Intelligent read-write terminal of unmanned storehouse

Abstract

The invention relates to the technical field of intelligent warehousing, in particular to an intelligent read-write terminal of an unmanned warehouse, which comprises a main control module, a face acquisition module, an Internet of Things sensing interface group, a sound-light alarm module, a voice interaction module, an RFID expansion system, a touch display module, a data storage unit and a wall-mounted mounting structure, four externally-connected circularly-polarized antennas are distributed in a T shape through an RFID expansion system, multi-label anti-collision reading is achieved through phase control, a millimeter wave radar, a pyroelectric infrared sensor and an electromagnetic electronic lock are integrated through a security linkage mechanism, a VESA standard support and a quick-disassembly buckle are supported by a modular wall-mounted structure, and the safety and reliability of the system are improved. According to edge calculation optimization, face recognition and offline voice instruction processing are accelerated through a main control module NPU. According to the terminal, the problems of large RFID blind area, slow security response and rigid deployment in the prior art are solved, and the warehouse management efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent warehousing, in particular to an intelligent read-write terminal of an unmanned warehouse. BACKGROUND

[0002] The intelligent read-write terminal of the unmanned warehouse is an automatic device applied to an unattended warehouse, which realizes automatic identification, data collection, storage management and information interaction of inventory items by integrating Internet of Things (IoT), RFID (Radio Frequency Identification), barcode scanning, computer vision or AI technology. The intelligent read-write terminal of the unmanned warehouse reads the information of goods quickly through RFID, a two-dimensional code or a sensor without manual operation, and then links with a warehouse management system (WMS) to update the inventory state (such as warehouse in / out, position change), while supporting unmanned inventory, sorting or goods tracking, improving efficiency and reducing human errors, and part of the terminal has AI capability to predict inventory demand or optimize storage path.

[0003] However, through analysis of most existing intelligent read-write terminals of unmanned warehouses, it is found that there are still some problems: (1) the RFID reading blind area is large, and the miss reading rate is high in a metal environment; (2) the security system is separated from human-computer interaction, and the abnormal response delay is more than 5 seconds; (3) the fixed installation leads to poor deployment flexibility.

[0004] In view of the above problems, the present application designs an intelligent read-write terminal of an unmanned warehouse. SUMMARY

[0005] The present application aims to provide an intelligent read-write terminal of an unmanned warehouse to solve the problems in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical scheme: An intelligent read-write terminal of an unmanned warehouse, comprising a main control module, a face collection module, an Internet of Things sensor interface group, an audible and visual alarm module, a voice interaction module, an RFID extension system, a touch display module, a data storage unit and a wall-mounted installation structure; The face collection module includes an infrared living body detection unit and a dynamic tracking camera, and the collected data is transmitted to the main control module for permission verification through an encrypted channel; The Internet of Things sensor interface group integrates RS485 / CAN / GPIO interfaces, and can be extended to connect millimeter wave radar, pyroelectric infrared sensors and electromagnetic electronic locks to build a three-dimensional protection network; The audible and visual alarm module 104 is connected with the main control module 101, and the audible and visual alarm module includes a three-color LED lamp strip and a high-loudness buzzer. The red light flashes and voice alarm are triggered in an abnormal state, and the green breathing light is displayed in normal operation. The connection and response logic of the three-color LED light strip, the high-loudness buzzer and the main control module are as follows: The three-color LED light strip is connected with the main control module through a PWM signal, and then runs through a constant current drive control red / green / blue independent channel, and the high-loudness buzzer is connected with the main control module through a digital IO, and then performs sound pressure output through a DC-DC boost, so as to realize the operation mode of sound and light alarm; Among them, the constant current drive adopts 16 channels, each channel has a maximum of 150mA, the high-loudness buzzer power amplifier adopts a class-D digital power amplifier, and the conversion efficiency can reach 92%, the three-color LED light strip is configured with 144 WS2812B three-in-one LEDs, and an IP68 protection level is adopted.

