI data of the fixed point;

[0092] d. Use the filtered RSSI value and the corresponding distance d data to build a ranging model;

[0093] The distance can be calculated using the following formula:

[0094]

[0095] Where A is the signal strength when the transmitter is 1 meter away, n is the environmental attenuation factor, and RSSI is the received signal strength.

[0096] Specifically, the A* algorithm realizes global path planning and navigation by combining heuristic methods with conventional methods, and its evaluation function formula is expressed as:

[0097] f(n)=g(n)+h(n)

[0098] Among them, f(n): is the comprehensive priority of node n. When we choose the next node to traverse, we always choose the node with the highest comprehensive priority (smallest value);

[0099] g(n): is the cost of node n from the starting point;

[0100] h(n): is the estimated cost of node n from the end point, which is also the heuristic function of the algorithm.

[0101] Algorithm flow:

[0102] Put the starting point into the "open list";

[0103] Repeat the process as follows:

[0104] Traverse the open list and calculate the evaluation function f(s) for each node s in the list. Find the node n with the smallest f(n) value and use it as the current node to be processed.

[0105] For all other nodes adjacent to the current node n, perform the following operations:

[0106] If node b is unreachable or in the closed list, ignore it. Otherwise, do the following.

[0107] If node c is not in the open list, add it to the open list and set the current node n as its parent node. Calculate the sum of f(n), g(b) and h(b) for node b.

[0108] If node d is already in the open list, we need to check whether this path (the path from node n to node d) is better. The reference index is the g value. If g is smaller, it means that this path is better. If this path is better, its parent node (set to node e) is set as the current node, and g(e) and f(e) are recalculated.

[0109] Move n to the "closed list". All elements in the closed list no longer need to be cared about.

[0110] If the end point is added to the open list, the path is found. If the end point cannot be found and the open list is empty, there is no path.

[0111] As can be seen from the above, smart shopping carts integrate a variety of advanced technologies, such as computer vision, artificial intelligence, machine learning, etc., which can accurately identify products and calculate prices, greatly improving shopping efficiency. Smart shopping carts can automatically list, automatically navigate, automatically check out and other convenient functions, greatly shortening consumers' shopping time and providing great convenience for consumers. Smart shopping carts can also connect with consumers' mobile phones, smart homes, etc., recommend personalized products, and provide consumers with professional shopping suggestions, thereby providing a better shopping experience.

[0112] Intelligent program settings can largely avoid accidental damage to shopping carts such as collisions and scratches, realize autonomous damage prevention of shopping carts during shopping, increase the service life of shopping carts, and reduce the loss cost of shopping carts in supermarkets. Indoor positioning technology is used to solve the problem that ordinary shopping carts cannot be accurately captured and located, and are difficult to recycle. This technology accurately locates the position of shopping carts, making the recycling management of shopping carts more convenient, helping supermarkets to achieve accurate and efficient recycling of shopping carts, and facilitating management and maintenance.

[0113] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0114] In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

[0115] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A SmartGo guided intelligent shopping cart recognition system, characterized in that: include: The vehicle body, as well as the motion control module, image recognition processing and display module, positioning and navigation module and cloud platform module built into the vehicle body.

2. The SmartGo guided intelligent shopping cart identification system according to claim 1, characterized in that: The vehicle body also includes a basket: the basket is used to place shopping items, and the center of gravity is reasonably and stably located; Trolley Armrests: Pull-rod armrests are used to adjust the height; Casters: High-performance casters ensure smooth and stable movement; Screen: integrated touch panel to enable interaction with users; Vehicle chassis: The design reserves space for placing large objects.

3. The SmartGo guided intelligent shopping cart identification system according to claim 1, characterized in that: The motion control module uses a dual-axis XY rocker for power control, and the control unit uses an STM32F103C8T6 chip for ADC conversion, outputs PWM values ​​to control the torque motor, and realizes intelligent control; The obstacle avoidance unit uses a diffuse reflection photoelectric switch with a detection distance of 0.2 to 0.3m. When an obstacle is detected, the vehicle will immediately slow down and brake to ensure driving safety.

