A main-vice-cabinet cascade multi-grid opening RFID archive self-service access system and method

Abstract

The application belongs to the technical field of intelligent file management equipment, and particularly relates to a main-vice-cabinet cascaded multi-compartment RFID file self-service storage and retrieval system and method, an identity authentication and interaction module, which is used for collecting user identity information and performing permission verification, receiving a file storage and retrieval request of a user and displaying operation guide information; a main-cabinet control unit, which is connected with the identity authentication and interaction module and used as a main control core of a cascade system, processes storage and retrieval business logic, issues control instructions to each vice cabinet and collects state data; the main-vice-cabinet cascaded multi-compartment RFID file self-service storage and retrieval system and method integrates all compartments of the main cabinet and the vice cabinets into a unified resource pool for virtualization management through a global compartment resource scheduling module, and realizes intelligent allocation of the compartments by combining a four-dimensional matching degree scoring algorithm of classification, space, convenience and load balancing, so as to not only ensure matching of the classification and space of the files and the compartments, but also effectively balance the utilization rate of each cabinet body, and the overall storage and retrieval efficiency is improved by more than 40%.

Description

Technical Field

[0001] This invention belongs to the technical field of intelligent archive management equipment, specifically relating to a self-service archive storage and retrieval system and method with cascaded main and auxiliary cabinets and multiple RFID slots. Background Technology

[0002] With the continuous advancement of information technology in archival management, RFID smart filing cabinets have gradually replaced traditional metal cabinets as core equipment for archival management departments at all levels. By affixing UHF RFID tags to the file boxes, and in conjunction with the internal reader array, smart filing cabinets can achieve functions such as automatic inventory, real-time on-site monitoring, and storage and retrieval operation recording, significantly improving the efficiency and accuracy of archival management. To meet the needs of scenarios with large file volumes, mainstream manufacturers generally adopt a cascaded expansion scheme of main and auxiliary cabinets. The main cabinet integrates a touch screen and industrial control host to handle control and interaction functions, while the auxiliary cabinets are only equipped with compartments and reader / writer modules. They receive unified scheduling from the main cabinet through cascaded communication. One system can support a one-main-multiple-auxiliary expansion architecture, and the file volume can increase linearly with the number of auxiliary cabinets.

[0003] The lack of unified scheduling for global storage compartments leads to low access efficiency: While existing cascaded filing cabinets with master and slave cabinets achieve physical expansion, compartment resource management remains fragmented, handled on a cabinet-by-cabinet basis. When storing files, the system often prioritizes allocating compartments to the master cabinet, only allocating to slave cabinets in turn once the master cabinet is full. This results in high-frequency use of the master cabinet while remote slave cabinets remain idle for extended periods, leading to severe uneven load distribution among cabinets. When borrowing multiple files, the target files may be scattered across different floors of various cabinets, requiring users to repeatedly search through multiple cabinets, resulting in a circuitous and lengthy operation path. Especially in large-scale cascaded scenarios with one master cabinet and eight slave cabinets, the average operation time for batch access increases by more than 60% compared to single-cabinet scenarios, resulting in a poor user experience and low efficiency.

[0004] Insufficient RFID cross-read suppression in cascaded multi-cabinet smart filing cabinets limits inventory accuracy: RFID cross-reading in multi-compartment smart filing cabinets has always been a challenge in the industry. In single-cabinet scenarios, cross-reading can be controlled within acceptable limits through methods such as electromagnetic shielding of compartments and adjustment of read / write power. However, in cascaded main and auxiliary cabinet scenarios, multiple cabinets are closely arranged, and the distance between corresponding compartments in adjacent cabinets is extremely short, making it easy for RFID signals to penetrate the cabinet side walls and cause cross-cabinet cross-reading. Most existing anti-cross-reading solutions are designed for single-cabinet internal use and do not consider the special characteristics of signal interference between cabinets in cascaded scenarios. During full-system inventory, if all compartments are working simultaneously, the cross-reading rate rises sharply, leading to unreliable data; if each cabinet is inventoried sequentially, the overall inventory time increases linearly with the number of auxiliary cabinets, resulting in low inventory efficiency.

[0005] The lack of process guidance in self-service file retrieval presents a high barrier to entry: Most existing smart filing cabinets only offer a simple "swipe card - open compartment - retrieve / place - close door" process. When users need to borrow or return multiple files located in different compartments or cabinets, the system only provides the cabinet and compartment numbers, leaving the user to manually find the correct order. For users unfamiliar with the cabinet layout, large cabinets with one main compartment and multiple secondary compartments can be overwhelming, requiring repeated searches and increasing the risk of misplacement or omission. This is especially problematic in strictly classified environments where file classification is crucial; misplacement can lead to serious consequences due to mismatched classification levels.

[0006] Insufficient reliability of cascaded communication, with single-point failures impacting the entire system: Existing cascaded main and auxiliary cabinets generally employ simple bus or chain-like communication structures, with auxiliary cabinets acting only as passive execution units, lacking data buffering and autonomous decision-making capabilities. When a communication failure occurs in one auxiliary cabinet within the cascaded link, all downstream auxiliary cabinets may lose connection, leading to partial functional paralysis of the entire cascaded system. Furthermore, existing solutions lack precise addressing capabilities at the grid level, resulting in overly coarse-grained commands that hinder flexible grid-level control and status feedback.

[0007] Therefore, there is an urgent need for a self-service retrieval system and method for cascaded multi-compartment RFID archives that can achieve unified scheduling of all compartments, cascaded anti-interference inventory, intelligent path guidance, and highly reliable cascaded communication. Therefore, a self-service retrieval system and method for cascaded multi-compartment RFID archives is proposed. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies by proposing a self-service retrieval system and method for cascaded main and auxiliary cabinets with multi-compartment RFID archives.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: a cascaded main and auxiliary cabinet multi-compartment RFID self-service archive storage and retrieval system, comprising:

[0010] The identity authentication and interaction module is used to collect user identity information and perform permission verification, receive user file access requests and display operation guidance information;

[0011] The main cabinet control unit, which is connected to the identity authentication and interaction module, serves as the main control core of the cascaded system, processes access business logic, issues control commands to each sub-cabinet, and summarizes status data.

[0012] The cascaded communication management module is connected to the main cabinet control unit and is used to maintain the communication links between the main cabinet and multiple auxiliary cabinets. It adopts a cascaded topology to realize instruction distribution and data feedback between the main cabinet and auxiliary cabinets, and between auxiliary cabinets.

[0013] The global grid resource scheduling module is connected to the main cabinet control unit and is used to include all grids of the main cabinet and all cascaded secondary cabinets into a unified resource pool for virtualized management. Based on the attribute information of the archive to be accessed, the user's permission level, the real-time status of each grid and the access path cost, the module allocates a target grid to each archive through a global optimal allocation algorithm.

[0014] The RFID read / write control and inventory module is connected to the main cabinet control unit and the cascaded communication management module. It is used to control the RFID readers in each compartment to perform file tag reading and writing operations, and to coordinate the inventory process of cascaded multi-cabinet through a time-slot scheduling algorithm, so as to achieve fast inventory of the whole system and no cross-reading between compartments and cabinets.

[0015] The access path optimization module, which is connected to the global grid resource scheduling module, is used to generate the optimal path and grid opening order for user access operations based on the cabinet number and location of the target grid, and provide step-by-step guidance to the user through the interaction module.

[0016] Multiple auxiliary cabinet control units are connected to the main cabinet control unit through the cascaded communication management module. They are used to receive control commands issued by the main cabinet, drive the electric locks of each compartment of the cabinet to perform opening and closing actions, and transmit the RFID inventory data of the cabinet back to the main cabinet.

[0017] Preferably, the specific process of the global grid resource scheduling module performing grid allocation is as follows:

[0018] Construct a global grid resource pool matrix G = [g11, g12, ..., gij, ..., gmn], where g_ij represents the status information of the j-th grid of the i-th cabinet, including occupancy status, remaining capacity, grid size, and security level attribute, m is the total number of cabinets, and n is the number of grids in a single cabinet;

[0019] For each file dp in the archive set D = [d1, d2, ..., dk], calculate its matching score S_pq with each available slot. The calculation formula is as follows:

[0020] Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·D_pq

[0021] Among them, Apq is the security level matching degree, which is 1 when the security level of the compartment is greater than or equal to the security level of the file, and 0 otherwise; Bpq is the space matching degree, which is calculated based on the ratio of the file box size to the remaining capacity of the compartment, and the value range is 0-1; Cpq is the access convenience degree, which is calculated based on the distance between the cabinet where the compartment is located and the user's operating position and the floor height, the closer the distance and the more suitable the height, the higher the score; Dpq is the load balancing degree, which is calculated based on the proportion of used compartments in the cabinet where the compartment is located, and priority is given to allocating to cabinets with lower usage rates; w1, w2, w3, and w4 are the weight coefficients of each item, and w1+w2+w3+w4=1, with w_1 having a maximum value of 0.4;

[0022] For each file, select the available cell with the highest matching score as the target cell, while ensuring that the files stored in the same batch meet the shortest path constraint in terms of cell distribution.