[0007] The RFID extension system supports four external circularly polarized antennas, the external circularly polarized antennas adopt a magnetic suction type rotatable base, and multi-tag anti-collision reading is realized through phase control. Among them, the four external circularly polarized antennas are distributed in a T shape, the phase control sends a phase offset instruction through the main control module, and the feeding phase difference of each external circularly polarized antenna is 0° / 90° / 180° / 270°.

[0008] The voice interaction module integrates a noise reduction microphone array and a class-D power amplifier sound box, supports offline voice command recognition and TTS broadcast.

[0009] Compared with the prior art, the beneficial effects of the present application are that through the multi-antenna cooperative RFID system: The four circularly polarized antennas are distributed in a T shape, and the phase difference is 0° / 90° / 180° / 270°; The magnetic suction base supports 360° rotation and adapts to complex shelf layout; The phase control instruction eliminates electromagnetic interference, and the metal environment missing reading rate is reduced to 1.3%.

[0010] Three-level security linkage mechanism: first-level detection: millimeter wave radar (60GHz) scans 0.2-8m area with an accuracy of ±3cm; second-level verification: pyroelectric infrared sensor receives radar coordinates and scans human body heat radiation; Third-level response: when the verification fails, the electromagnetic electronic lock is locked within 0.2 seconds, and the sound and light alarm is triggered synchronously.

[0011] Modular rapid deployment: VESA support wall-mounted fixing, and the device backboard is connected with the support through a spring clasp. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 It is a schematic diagram of an intelligent read-write terminal of an unmanned warehouse.

[0013] Figure 2 It is a schematic diagram of a sound and light alarm module in an intelligent read-write terminal of an unmanned warehouse.

[0014] Figure 3 It is a schematic diagram of a noise reduction microphone array for sound preprocessing in an unmanned warehouse intelligent read-write terminal. DETAILED DESCRIPTION

[0015] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0016] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application. In addition, the terms "first", "second" and the like are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined with "first", "second" and the like can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0017] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood through specific circumstances.

[0018] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0019] Please refer to Figure 1 An unmanned warehouse intelligent read-write terminal, comprising a main control module 101, a face collection module 102, an Internet of Things sensing interface group 103, an audible and visual alarm module 104, a voice interaction module 105, an RFID expansion system 106, a touch display module 107, a data storage unit 108 and a wall-mounted installation structure 109; Among them, the face collection module 102 contains an infrared living body detection unit 1021 and a dynamic tracking camera 1022, and the collected data is transmitted to the main control module 101 for permission verification through an encrypted channel; The Internet of Things sensing interface group 103 integrates RS485 / CAN / GPIO interfaces, and can be extended to connect millimeter wave radar 1031, pyroelectric infrared sensors 1032, and electromagnetic electronic locks 1033 to build a three-dimensional protection network. The sound and light alarm module 104 is connected with the main control module 101, and the sound and light alarm module 104 includes a three-color LED lamp strip 1041 and a high-loudness buzzer 1042. The red light flashes and the voice alarm are triggered in the abnormal state, and the green breathing light is displayed in the normal operation. Referring to Figure 2 The connection and response logic of the three-color LED lamp strip 1041 and the high-loudness buzzer 1042 with the main control module 101 are as follows: The three-color LED lamp strip 1041 is connected with the main control module 101 through a PWM signal, and then runs through a constant-current drive control red / green / blue independent channel. The high-loudness buzzer 1042 is connected with the main control module 101 through a digital IO, and then performs sound pressure output through a DC-DC boost, so as to realize the operation mode of sound and light alarm. Among them, the constant-current drive adopts 16 channels, each channel has a maximum of 150 mA, the high-loudness buzzer 1042 power amplifier adopts a class-D digital power amplifier, and the conversion efficiency can reach 92%. The three-color LED lamp strip 1041 is configured with 144 WS2812B three-in-one LEDs, and has an IP68 protection level.

