Intelligent warehouse AGV cooperates RFID aviation consumable automatic storage and taking method and system

Abstract

The application discloses an intelligent warehouse AGV collaborative RFID aviation consumable automatic storage and taking method and system, signals of a warehouse location RFID label are collected in parallel, and two-dimensional code information read by a visual camera is collected, and forced consistency verification is carried out, so that absolute accuracy of AGV positioning is ensured, a dynamic Q algorithm is used to process multi-label signal conflicts, and an upper limit of a label number of single identification is set, so that signal collision is avoided, a wave-absorbing material cover is additionally arranged on an AGV reader, interference is reduced from a physical layer, and antenna power is automatically adjusted during reading, so that a reading range is accurately limited to a target warehouse location, and adjacent warehouse locations of consumables are prevented from being misread.

Description

Technical Field

[0001] This invention relates to the field of warehouse management, and in particular to an intelligent warehouse AGV-assisted RFID automatic storage and retrieval method and system for aviation consumables. Background Technology

[0002] A smart warehouse is a digital, automated, and intelligent warehouse that relies on the Internet of Things (IoT), radio frequency identification (RFID), automated guided vehicles (AGV), sensor detection, warehouse management system (WMS / WCS), intelligent scheduling, and big data technology to achieve unmanned / minimal manpower management of the entire inbound and outbound process, including warehouse environment monitoring, automatic identification of goods, intelligent allocation of storage locations, automatic material handling, automatic storage and retrieval, automatic inventory counting, and unmanned / minimal manpower management of the entire inbound and outbound process.

[0003] In dense warehouse environments, the simultaneous presence of a large number of RFID tags can easily lead to signal conflicts and misreadings. When multiple AGVs work together in the same space, path conflicts and congestion are likely to occur. Traditional management methods make it difficult to conduct precise monitoring of high-value aviation consumables with shelf-life requirements.

[0004] In summary, there is a need for an intelligent warehouse AGV-assisted RFID-based automatic storage and retrieval method and system for aviation consumables to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for automatic storage and retrieval of aviation consumables using AGV-assisted RFID in intelligent warehouses, aiming to solve the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for automatic storage and retrieval of aviation consumables using AGV-assisted RFID in an intelligent warehouse, comprising the following steps:

[0007] Step S1: Equipment deployment and shelf identification settings. The warehouse is divided into grid-based spaces and functional zones. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information.

[0008] Step S2: Build a central management platform, establish an anti-interference identification mechanism, use the dynamic Q algorithm for anti-collision processing, and set an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the cargo location tag and the QR code information read by the vision camera are collected in parallel, and the consistency of the two is verified.

[0009] Step S3: Execute the storage task. The WMS system warehouse management system receives the storage instruction and automatically allocates the optimal storage location for aviation consumables according to the storage strategy. The RCS robot scheduling system schedules AGVs to transport the consumables to the target storage location. After the AGV confirms the storage location through the anti-interference identification mechanism, it puts the consumables on the shelf and writes the storage location information into the RFID tag of the consumables.

[0010] When the retrieval task is executed, the WMS system locks the target storage location according to the preset rules, the RCS system dispatches the AGV to the target storage location, and after the AGV confirms the storage location and consumable information through the anti-interference identification mechanism, it retrieves the consumables and transports them to the outbound port.

[0011] Step S4: Warehouse information update and alarm. The WMS system updates inventory data in real time and monitors the inventory status according to preset alarm rules, triggering alarm information of different levels.

[0012] Optionally, step S1 is implemented in the following manner:

[0013] Step S11: Deploy the physical environment of the warehouse, implement unified grid-based spatial marking of the internal shelving layout, passageways, work areas and functional points of the warehouse, establish a unique coordinate system for the entire area, divide the warehouse into zones according to the confidentiality level and physical and chemical properties of aviation consumables, implement physical isolation and access control between zones, and lay positioning QR code grids and physical landmarks on the ground at fixed intervals to build a visual navigation reference network;

[0014] Step S12: Location identification and binding. Read the unique EPC code of the anti-interference RFID anchor tag at each key node, bind it one-to-one with the system location code, and then upload it to the WMS system in batches to complete the binding and storage of location RFID anchors and location coordinate information.

[0015] Optionally, step S2 is implemented in the following manner:

[0016] Step S21: Hardware protection. Install a wave-absorbing material cover on the AGV's forklift reader / writer. Based on the physical characteristics of near-field communication, control the reading and writing distance within a threshold range to reduce physical interference.

[0017] Step S22: Algorithm anti-collision, using dynamic Q algorithm for anti-collision processing, setting only authorized activation of tags within the target area at the same time, and performing multi-tag signal collision processing;

[0018] Step S23: Dual verification. During each identification, the RSSI signal strength of the cargo location tag and the ground QR code information read by the vision camera are collected in parallel. The cargo location ID decoded from the tag with the strongest RSSI signal is subjected to a mandatory consistency verification with the cargo location ID read from the QR code.

[0019] Optionally, the anti-interference identification mechanism established in S2 also includes an environmental electromagnetic fingerprint and an adaptive power adjustment mechanism:

[0020] Environmental monitoring nodes are deployed inside the warehouse to sense the temperature, humidity and electromagnetic interference intensity in real time, forming an environmental electromagnetic fingerprint.

[0021] When environmental parameters change beyond a preset threshold, the system automatically adjusts the RFID reader's transmission power and receiving sensitivity threshold based on the current environmental electromagnetic fingerprint to offset the impact of environmental changes on RFID signal attenuation.

[0022] Optionally, after building the central management platform in step S2, the method further includes the step of establishing a warehouse-level digital twin system:

[0023] The system collects the location information, posture information, battery level, and cargo location status of the AGV in real time, and maps this information to the virtual warehouse model in milliseconds.

