AI-enabled kits for self-checkout devices

Abstract

The present invention provides an AI-enabled kit for self-checkout devices, which is attached to the self-checkout device and includes: a separate AI-enabled box module, which includes a first microcontroller and is configured to communicate with the self-checkout device; a checkout-end computer program product, which is selectively configured on the self-checkout device; and an enabler-end computer program product, which is selectively configured on the separate AI-enabled box module and configured to control the first microcontroller to perform peripheral device simulation.

Description

Technical Field

[0001] This invention relates to an AI-enabled kit for self-checkout equipment, and more particularly to an AI-enabled kit for self-checkout equipment that introduces AI functionality to traditional self-checkout equipment without modifying its hardware or software. Background Technology

[0002] Following the pandemic, the shortage of human resources has led to the rise of self-checkout (SCO) technology. However, with the rapid growth in demand for self-service, traditional SCO machines have also revealed their insufficient anti-theft and anti-missed transactions capabilities. Nevertheless, due to the rapid development of AI software and hardware in recent years, there has been significant progress in detecting and preventing theft and missed transactions.

[0003] However, existing SCO machines, due to their earlier development, often lack these AI anti-theft and anti-missed-swiping functions, and also lack the powerful computing power required to run AI calculations. If you want to introduce AI solutions into existing SCO machines, you will face many problems at both the software and hardware levels.

[0004] For example, most peripheral devices in the current SCO system, such as cameras, barcode scanners, card readers, and printers, need to be directly connected to the existing SCO machines and are controlled and exchange data based on protocols and instruction sets such as OPOS (OLE for Retail POS), JPOS (Java for Retail POS), and Command Set. The interfaces and data structures of these peripheral devices must fully comply with the original system design, and the system design is relatively closed. If AI functions are to be introduced, the communication logic or data flow design between the peripheral devices and the main system needs to be significantly modified. Furthermore, the newly introduced AI system and the original SCO system require a long period of testing and debugging to ensure the stability and security of the entire system.

[0005] Furthermore, since SCO devices were developed relatively early, their hardware architecture and software design are usually completed by different vendors, resulting in inconsistent interface standards between systems. Therefore, integrating different interfaces is also a major challenge. Moreover, the existing SCO hardware and software designs do not take into account direct support for AI solutions.

[0006] Therefore, if businesses want to seamlessly integrate AI recognition technology, they usually need to make significant modifications to their existing software systems, such as redesigning or updating existing applications and control logic to ensure that the AI ​​system can communicate and operate smoothly with peripheral devices. These modifications require a lot of manpower, time and money, and may lead to system instability or changes in operating procedures, further increasing the resistance to technology adoption.

[0007] Furthermore, since the SCO system involves cash flow and privacy, many merchants are unwilling to open up the data interfaces of their existing applications for modification or accept software and hardware modifications, based on privacy and system security considerations and in order to ensure transaction security. This further increases the challenges of introducing AI solutions.

[0008] However, with the rapid development of AI technology, including the improvement of AI chip computing power and the emergence of various AI models, many retailers, wholesalers and independent software vendors (ISVs) are now looking forward to upgrading traditional SCO systems to AI-SCO systems by introducing AI solutions, so as to use functions such as detecting missed transactions, preventing fraudulent transactions and performing product recognition, thereby improving the reliability of SCO systems.

[0009] Therefore, in view of the shortcomings of the existing technology, the inventor, through careful attempts and research, and with a persistent spirit, finally conceived the "AI-enabled kit for self-checkout equipment" in this case, which can overcome the above-mentioned shortcomings. The following is a brief description of the invention. Summary of the Invention

[0010] This invention relates to an AI-enabled kit for self-checkout equipment, and more particularly to an AI-enabled kit for self-checkout equipment that introduces AI functionality to traditional self-checkout equipment without altering its hardware or software.

[0011] This invention proposes an AI-enabled kit for self-checkout devices, which is attached to the self-checkout device and includes: a separate AI-enabled box module, which includes a first microcontroller and is configured to communicate with the self-checkout device; a checkout-end computer program product, which is selectively configured on the self-checkout device; and an enabler-end computer program product, which is selectively configured on the separate AI-enabled box module and configured to control the first microcontroller to perform peripheral device simulation.

[0012] The above-described invention description is intended to provide a simplified summary of the present disclosure so that the reader can have a basic understanding of the present disclosure. This invention description is not a complete description of the invention and is not intended to point out important / key elements of the embodiments of the invention or to define the scope of the invention. Attached Figure Description

[0013] Figure 1 A schematic diagram of the system architecture of the AI-enabled self-checkout device kit included in this invention is shown.