[0023] Preferably, the access path optimization module generates the optimal access path in the following specific way:

[0024] When a user submits an access request containing N files, obtain the target cell coordinates Pi = (cabineti, rowi, coli) for each file, where cabineti is the cabinet number, rowi is the row number, and col_i is the column number.

[0025] Starting with the main cabinet operation, construct the path weight matrix W between compartments, where the weight w_ij between any two compartments consists of two parts: the distance moved between cabinets and the distance moved within the cabinet.

[0026] wij = |cabineti - cabinetj|·L + √[(rowi - rowj) 2 + (coli - col_j) 2 ]·H

[0027] Where L is the unit movement cost between adjacent cabinets, and H is the unit movement cost between adjacent compartments within the cabinet, L > H;

[0028] A genetic algorithm is used to solve the optimal access order of N target grids, with the goal of minimizing the total path cost. The population size is 50, the number of iterations is 100, the crossover probability is 0.8, and the mutation probability is 0.1.

[0029] Based on the optimal access order obtained from the solution, operation guidance instructions for each step are generated sequentially to control the corresponding grids to open in sequence, and the current step and the next step prompts are displayed to the user through the identity authentication and interaction module.

[0030] Preferably, the time-slot scheduling algorithm of the RFID read / write control and inventory module specifically includes:

[0031] The entire cascaded system is divided into M time slot groups, each containing several spatially non-adjacent grids, such that the distance between any two grids in the same group is greater than the RFID signal serial reading threshold distance D_th.

[0032] The time slot allocation principle is as follows: adjacent compartments within the same cabinet are assigned to different time slot groups, and corresponding compartments in adjacent cabinets are assigned to different time slot groups;

[0033] During inventory, the RFID readers in each slot are started sequentially according to the time slot group number. The duration of each time slot is Tslot = Tread + Tswitch, where Tread is the tag reading time and T_switch is the reader switching time.

[0034] For the kth time slot group, the RFID readers in all slots of the group work simultaneously, and the reading and writing power is dynamically adjusted according to the slot position: the power of the slots near the edge of the cabinet is reduced by 10%-15%, and the middle slots use the rated power.

[0035] Once all time slot groups have completed an inventory check, the label data of each slot is summarized to form a complete system archive inventory list.

[0036] Preferably, the RFID read / write control and inventory module further includes a storage / access confirmation sub-process:

[0037] When a user performs an access operation and closes the grid, the RFID reader of that grid immediately initiates a quick inventory check, reads the EPC codes of all file tags in the grid, and obtains the current tag set S_current;

[0038] Compare Scurrent with the label set Sbefore before the operation, and calculate the new label set Sadd = Scurrent - Sbefore and the reduced label set Sremove = Sbefore - Scurrent;

[0039] Compare Sadd and Sremove with the list of files saved and retrieved by the user this time, respectively. If they match completely, the save and retrieve are confirmed to be successful.

[0040] If there is a mismatch, calculate the deviation Δ = |Sadd| - number of entries + |Sremove| - number of entries. When Δ ≠ 0, start an auxiliary scan of adjacent cells to check for possible cross-reads, and prompt the user to check on the interactive interface.

[0041] Preferably, the cascaded communication management module adopts a serial cascaded topology. The main cabinet is connected to the first-level sub-cabinet through the first Ethernet interface, the first-level sub-cabinet is connected to the second-level sub-cabinet through the second Ethernet interface, and so on to form a chain connection. Each sub-cabinet has a built-in cascaded control board with two network interfaces and data forwarding function. The cascade depth supports more than 1 main cabinet and 8 sub-cabinets.

[0042] The communication protocol of the cascaded communication management module adopts a custom binary protocol. The instruction frame format includes a frame header, address field, function code, data field, and check code. The high 8 bits of the address field are the cabinet number and the low 8 bits are the grid number, realizing precise addressing at the single grid level.

[0043] The cascaded communication management module also has a link fault detection and automatic reconnection mechanism. When the communication of a certain level sub-cabinet is interrupted, its downstream sub-cabinets automatically switch to the backup path or report the fault status, ensuring the partial availability of the cascaded system.

[0044] A method for self-service storage and retrieval of archives using RFID technology in a cascaded main and auxiliary cabinet system includes the following steps:

[0045] S1. Identity Authentication and Permission Verification: Users submit their identity credentials through the identity authentication module. The main cabinet control unit verifies the validity of the user's identity and the corresponding file operation permissions. After successful authentication, the user enters the storage and retrieval operation interface.

[0046] S2. Access Request Reception and Global Grid Search: When a user submits a file access request, the main cabinet control unit sends the request to the global grid resource scheduling module. This module traverses the status of all grids in the main cabinet and all cascaded secondary cabinets, and searches for the target grid that meets the conditions or locates the location of the file to be retrieved.

[0047] S3. Dynamic grid allocation and path planning: The global grid resource scheduling module allocates the optimal grid to the stored file based on a multi-dimensional matching degree scoring algorithm, and the access path optimization module calculates the optimal access order and guidance path according to the target grid distribution.

[0048] S4. Main and auxiliary cabinet linkage compartment opening: The main cabinet control unit sends an opening command to the cabinet where the target compartment is located through the cascaded communication management module. The corresponding auxiliary cabinet control unit drives the electric lock to open the target compartment. The compartment opening sequence is consistent with the path planning result.

[0049] S5. Real-time RFID identification and access confirmation: After the user completes the file retrieval and storage operation and closes the compartment, the RFID reader of the corresponding compartment immediately performs a quick inventory, reads all the tags in the current compartment and compares them with the status before the operation to confirm the consistency between the number of files stored and the file information.

[0050] S6. Time-slot cascaded inventory and status update: After the storage and retrieval operation is completed, the system starts a full system time-slot inventory, and sequentially starts the RFID readers of each compartment of each cascaded cabinet according to the time slot group to complete the synchronous update of the archive status of the entire system.

[0051] S7. Operation Completion and Data Reporting: Store the personnel information, time, file list, and grid change record of this access operation into the local database, and simultaneously upload them to the file management server, generate an operation record, and prompt the user that the operation is complete.

[0052] Preferably, in step S3, the multi-dimensional matching degree scoring algorithm for dynamic grid allocation is specifically as follows:

[0053] Construct a global grid resource pool matrix G = [g11, g12, ..., gij, ..., gmn], where gij represents the status information of the j-th grid of the i-th cabinet, including the occupancy status sij, the remaining capacity cij, the grid size szij, and the security level attribute cl_ij, m is the total number of cabinets, and n is the number of grids in a single cabinet;

[0054] For each file dp in the stored file set D = [d1, d2, ..., dk], extract its attributes: file size szp, file security level clp, and estimated access frequency f_p;

[0055] Calculate the four matching indices between the archive dp and the grid gij:

[0056] Security level matching degree Apq: Apq = 1 when clij ≥ clp, otherwise Apq = 0, and Apq is a veto item. When the value is 0, the cell is directly excluded.

[0057] Spatial matching degree Bpq = 1 - |szp - szij| / szij, when szp > szij B_pq = 0;

[0058] Access convenience Cpq = 1 - (α·|i-1| + β·|j - j0| / n), where |i-1| is the distance between the cabinet number and the main cabinet, j_0 is the optimal column corresponding to the user operation surface, α is the weight of the distance between cabinets, and β is the weight of the column position within the cabinet, both ranging from 0.2 to 0.4.

[0059] Load balancing degree Dpq = 1 - ui, where ui is the utilization rate of the compartments of the i-th cabinet, ui = number of occupied compartments / total number of compartments;

[0060] The formula for calculating the overall matching score is:

[0061] Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·D_pq

[0062] Where w1=0.4, w2=0.25, w3=0.2, w4=0.15;

[0063] For each file, select the available grid with the highest matching score. When multiple files are allocated at the same time, a greedy algorithm is used to allocate them sequentially, and the grid state matrix is ​​updated after each allocation.

