Intelligent warehouse management system

Abstract

The invention belongs to the technical field of warehouse management, and particularly discloses a smart warehouse management system, which comprises a model creation module, a label making module, a label deployment module, a robot deployment module, a robot management module and an information verification module. The model creation module is used for collecting warehouse data and constructing a current warehouse model according to the warehouse data; the label making module is used for making coding rules, making barcodes corresponding to goods during warehousing through a barcode printer, reading label information and writing the label information to an RFID label EPC area; the tag deployment module is used for deploying RFID tags on articles and deploying RFID reader-writers, antennas and splitters on goods shelves; the robot deployment module is used for installing a reader-writer on a robot; the robot management module is used for setting different goods shelf areas, and each goods shelf area is managed by a designated intelligent management robot; and the information verification module is used for checking and confirming assets.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to a smart warehouse management system. Background Art

[0002] As my country's economy enters a new normal of improving quality and efficiency, the industrial structure is constantly being optimized and adjusted, and the logistics industry also needs to transform and upgrade from traditional logistics to a modern logistics system. As an important part of modern logistics, warehousing not only undertakes the task of storing goods, but also needs to better connect with the upstream and downstream of the supply chain to improve the efficiency of warehousing operations and the quality of services. Therefore, information-based, automated and intelligent smart warehousing has become a future development goal.

[0003] The total number of warehouses in my country is large, but the management level is relatively low, and advanced warehouse management models are rarely used; the utilization rate of warehouse equipment is low, and the degree of automation is not high enough; the degree of informatization is low, the use of warehouse management information systems is relatively backward, and the processing and use of warehouse information is not effective enough. The utilization rate and circulation of information in the entire logistics process are poor.

[0004] Amazon, Alibaba, and Dong are competing to use intelligent warehousing robots, intelligent warehousing systems, and intelligent warehouses, laying the foundation for the development of intelligent warehousing. With the advent of the era of intelligent warehousing, experts and scholars at home and abroad are studying from different perspectives how to use information recognition technology, automatic control technology, wireless sensor network technology, communication technology, embedded technology, intelligent control technology, etc. to improve the efficiency of warehousing operations, so as to realize the informatization, automation, and intelligence of warehousing. Against the background of the rapid development of emerging technologies such as big data, digital Internet of Things, mobile Internet, cloud computing, artificial intelligence, and digital twins, consumption patterns have also undergone substantial changes. The speed of logistics is increasing year by year, and the throughput of express delivery services has increased rapidly year-on-year. Faced with the increasing delivery volume of goods, warehouse explosion has become one of the biggest problems of warehousing logistics. Traditional warehousing management and operation methods can no longer meet current needs. Summary of the invention

[0005] The purpose of the present invention is to provide a smart warehouse management system to solve the problems raised in the above background technology.

[0006] To achieve the above-mentioned object, the present invention provides the following technical solutions: a smart warehouse management system, comprising a model creation module, a label formulation module, a label deployment module, a robot deployment module, a robot management module, and an information verification module;

[0007] The model creation module is used to collect warehouse data and build a current warehouse model based on the warehouse data;

[0008] The label formulation module is used to formulate coding rules, and through the barcode printer, prepare the barcode corresponding to the goods when entering the warehouse and read the label information and write it to the RFID label EPC area;

[0009] The tag deployment module is used to deploy RFID tags on all items, and the shelves are deployed with RFID readers, antennas, and splitters to ensure that the entire shelf is fully sensed and the location of each item can be obtained in real time;

[0010] The robot deployment module is used to install a reader on the robot. The robot has a navigation and positioning function to locate and count assets, and plans the robot's preset working range according to the transportation task and the current warehouse model;

[0011] The robot management module is used to set different shelf areas, and each shelf area is managed by a designated intelligent management robot, which can improve the working efficiency of the robot;

[0012] The information verification module is used for asset inventory confirmation.