[0014] Figure 2 A block diagram illustrating a first embodiment of the present invention;

[0015] Figure 3A schematic diagram illustrating the operating principle of the checkout terminal computer program product included in this invention;

[0016] Figure 4 A schematic diagram of the data flow of the first embodiment of the present invention is disclosed.

[0017] Figure 5 A block diagram illustrating a second embodiment of the present invention;

[0018] Figure 6 A flowchart illustrating the implementation steps of the initialization procedure included in the peripheral device simulation proposed in this invention is provided.

[0019] Figure 7 A flowchart illustrating the implementation steps of the constant-running program included in the peripheral device simulation proposed in this invention;

[0020] Figure 8 A block diagram illustrating a third embodiment of the present invention; and

[0021] Figure 9 This invention discloses the circuit layout of the second microcontroller and the schematic diagram of the operating principle of the acquisition terminal computer program product.

[0022] Explanation of reference numerals in the attached figures

[0023] 200 self-checkout machines

[0024] 201 Communication Interface

[0025] 100 AI-powered self-checkout device kits

[0026] 101 Separable AI-enabled Box Module

[0027] 102 Data Acquisition Module

[0028] 103 First transmission path

[0029] 104 First transmission path

[0030] 111 First Microcontroller

[0031] 112 Second Microcontroller

[0032] 113 processor

[0033] 121 Checkout Terminal Computer Program Product

[0034] 122 Empowering PC Program Products

[0035] 123 Acquisition of terminal computer program products

[0036] 131 barcode scanner

[0037] 132 printer

[0038] 133 Card Reader

[0039] 134 Magnetic Card Machine

[0040] 141 Image Sensor

[0041] 142 motion sensors

[0042] 150 USB host function driver stack

[0043] 151 Functional Device Objects

[0044] 152 USB filter driver

[0045] 153 Solid Devices and Objects

[0046] 160 Serial Port Driver Stack

[0047] 161 Virtual COM Device

[0048] 162 Sequence Port Filter Driver

[0049] 163 Device COM Port

[0050] 10 Initialization Program

[0051] 11-17 Implementation Steps

[0052] 20. Always-on running programs

[0053] 21-28 Implementation Steps Detailed Implementation

[0054] The invention will be fully understood from the following embodiments, which enable those skilled in the art to carry out the invention. However, the implementation of the invention is not limited to the following embodiments. The accompanying drawings of the invention do not contain any limitations on size, dimensions and scale. The actual size, dimensions and scale of the invention are not limited by the accompanying drawings.

[0055] The term "preferred" as used herein is non-exclusive and should be understood as "preferred but not limited to". Any step described or recorded in any specification or claim may be performed in any order, but is not limited to the order stated in the claim. The scope of the invention should be determined only by the appended claims and their equivalents, and not by the embodiments of the examples of implementation. The term "comprising" and its variations appearing in the specification and claims herein are open-ended terms, not restrictive, and do not exclude other features or steps.

[0056] This invention aims to provide an AI upgrade solution for self-checkout (SCO) devices that do not have artificial intelligence (AI) capabilities. It can acquire the data required to implement AI functions without modifying the original SCO system's hardware and software architecture, integrate an external AI Box into the existing SCO device, introduce multiple AI functions into the existing SOC device, provide a simple and efficient AI upgrade path for the existing SCO system, and significantly improve the performance of the existing SCO system.

[0057] This invention proposes a complete system architecture for a separate SCO (Single-Site Container Registry) AI-enabled system. It employs a basic architecture that separates the "AI-enabled system" from the "existing SCO system," constructing an external AI module that connects to existing SCO machines. This module enables the integration of powerful AI functions into existing SCO equipment without modifying the existing hardware and software configurations, operating systems, and applications (Apps), nor requiring code modification or redesign of peripheral devices. This approach not only helps reduce the complexity of implementing AI functions but also provides different solutions to meet the varying needs of different vendors for the depth and breadth of AI applications.

[0058] This invention provides an AI solution that can be quickly implemented without modifying the existing SCO system's hardware and software configuration. It can be used by users with different needs, including but not limited to retailers, wholesalers, independent software vendors (ISVs), system integrators, and AI service providers, to smoothly and quickly implement AI systems of different levels while minimizing integration risks and development costs.