[0064] Preferably, in step S6, the time-slot cascaded inventory method specifically includes:

[0065] Time slot pre-allocation stage: Based on the number of cabinets M and the number of slots N in each cabinet of the cascaded system, construct the time slot allocation matrix T = [tij]{M×N}, where t_ij is the time slot group number of the j-th slot in the i-th cabinet; the time slot allocation satisfies the following constraints: adjacent slots in the same cabinet whose row number or column number differs by 1 have a time slot group number difference ≥ 2; corresponding slots in the same row and column of adjacent cabinets have a time slot group number difference ≥ 3;

[0066] Power configuration phase: Configure the transmission power P_ij for each RFID reader / writer in each grid, and the calculation formula is as follows:

[0067] Pij = Pmax·(1 - γ·e_ij)

[0068] Where Pmax is the rated transmission power, eij is the edge coefficient, eij=0 when the slot is located in the inner layer of the cabinet and there are other slots on all sides, eij is 0.1-0.2 when the slot is located on the edge of the cabinet or near the cabinet door, and γ is the adjustment coefficient of 0.5.

[0069] Inventory execution phase: Inventory is performed sequentially according to the time slot group number from smallest to largest. Each time slot lasts for T_slot = 2s, including 1.5s reading time and 0.5s switching time. Within each time slot, the readers of all slots in the group simultaneously transmit signals to read tags at their respective configured power.

[0070] Data aggregation phase: After all time slots are inventoried, the list of tags read from each grid is aggregated to generate a system-wide archive in-place list. This list is then compared with the archive information recorded in the database, and archives that are not in place are marked and alarms are triggered.

[0071] Preferably, the specific process for access confirmation and exception handling in step S5 is as follows:

[0072] After the user closes the grid, the RFID reader of the grid starts a rapid inventory check, with a reading time of ≤2s, to obtain the current tag set S_curr;

[0073] Before the query operation, the tag set Sprev of this cell is retrieved, and the difference is calculated: the tag set Sout = Sprev - Scurr is retrieved and stored in the tag set Sin = Scurr - S_prev;

[0074] Compare Sout with the list of files requested by the user to be retrieved (Lout), and compare Sin with the list of files requested by the user to be stored (Lin).

[0075] If Sout = Lout and Sin = Lin, then the access is considered consistent, and the operation is confirmed as successful.

[0076] If |Sout| < |Lout|, it is determined that there is a missing file. The system will prompt "There are still files that have not been retrieved" on the interactive interface and reopen the grid.

[0077] If Sextra = Sin - L_in is not empty, it is determined to be an incorrect file, and the system prompts "An unauthorized file was detected" and asks the user to retrieve the extra file.

[0078] If the data is inconsistent after three consecutive checks, the system will retain the on-site data and notify the administrator to intervene.

[0079] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0080] This cascaded multi-compartment RFID self-service archive storage and retrieval system and method utilizes a global compartment resource scheduling module to integrate all compartments of the main and auxiliary cabinets into a unified resource pool for virtualized management. It employs a four-dimensional matching algorithm based on security level, space, convenience, and load balancing to achieve intelligent compartment allocation. This ensures both security level matching and space adaptation between archives and compartments, effectively balancing the utilization rate of each cabinet, resulting in an overall storage and retrieval efficiency improvement of over 40%. Furthermore, a time-slot scheduling algorithm coordinates RFID inventory checks across the cascaded multi-cabinet system, assigning spatially non-adjacent compartments to the same time slot group for simultaneous operation, while ensuring adjacent compartments... The grid slots operate in staggered time slots, and with dynamic power adjustment of edge grid slots, the overall system inventory time is reduced to about one-third of the traditional cabinet-by-cabinet inventory method while ensuring zero cross-reading. The access path optimization module uses a genetic algorithm to solve the optimal access order of multi-target grid slots, and with step-by-step progressive grid opening and graphical path guidance, the operation complexity of users in multi-cabinet and multi-grid scenarios is significantly reduced, and the average operation time of batch access is shortened by more than 30%. The data forwarding and fault detection mechanism of the sub-cabinet cascade control board improves the reliability of cascade communication and system availability. Attached Figure Description

[0081] Figure 1 This is a structural block diagram of the main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system of the present invention;

[0082] Figure 2 This is a flowchart of the self-service storage and retrieval method for cascaded main and auxiliary cabinets with multi-compartment RFID files according to the present invention. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0084] like Figure 1 As shown, this embodiment provides a cascaded multi-compartment RFID self-service archive storage system, including: an identity authentication and interaction module, a main cabinet control unit, a cascaded communication management module, a global compartment resource scheduling module, an RFID read / write control and inventory module, an access path optimization module, and multiple secondary cabinet control units. The identity authentication and interaction module is connected to the main cabinet control unit for user authentication and operation interaction; the main cabinet control unit, as the system's main control core, connects to the cascaded communication management module, the global compartment resource scheduling module, and the RFID read / write control and inventory module; the cascaded communication management module connects to each secondary cabinet control unit via cascaded links to achieve instruction distribution and data feedback between the main and secondary cabinets; the global compartment resource scheduling module connects to the access path optimization module to complete intelligent compartment allocation and path planning; the RFID read / write control and inventory module controls the RFID readers in each compartment via cascaded links. All modules work collaboratively to form a complete self-service storage and retrieval closed loop from user authentication, intelligent allocation, path guidance, access confirmation to system-wide inventory.

[0085] The identity authentication and interaction module is integrated into the front panel of the main cabinet, including a 22-inch capacitive touchscreen, a binocular face recognition module, an IC card reader, an ID card reader, and a voice broadcast unit. The touchscreen has a resolution of 1920×1080 and runs the user interface software, providing users with access to functions such as file retrieval, storage / retrieval / return operations, and inventory checks. The binocular face recognition module uses a 2-megapixel binocular camera, supports liveness detection, has an accuracy rate of ≥99.5%, and a recognition time of ≤1 second. The IC card reader supports the ISO 14443A protocol and can read employee work cards. The ID card reader supports second-generation ID card reading and is used in confidential scenarios requiring strong identity verification. The voice broadcast unit uses synthesized Chinese voice to provide voice prompts at key operational points. The system supports multiple authentication mode combinations: in normal mode, single-factor authentication with either work card or face is sufficient; in confidential mode, dual authentication with both face and work card is required. After successful authentication, the system loads the range of file security levels and operable access permissions for that user from the permissions database based on their role, providing access services only within the authorized scope. By combining multi-factor authentication with fine-grained access control, the system ensures that only authorized personnel can access files of the corresponding security level, strengthening file security at the entry point and meeting the security management needs of different security levels.

[0086] The main control unit uses an x86 architecture industrial motherboard, equipped with an Intel i5 processor, 8GB of RAM, and a 128GB SSD, running Windows 10 or a domestic operating system. The main control unit is the core of the entire cascaded system, responsible for business logic processing, database management, human-machine interaction scheduling, and coordinated control of various functional modules. The local database stores core data such as file information tables, grid status tables, user permission tables, and operation log tables. The main control unit connects to the cascade communication management module via an Ethernet interface, to authentication peripherals via a USB interface, and to a touchscreen display via an HDMI interface. Centralized processing of business logic and data management through a high-performance industrial control platform ensures the cascaded system's response speed and data processing capabilities in multi-task concurrent scenarios, providing computational support for global scheduling and real-time interaction.

[0087] The cascaded communication management module consists of the main communication interface on the main cabinet side and the cascaded control boards built into each auxiliary cabinet, adopting a serial cascaded topology. The main cabinet connects to the uplink port of the first-level auxiliary cabinet via a 100Mbps Ethernet interface, the downlink port of the first-level auxiliary cabinet connects to the uplink port of the second-level auxiliary cabinet, and so on, forming a chain connection. Each cascaded control board uses an ARM Cortex-M4 core microcontroller with a main frequency of 168MHz, equipped with two 100Mbps Ethernet interfaces (RMII interfaces), supporting data forwarding. The cascade depth supports more than 1 main cabinet and 8 auxiliary cabinets, and the maximum number of slots in a single system can reach 88 (8 slots in the main cabinet + 10 slots in each of the 8 auxiliary cabinets). The communication protocol adopts a custom binary protocol, and the instruction frame format is: frame header (2 bytes, 0xAA0x55) + address field (2 bytes, high 8 bits cabinet number, low 8 bits slot number) + function code (1 byte) + data length (1 byte) + data field (N bytes) + checksum (2 bytes, CRC16). The address domain design enables the system to precisely address individual cells, achieving fine-grained control at the cell level. The cascading communication management module also features a link failure detection mechanism: a heartbeat frame is sent every 5 seconds; if no response is received after three consecutive heartbeats, the link is considered faulty. The main cabinet immediately marks the corresponding auxiliary cabinet as offline and displays an alarm on the interface, while ensuring that non-faulty auxiliary cabinets continue to operate normally. Through the chain-cascading topology and cell-level precise addressing communication protocol, large-scale cabinet expansion can be achieved with minimal wiring, while maintaining independent control over each cell, balancing deployment convenience and control flexibility.