[0013] As a preferred embodiment of the present invention, the warehouse data collected includes the geographical location and image database of the warehouse. The image database is fully covered by the video surveillance equipment of the preset gun camera and ball camera models in the warehouse. The video data is stored 7*24 hours a day and the panoramic free switching control is controlled; the warehouse geographical data includes: the area data, structural data and location division data of the warehouse; the image database also includes storage location data, the image database includes: shelf length / width / height, shelf quantity and robot coordinates; the warehouse data is collected to complete the warehouse 3D modeling through the digital twin algorithm to obtain the current warehouse model. The warehouse scene is truly restored, the cargo information is dynamically displayed, and the panoramic 3D warehouse visualization is realized by combining cameras, intelligent overhead cranes and other equipment.

[0014] As a preferred embodiment of the present invention, the specific steps of formulating the coding rules include: using a barcode printer to prepare the barcode corresponding to the goods when they are put into storage; reading the barcode or QR code information on the surface of the label through an RFID workstation, and automatically writing it to the EPC area of ​​the RFID label to complete the initialization of the label.

[0015] As a preferred embodiment of the present invention, in the tag deployment module, the shelf deployment of RFID readers, antennas, and splitters specifically includes: fixing a flat antenna on the top of each layer, deploying an 8-port antenna splitter on each layer, cascading all antennas to complete full coverage management of the entire shelf; real-time inventory can be performed; the open management rack is divided into areas, and identification indicator lights are deployed in each area; the current asset location is displayed in real time through a display and indicator lights.

[0016] As a preferred embodiment of the present invention, the intelligent inventory robot completes automatic inventory along the specified path, and the intelligent inventory robot automatically identifies the results and performs statistics in the material database, and can complete the inventory of all materials in one pass; in addition, the intelligent transport vehicle can identify the logistics terminal by itself, realize network operation and intelligent work through the Internet of Things recognition technology.

[0017] As a preferred embodiment of the present invention, the intelligent inventory robot performs material database statistics based on the identification results, including using data to dynamically manage and maintain the automatically identified material information, updating the material information in real time and obtaining the in-and-out information and inventory information through external connection, realizing automatic inventory verification and outputting inventory statistics, wherein the information includes category, location, and quantity information;

[0018] The material testing includes:

[0019] Image segmentation and spectral feature selection, including using Mean Shift algorithm to perform preliminary segmentation on video image data to obtain segmented grayscale images;

[0020] Spectral feature selection and threshold calculation, including selecting the difference BR component between the normalized R component and the B component in the color space as the precious metal detection feature, and calculating the optimal threshold of the precious metal detection feature according to the Otsu automatic threshold method;

[0021] Threshold segmentation, use the optimal threshold T_BR to binarize the video image data, and judge the pixels with BR<T_BR as metal area S as the precious metal binary image, and the metal area S is;

[0022] S={(i,j)|BR(i,j) <T_BR}

[0023] The pixels with BR≥T_BR are judged as the background area;

[0024] Among them, BR(i,j) is the characteristic value of pixel (i,j) BR, and T_BR is the optimal threshold of the metal detection feature BR component;

[0025] The Otsu automatic threshold method includes performing Otsu processing on the grayscale image to obtain the threshold T_MS, determining that the image pixels with grayscale values ​​greater than T_MS are added to the attribute histogram of the main detection feature BR, and using the optimal threshold T_B calculated based on Otsu for the BR component.

[0026] As a preferred embodiment of the present invention, the robot management module also includes a wireless sensor network. All intelligent robots share a wireless sensor network topology map. In order to avoid collision, a central scheduling server is required to design an algorithm. At the same time, all robots in the warehouse are monitored in real time by video surveillance equipment and mapped to the warehouse model for real-time monitoring.

[0027] As a preferred embodiment of the present invention, the central scheduling server needs to cooperate with the wireless sensor network. The central scheduling server controls all sensor nodes in the wireless sensor network. The sensor nodes of the central scheduling server include two types: video sensor nodes and vibration sensor nodes. A video sensor node and a vibration sensor node are combined into a task node. Through the collaboration between the video sensor node and the vibration sensor node, the video collaborative scheduling monitoring is completed and mapped to the warehouse model.