[0059] Figure 1 This invention discloses a schematic diagram of the system architecture of the AI-enabled self-checkout device kit included in this invention. The AI-enabled self-checkout device kit 100 of this invention includes one of the following: a separate AI-enabled box module 101, a data acquisition module 102, a checkout terminal computer program product 121, an enabler terminal computer program product 122, and an acquisition terminal computer program product 123. The separate AI-enabled box module 101 is preferably configured with a processor 113 and a first microcontroller (MCU) 111. The processor 113 is preferably configured to execute, for example, but not limited to, various AI algorithms. The data acquisition module 102 is preferably configured with a second microcontroller 112.

[0060] The self-checkout device 200 preferably also includes SCO peripheral devices such as barcode scanner 131, printer 132, card reader 133 and magnetic card reader 134, etc., and the artificial intelligence empowerment box module 101 preferably also includes AI peripheral devices such as image sensor 141 and motion sensor 142, etc.

[0061] In one embodiment, the separate AI-enabled box module 101 is preferably directly connected to the self-checkout device 200 via a first transmission path 103. In another embodiment, the separate AI-enabled box module 101 is preferably connected to the self-checkout device 200 via a second transmission path 104 including a data acquisition module 102. The first transmission path 103 and the second transmission path 104 preferably include various communication interfaces such as a USB communication interface or a serial port communication interface, a network such as a wired network or a wireless network, the Internet, a telecommunications network, or a combination thereof.

[0062] The checkout computer program product 121 is preferably configured and executed on the self-checkout device 200, the enabling computer program product 122 is preferably configured and executed on the separate artificial intelligence enabling box module 101, the acquisition computer program product 123 is preferably configured and executed on the data acquisition module 102, and the entire self-checkout device artificial intelligence enabling kit 100 is preferably attached to a self-checkout device 200 that does not have artificial intelligence capabilities and operates therein.

[0063] Figure 2 The block diagram illustrates a first embodiment of the present invention. In the first embodiment, depending on different hardware and software conditions and the client's authorization scope, the actual configurable conditions may be: it is acceptable for the SCO device to be directly connected to the AI ​​box and for external software to be installed, but SCO peripheral devices need to be connected to the SCO device.

[0064] In the first embodiment, the self-checkout device AI-enabled kit 100 selectively includes a separate AI-enabled box module 101 and a checkout terminal computer program product 121. The separate AI-enabled box module 101 is configured to connect to AI peripheral devices, including but not limited to image sensors 141 and motion sensors 142. The self-checkout device 200 is configured to connect to SCO peripheral devices, including but not limited to barcode scanners 131, printers 132, card readers 133, and magnetic card readers 134. The core task of the checkout terminal computer program product 121 is to intercept and acquire data streams and install and execute it on the self-checkout device 200 in the form of application software.

[0065] Preferably, the checkout terminal computer program product 121 is configured to insert a filter driver into a specified location in the communication interface driver stack by modifying the installation configuration file (such as the INF file) of the SCO peripheral device driver in the self-checkout device 200 or by modifying system settings. This allows the filter driver to monitor and intercept all data passing through the communication interface, filter out data from the SCO peripheral device, and then transmit the acquired raw data to the separate AI-enabled box module 101 for subsequent AI analysis, detection, interpretation, and recognition, including but not limited to: missed swipe detection, anti-theft swipe, product recognition, abnormal item detection, transaction information comparison, and consumer image data analysis.

[0066] Figure 3 This invention discloses a schematic diagram illustrating the operational principle of the checkout terminal computer program product. In one embodiment, SCO peripheral devices such as barcode scanner 131, printer 132, card reader 133, and magnetic card reader 134 are preferably connected to the self-checkout device 200 via a communication interface 201 configured on the self-checkout device 200, such as a USB or serial port communication interface. Taking the USB communication interface as an example, the checkout terminal computer program product 121 is configured to insert a USB filter driver 152 between a functional device object (FDO) 151 and a physical device object (PDO) 153, forming a USB function driver stack 150. This allows the USB function driver to monitor and intercept all data passing through the USB communication interface between the USB function driver and the bus driver, and to filter out data from the SCO peripheral devices.

[0067] Taking the serial port communication interface as an example, the checkout computer program product 121 is configured to insert the serial port filter driver 162 between the virtual COM device 161 and the device COM port 163, forming a serial port driver stack 160, so as to monitor and intercept all data passing through the serial port communication interface between the serial port main function driver and the bus driver, and obtain data from the SCO peripheral device.