[0088] The global grid resource scheduling module is one of the core modules of this invention. Running on the main cabinet control unit, it is responsible for incorporating all grids of the main cabinet and all cascaded auxiliary cabinets into a unified resource pool for virtual management, and allocating the optimal grid to the archives using a multi-dimensional matching degree scoring algorithm. The specific working process is as follows: First, a global grid resource pool matrix G = [g11,g12, ..., gij, ..., gmn] is constructed, where gij represents the status information structure of the j-th grid in the i-th cabinet, containing the following fields: occupancy status sij (0=idle, 1=occupied, 2=faulty), remaining capacity cij (unit: mm, total thickness of the archive boxes that can be accommodated), grid size szij (width × height × depth), and security classification clij (0=non-confidential, 1=internal, 2=secret, 3=confidential). m is the total number of cabinets (1 main cabinet + k auxiliary cabinets), and n is the number of grids in a single cabinet. For a set of files to be stored, D = [d1, d2,..., dk], each file dp has the following attributes: file size szp (thickness), file security classification clp, file type typep, and estimated access frequency fp. For each file, the system calculates a matching score Spq across all available slots using the following formula:

[0089] Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·D_pq

[0090] Among them, Apq is the security level matching degree, a decisive indicator: when clij ≥ clp, Apq = 1; otherwise, Apq = 0 and the compartment is directly excluded to ensure that low-security compartments are never used to store high-security files. Bpq is the spatial matching degree, Bpq = 1 - |szp - cij| / szij; when szp > cij, Bpq = 0, indicating insufficient compartment capacity. Cpq is the access convenience degree, Cpq = 1 - (α·dcab + β·drow + γ·dcol), where dcab is the cabinet distance (number of cabinets from the main cabinet), drow is the row position deviation (number of rows from the optimal line of sight), and dcol is the column position deviation (number of columns closest to the user's operating surface). α, β, and γ are the weights of cabinet, row, and column positions, respectively, with α = 0.4, β = 0.35, and γ = 0.25. Dpq is the load balancing degree, Dpq = 1 - ui represents the utilization rate of the compartments in the i-th cabinet (number of occupied compartments / total number of compartments). Weighting coefficients w1=0.4, w2=0.25, w3=0.2, w4=0.15, with a sum of 1. After scoring, the system selects the highest-scoring available compartment as the target compartment for each file. When multiple files are allocated simultaneously, a greedy algorithm is used for sequential allocation, updating the compartment status matrix immediately after each allocation to ensure subsequent allocations are based on the latest status. By abstracting the compartments scattered across various cabinets into a unified resource pool and performing multi-dimensional intelligent matching, the limitations of allocation based on cabinet units are broken, achieving globally optimal allocation of compartment resources. This ensures both security level and reasonable space utilization, while load balancing avoids overuse of main cabinets and long-term idleness of secondary cabinets, improving overall resource utilization by approximately 25%.

[0091] The access path optimization module works in conjunction with the global grid resource scheduling module to calculate the optimal path for user operations and the grid opening order for scenarios involving batch access of multiple files. Specifically, when a user submits an access request containing N files, the system first obtains the target grid location coordinates Pi = (ci, ri, li) for each file, where ci is the cabinet number (0 represents the main cabinet, 1-k represent auxiliary cabinets), ri is the row number (1-5 from top to bottom), and li is the column number (1-2 from left to right). Then, a path cost function is defined, considering the differences in unit cost for three types of movement: inter-cabinet movement has the highest cost because the user needs to move between different cabinets; intra-cabinet row movement is the next most expensive, requiring vertical movement; and intra-cabinet column movement has the lowest cost, requiring only horizontal movement. The formula for calculating the path cost between two grids pi and p_j is as follows:

[0092] cost(pi, pj) = Wc·|ci - cj| + Wr·|ri - rj| + Wl·Δlij

[0093] Where Wc is the cost per unit of movement between cabinets, with a value of 10; Wr is the cost per unit of movement within a cabinet (row), with a value of 2; Wl is the cost per unit of movement within a cabinet (column), with a value of 1; Δlij is the column distance, where Δlij = |li - lj| when two compartments are in the same cabinet, and Δlij = |li - 1.5| + |lj - 1.5| when they are in different cabinets (assuming the center of the cabinet is the passageway). After obtaining the positions of all target compartments, a genetic algorithm is used to solve for the optimal access sequence. This problem is essentially a variant of the Traveling Salesman Problem (TSP). The algorithm parameters are set as follows: population size 50, number of generations 100, crossover probability 0.8, mutation probability 0.1. Chromosomes are encoded as permutations of N compartment numbers. The selection operation uses a roulette wheel method, where the probability of an individual being selected is proportional to its fitness, and the fitness function is the reciprocal of the total path cost: fitness = 1 / total_cost. The crossover operation uses the order crossover method, randomly selecting two crossover points and copying the gene segment between the crossover points in parent generation 1 to the offspring. The remaining genes are filled in according to the order in parent generation 2, ensuring that each slot appears only once. The mutation operation uses exchange mutation, randomly swapping genes at two positions in the chromosome. After the algorithm runs, the access sequence with the minimum total path cost is output as the optimal path. Based on the optimal access sequence, the system generates corresponding operation guidance instructions and opens each target slot in sequence. Through multi-target slot path planning based on an improved genetic algorithm, combined with differentiated cost models for inter-cabinet, intra-cabinet rows, and intra-cabinet columns, the shortest path for user operations can be found, reducing back-and-forth travel between multiple cabinets and slots. Especially in the scenario of batch storage and retrieval of more than 10 files, the path optimization effect can achieve a time saving of more than 30%.

[0094] The RFID read / write control and inventory module is responsible for controlling the read / write operations of the UHF RFID readers in each compartment and achieving cross-read-free rapid inventory of cascaded multi-cabinet systems through a time-slot scheduling algorithm. The system adopts the ISO 18000-6C (EPC C1 G2) protocol and has an adjustable operating frequency of 840-960MHz. Each compartment is equipped with one UHF RFID antenna, which is connected to the corresponding RFID read / write module through a power divider. The time-slot scheduling algorithm is the core innovation of this module, and its specific implementation is as follows: First, time slot pre-allocation. Based on the number of cabinets M and the number of compartments N in each cabinet of the cascaded system, a time slot allocation matrix T = [tij]{M×N} is constructed, where tij is the time slot group number to which the j-th compartment of the i-th cabinet belongs. The time slot allocation satisfies three constraints:

[0095] (1) For adjacent compartments in the same cabinet whose row number differs by 1 or whose column number differs by 1, the difference in time slot group number is ≥2 (to ensure they are not adjacent).

[0096] (2) The time slot group number of corresponding compartments in the same row and column of adjacent cabinets is ≥3 (to prevent cross-cabinet cross-reading);

[0097] (3) Minimize the total number of time slot groups. Using a graph coloring algorithm, for a standard 5x2 grid layout, a single grid requires 4 time slot groups, and a total of 6 time slot groups are needed in a cascaded scenario. Second, dynamic power configuration. The transmit power Pij is configured for the RFID reader / writer in each grid, calculated using the following formula:

[0098] Pij = Pmax·(1 - γ·e_ij)

[0099] Wherein, Pmax is the rated transmit power (usually 30dBm), eij is the edge coefficient: when the grid is surrounded by other grids or metal walls, eij=0 (fully shielded, full power); when the grid is close to the cabinet door glass or cabinet side panel, eij is 0.15 (prone to leakage, power reduced); γ is the adjustment coefficient, taken as 0.5. By appropriately reducing the power at the edge grids, the possibility of signal leakage from cabinet door gaps or cabinet joints is reduced. Third, inventory execution. Inventory is executed sequentially according to the time slot group number from smallest to largest. Each time slot lasts T_slot = 2s, of which the tag reading time is 1.5s and the reader switching time is 0.5s. Within each time slot, the readers of all grids in that time slot group simultaneously transmit signals at their respective configured power to read tags, and the read tag EPC code and RSSI value are transmitted back in real time. Fourth, data aggregation. After all time slots are inventoried, the list of tags read from each grid is compiled, duplicates (the same tag read multiple times from the same grid) are removed, and a system-wide file inventory list is generated. This list is then compared with the records that should be in the database, marking missing files and triggering alarms. This inventory scheme, combining time slot scheduling and dynamic power adjustment, allows as many grids as possible to work in parallel within the same time slot while ensuring zero cross-reading between adjacent grids and cabinets. The overall system inventory time is reduced by approximately 65% ​​compared to traditional cabinet-by-cabinet inventory methods, while maintaining an accuracy rate of over 99.9%.