[0028] After the semi-activated vibration sensor detects an intruder, the central dispatch server performs the following steps:

[0029] If the intrusion target appears in the target monitoring area of ​​the wireless sensor network for the first time, the vibration sensor node with the largest amplitude A is selected as the central task node, and the adjacent nodes within the nearest one-hop range from the central task node are formed into a dynamic group Gi. The wireless sensor nodes in group Gi cooperate to monitor the intrusion target. At this time, all nodes in group Gi are set to a semi-activated state; in group Gi, when the amplitude A of the vibration sensor is greater than or equal to the set threshold, the video sensor node with the smallest sending energy consumption is selected from group Gi and set to an activated state, and the collected video is sent to the central scheduling server;

[0030] The calculation formula for sending energy consumption is:

[0031]

[0032] Where E Tx(k,d) represents the energy consumed by transmitting k bits of data at a distance of d; E emit(k) represents the energy consumed by the transmitting module for transmitting k bits of data; E emit represents the energy consumed by the transmitting module for transmitting 1 bit of data; E amp(k) represents the energy consumed by the power amplification for transmitting k bits of data; εfs represents the acceptable bit error rate for free space transmission, and εmp represents the acceptable bit error rate for multi-path attenuation transmission; d represents the propagation distance, d 0 represents the distance critical value, and the size of d 0 is:

[0033] When the amplitude A of the central task node is less than the set amplitude threshold, a node that meets the two performance indicators of amplitude threshold and transmission energy consumption is reselected as the central task node to re-form a new dynamic group;

[0034] When the intrusion target leaves the group, the next central task node is reselected to form a new dynamic group;

[0035] In a dynamic group, a network coverage node scheduling algorithm is used to determine redundant nodes and put the redundant nodes into a dormant state, including the following steps:

[0036] Each node multicasts messages in the dynamic group to obtain information about neighboring nodes; the information includes the node ID, the node's current remaining energy Ej, and the node's coordinate position (xj, yj);

[0037] Each node updates its neighbor information list based on the received reply message;

[0038] The weight of each node is calculated according to the formula, and the node with the largest weight is first judged for redundant coverage; if the weights of the nodes are the same, the ID values ​​are compared, and the node with the larger ID is judged for redundant coverage first; where λ(j) represents the weight of node j, a, b are proportional factors and a+b=1, Ej is the current residual energy of node j, E0 is the initial energy of node j, N(j) is the set of active neighbor nodes within the perception range of node j, and N is the total number of nodes in the dynamic group;

[0039] If the node does not meet the redundant coverage condition, the original state remains unchanged;

[0040] If the node meets the redundant coverage conditions, it will multicast to neighboring nodes and enter the dormant state;

[0041] The redundant coverage condition is that the coverage ranges of two or more task nodes overlap.

[0042] As a preferred embodiment of the present invention, the method for asset inventory confirmation is: the administrator holds an RFID handheld terminal, uploads the inventory information, and the system automatically verifies the information. The inventory results are uploaded to the computer through the wireless sensor network for data collation and statistical storage. The wireless sensor network is composed of a large number of micro sensor nodes deployed in the detection area. It is a self-organizing network system formed by wireless communication. Its purpose is to collaboratively sense, collect and process the information of the sensed objects in the network coverage area, and send it to the observer.

[0043] As a preferred embodiment of the present invention, the information verification module also includes an identity recognition module, which is used to register and manage personnel entering and leaving the management area, and set an ETC smart channel door. The smart channel door can only identify the tags of staff and assets. When unauthorized personnel and assets enter and leave the warehouse, the smart channel door will automatically alarm.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: the intelligent warehouse management system based on the Internet of Things technology of the present invention applies the Internet of Things technology to the warehouse management system, improves the recognition rate of materials in and out of the warehouse, and simultaneously identifies and manages multiple materials, thereby improving the efficiency of entering and leaving the warehouse; shortens the inventory cycle, improves the real-time nature of data, and dynamically grasps the inventory status in real time, thereby realizing the visual management of inventory materials; effectively improves the efficiency and accuracy of the picking and distribution process, speeds up the delivery, and reduces labor costs; realizes the standardized management, refined management, and full life cycle management goals of materials, and can better adapt to the urgent needs of warehouse management under the modern logistics management model. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0046] 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.