[0068] Figure 4A schematic diagram of the data flow of the first embodiment of the present invention is shown. After receiving raw data from filter drivers, including but not limited to USB filter driver 152 and serial port filter driver 162, the checkout computer program product 121 checks and classifies the raw data. After the raw data passes the check and classification, it is transmitted to the separate AI-enabled box module 101. In one embodiment, the checkout computer program product 121 preferably supports intercepting and processing instruction sets and data streams based on, for example, but not limited to, OPOS and JPOS.

[0069] After receiving raw data from the SCO peripheral device, the split AI-enabled box module 101 will perform subsequent AI analysis, detection, interpretation and recognition based on the raw data from the SCO peripheral device and the image data and detection data obtained from the image sensor 141 and motion sensor 142.

[0070] In this embodiment, since the filter driver is deployed between the software layer functional objects and the underlying hardware objects of the self-checkout device 200, the data streams from SCO peripheral devices are intercepted before reaching the operating system or application. Therefore, during the data acquisition process, the self-checkout device 200 does not need to know about the existence of the separate AI enabler module 101 and assumes that its own software and hardware configuration has not changed. Thus, without modifying the existing software process of the self-checkout device 200, without modifying any source code, and without modifying the hardware, the separate AI enabler module 101 can still acquire the data streams from all SCO peripheral devices for subsequent AI interpretation.

[0071] In this embodiment, for the client, the client only needs to simply install the checkout terminal computer program product 121 on the self-checkout device 200 and connect the self-checkout device 200 to the separate artificial intelligence empowerment box module 101. With very few changes, the artificial intelligence upgrade of the self-checkout device 200 can be completed on the spot, giving the original SCO system AI capabilities. The original checkout process does not need to be changed, and the original SCO system functions are all working normally without any changes.

[0072] Figure 5 The block diagram illustrates a second embodiment of the present invention. In the second embodiment, depending on the hardware and software conditions or the scope of authorization of the client, the actual configurable conditions may be as follows: it is acceptable for the SCO device to be directly connected to the AI ​​box and for all SCO peripheral devices to be connected to the AI ​​box, but no external software can be installed on the SCO device.

[0073] In the second embodiment, the self-checkout device AI empowerment kit 100 selectively includes a separate AI empowerment box module 101 and an empowerment terminal computer program product 122. The separate AI empowerment box module 101 includes a first microcontroller 111 and connects to SCO peripheral devices and AI peripheral devices such as barcode scanner 131, printer 132, card reader 133 and magnetic card reader 134, image sensor 141 and motion sensor 142, etc. The first microcontroller 111 is configured to perform peripheral device emulation.

[0074] The core tasks of the enabling computer program product 122 include relaying two-way communication and simulating peripheral devices. It is executed as firmware attached to the separate artificial intelligence enabling box module 101. The enabling computer program product 122 is configured to control the first microcontroller 111 to simulate peripheral devices.

[0075] In this embodiment, since the SCO peripheral device is changed from being connected to the self-checkout device 200 to being connected to the separate AI-enabled box module 101, it is necessary to simulate the peripheral device from the separate AI-enabled box module 101 so that the self-checkout device 200 thinks that its own hardware and software configuration has not changed. This achieves the effect of acquiring data and giving the SCO device AI capabilities without modifying the SCO device's hardware and software.

[0076] In one embodiment, the peripheral device simulation includes at least an initialization program and a standby program. The initialization program triggers a hot-plug event on the separate AI-enabled box module 101, causing the self-checkout device 200 to trigger the device enumeration service, thereby realizing the peripheral device simulation task. The standby program realizes the bidirectional data relay task by forming a data flow loop between the separate AI-enabled box module 101 and the self-checkout device 200.

[0077] In one embodiment, the separate AI-enabled box module 101 preferably provides a USB communication interface and a serial port communication interface for peripheral devices to communicate with each other, and provides a USB communication interface to communicate with the self-checkout device 200.

[0078] Figure 6 A flowchart illustrating the implementation steps of the initialization procedure included in the peripheral device simulation proposed in this invention is provided. In one embodiment, the initialization procedure 10 includes the following steps:

[0079] Step 11: The enabling computer program product 122 detects the peripheral devices connected to the separate artificial intelligence enabling box module 101.

[0080] Step 12: The enabling computer program product 122 obtains the device description information (descriptor) of the connected peripheral device, such as the USB description and other information required by the device driver.

[0081] Step 13: The enabling computer program product 122 transmits the acquired device description information to the first microcontroller 111.