[0100] The RFID read / write control and inventory module also features a storage / retrieval confirmation function, used to immediately verify the accuracy of the storage / retrieval results after each user operation. The specific process is as follows: When a user operates on a specific compartment and closes the cabinet door, the RFID reader for that compartment immediately initiates a rapid inventory check, with a reading time ≤ 2 seconds, obtaining the complete tag set Scurr within the current compartment. The system queries the tag set Sprev of that compartment before the operation (retrieved from the database) and calculates the difference using set operations: retrieved tag set Sout = Sprev - Scurr, stored tag set Sin = Scurr - Sprev. Then, Sout is compared with the list of files retrieved by the user (Lout), and Sin is compared with the list of files stored by the user (Lin). If Sout and Lout are completely identical and Sin and Lin are completely identical, the storage / retrieval is considered consistent, the operation is successfully confirmed, and the compartment tag record in the database is updated. If discrepancies exist, they are handled as follows: If |Sout| < |Lout| (the number retrieved is less than the requested number), it is considered a missed retrieval, and the system displays a message on the touchscreen saying "XX files remain to be retrieved," and reopens the slot; if Sextra = Sin - Lin is not empty (an unauthorized file has been stored), it is considered an incorrect storage, and the system prompts "Unauthorized file detected, please retrieve it"; if |Sin| < |L_in| (insufficient number stored), it is considered a missed storage, and the user is prompted to continue storing. If inconsistencies persist after three consecutive closed checks, the system saves the current operation data (including all tag reading records, operation video, and timestamps) and notifies the file administrator to intervene and investigate. Through the instant confirmation of storage and retrieval and the abnormal classification and handling mechanism, storage and retrieval errors can be detected and corrected in time before the user leaves the cabinet, avoiding the problem of missing files only being discovered during periodic inventory checks in traditional solutions, and significantly improving the accuracy and security of file storage and retrieval.

[0101] The auxiliary cabinet control unit is the local control core of each auxiliary cabinet, employing an ARM Cortex-M4 core microcontroller. Peripherals include: 8-10 channels of electric lock driver interfaces, 8-10 channels of door magnetic detection interfaces, multiple RFID reader / writer module interfaces, two Ethernet interfaces, and a spare RS485 interface. The main functions of the auxiliary cabinet control unit include: receiving control commands from the main cabinet via the cascading link; parsing the cabinet number and compartment number in the commands; executing the corresponding action if it matches the compartment number of this cabinet; driving the electric locks to control the opening and closing of the compartment doors, supporting both pulse unlocking and continuous power-on modes; collecting the door magnetic status of each compartment and monitoring the cabinet door opening and closing status in real time; controlling the RFID readers / writers of each compartment to perform inventory operations; and transmitting the status and inventory data of this cabinet back to the main cabinet. The auxiliary cabinet control unit has a built-in data buffer that can cache the most recent 1000 operation records. When cascading communication is temporarily interrupted, local operations can still be executed and data can be temporarily stored; automatic retransmission occurs after communication is restored. All auxiliary cabinets have identical hardware designs, supporting plug-and-play expansion. Users can easily expand the system by adding auxiliary cabinets and configuring their cabinet numbers through the main cabinet interface. The intelligent auxiliary cabinet control unit design gives each cabinet a certain degree of autonomous processing and data caching capabilities, reducing real-time dependence on the main cabinet and improving the overall reliability and scalability of the cascaded system.

[0102] In this invention, the global grid resource scheduling module, the access path optimization module, and the RFID read / write control and inventory module work closely together. Specifically, when allocating grids, the global grid resource scheduling module considers not only the matching degree across four dimensions—security level, space, convenience, and load balancing—but also incorporates the path cost calculated by the access path optimization module as a constraint in the allocation decision. When multiple files are stored in the same batch, the system prioritizes grid combinations with more concentrated spatial locations, while ensuring a high matching degree for each file, to reduce the path overhead for subsequent user access. Simultaneously, the time-slot scheduling results of the RFID read / write control and inventory module provide the global grid resource scheduling module with real-time grid occupancy status and inventory data, ensuring that the status information in the resource pool is always accurate and reliable. The allocation results of the global grid resource scheduling module directly determine the input and target grid distribution of the access path optimization module. The three modules support each other and communicate data, forming a complete closed loop from resource allocation to path optimization to inventory verification. The synergistic effect is far beyond the simple sum of the functions of the three modules.

[0103] Taking a "1 main, 3 auxiliary" configuration as an example, the working effect of the present invention is verified. The system configuration is as follows: main cabinet (number 0) with 8 compartments, auxiliary cabinet 1 (number 1) with 10 compartments, auxiliary cabinet 2 (number 2) with 10 compartments, and auxiliary cabinet 3 (number 3) with 10 compartments, for a total of 38 compartments. The compartment layout is 5 rows and 2 columns. The security classification of the main cabinet compartments is as follows: compartments 1-4 are classified as secret, and compartments 5-8 are classified as confidential; compartments 1 and 2 are classified as secret; and compartment 3 is classified as internal. The RFID operating frequency is 920MHz, and the rated power is 30dBm. The time-slot inventory configuration consists of 6 time slot groups, each time slot is 2s, and the total inventory time of the entire system is approximately 12s. Weight parameters: w1=0.4, w2=0.25, w3=0.2, w4=0.15. Path cost parameters: Wc=10, Wr=2, W_l=1.

[0104] Example 1: Grid allocation and path planning for storing 10 classified files.

[0105] Assume a user is a departmental records administrator with confidential access and needs to store 10 confidential-level files, each file box approximately 30mm thick. Upon receiving the storage request, the global grid resource scheduling module begins allocation. First, it filters grids matching the confidentiality level: grids 5-8 in the main cabinet (confidential) and all grids in secondary cabinet 3 (internal) are excluded. Available grids are grids 1-4 in the main cabinet and all grids in secondary cabinets 1 and 2, totaling 24 empty grids. Then, it calculates the matching score between each file and each available grid. Taking the first file as an example: the accessibility Cpq of the first compartment (compartment 01) in the first row and first column of the main cabinet is 1 - (0.4×0 + 0.35×0.6 + 0.25×0.5) = 1 - (0+0.21+0.125) = 0.665, the load balance Dpq is 1 -0 / 8 = 1.0 (the main cabinet is initially empty), the space matching Bpq is 1 - |30-340| / 340 ≈ 0.912, the security level matching Apq=1, and the overall score S is 0.4×1 + 0.25×0.912 + 0.2×0.665 + 0.15×1.0 = 0.4 + 0.228 +0.133 + 0.15 = 0.911. The accessibility Cpq of the 3rd row, 1st column compartment (compartment 25) in secondary cabinet 2 is 1 - (0.4×2 + 0.35×0.2 + 0.25×0.5) = 1 - (0.8+0.07+0.125) = 0.005, and the load balancing Dpq is 1.0 (secondary cabinet 2 is completely empty). The overall score S = 0.4×1 + 0.25×0.912 + 0.2×0.005 + 0.15×1.0 = 0.4 + 0.228 + 0.001 + 0.15 = 0.779. The main cabinet compartment has a higher score (due to its higher accessibility), but considering the load balancing constraint, the system will not allocate all 10 files to the main cabinet. The final allocation result is: 4 slots in the main cabinet (accounting for 100% of the secret-level compartments in the main cabinet), 4 slots in secondary cabinet 1 (accounting for 40% of the compartments in secondary cabinet 1), and 2 slots in secondary cabinet 2 (accounting for 20% of the compartments in secondary cabinet 2), basically achieving the strategy goal of "prioritizing nearby cabinets + balanced distribution". After the allocation is completed, the storage path optimization module calculates the storage order. The 10 target compartments are distributed across 3 cabinets. A genetic algorithm is used to solve for the optimal sequence. After 100 iterations, it converges to obtain the optimal path: 4 compartments in the main cabinet (in top-down order) → 4 compartments in secondary cabinet 1 (in top-down order) → 2 compartments in secondary cabinet 2. The total path cost is approximately 10×2 (from main cabinet to secondary cabinet 1) + 10×1 (from secondary cabinet 1 to secondary cabinet 2) + 2×4×4 (movement within the cabinet) = 20 + 10 + 32 = 62. Compared with the approximately 95 path cost of a random order, the optimization reaches 35%.The system opens the grid slots sequentially and highlights the current grid slot position graphically on the touchscreen, while providing voice prompts for the next step. This example demonstrates how the synergy between global grid slot scheduling and path optimization can significantly reduce the user's operational path length while ensuring reasonable file allocation.