[0047] In the description of the present invention, it should be noted that the terms "vertical", "up", "down", "horizontal", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0048] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] See also Figure 1 ,The present invention provides a technical solution: a smart warehouse management system, including a model creation module, a label formulation module, a label deployment module, a robot deployment module, a robot management module, and an information verification module;

[0050] The model creation module is used to collect warehouse data and build the current warehouse model based on the warehouse data;

[0051] The label making module is used to make coding rules. Through the barcode printer, the corresponding barcode of the goods is prepared when entering the warehouse and the label information is read and written into the EPC area of ​​the RFID label;

[0052] The tag deployment module is used to deploy RFID tags on all items, and the shelves are equipped with RFID readers, antennas, and splitters to ensure that the entire shelf is fully sensed and the location of each item can be obtained in real time;

[0053] The robot deployment module is used to install readers on the robots. The robots have navigation and positioning functions to locate and count assets, and plan the preset working range of the robots according to the transportation tasks and the current warehouse model.

[0054] The robot management module is used to set up different shelf areas. Each shelf area is managed by a designated intelligent management robot, which can improve the robot's work efficiency;

[0055] The information verification module is used for asset inventory confirmation.

[0056] As a further feature of the present invention, the warehouse data collected includes the geographical location and image database of the warehouse. The image database is fully covered by the video surveillance equipment of the preset gun and ball models in the warehouse. The video data is stored 24 hours a day and the panoramic free switching control is carried out. The warehouse geographical data includes: the area data, structural data and location division data of the warehouse; the image database also includes storage location data, including: shelf length / width / height, number of shelves and robot coordinates; the warehouse data is collected to complete the warehouse 3D modeling through the digital twin algorithm to obtain the current warehouse model. The warehouse scene is truly restored, the cargo information is dynamically displayed, and the panoramic 3D warehouse visualization is realized by combining cameras, intelligent overhead cranes and other equipment.

[0057] As a further embodiment of the present invention, the specific steps of formulating the coding rules include: using a barcode printer to prepare the barcode corresponding to the goods when they are put into storage; reading the barcode or QR code information on the surface of the label through an RFID workstation, and automatically writing it to the EPC area of ​​the RFID label to complete the initialization of the label.

[0058] As a further feature of the present invention, in the tag deployment module, the shelf deployment of RFID readers, antennas, and splitters specifically includes: fixing a flat antenna on the top of each layer, deploying an 8-port antenna splitter on each layer, cascading all antennas to complete full coverage management of the entire shelf; real-time inventory can be performed; the open management rack is divided into areas, and identification indicator lights are deployed in each area; the current asset location is displayed in real time through a display and indicator lights.

[0059] As a further feature of the present invention, the robots include intelligent inventory robots and intelligent transport robots. The working range of RFID can be set: 1 meter to 3.5 meters. Tags within this area are activated and data is reported. The intelligent inventory robot automatically identifies the results and performs statistics on the material database. All material inventories can be completed in one pass. In addition, the intelligent transport vehicle can identify the logistics terminal by itself, realize network operation and intelligent work through the Internet of Things identification technology.

[0060] The intelligent inventory robot performs material database statistics based on the identification results, including using data to dynamically manage and maintain the automatically identified material information, updating the material information in real time, and obtaining the in-and-out information and inventory information through external connection, so as to realize automatic inventory verification and output the inventory statistics results, and the information includes category, location, and quantity information;

[0061] The material testing includes:

[0062] Image segmentation and spectral feature selection, including using the Mean Shift algorithm to perform preliminary segmentation on video image data to obtain segmented grayscale images;

[0063] Spectral feature selection and threshold calculation, including selecting the difference BR component between the normalized R component and the B component in the color space as the precious metal detection feature, and calculating the optimal threshold of the precious metal detection feature according to the Otsu automatic threshold method;