[0024] Before executing batch storage and retrieval tasks, a task rehearsal is performed in the virtual warehouse model to detect whether there is a risk of deadlock or congestion in the path planning.

[0025] Only when there are no conflicts in the pre-rehearsal results will the central management platform issue execution instructions to the physical AGVs to achieve synchronous control between the virtual and physical systems.

[0026] Optionally, the storage task in step S3 is performed in the following manner:

[0027] Step S31: Task generation and storage location allocation. After scanning the consumables label, the WMS system parses the consumables attributes and uses a three-dimensional weighted scoring method to comprehensively consider three dimensions: shortest path, highest AGV efficiency, and centralized storage of similar items. Based on the storage strategy, the optimal storage location is automatically calculated and allocated.

[0028] Step S32: AGV handling and navigation. The RCS system dispatches the AGV to the warehouse entrance. The AGV performs a second check on the consumable label and picks up the material. While traveling along the planned path to the target storage location, the vehicle-mounted vision camera continuously scans the ground QR code and compares it with the planned path. If the deviation exceeds the threshold, it will automatically correct the deviation.

[0029] Step S33: Precise placement and binding. After the AGV arrives at the target storage location, it reads the RFID anchor tag of the storage location and verifies the location with the visual QR code. It then places the consumables smoothly into the target storage location, writes the current storage location number into the user area of ​​the consumable RFID tag, completes the binding of the consumables to the storage location, and updates the inventory data to the WMS system.

[0030] Optionally, the optimal storage location is automatically calculated in step S31 using the following method:

[0031] WMS iterates through all legal candidate storage locations, calculates Score(i) for each storage location, selects the storage location with the largest Score, and outputs the X, Y, and Z three-dimensional coordinates of the storage location, which are the optimal storage location coordinates.

[0032] Score(i)=α×S_path(i)+β×S_eff(i)+γ×S_cluster(i);

[0033] α+β+γ=1;

[0034] In the formula, Score(i) is the overall score of storage location i, S_path(i) is the path distance score, S_eff(i) is the AGV efficiency score, S_cluster(i) is the score of centralized storage of the same type, α is the path distance weight coefficient, β is the AGV efficiency weight coefficient, and γ is the centralized storage weight coefficient of the same type.

[0035] Optionally, before step S33 is accurately placed and bound, an AI vision-based consumable appearance defect detection step is also included:

[0036] Before the AGV places the consumables to the target location, a high-definition industrial camera mounted on the AGV is used to capture surface images of the aviation consumables.

[0037] The surface image is input into a pre-trained convolutional neural network model to analyze whether the consumables have defects such as rust, damaged packaging, or blurred labels.

[0038] If the test results show that the consumables have an abnormal appearance, the AGV will automatically stop the shelving action and transfer the consumables to the quality inspection isolation area, while sending an abnormal appearance alarm message to the WMS system.

[0039] Optionally, the retrieval task in step S3 is performed in the following manner:

[0040] Step S41: Task locking and scheduling. The WMS system prioritizes locking the storage locations of consumables that are close to their expiration date based on the outbound instructions and preset rules, generates an outbound task and sends it to the RCS system. The RCS system uses a time window and route reservation mechanism to avoid conflicts between multiple vehicles, and schedules the optimal AGV to the target storage location based on the task instructions and the real-time status of the AGV.

[0041] Step S42: Accurate identification and extraction. After the AGV arrives at the target location, it confirms that the location and consumable information are correct through an anti-interference identification mechanism, and then performs the extraction action. If the AGV senses multiple tags at the same time when extracting consumables, it will automatically reduce the antenna power and limit the reading range to the target location.

[0042] Step S43: Handling and Outbound. After the AGV retrieves the consumables, it continuously scans the QR code on the ground with a vision camera to correct its path. Upon reaching the outbound gate, the AGV places the consumables in the designated area and reports the task completion status to the WMS system.

[0043] Optionally, step S4 is implemented in the following manner:

[0044] Step S51: Warehouse information update. After each storage and retrieval task is completed, the WMS system updates the inventory data in real time. The updated data includes the current status of consumables, storage location information, quantity, and operation timestamp.

[0045] Step S52: Multi-level alarm monitoring. The WMS system continuously monitors the inventory status according to preset alarm rules and triggers alarm information of different levels when the conditions are met.

[0046] Optionally, step S51, updating warehouse information, is recorded using blockchain-based distributed ledger technology.

[0047] The key operation data for each task deposit, retrieval, and alarm information is hashed and uploaded to the consortium blockchain.

[0048] Smart contracts are used to verify permissions to modify inventory data. Changes to inventory data are verified on-chain and the operator's identity is recorded.

[0049] Optionally, in step S52, the alarm levels are divided into three categories: low risk, medium risk, and high risk.

[0050] The low-risk alarm is triggered when the consumables are nearing their expiration date;

[0051] The trigger condition for a medium-risk alert is that the inventory quantity is lower than the lower limit or higher than the upper limit;

[0052] The high-risk alarm is triggered when consumables have expired, are stored in the wrong area, or are mixed up.

[0053] An intelligent warehouse AGV collaborative RFID aviation consumables automatic storage and retrieval system, employing the aforementioned intelligent warehouse AGV collaborative RFID aviation consumables automatic storage and retrieval method, includes a physical environment and identification module, a central management module, a warehouse management module, an AGV scheduling module, an automatic guided transportation module, and an update and alarm module;

[0054] The physical environment and identification module is used to perform grid-based spatial marking and functional zoning of the warehouse. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information.

[0055] The central management module is used to establish an anti-interference identification mechanism, adopts the dynamic Q algorithm for anti-collision processing, and sets an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the cargo location tag and the QR code information read by the vision camera are collected in parallel, and the consistency of the two is verified.

[0056] The warehouse management module is used to receive inbound and outbound instructions, automatically allocate the optimal storage location for aviation consumables according to the preset storage strategy, or lock the target storage location according to the rules.