[0106] Example 2: Full system time-slot inventory and serial read suppression verification.

[0107] The system initiates a full system inventory check, with six time slot groups working sequentially. Time slot 1: The working slots are the first, third, and fifth rows and first column slots of each cabinet, a total of 12 slots reading tags simultaneously. These slots have row numbers differing by 2 (non-adjacent rows), column numbers are the same but rows are not adjacent, and the time slot difference for the same slot in adjacent cabinets is ≥3 (cabinet 0, first row, first column is time slot 1, cabinet 1, first row, first column is time slot 4, cabinet 2, first row, first column is time slot 1, and so on). Time slot 2: The working slots are the second and fourth rows and second column slots of each cabinet. Time slot 3: The working slots are the first, third, and fifth rows and second column slots of each cabinet. And so on. Each time slot lasts 2 seconds, and the six time slots, totaling 12 seconds, complete the inventory check of all 38 slots in the system. Compared to traditional cabinet-by-cabinet inventory checks (5 seconds per cabinet, 20 seconds for 4 cabinets), the time is reduced by 40%, and because the compartments are not adjacent within the time slots and the power is dynamically adjusted, the cross-read rate is controlled below 0.1%. After the inventory is completed, the system summarizes all compartment label data and compares it with the EPC codes of 10 newly stored files in the database. All are in place, confirming successful storage. This example verifies the good balance achieved between inventory efficiency and cross-read prevention by the time-slot scheduling algorithm.

[0108] like Figure 2 As shown, this embodiment also provides a self-service retrieval method for cascaded main and auxiliary cabinets with multi-compartment RFID archives, including seven steps: identity authentication and authorization verification S1, retrieval request reception and global compartment retrieval S2, dynamic compartment allocation and path planning S3, main and auxiliary cabinet linked compartment opening S4, RFID real-time identification and retrieval confirmation S5, time-slot cascaded inventory and status update S6, and operation completion and data reporting S7. The specific implementation process of each step is described in detail below.

[0109] Step S1: Identity Authentication and Access Verification. Users approach the main counter and can choose to authenticate their identity by swiping their employee card or using facial recognition. In normal scenarios, swiping the employee card is sufficient; the system reads the employee card serial number and compares it with the user database to query the corresponding user information and access level. In confidential scenarios, dual authentication of facial recognition and employee card is required. A binocular camera captures facial images for liveness detection and facial recognition; only after successful verification can the user enter the system. After successful authentication, the system loads the user's file operation permissions: including the scope of accessible file security levels (internal / secret / confidential), the types of files that can be manipulated, and the access approval status. After the permissions are loaded, the main counter touchscreen displays the main operation interface, including function entries such as "Store Files," "Retrieve Files," "Query Files," and "My Borrowings." Through multi-level authentication and fine-grained access management, it ensures that only authorized personnel can perform file operations at the corresponding level, guaranteeing file security from the source.

[0110] Step S2: Receiving Access Requests and Global Compartment Retrieval. The user selects the operation type. If "Retrieve File," the user can retrieve the required file by file name, file number, QR code scanning, etc. The system queries the database for matching files, obtains the current cabinet number and compartment number of each file, and verifies whether the user has permission to retrieve the file (security level matching, approval, etc.). If "Store File," the user adds file information by scanning a code or manually entering it. After verifying the file's security level and other attributes, the system sends the storage request to the global compartment resource scheduling module. This module traverses the status of all compartments in the main cabinet and all cascaded auxiliary cabinets, filtering out idle compartments from the global compartment resource pool matrix that meet the following conditions: security level not lower than the file's security level, remaining capacity greater than the file's thickness, and compartment size matching. The filtering results serve as the candidate compartment set for subsequent intelligent allocation. By conducting a global compartment search rather than being limited to a single cabinet, the system ensures global optimization in subsequent allocation.

[0111] Step S3: Dynamic Grid Allocation and Path Planning. For storage operations, the global grid resource scheduling module selects the optimal grid from the candidate grid set for each file to be stored based on a multi-dimensional matching degree scoring algorithm. The scoring algorithm includes four dimensions: security level matching, spatial matching, access convenience, and load balancing. The comprehensive scoring formula is Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·Dpq, where w1=0.4, w2=0.25, w3=0.2, and w4=0.15. When multiple files are allocated simultaneously, a greedy strategy is adopted: files are sorted from highest to lowest security level, with higher-security files allocated first to ensure security. After allocation, the access path optimization module calculates the optimal access order for all target grids. An improved genetic algorithm is employed, taking the positions of N target compartments as input. The path cost function considers differentiated weights for inter-compartment movement (unit cost 10), intra-compartment row movement (unit cost 2), and intra-compartment column movement (unit cost 1). An approximate optimal solution is obtained through 50 populations and 100 generations of evolution. The optimal access sequence output by the algorithm represents the user's operation order. Through the synergy of intelligent allocation and path optimization, it ensures that each file is stored in the most suitable compartment while minimizing the user's operation path.

[0112] Step S4: The main and auxiliary cabinets work together to open the corresponding compartments. The main cabinet control unit sends opening commands to each target compartment sequentially according to the optimal access order planned by the path planning. Compartment opening adopts a step-by-step progressive strategy: instead of opening all target compartments at once, they are opened one by one. The system automatically opens the next target compartment only after the operation of the current compartment is completed and the cabinet door is closed. The opening command is issued through the cascading communication management module, and the command frame format is: frame header (0xAA 0x55) + address field (high 8 bits of cabinet number + low 8 bits of compartment number) + function code (0x01 indicates unlocking) + data length + unlocking duration (2 bytes, in seconds) + CRC checksum. After receiving the command, the auxiliary cabinet control unit determines whether it is a compartment in its own cabinet by checking the address field. If so, it drives the corresponding electric lock and returns a successful opening confirmation frame. For multiple target compartments within the same cabinet, the system adopts a priority opening strategy for compartments in the same row or column: when multiple target compartments are in the same row or column, they are opened at once, allowing users to operate continuously and reducing the number of times the door is opened and closed. As the compartment opens, the touchscreen displays the current cabinet layout, highlighting the compartment being opened with a flashing indicator, and announces, "Please retrieve or place the file in compartment X, row X, column X of cabinet XX." Through step-by-step opening and graphical guidance, users simply follow the system prompts, eliminating the need for manual planning and significantly reducing operational complexity in multi-cabinet, multi-compartment scenarios.

[0113] Step S5: Real-time RFID Identification and Access Confirmation. After the user completes the retrieval or placement of files in the open compartment and closes the cabinet door, the system immediately initiates the access confirmation process for that compartment. The RFID reader / writer for that compartment performs a quick inventory check, with a reading time set to 2 seconds (enough to read dozens or hundreds of tags). After reading the EPC codes of all tags, the current tag set Scurr is formed. The system retrieves the tag set Sprev of that compartment from the database before the operation and performs a set difference operation: retrieved tag set Sout = Sprev - Scurr, stored tag set Sin = Scurr - Sprev. Then, Sout is compared one by one with the EPC codes of the file list Lout requested by the user, and Sin is compared one by one with the file list Lin requested by the user. If the retrieved set and the requested set are completely consistent, and the stored set and the stored list are also completely consistent, the access is considered successful, and the tag record of that compartment and the location information of the corresponding file in the database are updated. If discrepancies occur, the system provides different prompts based on the type of deviation: For missed retrieval, the system prompts "X files have not been retrieved; please retrieve them and close the cabinet door"; for incorrect storage, the system prompts "Unauthorized file X detected; please retrieve it and close the cabinet door." The system allows a maximum of three retries. If discrepancies persist after three attempts, the system saves the data and notifies the administrator. This real-time access confirmation mechanism allows for the detection and correction of operational errors before the user leaves the cabinet, preventing the reactive situation of discovering errors only during post-incident inventory checks.