[0082] Step 14: The first microcontroller 111 modifies or reorganizes the device description information, triggers a hot-plug event, and sends a hot-plug event notification to the self-checkout device 200.

[0083] Step 15: In response to a hot-plug event, the self-checkout device 200 triggers the device enumeration service and requests device description information from the first microcontroller 111.

[0084] Step 16: The first microcontroller 111 transmits the device description information to the self-checkout device 200.

[0085] Step 17: The self-checkout device 200 loads the correct drivers and application processes according to the device description information.

[0086] After the initialization process is successfully executed, the self-checkout device 200 will treat the first microcontroller 111 as a real SCO peripheral device and assume that it is still connected to these SCO peripheral devices. It will not know that the SCO peripheral devices have actually been plugged into the separate AI-enabled box module 101.

[0087] Figure 7 A flowchart illustrating the implementation steps of the always-running program included in the peripheral device simulation proposed in this invention is provided. In one embodiment, the always-running program 20 includes the following steps:

[0088] Step 21: Peripheral devices such as barcode scanner 131, printer 132, card reader 133 and magnetic card reader 134 send out packet data.

[0089] Step 22: The enabling computer program product 122 receives packet data generated by the peripheral device.

[0090] Step 23: The enabling computer program product 122 transmits packet data to the first microcontroller 111 through the software driver (SW driver), so that the first microcontroller 111 pretends to be a peripheral device.

[0091] Step 24: The first microcontroller 111 transmits the packet data to the self-service checkout device 200 according to the device instructions and data format corresponding to the peripheral device.

[0092] Step 25: The self-checkout device 200 sends an ACK back to the first microcontroller 111.

[0093] Step 26: The first microcontroller 111 acquires packet data transmitted from the self-checkout device 200 to the peripheral device.

[0094] Step 27: The first microcontroller 111 transmits the packet data to the enabling computer program product 122.

[0095] Step 28: The enabling computer program product 122 transmits the packet data to the corresponding peripheral device.

[0096] This creates a loop of data transmission and reception between the separate AI-enabled box module 101 and the self-checkout device 200, thus achieving the task of bidirectional data relay.

[0097] In this embodiment, the SCO peripheral device is actually connected to the separate AI-enabled box module 101. However, after the peripheral device is simulated by the first microcontroller 111 to disguise itself as an SCO peripheral device, the self-checkout device 200 does not perceive the change in the peripheral device configuration at the hardware level and assumes that the peripheral device is always connected to the self-checkout device 200.

[0098] In this embodiment, although the SCO peripheral devices are plugged into the separate AI-enabled box module 101, since no external programs are installed on the existing SCO machine and the existing SCO machine also believes that there is no change in hardware configuration, the self-checkout device 200 will think that there is no change in its own software and hardware configuration. Therefore, it is possible to achieve the effect of obtaining the data stream of all SCO peripheral devices for subsequent AI interpretation without modifying the software process of the existing self-checkout device 200, without modifying any source code, and without modifying the hardware device.

[0099] In this embodiment, for the client, the client only needs to change the peripheral device originally connected to the self-checkout device 200 to the separate AI-enabled box module 101, and then connect the separate AI-enabled box module 101 to the self-checkout device 200. With very few changes, the AI ​​upgrade of the self-checkout device 200 can be completed on the spot, giving the original SCO system AI capabilities. The original checkout process does not need to be changed, and the original SCO system functions are all working normally without any changes.

[0100] Figure 8The block diagram illustrates a third embodiment of the present invention. In the third embodiment, depending on the hardware and software conditions or the scope of authorization of the client, the actual configurable conditions may be as follows: the SCO device cannot be directly connected to the AI ​​box, nor can external software be installed, but it is acceptable for the SCO device to be indirectly connected to the AI ​​box through a hardware device in the form of a hub or dongle, and all SCO peripheral devices can be connected to the hub module.

[0101] In the third embodiment, the self-checkout device AI-enabled kit 100 selectively includes a separate AI-enabled box module 101, a data acquisition module 102, and an acquisition-end computer program product 123. The data acquisition module 102 is preferably made in the form of a hub or plug-and-play device and includes a second microcontroller 112. The data acquisition module 102 connects to SCO peripheral devices including, but not limited to, a barcode scanner 131, a printer 132, a card reader 133, and a magnetic card reader 134. The separate AI-enabled box module 101 is configured to connect to AI peripheral devices including, but not limited to, an image sensor 141 and a motion sensor 142. The self-checkout device 200 is connected to the separate AI-enabled box module 101 through the data acquisition module 102.