[0057] The AGV scheduling module is used to schedule the optimal AGV to the designated location based on the task instructions issued by WMS and the real-time status of the AGV. It adopts a time window and path reservation mechanism to plan the AGV's travel path, effectively avoiding conflicts or congestion when multiple vehicles are running.

[0058] The automated guided transport module is responsible for moving consumables from the inbound entrance to the target storage location, or retrieving consumables from the storage location and transporting them to the outbound entrance. It continuously scans the QR code grid laid on the ground with the vehicle-mounted vision camera and compares it with the planned route. If a deviation is found, it will automatically correct the deviation to ensure accurate driving.

[0059] The update and alarm module is used to update inventory data in real time after the storage and retrieval tasks are completed, continuously monitor the inventory status, and trigger alarm information of different levels according to preset rules.

[0060] The beneficial effects of this invention are:

[0061] 1. In this invention, the signal strength of the RFID tags in the cargo location and the QR code information read by the vision camera are collected in parallel, and a mandatory consistency check is performed to ensure the absolute accuracy of AGV positioning. The dynamic Q algorithm is used to handle signal conflicts of multiple tags, and an upper limit is set for the number of tags that can be identified at one time to avoid signal collisions. A wave-absorbing material cover is added to the AGV reader to reduce interference from a physical level, and the antenna power is automatically adjusted during reading to accurately limit the reading range to the target cargo location and prevent misreading of consumables in adjacent cargo locations.

[0062] 2. In this invention, the RCS system adopts a time window and path reservation mechanism to plan a conflict-free driving route for each AGV, effectively avoiding multi-vehicle conflicts. When the WMS system enters the warehouse, it adopts a three-dimensional weighted scoring method, comprehensively considering three dimensions: shortest path, highest AGV efficiency, and centralized storage of similar items, to automatically calculate the optimal storage location, thereby improving the overall operational efficiency from the source. The AGV continuously scans the ground QR code while driving, compares it with the planned path, and automatically corrects deviations to ensure driving accuracy.

[0063] 3. In this invention, consumable information is bound to RFID tags at the storage location, and the WMS system updates inventory data in real time, realizing full traceability of consumables from warehousing to outbound. The system can continuously monitor the inventory status according to preset rules and trigger alarms of different levels, thereby proactively preventing management risks. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of a method flow of the present invention.

[0065] Figure 2 This is a schematic diagram of step S1 of the present invention.

[0066] Figure 3 This is a schematic diagram of step S2 of the present invention.

[0067] Figure 4 This is a schematic diagram of step S3 of the present invention.

[0068] Figure 5 This is a schematic diagram of step S4 of the present invention.

[0069] Figure 6 This is a schematic diagram of step S5 of the present invention.

[0070] Figure 7 This is a schematic diagram of a system task scheduling according to the present invention.

[0071] Figure 8 This is a schematic diagram of a real-time system monitoring according to the present invention.

[0072] Figure 9 This is a schematic diagram of an intelligent computing system according to the present invention. Detailed Implementation

[0073] To more clearly illustrate the technical solutions in the embodiments of the invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0074] like Figures 1 to 9 As shown, a method for automatic storage and retrieval of aviation consumables in a smart warehouse using AGV and RFID includes the following:

[0075] Step S1: Equipment deployment and shelf identification settings. The warehouse is divided into grid-based spaces and functional zones. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information.

[0076] Step S2: Build a central management platform, establish an anti-interference identification mechanism, use the dynamic Q algorithm for anti-collision processing, and set an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the cargo location tag and the QR code information read by the vision camera are collected in parallel, and the consistency of the two is verified.

[0077] It also includes environmental electromagnetic fingerprinting and adaptive power regulation mechanisms:

[0078] Environmental monitoring nodes are deployed inside the warehouse to sense the temperature, humidity and electromagnetic interference intensity in real time, forming an environmental electromagnetic fingerprint.

[0079] When environmental parameters change beyond a preset threshold, the system automatically adjusts the RFID reader's transmission power and receiving sensitivity threshold based on the current environmental electromagnetic fingerprint to offset the impact of environmental changes on RFID signal attenuation.

[0080] After building the central management platform, the next step is to establish a warehouse-level digital twin system:

[0081] The system collects the location information, posture information, battery level, and cargo location status of the AGV in real time, and maps this information to the virtual warehouse model in milliseconds.

[0082] Before executing batch storage and retrieval tasks, a task rehearsal is performed in the virtual warehouse model to detect whether there is a risk of deadlock or congestion in the path planning.

[0083] Only when there are no conflicts in the pre-rehearsal results will the central management platform issue execution instructions to the physical AGVs to achieve synchronous control between the virtual and physical systems.

[0084] Step S3: Execute the storage task. The WMS system warehouse management system receives the storage instruction and automatically allocates the optimal storage location for aviation consumables according to the storage strategy. The RCS robot scheduling system schedules AGVs to transport the consumables to the target storage location. After the AGV confirms the storage location through the anti-interference identification mechanism, it puts the consumables on the shelf and writes the storage location information into the RFID tag of the consumables.

[0085] When the retrieval task is executed, the WMS system locks the target storage location according to the preset rules, the RCS system dispatches the AGV to the target storage location, and after the AGV confirms the storage location and consumable information through the anti-interference identification mechanism, it retrieves the consumables and transports them to the outbound port.

[0086] Before accurate shelf placement and binding, there is also a consumable appearance defect detection step based on AI vision:

[0087] Before the AGV places the consumables to the target location, a high-definition industrial camera mounted on the AGV is used to capture surface images of the aviation consumables.

[0088] The surface image is input into a pre-trained convolutional neural network model to analyze whether the consumables have defects such as rust, damaged packaging, or blurred labels.