[0114] Step S6: Time-slot Cascaded Inventory and Status Update. After a single access operation is successfully confirmed, the system can choose to initiate a full system inventory to ensure complete data synchronization. The full system inventory uses a time-slot scheduling algorithm: First, time slot groups are pre-allocated according to the grid layout of the cascaded system. The time slot difference between adjacent grids within the same cabinet is ≥2, and the time slot difference between corresponding grids in adjacent cabinets is ≥3. Then, the inventory is started sequentially according to the time slot group number from smallest to largest, with each time slot lasting 2 seconds (1.5 seconds for reading + 0.5 seconds for switching). Within each time slot, the RFID readers of all grids in that group work simultaneously, and the transmission power of edge grids is reduced by 10%-15% to reduce cross-reading. After all time slots are inventoried, the tag data of each grid is summarized to generate a full system archive in-place list, which is compared with the records that should be in place in the database, and the location and in-place status of each archive are updated. For cases of inconsistency (such as an archive record being in grid 1 but actually in grid 2), the system marks it as abnormal and prompts the administrator to check. By using time-slot parallel inventory counting, the traditional serial inventory counting is transformed into limited parallel counting, which not only ensures the security against cross-reading but also significantly improves the inventory counting efficiency in cascaded multi-cabinet scenarios.

[0115] Step S7: Operation Completion and Data Reporting. After all target grid operations are completed, the system stores a complete record of the operation in the local database, including: operator ID and name, operation time, operation type (save / retrieve), list of involved files, list of involved grids, operation time, and any exceptions. Simultaneously, the operation record is uploaded to the file management server via the network to ensure data consistency between the local and backend systems. Upon completion, the main interface displays a "Operation Successful" message and shows the list of files accessed for user confirmation. The touchscreen returns to the main operation interface, awaiting the next user. Through complete operation records and data synchronization, every file access is traceable and verifiable, providing comprehensive data support for auditing and tracing file management.

[0116] The overall architecture upon which the aforementioned system and methods are implemented mainly includes the following functional modules: The identity authentication and interaction module is responsible for user identity verification and operation guidance, serving as the system's human-machine interface; the main cabinet control unit is the core of control and computation, carrying the main business logic; the cascaded communication management module is responsible for reliable communication and command distribution between the main and auxiliary cabinets; the global grid resource scheduling module realizes virtualized unified management and intelligent allocation of multiple cabinet grids; the RFID read / write control and inventory module realizes tag reading / writing and time-slot inventory; the storage path optimization module realizes optimal path planning for multiple target grids; and the auxiliary cabinet control unit is responsible for local grid control and data acquisition in the auxiliary cabinet. Through close data interaction and functional collaboration, these modules constitute a complete cascaded main and auxiliary cabinet multi-grid RFID self-service archive storage and retrieval system.

[0117] It will be understood by those skilled in the art that all or part of the steps in the above embodiments can be implemented by hardware related to computer program instructions. The computer program instructions can be stored in a computer-readable storage medium, and these instructions cause the computer to execute the methods described in the above embodiments. The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the above embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the scope of the technical solutions of the embodiments of the present invention.

[0118] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A cascaded main and auxiliary cabinet multi-compartment RFID self-service file storage and retrieval system, characterized in that, include: The identity authentication and interaction module is used to collect user identity information and perform permission verification, receive user file access requests and display operation guidance information; The main cabinet control unit, which is connected to the identity authentication and interaction module, serves as the main control core of the cascaded system, processes access business logic, issues control commands to each sub-cabinet, and summarizes status data. The cascaded communication management module is connected to the main cabinet control unit and is used to maintain the communication links between the main cabinet and multiple auxiliary cabinets. It adopts a cascaded topology to realize instruction distribution and data feedback between the main cabinet and auxiliary cabinets, and between auxiliary cabinets. The global grid resource scheduling module is connected to the main cabinet control unit and is used to include all grids of the main cabinet and all cascaded secondary cabinets into a unified resource pool for virtualized management. Based on the attribute information of the archive to be accessed, the user's permission level, the real-time status of each grid and the access path cost, the module allocates a target grid to each archive through a global optimal allocation algorithm. The RFID read / write control and inventory module is connected to the main cabinet control unit and the cascaded communication management module. It is used to control the RFID readers in each compartment to perform file tag reading and writing operations, and to coordinate the inventory process of cascaded multi-cabinet through a time-slot scheduling algorithm, so as to achieve fast inventory of the whole system and no cross-reading between compartments and cabinets. The access path optimization module, which is connected to the global grid resource scheduling module, is used to generate the optimal path and grid opening order for user access operations based on the cabinet number and location of the target grid, and provide step-by-step guidance to the user through the interaction module. Multiple auxiliary cabinet control units are connected to the main cabinet control unit through the cascaded communication management module. They are used to receive control commands issued by the main cabinet, drive the electric locks of each compartment of the cabinet to perform opening and closing actions, and transmit the RFID inventory data of the cabinet back to the main cabinet.

2. The main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system according to claim 1, characterized in that: The specific process by which the global grid resource scheduling module performs grid allocation is as follows: Construct a global grid resource pool matrix G = [g11, g12, ..., gij, ..., gmn], where g_ij represents the status information of the j-th grid of the i-th cabinet, including occupancy status, remaining capacity, grid size, and security level attribute, m is the total number of cabinets, and n is the number of grids in a single cabinet; For each file dp in the archive set D = [d1, d2, ..., dk], calculate its matching score S_pq with each available slot. The calculation formula is as follows: Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·D_pq Among them, Apq is the security level matching degree, which is 1 when the security level of the compartment is greater than or equal to the security level of the file, and 0 otherwise; Bpq is the space matching degree, which is calculated based on the ratio of the file box size to the remaining capacity of the compartment, and the value range is 0-1; Cpq is the access convenience degree, which is calculated based on the distance between the cabinet where the compartment is located and the user's operating position and the floor height, the closer the distance and the more suitable the height, the higher the score; Dpq is the load balancing degree, which is calculated based on the proportion of used compartments in the cabinet where the compartment is located, and priority is given to allocating to cabinets with lower usage rates; w1, w2, w3, and w4 are the weight coefficients of each item, and w1+w2+w3+w4=1, with w_1 having a maximum value of 0.4; For each file, select the available cell with the highest matching score as the target cell, while ensuring that the files stored in the same batch meet the shortest path constraint in terms of cell distribution.

3. The main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system according to claim 1, characterized in that: The specific method by which the access path optimization module generates the optimal access path is as follows: When a user submits an access request containing N files, obtain the target cell location coordinates Pi = (cabineti, rowi, coli) for each file, where cabineti is the cabinet number, rowi is the row number, and col_i is the column number. Starting with the main cabinet operation, construct the path weight matrix W between compartments, where the weight w_ij between any two compartments consists of two parts: the distance moved between cabinets and the distance moved within the cabinet. head = |cabinets - cabinetj|·L + √[(row - row) 2 + (col - col_j) 2 ]·H Where L is the unit movement cost between adjacent cabinets, and H is the unit movement cost between adjacent compartments within the cabinet, and L > H; A genetic algorithm is used to solve the optimal access order of N target grids, with the goal of minimizing the total path cost. The population size is 50, the number of iterations is 100, the crossover probability is 0.8, and the mutation probability is 0.

1. Based on the optimal access order obtained from the solution, operation guidance instructions for each step are generated sequentially to control the corresponding grids to open in sequence, and the current step and the next step prompts are displayed to the user through the identity authentication and interaction module.

4. The main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system according to claim 1, characterized in that: The time-slot scheduling algorithm for the RFID read / write control and inventory module specifically includes: The entire cascaded system is divided into M time slot groups, each containing several spatially non-adjacent grids, such that the distance between any two grids in the same group is greater than the RFID signal serial reading threshold distance D_th. The time slot allocation principle is as follows: adjacent compartments within the same cabinet are assigned to different time slot groups, and corresponding compartments in adjacent cabinets are assigned to different time slot groups; During inventory, the RFID readers in each slot are started sequentially according to the time slot group number. The duration of each time slot is Tslot = Tread + Tswitch, where Tread is the tag reading time and T_switch is the reader switching time. For the kth time slot group, the RFID readers in all slots of the group work simultaneously, and the reading and writing power is dynamically adjusted according to the slot position: the power of the slots near the edge of the cabinet is reduced by 10%-15%, and the middle slots use the rated power. Once all time slot groups have completed an inventory check, the label data of each slot is summarized to form a complete system archive inventory list.