4. The SmartGo guided intelligent shopping cart identification system according to claim 1, characterized in that: The image recognition processing and display module includes a K210 visual recognition unit, a Scanner v3.0 barcode recognition unit and a GUI graphical user interface unit; The K210 visual recognition unit includes a main control board, which uses K210, has visual and auditory processing capabilities, low power consumption and high performance, and is suitable for the fields of Internet of Things and artificial intelligence; Neural network processor KPU, which implements convolutional neural network calculations and obtains product information in real time; Model training: Use the Maixhub platform to train product image datasets, and load the model to K210 for product recognition and classification; The Scanner v3.0 barcode recognition unit includes a barcode recognition module, which uses a professional image processing chip, has high-performance code reading in complex environments, and can quickly read various barcodes and QR codes; Serial communication, by configuring the baud rate and recognition mode, realizes communication with the microcontroller to achieve efficient barcode recognition effect; The GUI graphical user interface unit includes a main control board, which adopts TKM32F499 to realize GUI design on an RGB touch screen, display product information, shopping lists, and multi-mode selection, and realize simple and comfortable human-computer interaction.

5. The SmartGo guided intelligent shopping cart identification system according to claim 1, characterized in that: The positioning and navigation module includes a product positioning and shopping cart positioning unit and a path planning and navigation unit; The product positioning and shopping cart positioning unit includes positioning technology, adopts Beacon Bluetooth positioning, and realizes the positioning of products and shopping carts through RSSI field strength information; The path planning and navigation unit adopts the A* algorithm to realize global path planning and navigation.

6. The SmartGo guided intelligent shopping cart identification system according to claim 1, characterized in that: The cloud platform module includes a visual management platform and mini-program development and management based on the IoT cloud platform; The visualization management platform based on the IoT cloud platform includes big data visualization, which displays complex data through geographic space, time series, and logical relationship dimensions to improve user decision-making ability and brand display effect; OneNET platform realizes visual management and information transmission of shopping carts. It uploads data to Huawei Cloud Platform through NB-IOT transmission protocol to realize real-time monitoring of shopping cart location, power consumption and faults. The mini program is developed and managed using WeChat mini program, the user end is developed using WeChat mini program, the cloud platform server processes data, and transmits data to the WeChat mini program server through the MQTT protocol to realize the front-end display of data.

7. The SmartGo guided intelligent shopping cart identification system according to claim 2, characterized in that: The center of gravity position calculation is to determine the center of gravity position of the entire shopping cart by calculating the mass and position of the cart basket, cart armrests, casters and screen, specifically: Among them, mi is the mass of each part, xi, yi, zi are the coordinates of each part.

8. The SmartGo guided intelligent shopping cart identification system according to claim 4, characterized in that: The convolutional neural network calculation uses the neural network processor KPU to perform convolution operations to identify product information. The convolution operation is expressed as: Among them, I is the input image and K is the convolution kernel.

9. The SmartGo guided intelligent shopping cart identification system according to claim 5, characterized in that: RSSI field strength ranging positioning is divided into four stages: data collection, data correction, distance fax, and position determination, specifically: a. Initialize the anchor node and send Beacon broadcast packets regularly; b. At fixed intervals from the anchor node, the onboard controller collects RSSI data and sends it to the host position together with the distance; c. After receiving the data, the upper station performs KGMM hybrid filtering on the RSSI data of the fixed point; d. Use the filtered RSSI value and the corresponding distance d data to build a ranging model; The distance is calculated using the following formula: Where A is the signal strength when the transmitter is 1 meter away, n is the environmental attenuation factor, and RSSI is the received signal strength.

10. The SmartGo guided intelligent shopping cart identification system according to claim 5, characterized in that: The A* algorithm achieves global path planning and navigation by combining heuristic methods with conventional methods. Its evaluation function formula is expressed as: f(n)=g(n)+h(n) Among them, f(n) is the comprehensive priority of node n, g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the end point.

Images