[0089] If the test results show that the consumables have an abnormal appearance, the AGV will automatically stop the shelving action and transfer the consumables to the quality inspection isolation area, while sending an abnormal appearance alarm message to the WMS system.

[0090] Step S4: Warehouse information update and alarm. The WMS system updates inventory data in real time and monitors the inventory status according to preset alarm rules, triggering alarm information of different levels.

[0091] Warehouse information updates are recorded using blockchain-based distributed ledger technology.

[0092] The key operation data for each task deposit, retrieval, and alarm information is hashed and uploaded to the consortium blockchain.

[0093] Smart contracts are used to verify permissions to modify inventory data. Changes to inventory data are verified on-chain and the operator's identity is recorded.

[0094] Specifically, it will be implemented in the following ways:

[0095] Step 1: Equipment Deployment and Shelf Identification: Implement unified grid-based spatial marking for the warehouse's internal shelving layout, passageways, work areas, and functional points. Establish a unique coordinate system for the entire area, and accurately locate key nodes such as storage locations, passageways, connection points, and outbound exits. Based on the confidentiality level, physical and chemical properties, usage scenarios, and management requirements of aviation consumables, the warehouse is divided into four functional zones: a confidential area, a general area, a precision components area, and a hazardous chemicals area. Physical isolation and access control are implemented between each zone, forming a storage pattern with clear zoning, classification, and responsibilities. Anti-interference aviation-specific RFID tags (working frequency band 920-925MHz, protection level IP67) are uniformly affixed to each zone, shelf layer, standard storage location, and key passage nodes as storage location identification anchors and area markers. A positioning QR code grid and physical landmarks are laid on the ground at fixed intervals of 1.5m × 1.5m. A visual navigation reference network is constructed by integrating the ground QR code grid, physical landmarks, vehicle-mounted vision system, and inertial navigation.

[0096] Step 2: Construct a central management platform: Establish an anti-interference identification mechanism, perform pre-initialization of the platform, manage consumable information, inventory accounts and alarm rules in the warehouse through the WMS warehouse management system, receive WMS instructions through the RCS robot scheduling system, perform storage / retrieval task parsing, automatically generate AGV paths based on the parsing results, and coordinate multi-vehicle scheduling.

[0097] When establishing an anti-interference identification mechanism, the following should be included:

[0098] A wave-absorbing material cover is installed on the AGV forklift reader, and the reading and writing distance is controlled to be ≤30cm based on the physical characteristics of near-field communication.

[0099] The dynamic Q algorithm is used to prevent collisions. When Q=2, there are 4 time slots, which can simultaneously count up to 4 tags. In order to cope with the sharp increase in the probability of signal collision when more than 3 tags are identified at a time under extreme metal reflection environment, only 3 tags in the target area are authorized to be activated at the same time.

[0100] During each identification, the RSSI signal strength of the tag is captured simultaneously, and the visual camera is aligned with the QR code to confirm the location number. The tag with the strongest signal is the one that the forklift is facing, and the location ID decoded from it must be completely consistent with the QR code here. If the RSSI and location mapping are contradictory, such as a strong signal but an incorrect location code, a reread is triggered.

[0101] During each identification process, the AGV simultaneously collects the RSSI signal strength of the location tag and the location number read by the vision camera from the QR code on the ground. The system performs a mandatory consistency check between the location ID decoded from the tag with the strongest RSSI signal and the location ID read from the QR code. Only when the two are completely consistent is the current location identification considered valid. If there is a discrepancy (e.g., strong signal but incorrect location code) or any missing information that prevents mapping, the rereading mechanism is immediately triggered to re-execute the identification process until the verification passes or an anomaly is reported.

[0102] The platform's initial setup includes setting up WMS, RCS initialization, RFID point deployment, warehouse map modeling, scanning and binding all RFID anchor points for inbound storage, AGV path calibration, alarm threshold configuration (expiration advance notice days, inventory upper and lower limits), and permission role settings.

[0103] WMS Setup: Enter the unique RFID tag IDs for all aviation consumables, bind consumable types, specifications, batches, expiration dates, prohibited storage areas, priority access rules, and inventory upper and lower limit alarm thresholds. Draw an electronic map of the warehouse, dividing it into storage locations, aisles, and work points. Bind a corresponding RFID tag to each storage location. When setting alarm rules, set expired, out-of-stock, over-storage, incorrect storage area, and mixed storage as alarm trigger conditions.

[0104] RCS system initialization: Import warehouse map, aisle paths, restricted areas, docking points, and workstation coordinates; RCS automatically / manually plans fixed patrol paths for the entire site; input all AGV numbers, equipment status, load, and travel parameters; establish an AGV resource pool; configure multi-AGV avoidance rules, intersection priorities, congestion strategies, and anti-collision safety distances; complete the interface connection between WMS and RCS to achieve two-way communication of tasks, status, and commands.

[0105] RFID point deployment: RFID readers are deployed at each storage location, entrance / exit, and connection point to identify consumable labels and storage location labels.

[0106] When modeling the warehouse map, the warehouse floor plan CAD drawings are drawn based on the actual measured dimensions through on-site surveys. The warehouse area, zones, shelf rows, standard storage locations, main and auxiliary aisles, turning nodes, intersections, connecting platforms, charging locations, and restricted areas are marked. The CAD drawings are imported into the WMS system map editor to build the warehouse model. The RSC reuses the WMS warehouse drawings to supplement the movement paths. The actual AGV movement points are calibrated on-site to correct map coordinate deviations and generate a unified electronic map.