[0102] The core task of the acquisition end computer program product 123 is to perform signal sniffing and analysis as well as data stream interception and acquisition. In one embodiment, the core program of the acquisition end computer program product 123 includes, for example, but not limited to, a USB protocol analyzer. The acquisition end computer program product 123 is preferably attached to the second microcontroller 112 in the form of firmware and executed.

[0103] Figure 9 This invention discloses the circuit layout of the second microcontroller and the schematic diagram of the operating principle of the acquisition terminal computer program product. In one embodiment, the second microcontroller 112 is preferably a high-frequency microcontroller, and is configured in the data acquisition module 102 with a circuit layout that bypasses the main circuit of the communication interface 201 configured with the self-checkout device 200 and the SCO peripheral device such as the barcode scanner 131. It sniffs and parses the signals passing through the main circuit, i.e., the communication interface, from the data stream flowing between the communication interface 201 and the SCO peripheral device, and obtains the required data packets from them.

[0104] In one embodiment, a signal enters the communication interface 201 from a peripheral device such as a barcode scanner 131. Assuming the communication interface is a USB communication interface, under the control of the acquisition end computer program product 123, the second microcontroller 112 is configured to parse the USB D+ / D- signal and then transmit the parsed signal to the separate AI-enabled box module 101. If the communication interface is a serial port communication interface, under the control of the acquisition end computer program product 123, the second microcontroller 112 is configured to parse the SerialPort TX / RX signal and then transmit the parsed signal to the separate AI-enabled box module 101.

[0105] In this embodiment, since the data acquisition module 102 acquires the data stream by means of bypass signal parsing, the existing SCO machine will think that there is no change in the hardware configuration. Since no external program is installed on the existing SCO machine, the self-checkout device 200 will think that there is no change in the hardware and software configuration of the machine. Therefore, without modifying the software process of the existing self-checkout device 200, without modifying any source code, and without modifying the hardware, the separate AI empowerment box module 101 can still acquire the data stream of all SCO peripheral devices through the data acquisition module 102 for subsequent AI interpretation.

[0106] In this embodiment, the separate AI-enabled box module 101 preferably allows the use of third-party brand products from suppliers different from those of the data acquisition module 102 and the self-checkout device 200, or products from any brand.

[0107] In this embodiment, for the client, the client only needs to reconnect the peripheral device originally connected to the self-checkout device 200 to the data acquisition module 102, and then connect the data acquisition module 102 to the separate AI-enabled box module 101 and the self-checkout device 200 respectively. With very few changes, the AI ​​upgrade of the self-checkout device 200 can be completed on the spot, giving the original SCO system AI capabilities. The original checkout process does not need to be changed, consumers do not need to change their established checkout habits, and the original SCO system functions are completely normal and do not need to be changed. No software needs to be installed in the original SCO machine. The SCO machine still receives peripheral data and performs transaction operations as usual, and consumers still perform self-checkout according to their previous habits. However, the separate AI-enabled box module 101 has simultaneously obtained all the necessary raw data.

[0108] This invention provides an AI upgrade and integration solution for existing self-checkout devices that lack AI capabilities. The solution includes a separate AI-enabled box module, a data acquisition module, a checkout terminal computer program product, an enabled terminal computer program product, and an acquisition terminal computer program product. This allows existing self-checkout devices with insufficient computing power and no AI capabilities to be easily upgraded into intelligent self-checkout devices with multiple AI capabilities.

[0109] This invention employs a modular hardware and software design, which can be flexibly applied to different types of SCO equipment. Whether it is a machine that allows the installation of external software or a closed system that only allows external hardware, it can be easily deployed, significantly reducing the integration difficulty of introducing AI technology into existing SCO systems.

[0110] The solution proposed in this invention can add AI judgment, detection and analysis functions to the existing SCO environment, while maintaining the normal operation of the original SCO system and peripheral devices, such as cameras, barcode scanners, card readers, printers, etc., so that merchants do not have to change the original checkout process that consumers are already used to, but can deploy AI functions such as preventing missed transactions and preventing theft.

[0111] The technical architecture proposed in this invention allows AI analysis functions to be realized through an external, separate AI enablement box module without changing the existing SCO machine hardware configuration. Its characteristic of not changing the existing SCO machine hardware configuration can ensure the stability of the original system and reduce the risks and interference in the technology deployment process.