[0064] Threshold segmentation, use the optimal threshold T_BR to binarize the video image data, and judge the pixels with BR<T_BR as metal area S as the precious metal binary image, and the metal area S is;

[0065] S={(i,j)|BR(i,j) <T_BR}

[0066] The pixels with BR≥T_BR are judged as the background area;

[0067] Among them, BR(i,j) is the characteristic value of pixel (i,j) BR, and T_BR is the optimal threshold of the metal detection feature BR component;

[0068] The Otsu automatic threshold method includes performing Otsu processing on the grayscale image to obtain the threshold T_MS, determining that the image pixels with grayscale values ​​greater than T_MS are added to the attribute histogram of the main detection feature BR, and using the optimal threshold T_B calculated based on Otsu to calculate the BR component.

[0069] As a further aspect of the present invention, the robot management module also includes a wireless sensor network, and all intelligent robots share a wireless sensor network topology map. In order to avoid collisions, a central scheduling server is required to design an algorithm. At the same time, all robots in the warehouse are monitored in real time by video surveillance equipment and mapped to the warehouse model for real-time monitoring;

[0070] The central dispatch server cooperates with the wireless sensor network. The central dispatch server controls all sensor nodes in the wireless sensor network. The sensor nodes of the central dispatch server include two types: video sensor nodes and vibration sensor nodes. A video sensor node and a vibration sensor node are combined into a task node. Through the collaboration between the video sensor node and the vibration sensor node, the video collaborative dispatch monitoring is completed and mapped to the warehouse model.

[0071] After the semi-activated vibration sensor detects an intruder, the central dispatch server performs the following steps:

[0072] If the intrusion target appears in the target monitoring area of ​​the wireless sensor network for the first time, the vibration sensor node with the largest amplitude A is selected as the central task node, and the adjacent nodes within the nearest one-hop range from the central task node are formed into a dynamic group Gi. The wireless sensor nodes in group Gi cooperate to monitor the intrusion target. At this time, all nodes in group Gi are set to a semi-activated state; in group Gi, when the amplitude A of the vibration sensor is greater than or equal to the set threshold, the video sensor node with the smallest sending energy consumption is selected from group Gi and set to an activated state, and the collected video is sent to the central scheduling server;

[0073] The calculation formula for sending energy consumption is:

[0074]

[0075] Where E Tx(k,d) represents the energy consumed by transmitting k bits of data at a distance of d; E emit(k) represents the energy consumed by the transmitting module for transmitting k bits of data; E emit represents the energy consumed by the transmitting module for transmitting 1 bit of data; E amp(k) represents the energy consumed by the power amplification for transmitting k bits of data; εfs represents the acceptable bit error rate for free space transmission, and εmp represents the acceptable bit error rate for multi-path attenuation transmission; d represents the propagation distance, d 0 represents the distance critical value, and the size of d 0 is:

[0076] When the amplitude A of the central task node is less than the set amplitude threshold, a node that meets the two performance indicators of amplitude threshold and transmission energy consumption is reselected as the central task node to re-form a new dynamic group;

[0077] When the intrusion target leaves the group, the next central task node is reselected to form a new dynamic group;

[0078] In a dynamic group, a network coverage node scheduling algorithm is used to determine redundant nodes and put the redundant nodes into a dormant state, including the following steps:

[0079] Each node multicasts messages in the dynamic group to obtain information about neighboring nodes; the information includes the node ID, the node's current remaining energy Ej, and the node's coordinate position (xj, yj);

[0080] Each node updates its neighbor information list based on the received reply message;

[0081] The weight of each node is calculated according to the formula, and the node with the largest weight is first judged for redundant coverage; if the weights of the nodes are the same, the ID values ​​are compared, and the node with the larger ID is judged for redundant coverage first; where λ(j) represents the weight of node j, a, b are proportional factors and a+b=1, Ej is the current residual energy of node j, E0 is the initial energy of node j, N(j) is the set of active neighbor nodes within the perception range of node j, and N is the total number of nodes in the dynamic group;

[0082] If the node does not meet the redundant coverage condition, the original state remains unchanged;

[0083] If the node meets the redundant coverage conditions, it will multicast to neighboring nodes and enter the dormant state;

[0084] The redundant coverage condition is that the coverage ranges of two or more task nodes overlap.