[0107] When all RFID anchor points are scanned and bound to the warehouse, aviation-grade RFID anchor tags are pre-assigned to each standard warehouse location. Each tag has a unique EPC code. The tag ID is pre-entered in the background and a blank code file is preset to complete the tag prefabrication and coding. The tags are then affixed to each warehouse location on-site according to the shelf layer and warehouse location number. Using an industrial RFID handheld device, the RFID tag ID of each warehouse location is read, and the corresponding warehouse location code is matched with the system warehouse location code on the handheld interface to complete the one-to-one binding. After binding, the data is uploaded to the warehouse management system (WMS) in batches. The system automatically associates the types of consumables allowed to be stored in that warehouse location, prohibited materials, and inventory upper and lower limit rules. By using a fixed RFID reader in the warehouse area to read the cards, the entire warehouse location is traversed to check for missing bindings, duplicate bindings, and invalid tags. Abnormal points are highlighted to prompt rebinding, and the verification is completed.

[0108] During the storage / retrieval task parsing, staff enter aviation consumables warehousing / retrieval documents, which are then identified by the WMS as storage / retrieval tasks. If consumables need to be stored, the WMS automatically allocates target storage locations according to the storage strategy, and determines the release order of consumables in the same area according to the first-in, first-out (FIFO) or expiry date priority (FEFO) rules. If consumables need to be retrieved, the WMS selects the only available storage location according to FEFO / FIFO, generates an outbound task, and breaks down the overall task into a series of executable sub-instructions. The WMS then calls the RCS task distribution interface through a standardized API to transmit the handling instructions to the RCS.

[0109] After receiving the work instructions from the WMS, the RCS parses the instructions, extracts the location information of the start and end points, and determines the location of the goods and vehicles to be transported. The RCS further breaks down the individual work tasks issued by the WMS (such as picking up goods from storage location A-15 and moving them to the outbound gate) into specific action sequences that the AGV can understand. The RCS comprehensively considers the urgency of the task, the current position of the AGV, the load status, the remaining power, the estimated execution time, and the path congestion, and automatically matches the optimal execution vehicle. The RCS plans the path from the current position to the start point for the selected AGV, and then plans the path from the start point to the end point. When multiple AGVs are scheduled, dynamic avoidance adopts a time window and path reservation mechanism, allowing only one AGV to enter each lane at the same time, and introducing traffic control points.

[0110] The storage strategy is as follows: warehouse area compliance, location availability, FEFO / FIFO sorting, and optimal coordinate calculation. Among them, warehouse area attribute matching is based on confidential / ordinary / precision / hazardous chemical partitioning, and consumables are placed in the warehouse area that can be stored. Full, fault, and prohibited storage areas are automatically skipped. WMS calculates the optimal storage location based on coordinates, path, and AGV efficiency. At the same time, storage locations are bound to RFID anchor points to prevent mixed or misplaced storage.

[0111] When automatically allocating storage locations, the WMS reads the consumable RFID tags to obtain the type, specifications, batch, expiration date, and prohibited storage areas of the consumables. It first filters out prohibited and faulty locations. The WMS calculates the optimal storage location coordinates according to rules such as shortest path, highest AGV efficiency, and concentration of similar types. The WMS locks the storage location, marks it for storage, and sends the coordinates to the RCS to schedule the AGV.

[0112] When calculating the optimal storage location coordinates, the candidate storage locations are scored according to three dimensions: shortest path, highest AGV efficiency, and similar type. The one with the highest total score is selected. The WMS traverses all legal candidate storage locations, calculates Score(i) for each storage location, selects the storage location with the largest Score, and outputs the X, Y, and Z three-dimensional coordinates of the storage location, which are the optimal storage location coordinates.

[0113] Score(i)=α×S_path(i)+β×S_eff(i)+γ×S_cluster(i);

[0114] α+β+γ=1;

[0115] In the formula, Score(i) is the comprehensive score of storage location i, S_path(i) is the path distance score, S_eff(i) is the AGV efficiency score, S_cluster(i) is the score of centralized storage of the same type, α is the path distance weight coefficient, β is the AGV efficiency weight coefficient, γ is the centralized storage weight coefficient of the same type, and α is 0.4, β is 0.3, and γ is 0.3.

[0116] Step 3, Storage Execution: The WMS warehouse management strategy in Step 2 is invoked to complete the intelligent allocation of storage locations. The RCS robot scheduling system issues instructions to schedule AGVs to complete the automatic storage operation of aviation consumables.

[0117] When performing a save task:

[0118] Operators scan the RFID tags on the outer packaging of consumables, or automatically identify consumable information through the RFID reader at the warehouse access control. The WMS system parses the consumable type, attributes, and storage requirements, and matches them with the corresponding permitted storage areas. The system simultaneously reads the production date and expiration date of the consumables, sorts them according to the expiration date priority rule, uses the WMS to calculate the optimal storage location coordinates, automatically excludes full, faulty, and prohibited storage locations, and sends the storage location information and operation instructions to the RCS.

[0119] After receiving the instruction, the AGV travels to the warehouse entrance and uses its built-in anti-interference reader to read the consumable tags for secondary verification. If the tag information does not match the system instruction, an alarm is immediately triggered and the goods are refused. After the information is verified, the AGV's forks pick up the material and navigate to the target storage location along the planned path. During navigation, a preset threshold for lateral or heading angle deviation is used. The AGV continuously scans the ground QR code grid with its onboard vision camera, reading the QR code coordinates of its current position in real time and comparing them with the expected position in the planned path. If the detected lateral or heading angle deviation exceeds the set threshold, the AGV control system automatically adjusts the driving parameters to correct the deviation, ensuring that the AGV always runs along the predetermined trajectory. After reaching the designated location, the AGV reads the fixed RFID anchor tag on the storage location for location confirmation. If the anchor ID does not match the instruction storage location, the AGV immediately stops and reports an anomaly to the platform. If the anchor ID matches but multiple QR code recognition failures occur during the visual correction process (e.g., three consecutive QR codes are not read), the AGV reduces its driving speed and reports a warning to the platform.