[0020] The RFID expansion system 106 supports four external circularly polarized antennas 1061. The external circularly polarized antennas 1061 adopt a magnetic type rotatable base, and realize multi-tag anti-collision reading through phase control. Among them, the four external circularly polarized antennas 1061 are distributed in a T shape, and the phase control sends a phase offset instruction through the main control module 101, and the feeding phase difference of each external circularly polarized antenna 1061 is 0° / 90° / 180° / 270°.

[0021] The voice interaction module 105 integrates a noise reduction microphone array 1051 and a class-D power amplifier sound box 1052, supports offline voice command recognition and TTS broadcast. The noise reduction microphone array 1051 pre-processes the sound signal, and then transmits the instruction signal to the class-D power amplifier sound box 1052 for sound generation through the main control module 101 for voice recognition. Among them, the noise reduction microphone array 1051 adopts a four-microphone linear array, and the MVDR algorithm suppresses ≥25 dB environmental noise. The class-D power amplifier sound box 1052 adopts a TI TPA3255 chip, with an efficiency of 92%, a THD of <0.03%, a wake-up word recognition rate of 98.7%, and a voice broadcast delay of <200 ms.

[0022] The touch display module 107 is a 10.1-inch IPS capacitive screen with a resolution of 1920*1200, and is used to dynamically display a material position heat map, a temperature and humidity curve and a gate access state. The material position heat map is realized by using an RFID positioning+Kriging interpolation algorithm, and is used for quickly positioning a high-frequency access area and optimizing a warehouse layout.

[0023] The temperature and humidity curve is realized by using a BME680 sensor+exponential smoothing prediction, and is used for early warning of environmental abnormalities and preventing material deterioration.

[0024] The gate access state is realized by using an electronic lock state machine model+real-time video stream, and is used for realizing an in-and-out warehouse behavior traceability. The RFID positioning+Kriging interpolation algorithm, the BME680 sensor+exponential smoothing prediction and the electronic lock state machine model+real-time video stream all belong to existing conventional technologies, and will not be described here.

[0025] The data storage unit 108 is equipped with a Linux system, uses LUKS disk encryption, stores logs, face features and an RFID tag database, and is used for preventing original data from being read after physical disassembly, so as to increase the data protection strength.

[0026] The wall-mounted installation structure 109 includes a VESA standard support 1091 and a quick dismounting buckle 1092, and is provided with a hidden cable management channel at the back; The VESA standard support 1091 is mounted on a wall-mounted plate by screws, and the equipment back plate is connected to the VESA standard support 1091 by using a spring clasp through the quick dismounting buckle 1092, so as to improve the installation and dismounting convenience of the equipment. The VESA standard support 1091 adopts a 200*200mm standard and can bear ≥15kg The quick dismounting buckle 1092 is made of 304 stainless steel and has a plug-in life of >10,000 times.

[0027] In the embodiment of the application, the main control module 101 adopts a RK3566 quad-core processor of Ruiyi Micro, and an NPU coprocessor is connected to accelerate face recognition. The infrared living body detection unit 1021 adopts a 940nm living body detection LED array, and the dynamic tracking camera 1022 adopts a 2 million pixel global shutter camera, wherein the hardware of the encryption channel adopts an ATECC608A encryption chip, a TLS 1.3 encryption tunnel is established, original pictures are desensitized in the dynamic tracking camera 1022, and then feature extraction is performed, and then the pictures are uploaded to the main control module 101 through the encryption channel. RS485 / CAN / GPIO interface built-in TVS tube + magnetic isolation circuit, millimeter wave radar 1031 adopts 60GHz, its detection distance is expanded to 0.2-8m, and the detection accuracy is reduced to ±3cm, when the millimeter wave radar 1031 detects that the moving target time is greater than 0.5 seconds, the pyroelectric infrared sensor 1032 can be awakened, the pyroelectric infrared sensor 1032 receives the radar coordinates, starts the 38*28 fan-shaped, scans the human body feature verification: the thermal radiation intensity is greater than 100uW / cm2, and then is linked with the electromagnetic electronic lock 1033, the electromagnetic electronic lock 1033 adopts GPIO control, when being in high level (3.3V) unlocking, low level locking, at the same time, a safety protocol is set, that is, when more than three times of verification fails, locking for five minutes, through the cooperation between the above elements, the response time of security can be fully improved, that is, from intrusion detection to locking ≤0.8 seconds, and the power consumption is reduced.