[0085] As a further step of the present invention, the method for asset inventory confirmation is as follows: the administrator holds an RFID handheld terminal, and the inventory information is uploaded, the system automatically verifies the information, and the inventory results are uploaded to the computer through the wireless sensor network for data collation and statistical storage. The wireless sensor network is composed of a large number of micro sensor nodes deployed in the detection area, and a multi-line self-organizing network system is formed by wireless communication. Its purpose is to collaboratively sense, collect and process the information of the sensed objects in the network coverage area, and send it to the observer.

[0086] As a further feature of the present invention, the information verification module also includes an identity recognition module, which is used to register and manage personnel entering and leaving the management area, and to set up an ETC smart channel door, which can only identify the tags of staff and assets. When unauthorized personnel and assets enter and leave the warehouse, the smart channel door will automatically alarm.

[0087] In summary, the entire smart logistics solution will be based on the following intelligent hardware devices to achieve automatic positioning, automatic inventory, automatic handling, automatic warning and automatic management of the warehouse: Collect warehouse data through the model creation module and build the current warehouse model based on the warehouse data. Formulate coding rules through the label formulation module, and prepare the corresponding barcodes of the goods when entering the warehouse through the barcode printer. Read the barcode or QR code information on the surface of the label through the RFID workstation, and automatically write it to the RFID label EPC area to complete the label initialization. Through the label deployment module, all items are deployed with RFID tags, and the smart shelf deploys RFID readers, antennas, splitters, etc. to ensure full perception of the entire shelf and obtain the location of each item in real time. Fix the flat antenna on the top of each layer, deploy 8-port antenna splitters on each layer, cascade all antennas, and complete the full coverage management of the entire shelf. Real-time inventory can be performed. Divide the open management shelf into areas, and deploy identification indicators in each area. Use a display and indicator light to display the current asset location in real time. Through the robot deployment module, install a reader on the robot, because the robot has navigation and positioning functions, it can complete the positioning and inventory of assets. The working range of RFID can be set from 1 meter to 3.5 meters. The tags in this area are activated and report data. The inventory robot completes the automatic inventory according to the specified path, and can complete the inventory of all materials in one pass. In addition, the intelligent transporter can identify the logistics terminal by itself, realize network operation and intelligent work through the Internet of Things identification technology. Different shelf areas are set through the robot management module, and each shelf area is managed by a designated intelligent management robot, which can improve the work efficiency of the robot. All intelligent robots share a network topology map. In order to avoid collision, the central scheduling server needs to design an algorithm. The information verification module is used for asset inventory confirmation. During inventory, the administrator holds the RFID handheld terminal, and the inventory information is uploaded. The system will automatically verify the information, and the inventory results are uploaded to the computer for data collation and statistics to save the results. Set up an ETC intelligent channel door, which can only identify the tags of staff and assets. When unauthorized personnel and assets enter and leave the warehouse, the intelligent channel door will automatically alarm.

[0088] It is worth noting that the entire device is controlled by a main control button. Since the device matched with the control button is a common device and belongs to the existing mature technology, its electrical connection relationship and specific circuit structure will not be described in detail here.

[0089] 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 smart warehouse management system, characterized by: It includes model creation module, label formulation module, label deployment module, robot deployment module, robot management module and information verification module; The model creation module is used to collect warehouse data and build a current warehouse model based on the warehouse data; The label formulation module is used to formulate coding rules, and through the barcode printer, prepare the barcode corresponding to the goods when entering the warehouse and read the label information and write it to the RFID label EPC area; The tag deployment module is used to deploy RFID tags on all items, and the shelves are deployed with RFID readers, antennas, and splitters to ensure that the entire shelf is fully sensed and the location of each item can be obtained in real time; The robot deployment module is used to install a reader on the robot. The robot has a navigation and positioning function to locate and count assets, and plans the robot's preset working range according to the transportation task and the current warehouse model; The robot management module is used to set different shelf areas, and each shelf area is managed by a designated intelligent management robot; The information verification module is used for asset inventory confirmation.