[0112] The AI ​​upgrade solution proposed in this invention does not require modification of the existing application or operating system source code of the SCO system, nor does it require rewriting the original driver logic. Through the combination of filter driver, simulated firmware and bypass acquisition module, this invention can intercept and process the required data while maintaining the normal operation of the original SCO system.

[0113] The three solutions provided by this invention can be flexibly configured for different software and hardware configuration conditions and customer needs. This invention is applicable to SCO systems that allow the installation of external software, closed systems that only allow external hardware, or special needs that require a bypass acquisition method.

[0114] The embodiments of this invention described above can be arbitrarily combined or substituted with each other to generate more implementation methods, but none of them shall depart from the scope of protection intended for this invention. Further examples of implementations of this invention are provided below:

[0115] Example 1: An AI-enabled kit for a self-checkout device, attached to the self-checkout device, includes: a separate AI-enabled box module containing a first microcontroller configured to communicate with the self-checkout device; a checkout-end computer program product selectively configured on the self-checkout device; and an enabler-end computer program product selectively configured on the separate AI-enabled box module, configured to control the first microcontroller to simulate peripheral devices.

[0116] Example 2: The AI-enabled kit as described in Example 1 further includes one of the following: a data acquisition module, selectively configured between the self-checkout device and the separate AI-enabled box module, and including a second microcontroller and a communication interface, the communication interface connecting the peripheral device and the self-checkout device, the second microcontroller being configured to connect the communication interface and the separate AI-enabled box module in a bypass configuration; and an acquisition-end computer program product, configured in the data acquisition module, to control the second microcontroller to perform bypass signal parsing to acquire data generated by the peripheral device and transmit it to the separate AI-enabled box module.

[0117] Example 3: The AI-enabled kit as described in Example 1, wherein the separate AI-enabled box module further includes: a processor configured to execute AI algorithms, wherein the separate AI-enabled box module endows the self-checkout device with AI capabilities.

[0118] Example 4: The AI-enabled kit as described in Example 2 further includes one of the following: multiple self-checkout peripheral devices, including one of a barcode scanner, a printer, a card reader, and a magnetic card reader; and multiple AI peripheral devices, including one of an image sensor and a motion sensor.

[0119] Example 5: The AI-enabled kit as described in Example 4, wherein the checkout terminal computer program product is configured to include a filter driver and insert the filter driver between the main function driver and the bus driver of the communication interface included in the self-checkout device, so as to monitor and intercept the data stream through the communication interface included in the self-checkout device after the first self-checkout peripheral device is inserted into the self-checkout device, so as to obtain the raw data from the first self-checkout peripheral device.

[0120] Example 6: As described in Example 4, the AI-enabled kit is configured to control the first microcontroller to simulate the peripheral device after the separate AI-enabled box module is inserted into the first self-checkout peripheral device included in the self-checkout peripheral device. The first microcontroller simulates the first self-checkout peripheral device so that the self-checkout device can perform corresponding hardware configuration to obtain raw data from the first self-checkout peripheral device.

[0121] Example 7: The AI-enabled kit as described in Example 6, wherein the peripheral device simulation includes an initialization program and a runtime program. The initialization program includes one of the following: obtaining device description information of the first self-checkout peripheral device; triggering a hot-plug event in the separate AI-enabled box module based on the device description information; sending a notification to the self-checkout device to notify of the hot-plug event; upon receipt of the notification, configuring the self-checkout device to request the device description information from the separate AI-enabled box module; and configuring the self-checkout device to trigger a device enumeration service based on the device description information.

[0122] Example 8: The AI-enabled kit as described in Example 7, wherein the always-on program includes one of the following: receiving first packet data generated by the first self-checkout peripheral device; transmitting the first packet data to the first microcontroller via a software driver; controlling the first microcontroller to transmit the first packet data to the self-checkout device using corresponding device instructions and data formats according to the device description information; receiving second packet data from the self-checkout device to the first self-checkout peripheral device via the first microcontroller; transmitting the second packet data to the enabling computer program product via the first microcontroller; and transmitting the second packet data to the first self-checkout peripheral device via the enabling computer program product.

[0123] Example 9: The AI-enabled kit as described in Example 4, wherein the second microcontroller is configured in the data acquisition module with a circuit layout that bypasses the main circuit of the communication interface included in the self-checkout device and the first self-checkout peripheral device included in the self-checkout peripheral device.