[0120] The AGV smoothly places the consumables into the target storage location, writes the current storage location number into the user area of ​​the consumable's RFID tag, and completes the binding between the consumable and the storage location. After the placement is completed, the AGV returns to its standby position, and the WMS system synchronously updates the storage location occupancy status and inventory data, and fully records the log of this inbound operation.

[0121] Retrieve and execute: Call the WMS management strategy in step two to lock the target storage location, and the RCS robot scheduling system issues instructions to schedule AGVs to complete the automatic outbound operation of aviation consumables;

[0122] When executing the retrieval task:

[0123] The operator enters the name or code of the consumables to be picked up. The WMS system queries the inventory list in real time, locks the only available storage location according to the first-in-first-out or expiration date priority rule, and marks the storage location as "outgoing" to avoid duplicate outgoing. After receiving the task, the RCS generates the optimal picking route based on the real-time road conditions in the warehouse. When there are other tasks in the route, the route is automatically adjusted and dynamically avoided to ensure the safety of multi-vehicle collaboration.

[0124] When RCS performs multi-vehicle scheduling, a time window and path reservation mechanism is used to achieve dynamic avoidance: a time window constraint is set for each roadway, allowing only one AGV to enter at the same time; at the same time, traffic control points are introduced to uniformly schedule key areas such as intersections and narrow passages. When multiple AGVs have conflicting paths, RCS automatically adjusts the path or arranges a waiting sequence according to task priority, vehicle status and path occupancy, thereby avoiding collisions and deadlocks and ensuring the safe operation of multi-vehicle collaboration.

[0125] The AGV travels along the planned path to the target storage location. It first reads the fixed RFID anchor tag at the location to confirm its position. After the anchor information matches, it reads the consumable tags on the storage location. If multiple tags are detected simultaneously, it's considered adjacent interference. The AGV automatically reduces its antenna power, limiting the reading range to its own storage location to ensure only the target consumable is identified. After confirming the information is correct, the AGV retrieves the consumable and reads the tag again for verification. If the consumable ID does not match the WMS command, an alarm is immediately triggered and operation is paused, awaiting manual intervention.

[0126] The AGV transports consumables to the outbound gate. The RFID reader at the outbound gate automatically identifies the tag and confirms that the consumables have left the warehouse normally. The WMS system simultaneously deducts the corresponding inventory, clears the occupancy status of the target storage location, and completes the outbound process. If the access control fails to read the consumable tag, it is determined to be an abnormal departure. The system automatically starts the AGV to return to the storage location for re-verification to avoid loss or misdelivery of consumables.

[0127] Step 4: Warehouse Information Update and Alarms: After each successful / failed access to consumables, the WMS immediately updates the database and synchronizes it to the AGV task queue (RCS is a list of tasks to be executed maintained by each AGV). The AGV periodically patrols the warehouse at low speed according to the path planned during the initialization of the RCS system in Step 2. When passing a storage location, it reads the storage location tag and consumable tag in anti-interference mode. If it finds that the consumables are expired, out of stock, overstocked, stored in the wrong area, or mixed, it sends an alarm message to the central management platform, triggering different levels of alarms. The platform displays the alarm information and takes different remedial measures.

[0128] Alarm levels are divided into three categories: low risk, medium risk, and high risk.

[0129] The low-risk trigger condition is: the current date of the consumable is greater than or equal to the expiration date of the consumable minus the number of days in advance for the warning set in the background. The WMS traverses the inventory of consumables daily / in real time, compares the current date of the consumable with the expiration date, and when an alarm is triggered, the WMS marks it as "near expiration", a pop-up window is displayed on the central control screen, and the AGV only reports the alarm without stopping. When the goods are issued, this batch is given priority. The warehouse manager checks the warning list and contacts the user department to collect the goods in advance or transfer them to another warehouse. If it is determined that the goods cannot be used up before the expiration date, the process can be manually switched to scrapping or the supplier can be contacted for return or replacement. After the process is completed, the warning is automatically lifted after confirmation in the WMS.

[0130] The trigger conditions for medium risk are: inventory quantity is lower than the lower limit (out of stock), or the inventory of a storage location / category is greater than the upper limit (overstock). When the current inventory level triggers an alarm, the central control screen will highlight it, suspend the inbound of the storage location (overstock) or suspend the outbound of the consumable (out of stock), and issue a short prompt sound without sounding the alarm siren. When handling manually, if there is an out of stock, verify the inventory and initiate procurement / transfer; if there is an overstock, manually move the storage or adjust the capacity limit; after handling, turn off the alarm in the WMS.

[0131] The high-risk trigger conditions are: expired consumables, incorrect storage area, and mixed storage. If the type of consumables read on-site does not match the zone attribute of the current storage location, the system determines it is an incorrect storage area. If two or more different types of consumables are identified in the same storage location, it is determined to be mixed storage. When a high-risk alarm is triggered, the consumables and storage location are locked, access is prohibited, a full-area audible and visual alarm is activated, the AGV immediately stops and reports coordinates, photos, and a list of tags; a pop-up window on the central control screen flashes, and the warehouse manager arrives to classify and handle the situation. Expired materials are placed in the isolation area for disposal, misplaced materials are moved to the correct zone, mixed materials are separated and returned to their original locations, and the alarm is deactivated by the management personnel after the WMS information is updated.

[0132] An intelligent warehouse AGV collaborative RFID aviation consumables automatic storage and retrieval system, employing the aforementioned intelligent warehouse AGV collaborative RFID aviation consumables automatic storage and retrieval method, includes a physical environment and identification module, a central management module, a warehouse management module, an AGV scheduling module, an automatic guided transportation module, and an update and alarm module;

[0133] The physical environment and identification module is used to perform grid-based spatial marking and functional zoning of the warehouse. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information.