[0028] In an example of the present application, for desensitization processing, it adopts a real-time desensitization processing mode, that is, the camera built-in chip or algorithm directly processes at the hardware end, which specifically includes: Face area blur / pixelization: code specific areas after recognizing faces; Feature extraction and deletion of original image: only output irreversible face feature vector (such as 128-dimensional code), and the original image is not stored or transmitted; Background stripping: only keep the face area and eliminate environmental information (such as location, other pedestrians); For desensitization processing in the camera, the chip or algorithm involved belongs to the prior art, which will not be described in detail here, nor will it be limited.

[0029] In an example of the present application, referring to Figure 3 , the preprocessing includes beamforming, spectral subtraction noise reduction, and endpoint detection; beamforming uses the spatial position difference of the four microphone array to enhance the sound in the target direction through phase interference, which is used to enhance the sound in the target direction (i.e. the sound emitted by the user's mouth) and eliminate noise in other directions; Spectral subtraction noise reduction is used to eliminate steady-state noise (such as fan sound / current sound) and retain transient voice features; Endpoint detection uses a double-threshold method to decide, filter invalid audio segments, and reduce the computational load of the subsequent recognition module; the operation principle and algorithm of beamforming, spectral subtraction noise reduction, and endpoint detection can be realized according to the existing conventional technology, which will not be described in detail here, nor will it be limited.

[0030] The working principle of the present application is that all the driving elements, i.e. power elements, electrical devices and adapted power sources are connected by wires at the idle place of the device, and the electrical devices are sequentially connected in working order, and the detailed connection means are well-known technologies in the art, and the working principle and process are mainly introduced below, and the electrical control is not described, 1. Warehouse operation Identity verification: the operator approaches the terminal, the dynamic tracking camera 1022 captures the face, the NPU extracts the feature vector and compares it with the encrypted database; Permission through: the sound and light alarm module 104 lights green, the electromagnetic electronic lock 1033 is powered on and unlocked; Material scanning: the RFID antenna 1061 reads the cargo label, and the touch screen 107 updates the inventory heat map.

[0031] 2. Abnormal processing Intrusion detection: millimeter wave radar 1031 detects moving targets for 0.5 seconds → wakes up pyroelectric infrared sensor 1032; Pyroelectric sensor fails to verify human features → main control module 101 outputs low level to lock, triggers sound and light alarm (red light flashing + 110dB alarm).

[0032] Voice intervention: noise reduction microphone array 1051 captures instructions → beamforming enhances human voice → endpoint detection filters noise → recognizes instructions → alarm is lifted.

[0033] 3. Environmental monitoring The temperature and humidity sensor data is predicted by exponential smoothing → the touch display module 107 displays the trend curve; When the standard is exceeded, the sound and light alarm is triggered (blue light gradient + voice broadcast).

[0034] 4. Maintenance and upgrade Quick release buckle 1092 separates the device and the bracket, and replaces the module or upgrades the hardware; Linux system remote OTA update algorithm library.

[0035] It should be understood that in the present application, each rotating, sliding, engaging, belt driving and other moving parts are well lubricated and not easy to slip or wear, and each moving part is provided with a corresponding protective shell, but in the drawings of the present application, the connection state of each moving part is not shown, and it can be understood that each part in the present application is made of metal or plastic material with suitable strength in the art to ensure that its structural rigidity meets the actual demand.

[0036] The preferred embodiments of the present application have been described in detail, but the present application is not limited to the above-described embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present application.