2. The intelligent warehouse management system according to claim 1, characterized in that: The warehouse data collected includes the geographical location and image database of the warehouse. The image database is fully covered by the video surveillance equipment of the preset gun camera and ball camera models in the warehouse. The video data is stored 24 hours a day and the panoramic free switching control is available. Warehouse geographic data includes: warehouse area data, structural data and location division data; the image database also includes storage location data, the image database includes: shelf length / width / height, shelf quantity and robot coordinates; Collect warehouse data and complete warehouse 3D modeling through digital twin algorithms to obtain the current warehouse model. Restore the warehouse scene realistically, dynamically display cargo information, and combine cameras, smart overhead cranes and other equipment to achieve panoramic 3D warehouse visualization.

3. The intelligent warehouse management system according to claim 1, characterized in that: The specific steps of formulating the coding rules include: using a barcode printer to prepare the barcode corresponding to the goods when entering the warehouse; reading the barcode or QR code information on the surface of the label through an RFID workstation, automatically writing it to the EPC area of ​​the RFID label, and completing the initialization of the label.

4. The intelligent warehouse management system according to claim 1, characterized in that: In the tag deployment module, the shelf deployment of RFID readers, antennas, and splitters specifically includes: A flat antenna is fixed on the top of each layer, and an 8-port antenna splitter is deployed on each layer. All antennas are cascaded to complete full coverage management of the entire rack; real-time inventory can be performed; the open management rack is divided into areas, and identification indicators are deployed in each area; the current asset location is displayed in real time through a display and indicator light.

5. The intelligent warehouse management system according to claim 1, characterized in that: The robots include intelligent inventory robots and intelligent handling robots, and the RFID working range can be set to: 1 meter to 3.5 meters, and the tags in this area will be activated and report data; The intelligent inventory robot completes automatic inventory along the designated path, and automatically identifies the results for material database statistics. It can complete the inventory of all materials in one pass. In addition, the intelligent transport vehicle can identify the logistics terminal by itself and realize network operation through the Internet of Things recognition technology.

6. The intelligent warehouse management system according to claim 5, characterized in that: The intelligent inventory robot performs material database statistics based on the identification results, including using data to dynamically manage and maintain the automatically identified material information, updating the material information in real time, and obtaining the in-and-out information and inventory information through external connection, so as to realize automatic inventory verification and output the inventory statistics results, and the information includes category, location, and quantity information; The material testing includes: Image segmentation and spectral feature selection, including using Mean Shift algorithm to perform preliminary segmentation on video image data to obtain segmented grayscale images; Spectral feature selection and threshold calculation, including selecting the difference BR component between the normalized R component and the B component in the color space as the precious metal detection feature, and calculating the optimal threshold of the precious metal detection feature according to the Otsu automatic threshold method; Threshold segmentation, use the optimal threshold T_BR to binarize the video image data, and judge the pixels with BR<T_BR as metal area S as the precious metal binary image, and the metal area S is; S={(i,j)|BR(i,j) <T_BR} The pixels with BR≥T_BR are judged as the background area; Among them, BR(i,j) is the characteristic value of pixel (i,j) BR, and T_BR is the optimal threshold of the metal detection feature BR component; The Otsu automatic threshold method includes performing Otsu processing on the grayscale image to obtain the threshold T_MS, determining that the image pixels with grayscale values ​​greater than T_MS are added to the attribute histogram of the main detection feature BR, and using the optimal threshold T_B calculated based on Otsu for the BR component.