[0134] The central management module is used to establish an anti-interference identification mechanism, adopts the dynamic Q algorithm for anti-collision processing, and sets an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the cargo location tag and the QR code information read by the vision camera are collected in parallel, and the consistency of the two is verified.

[0135] The warehouse management module is used to receive inbound and outbound instructions, automatically allocate the optimal storage location for aviation consumables according to the preset storage strategy, or lock the target storage location according to the rules.

[0136] The AGV scheduling module is used to schedule the optimal AGV to the designated location based on the task instructions issued by WMS and the real-time status of the AGV. It adopts a time window and path reservation mechanism to plan the AGV's travel path, effectively avoiding conflicts or congestion when multiple vehicles are running.

[0137] The automated guided transport module is responsible for moving consumables from the inbound entrance to the target storage location, or retrieving consumables from the storage location and transporting them to the outbound entrance. It continuously scans the QR code grid laid on the ground with the vehicle-mounted vision camera and compares it with the planned route. If a deviation is found, it will automatically correct the deviation to ensure accurate driving.

[0138] The update and alarm module is used to update inventory data in real time after the storage and retrieval tasks are completed, continuously monitor the inventory status, and trigger alarm information of different levels according to preset rules.

[0139] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions or improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for automatic storage and retrieval of aviation consumables in a smart warehouse using AGV and RFID, characterized in that, Includes the following steps: Step S1: Equipment deployment and shelf identification settings. The warehouse is marked with a grid space and functional zones. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information. Step S2: Construct a central management platform, establish an anti-interference identification mechanism, adopt the dynamic Q algorithm for anti-collision processing, and set an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the storage location tag and the storage location number read by the vision camera pointing at the ground QR code are collected in parallel. The storage location ID decoded from the tag with the strongest RSSI signal is forced to be consistent with the storage location ID read from the QR code. Step S3: Execute the storage task. The WMS system warehouse management system receives the storage instruction and automatically allocates the optimal storage location for aviation consumables according to the storage strategy. The RCS robot scheduling system schedules AGVs to transport the consumables to the target storage location. After the AGV confirms the storage location through the anti-interference identification mechanism, it puts the consumables on the shelf and writes the storage location information into the RFID tag of the consumables. When the retrieval task is executed, the WMS system locks the target storage location according to the preset rules, the RCS system dispatches the AGV to the target storage location, and after the AGV confirms the storage location and consumable information through the anti-interference identification mechanism, it retrieves the consumables and transports them to the outbound port. Step S4: Warehouse information update and alarm. The WMS system updates inventory data in real time and monitors the inventory status according to preset alarm rules, triggering alarm information of different levels.

2. The intelligent warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, Step S1 is implemented in the following manner: Step S11: Deploy the physical environment of the warehouse, implement unified grid-based spatial marking of the internal shelving layout, passageways, work areas and functional points of the warehouse, establish a unique coordinate system for the entire area, divide the warehouse into zones according to the confidentiality level and physical and chemical properties of aviation consumables, implement physical isolation and access control between zones, and lay positioning QR code grids and physical landmarks on the ground at fixed intervals to build a visual navigation reference network; Step S12: Location identification and binding. Read the unique EPC code of the anti-interference RFID anchor tag at each key node, bind it one-to-one with the system location code, and then upload it to the WMS system in batches to complete the binding and storage of location RFID anchors and location coordinate information.

3. The intelligent warehouse AGV-assisted RFID-based automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, Step S2 is implemented in the following manner: Step S21: Hardware protection. Install a wave-absorbing material cover on the AGV's forklift reader / writer to control the reading and writing distance within a threshold range based on the physical characteristics of near-field communication, thereby reducing physical interference. Step S22: Algorithm anti-collision, using dynamic Q algorithm for anti-collision processing, setting only authorized activation of tags within the target area at the same time, and performing multi-tag signal collision processing; Step S23: Dual verification. During each identification, the RSSI signal strength of the cargo location tag and the ground QR code information read by the vision camera are collected in parallel. The cargo location ID decoded from the tag with the strongest RSSI signal is subjected to a mandatory consistency verification with the cargo location ID read from the QR code.

4. The intelligent warehouse AGV-assisted RFID-based automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, The anti-interference identification mechanism established in S2 also includes an environmental electromagnetic fingerprint and an adaptive power adjustment mechanism: Environmental monitoring nodes are deployed inside the warehouse to sense the temperature, humidity and electromagnetic interference intensity in real time, forming an environmental electromagnetic fingerprint. When the change in environmental parameters exceeds the preset threshold, the system automatically fine-tunes the transmission power and receiving sensitivity threshold of the RFID reader based on the current environmental electromagnetic fingerprint in order to offset the impact of environmental changes on RFID signal attenuation. After step S2 builds the central management platform, the next step is to establish a warehouse-level digital twin system: The system collects the location information, posture information, battery level, and cargo location status of the AGV in real time, and maps this information to the virtual warehouse model in milliseconds. Before executing batch storage and retrieval tasks, a task rehearsal is performed in the virtual warehouse model to detect whether there is a risk of deadlock or congestion in the path planning. Only when there are no conflicts in the pre-rehearsal results will the central management platform issue execution instructions to the physical AGVs to achieve synchronous control between the virtual and physical systems.

5. The intelligent warehouse AGV-assisted RFID-based automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, The storage task in step S3 is performed in the following manner: Step S31: Task generation and storage location allocation. After scanning the consumables label, the WMS system parses the consumables attributes and uses a three-dimensional weighted scoring method to comprehensively consider three dimensions: shortest path, highest AGV efficiency, and centralized storage of similar items. Based on the storage strategy, the optimal storage location is automatically calculated and allocated. Step S32: AGV handling and navigation. The RCS system dispatches the AGV to the warehouse entrance. The AGV performs a second check on the consumable label and picks up the material. While traveling along the planned path to the target storage location, the vehicle-mounted vision camera continuously scans the ground QR code and compares it with the planned path. If the deviation exceeds the threshold, it will automatically correct the deviation. Step S33: Precise placement and binding. After the AGV arrives at the target storage location, it reads the RFID anchor tag of the storage location and verifies the location with the visual QR code. It then places the consumables smoothly into the target storage location, writes the current storage location number into the user area of ​​the consumable RFID tag, completes the binding of the consumables to the storage location, and updates the inventory data to the WMS system.