Claims

1. An intelligent reading and writing terminal for unmanned warehouse, characterized in that: include: The main control module (101) uses a Rockchip RK3566 quad-core processor and is connected to an NPU coprocessor; A face acquisition module (102), comprising an infrared living body detection unit (1021) and a dynamic tracking camera (1022), is connected to the main control module (101) via an encrypted channel; An Internet of Things sensor interface group (103) integrates RS485 / CAN / GPIO interfaces, and is extended to connect a millimeter wave radar (1031), a pyroelectric infrared sensor (1032), and an electromagnetic electronic lock (1033) to form a three-dimensional protection network; An audible and visual alarm module (104), comprising a three-color LED light strip (1041) and a high-volume buzzer (1042), responds to abnormal status instructions; A voice interaction module (105) integrating a noise reduction microphone array (1051) and a Class D amplifier speaker (1052); An RFID expansion system (106) supports four external circularly polarized antennas (1061) arranged in a T-shape; A touch display module (107) dynamically displays a thermal map of material locations, temperature and humidity curves, and access control status; The data storage unit (108) is equipped with a Linux system and uses LUKS disk encryption; The wall mounting structure (109) includes a VESA standard bracket (1091) and a quick-release buckle (1092).

2. The intelligent reading and writing terminal for unmanned warehouse according to claim 1, characterized in that: The external circularly polarized antenna (1061) of the RFID extension system (106) is mounted via a magnetically rotatable base, and the feeding phase difference of each antenna is 0° / 90° / 180° / 270°. The main control module (101) realizes multi-tag anti-collision reading through phase offset instructions.

3. The intelligent reading and writing terminal for unmanned warehouse according to claim 1, characterized in that: The execution process of the three-dimensional protection network is as follows: The millimeter wave radar (1031) wakes up the pyroelectric infrared sensor (1032) after detecting the moving target for 0.5 seconds; The pyroelectric infrared sensor (1032) scans a sector-shaped area where the human body thermal radiation intensity is greater than 100 μW / cm²; When the verification fails, the main control module (101) outputs a low level to the electromagnetic electronic lock (1033) to lock it, and simultaneously triggers the sound and light alarm module (104).

4. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The three-color LED light strip (1041) of the sound and light alarm module (104) is controlled by a 16-channel constant current driver, with a maximum current of 150mA per channel; the high-volume buzzer (1042) is driven by a Class D digital amplifier, with a sound pressure output of 110dB.

5. The intelligent reading and writing terminal for unmanned warehouse according to claim 1, characterized in that: The noise reduction microphone array (1051) of the voice interaction module (105) adopts a four-microphone linear arrangement and suppresses ≥25dB ​​of environmental noise through the MVDR algorithm; the Class D power amplifier speaker (1052) supports offline voice command recognition, and the broadcast delay is less than 200ms.

6. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The encryption channel of the face acquisition module (102) uses the ATECC608A encryption chip to establish a TLS 1.3 tunnel, and the dynamic tracking camera (1022) performs real-time desensitization processing, including blurring the face area and deleting the original image after feature extraction.

7. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The VESA standard bracket (1091) of the wall-mounted mounting structure (109) has a specification of 200×200 mm and a load capacity of ≥15 kg; the quick-disassembly buckle (1092) is made of 304 stainless steel and has a plug-in life of >10,000 times.

8. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The NPU coprocessor of the main control module (101) accelerates the face recognition algorithm, and the infrared liveness detection unit (1021) adopts a 940nm liveness detection LED array.

9. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The data storage unit (108) uses LUKS disk encryption to ensure that the original data cannot be read after physical disassembly.

10. The intelligent reading and writing terminal for an unmanned warehouse according to claim 1, characterized in that: The material location heat map of the touch display module (107) is generated using RFID positioning and Kriging interpolation algorithm, and the temperature and humidity curve is realized by using BME680 sensor combined with exponential smoothing prediction.

Images