[0124] Example 10: The AI-enabled kit as described in Example 9, wherein the acquisition-end computer program product is configured to include a protocol parser to control the second microcontroller to perform bypass signal parsing after the data acquisition module is inserted into the first self-checkout peripheral device included in the self-checkout peripheral device, so as to parse the data stream through the communication interface in order to obtain the raw data about the first self-checkout peripheral device.

[0125] The various embodiments of the invention can be arbitrarily combined or substituted with each other to generate more implementation methods, but none of them shall deviate from the scope of protection of the invention. The scope of protection of the invention shall be determined by the claims of the invention.

Claims

1. An AI-enabled kit attached to a self-checkout device, comprising: A separate AI-enabled box module, which includes a first microcontroller and is configured to communicate with the self-checkout device; A checkout terminal computer program product, which is selectively configured in the self-checkout device; and The enabling computer program product is selectively configured in the separate artificial intelligence enabling box module and configured to control the first microcontroller to simulate peripheral devices.

2. The AI-enabled kit according to claim 1 further comprises one of the following: A data acquisition module, selectively configured between the self-checkout device and the separate AI-enabled box module, includes a second microcontroller configured to be connected to the self-checkout device in a bypass-configured circuit layout; and The acquisition terminal computer program product is configured in the data acquisition module to control the second microcontroller to perform bypass signal analysis in order to acquire the data generated by the peripheral device and transmit it to the separate artificial intelligence enabling box module.

3. The AI-enabled kit according to claim 1, wherein the separate AI-enabled box module further comprises: The processor is configured to execute artificial intelligence algorithms. The separate AI-enabled box module provides the self-checkout device with AI capabilities.

4. The AI-enabled kit according to claim 2 further comprises one of the following: Multiple self-checkout peripheral devices, including one of the following: barcode scanner, printer, card reader, and magnetic card reader; and Multiple artificial intelligence peripheral devices, including one of the image sensor and one of the motion sensor.

5. The AI-enabled kit of claim 4, wherein the checkout terminal computer program product is configured to include a filter driver and insert the filter driver between the main function driver and the bus driver of the communication interface included in the self-checkout device, so as to monitor and intercept the data stream through the communication interface included in the self-checkout device after the first self-checkout peripheral device included in the plurality of self-checkout peripheral devices is inserted into the self-checkout device, so as to obtain the raw data from the first self-checkout peripheral device.

6. The AI-enabled kit according to claim 4, wherein the enabling computer program product is configured to control the first microcontroller to simulate the peripheral device after the first self-checkout peripheral device included in the plurality of self-checkout peripheral devices is inserted, so as to simulate the first self-checkout peripheral device through the first microcontroller for the self-checkout device to perform corresponding hardware configuration, so as to obtain raw data from the first self-checkout peripheral device.

7. The AI-enabled kit of claim 6, wherein the peripheral device simulation includes an initialization program and a runtime program, the initialization program including one of the following: Obtain the device description information of the first self-checkout peripheral device; Based on the device description information, a hot-plug event is triggered in the separate AI-enabled box module. Send a notification to the self-checkout device to inform it of the hot-plug event; In response to the receipt of the notification, the self-checkout device is configured to request the device description information from the separate AI-enabled box module; as well as Based on the device description information, the self-checkout device is configured to trigger the device listing service.

8. The AI-enabled suite of claim 7, wherein the always-running program comprises one of the following: Receive the first packet data generated by the first self-service checkout peripheral device; The first packet data is transmitted to the first microcontroller via a software driver. The first microcontroller is controlled to transmit the first packet data to the plurality of self-service checkout devices using the corresponding device instructions and data format according to the device description information; The first microcontroller receives second packet data from the self-checkout device and sends it to the first self-checkout peripheral device. The second packet data is transmitted to the enabling computer program product via the first microcontroller. as well as The second packet data is transmitted to the first self-checkout peripheral device via the enabling terminal computer program product.

9. The AI-enabled kit according to claim 4, wherein the second microcontroller is configured in the data acquisition module with a circuit layout in which the main circuit between the communication interface included in the self-checkout device and the first self-checkout peripheral device included in the plurality of self-checkout peripheral devices is bypassed.

10. The AI-enabled kit of claim 9, wherein the acquisition-end computer program product is configured to include a protocol parser to control the second microcontroller to perform the bypass signal parsing after the data acquisition module is inserted into the first self-checkout peripheral device included in the plurality of self-checkout peripheral devices, in order to parse the data stream through the communication interface to obtain raw data about the first self-checkout peripheral device therefrom.

Images