6. The intelligent warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables according to claim 5, characterized in that, In step S31, the optimal storage location is automatically calculated using the following method: WMS iterates through all legal candidate storage locations, calculates Score(i) for each storage location, selects the storage location with the largest Score, and outputs the X, Y, and Z three-dimensional coordinates of the storage location, which are the optimal storage location coordinates. Score(i)=α×S_path(i)+β×S_eff(i)+γ×S_cluster(i); α+β+γ=1; In the formula, Score(i) is the overall score of storage location i, S_path(i) is the path distance score, S_eff(i) is the AGV efficiency score, S_cluster(i) is the score of centralized storage of the same type, α is the path distance weight coefficient, β is the AGV efficiency weight coefficient, and γ is the centralized storage weight coefficient of the same type. Before step S33, which involves precise shelf placement and binding, there is also an AI-based vision-based consumable appearance defect detection step: Before the AGV places the consumables to the target location, a high-definition industrial camera mounted on the AGV is used to capture surface images of the aviation consumables. The surface image is input into a pre-trained convolutional neural network model to analyze whether the consumables have defects such as rust, damaged packaging, or blurred labels. If the test results show that the consumables have an abnormal appearance, the AGV will automatically stop the shelving action and transfer the consumables to the quality inspection isolation area, while sending an abnormal appearance alarm message to the WMS system.

7. The intelligent warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, The retrieval task in step S3 is performed in the following manner: Step S41: Task locking and scheduling. The WMS system prioritizes locking the storage locations of consumables that are close to their expiration date based on the outbound instructions and preset rules, generates an outbound task and sends it to the RCS system. The RCS system uses a time window and route reservation mechanism to avoid conflicts between multiple vehicles, and schedules the optimal AGV to the target storage location based on the task instructions and the real-time status of the AGV. Step S42: Accurate identification and extraction. After the AGV arrives at the target location, it confirms that the location and consumable information are correct through an anti-interference identification mechanism, and then performs the extraction action. If the AGV senses multiple tags at the same time when extracting consumables, it will automatically reduce the antenna power and limit the reading range to the target location. Step S43: Handling and Outbound. After the AGV retrieves the consumables, it continuously scans the QR code on the ground with a vision camera to correct its path. Upon reaching the outbound port, the AGV places the consumables in the designated area and reports the task completion status to the WMS system.

8. The intelligent warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables according to claim 1, characterized in that, Step S4 is implemented in the following manner: Step S51: Warehouse information update. After each storage and retrieval task is completed, the WMS system updates the inventory data in real time. The updated data includes the current status of consumables, storage location information, quantity, and operation timestamp. Records are kept using blockchain-based distributed ledger technology: The key operation data for each task deposit, retrieval, and alarm information is hashed and uploaded to the consortium blockchain. Smart contracts are used to verify permissions to modify inventory data, and changes to inventory data are verified on-chain and the operator's identity is recorded. Step S52: Multi-level alarm monitoring. The WMS system continuously monitors the inventory status according to preset alarm rules and triggers alarm information of different levels when the conditions are met.

9. The intelligent warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables according to claim 8, characterized in that, In step S52, the alarm levels are divided into three categories: low risk, medium risk, and high risk. The low-risk alarm is triggered when the consumables are nearing their expiration date; The trigger condition for a medium-risk alert is that the inventory quantity is lower than the lower limit or higher than the upper limit; The high-risk alarm is triggered when consumables have expired, are stored in the wrong area, or are mixed up.

10. A smart warehouse AGV-assisted RFID automatic storage and retrieval system for aviation consumables, employing the smart warehouse AGV-assisted RFID automatic storage and retrieval method for aviation consumables as described in any one of claims 1-9, characterized in that, It includes a physical environment and identification module, a central management module, a warehouse management module, an AGV scheduling module, an automated guided transportation module, and an update and alarm module; The physical environment and identification module is used to perform grid-based spatial marking and functional zoning of the warehouse. Physical isolation and access control are implemented between the functional zones. A unique RFID anchor tag is set for each storage location and bound to the storage location coordinate information. The central management module is used to establish an anti-interference identification mechanism, adopts the dynamic Q algorithm for anti-collision processing, and sets an upper limit on the number of tags authorized to be activated in a single identification. During each identification, the RSSI signal strength of the location tag and the location number read by the vision camera pointing at the ground QR code are collected in parallel. The location ID decoded from the tag with the strongest RSSI signal is forced to be consistent with the location ID read from the QR code. The warehouse management module is used to receive inbound and outbound instructions, automatically allocate the optimal storage location for aviation consumables according to the preset storage strategy, or lock the target storage location according to the rules. The AGV scheduling module is used to schedule the optimal AGV to the designated location based on the task instructions issued by WMS and the real-time status of the AGV. It adopts a time window and path reservation mechanism to plan the AGV's travel path, effectively avoiding conflicts or congestion when multiple vehicles are running. The automated guided transport module is responsible for moving consumables from the inbound entrance to the target storage location, or retrieving consumables from the storage location and transporting them to the outbound entrance. It continuously scans the QR code grid laid on the ground with the vehicle-mounted vision camera and compares it with the planned route. If a deviation is found, it will automatically correct the deviation to ensure accurate driving. The update and alarm module is used to update inventory data in real time after the storage and retrieval tasks are completed, continuously monitor the inventory status, and trigger alarm information of different levels according to preset rules.

Images