7. The intelligent warehouse management system according to claim 1, characterized in that: The robot management module also includes a wireless sensor network. All intelligent robots share a wireless sensor network topology. In order to avoid collisions, a central scheduling server is required to design an algorithm. At the same time, all robots in the warehouse are monitored in real time by video surveillance equipment and mapped to the warehouse model; The central dispatch server needs to cooperate with the wireless sensor network. The central dispatch server controls all sensor nodes in the wireless sensor network. The sensor nodes of the central dispatch server include video sensor nodes and vibration sensor nodes. A video sensor node and a vibration sensor node are combined into a task node. Through the collaboration between the video sensor node and the vibration sensor node, the video collaborative dispatch monitoring is completed and mapped to the warehouse model. After the semi-activated vibration sensor detects an intruder, the central dispatch server performs the following steps: If the intrusion target appears in the target monitoring area of ​​the wireless sensor network for the first time, the vibration sensor node with the largest amplitude A is selected as the central task node, and the adjacent nodes within the nearest one-hop range from the central task node are formed into a dynamic group Gi. The wireless sensor nodes in group Gi cooperate to monitor the intrusion target. At this time, all nodes in group Gi are set to a semi-activated state; in group Gi, when the amplitude A of the vibration sensor is greater than or equal to the set threshold, the video sensor node with the smallest sending energy consumption is selected from group Gi and set to an activated state, and the collected video is sent to the central scheduling server; The calculation formula for sending energy consumption is: Where E Tx(k,d) represents the energy consumed by transmitting k bits of data at a distance of d; E emit(k) represents the energy consumed by the transmitting module for transmitting k bits of data; E emit represents the energy consumed by the transmitting module for transmitting 1 bit of data; E amp(k) represents the energy consumed by the power amplification for transmitting k bits of data; εfs represents the acceptable bit error rate for free space transmission, and εmp represents the acceptable bit error rate for multi-path attenuation transmission; d represents the propagation distance, d 0 represents the distance critical value, and the size of d 0 is: When the amplitude A of the central task node is less than the set amplitude threshold, a node that meets the two performance indicators of amplitude threshold and transmission energy consumption is reselected as the central task node to re-form a new dynamic group; When the intrusion target leaves the group, the next central task node is reselected to form a new dynamic group; In a dynamic group, a network coverage node scheduling algorithm is used to determine redundant nodes and put the redundant nodes into a dormant state, including the following steps: Each node multicasts messages in the dynamic group to obtain information about neighboring nodes; the information includes the node ID, the current remaining energy Ej of the node, and the node coordinate position (xj, yj); Each node updates its neighbor information list based on the received reply message; The weight of each node is calculated according to the formula, and the node with the largest weight is first judged for redundant coverage; if the weights of the nodes are the same, the ID values ​​are compared, and the node with the larger ID is judged for redundant coverage first; where λ(j) represents the weight of node j, a, b are proportional factors and a+b=1, Ej is the current residual energy of node j, E0 is the initial energy of node j, N(j) is the set of active neighbor nodes within the perception range of node j, and N is the total number of nodes in the dynamic group; If the node does not meet the redundant coverage condition, the original state remains unchanged; If the node meets the redundant coverage conditions, it will multicast to neighboring nodes and enter the dormant state; The redundant coverage condition is that the coverage ranges of two or more task nodes overlap.

8. The intelligent warehouse management system according to claim 1, characterized in that: The method for asset inventory confirmation is as follows: the administrator holds an RFID handheld terminal, uploads inventory information, and the system automatically verifies the information. The inventory results are uploaded to the computer through the wireless sensor network for data collation and statistical storage; A wireless sensor network is composed of a large number of micro sensor nodes deployed in the detection area. It is a self-organizing network system formed by wireless communication. Its purpose is to collaboratively sense, collect and process information of perceived objects in the network coverage area.

9. The intelligent warehouse management system according to claim 1, characterized in that: The information verification module also includes an identity recognition module, which is used to register and manage personnel entering and leaving the management area, and set up an ETC smart channel door, which can only identify the tags of staff and assets. When unauthorized personnel and assets enter and leave the warehouse, the smart channel door will automatically alarm.

Images