5. The main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system according to claim 4, characterized in that: The RFID read / write control and inventory module also includes a storage confirmation sub-process: When a user performs an access operation and closes the grid, the RFID reader of that grid immediately initiates a quick inventory check, reads the EPC codes of all file tags in the grid, and obtains the current tag set S_current; Compare Scurrent with the label set Sbefore before the operation, and calculate the new label set Sadd = Scurrent - Sbefore and the reduced label set Sremove = Sbefore - Scurrent; Compare Sadd and Sremove with the list of files saved and retrieved by the user this time, respectively. If they match completely, the save and retrieve are confirmed to be successful. If there is a mismatch, calculate the deviation Δ = |Sadd| - number of entries + |Sremove| - number of entries. When Δ ≠ 0, start an auxiliary scan of adjacent cells to check for possible cross-reads, and prompt the user to check on the interactive interface.

6. The main and auxiliary cabinet cascaded multi-compartment RFID file self-service storage and retrieval system according to claim 1, characterized in that: The cascaded communication management module adopts a serial cascaded topology. The main cabinet is connected to the first-level sub-cabinet through the first Ethernet interface, the first-level sub-cabinet is connected to the second-level sub-cabinet through the second Ethernet interface, and so on to form a chain connection. Each sub-cabinet has a built-in cascaded control board with two network interfaces and data forwarding function. The cascade depth supports more than 1 main cabinet and 8 sub-cabinets. The communication protocol of the cascaded communication management module adopts a custom binary protocol. The instruction frame format includes a frame header, address field, function code, data field, and check code. The high 8 bits of the address field are the cabinet number and the low 8 bits are the grid number, realizing precise addressing at the single grid level. The cascaded communication management module also has a link fault detection and automatic reconnection mechanism. When the communication of a certain level sub-cabinet is interrupted, its downstream sub-cabinets automatically switch to the backup path or report the fault status, ensuring the partial availability of the cascaded system.

7. A method for self-service storage and retrieval of archives using RFID technology in a cascaded main and auxiliary cabinet system, characterized in that... Includes the following steps: S1. Identity Authentication and Permission Verification: Users submit their identity credentials through the identity authentication module. The main cabinet control unit verifies the validity of the user's identity and the corresponding file operation permissions. After successful authentication, the user enters the storage and retrieval operation interface. S2. Access Request Reception and Global Grid Search: When a user submits a file access request, the main cabinet control unit sends the request to the global grid resource scheduling module. This module traverses the status of all grids in the main cabinet and all cascaded secondary cabinets, and searches for the target grid that meets the conditions or locates the location of the file to be retrieved. S3. Dynamic grid allocation and path planning: The global grid resource scheduling module allocates the optimal grid to the stored file based on a multi-dimensional matching degree scoring algorithm, and the access path optimization module calculates the optimal access order and guidance path according to the target grid distribution. S4. Main and auxiliary cabinet linkage compartment opening: The main cabinet control unit sends an opening command to the cabinet where the target compartment is located through the cascaded communication management module. The corresponding auxiliary cabinet control unit drives the electric lock to open the target compartment. The compartment opening sequence is consistent with the path planning result. S5. Real-time RFID identification and access confirmation: After the user completes the file retrieval and storage operation and closes the compartment, the RFID reader of the corresponding compartment immediately performs a quick inventory, reads all the tags in the current compartment and compares them with the status before the operation to confirm the consistency between the number of files stored and the file information. S6. Time-slot cascaded inventory and status update: After the storage and retrieval operation is completed, the system starts a full system time-slot inventory, and sequentially starts the RFID readers of each compartment of each cascaded cabinet according to the time slot group to complete the synchronous update of the archive status of the entire system. S7. Operation Completion and Data Reporting: Store the personnel information, time, file list, and grid change record of this access operation into the local database, and simultaneously upload them to the file management server, generate an operation record, and prompt the user that the operation is complete.

8. The self-service retrieval method for multi-compartment RFID archives via cascaded main and auxiliary cabinets according to claim 7, characterized in that: In step S3, the multi-dimensional matching degree scoring algorithm for dynamic grid allocation is specifically as follows: Construct a global grid resource pool matrix G = [g11, g12, ..., gij, ..., gmn], where gij represents the status information of the j-th grid of the i-th cabinet, including the occupancy status sij, the remaining capacity cij, the grid size szij, and the security level attribute cl_ij, m is the total number of cabinets, and n is the number of grids in a single cabinet; For each file dp in the stored file set D = [d1, d2, ..., dk], extract its attributes: file size szp, file security level clp, and estimated access frequency f_p; Calculate the four matching indices between the archive dp and the grid gij: Security level matching degree Apq: Apq = 1 when clij ≥ clp, otherwise Apq = 0, and Apq is a veto item. When the value is 0, the cell is directly excluded. Spatial matching degree Bpq = 1 - |szp - szij| / szij, when szp > szij B_pq = 0; Access convenience Cpq = 1 - (α·|i-1| + β·|j - j0| / n), where |i-1| is the distance between the cabinet number and the main cabinet, j_0 is the optimal column corresponding to the user operation surface, α is the weight of the distance between cabinets, and β is the weight of the column position within the cabinet, both ranging from 0.2 to 0.

4. Load balancing degree Dpq = 1 - ui, where ui is the utilization rate of the compartments in the i-th cabinet, ui = number of occupied compartments / total number of compartments; The formula for calculating the overall matching score is: Spq = w1·Apq + w2·Bpq + w3·Cpq + w4·D_pq Where w1=0.4, w2=0.25, w3=0.2, w4=0.15; For each file, select the available grid with the highest matching score. When multiple files are allocated at the same time, a greedy algorithm is used to allocate them sequentially, and the grid state matrix is ​​updated after each allocation.

9. The self-service retrieval method for multi-compartment RFID archives via cascaded main and auxiliary cabinets according to claim 7, characterized in that: In step S6, the time-slot cascaded inventory method specifically includes: Time slot pre-allocation stage: Based on the number of cabinets M and the number of slots N in each cabinet of the cascaded system, construct the time slot allocation matrix T = [tij]{M×N}, where t_ij is the time slot group number of the j-th slot in the i-th cabinet; the time slot allocation satisfies the following constraints: adjacent slots in the same cabinet whose row number or column number differs by 1 have a time slot group number difference ≥ 2; corresponding slots in the same row and column of adjacent cabinets have a time slot group number difference ≥ 3; Power configuration phase: Configure the transmission power P_ij for each RFID reader / writer in each grid, and the calculation formula is as follows: Pij = Pmax·(1 - γ·e_ij) Where Pmax is the rated transmission power, eij is the edge coefficient, eij=0 when the slot is located in the inner layer of the cabinet and there are other slots on all sides, eij is 0.1-0.2 when the slot is located on the edge of the cabinet or near the cabinet door, and γ is the adjustment coefficient of 0.

5. Inventory execution phase: Inventory is performed sequentially according to the time slot group number from smallest to largest. Each time slot lasts for T_slot = 2s, including 1.5s reading time and 0.5s switching time. Within each time slot, the readers of all slots in the group simultaneously transmit signals to read tags at their respective configured power. Data aggregation phase: After all time slots are inventoried, the list of tags read from each grid is aggregated to generate a system-wide archive in-place list. This list is then compared with the archive information recorded in the database, and archives that are not in place are marked and alarms are triggered.

10. The self-service retrieval method for multi-compartment RFID archives via cascaded main and auxiliary cabinets according to claim 7, characterized in that: In step S5, the specific process for access confirmation and exception handling is as follows: After the user closes the grid, the RFID reader of the grid starts a rapid inventory check, with a reading time of ≤2s, to obtain the current tag set S_curr; Before the query operation, the tag set Sprev of this cell is retrieved, and the difference is calculated: the tag set Sout = Sprev - Scurr is retrieved and stored in the tag set Sin = Scurr - S_prev; Compare Sout with the list of files requested by the user to be retrieved (Lout), and compare Sin with the list of files requested by the user to be stored (Lin). If Sout = Lout and Sin = Lin, then the access is considered consistent, and the operation is confirmed as successful. If |Sout| < |Lout|, it is determined that there is a missing file. The system will prompt "There are still files that have not been retrieved" on the interactive interface and reopen the grid. If Sextra = Sin - L_in is not empty, it is determined to be an incorrect file, and the system prompts "An unauthorized file was detected" and asks the user to retrieve the extra file. If the data is inconsistent after three consecutive checks, the system will retain the on-site data and notify the administrator to intervene.

Images