Multi-protocol converged internet of things device intelligent gateway data conversion method and system

Abstract

The application relates to the technical field of data processing, and discloses a multi-protocol fusion Internet of Things device intelligent gateway data conversion method and system. The method comprises the following steps: collecting multi-protocol device data packets, extracting protocol features to construct a vector library; identifying protocol types and analyzing data based on the vector library to construct a data object set containing semantic labels; analyzing semantic correlation through a self-adaptive learning algorithm to establish a dynamic protocol semantic mapping matrix; converting data into a standard format according to the mapping matrix and recording matching degrees to form a target data pool; and extracting fusion data from the data pool, re-encoding output data frames according to a target protocol format. The application solves the problem of lacking protocol semantic understanding and self-adaptive learning capability in the existing multi-protocol fusion data conversion method, and improves the semantic consistency and conversion quality of data conversion between multi-protocol devices.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data conversion method and system for a multi-protocol integrated IoT device intelligent gateway. Background Art

[0002] Existing data conversion technology for intelligent gateways for IoT devices primarily utilizes static protocol mapping tables to handle multi-protocol convergence scenarios, implementing data format conversion between different protocols through pre-configured protocol conversion rules. Traditional methods rely on manually configured fixed mapping relationships for packet parsing and format conversion when interconnecting multi-protocol devices such as AGV controllers, intelligent warehouse equipment, and robotic scheduling systems. Protocol identification relies on simple header field matching, and data conversion utilizes a one-to-one static mapping model. This technology approach has been widely adopted in industrial IoT applications such as automotive manufacturing and intelligent warehousing, meeting basic multi-protocol data interoperability requirements.

[0003] However, the existing technology has significant shortcomings. First, the protocol scalability is poor. When new equipment is introduced into the industrial site or the protocol version is upgraded, the protocol adapter module needs to be redeveloped and deployed, and it cannot dynamically adapt to protocol changes. Second, the semantic consistency of data conversion cannot be guaranteed. The static mapping rules lack an understanding of the semantic correlation between protocols, which easily leads to the loss or offset of semantic information during the data conversion process. Third, there is a lack of an effective evaluation mechanism for conversion quality. The existing methods cannot quantitatively evaluate the accuracy and reliability of data conversion, making it difficult to detect and correct conversion errors in a timely manner.

[0004] Based on the aforementioned technical deficiencies, further analysis revealed that the problem with existing technologies lies in their lack of intelligent protocol semantic understanding and adaptive learning capabilities. Because static mapping rules cannot understand the semantic meaning of data from different protocols, when AGV position data needs to be associated and converted with warehouse storage location information, existing technologies can only perform format-level conversions and cannot establish semantic-level mapping relationships. This limitation results in a lack of intelligence and adaptability in inter-protocol data conversion. A deeper issue is that existing technologies lack dynamic learning and optimization mechanisms, and are unable to automatically adjust and optimize conversion strategies based on the data flow characteristics of actual application scenarios, limiting the application effectiveness and reliability of multi-protocol fusion systems in complex industrial environments. Summary of the Invention

[0005] The present application provides a multi-protocol fusion IoT device intelligent gateway data conversion method and system, which is used to solve the problem of lack of protocol semantic understanding and adaptive learning ability in existing multi-protocol fusion data conversion methods, and improve the semantic consistency and conversion quality of data conversion between multi-protocol devices.

[0006] In a first aspect, the present application provides a multi-protocol fusion IoT device intelligent gateway data conversion method, the multi-protocol fusion IoT device intelligent gateway data conversion method comprising:

[0007] Collect data packets from AGV controllers, intelligent warehouse equipment, and robot scheduling systems, extract protocol header features, data structure patterns, and timing features, and build a protocol feature vector library;

[0008] Identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags;

[0009] Analyzing the semantic relevance of different protocol data in the original data object set through a protocol semantic adaptive learning algorithm, automatically learning and establishing a dynamically updated protocol semantic mapping matrix;

[0010] Convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record the semantic matching value to obtain a target data pool;

[0011] The fused data is extracted from the target data pool according to the target device protocol specification, and the output protocol data frame is re-encoded according to the target protocol data frame format.

[0012] In a second aspect, the present application provides a multi-protocol integrated IoT device intelligent gateway data conversion system, the multi-protocol integrated IoT device intelligent gateway data conversion system comprising:

[0013] A construction module is used to collect data packets from AGV controllers, intelligent stereoscopic warehouse equipment, and robot scheduling systems, extract protocol header features, data structure patterns, and timing features, and build a protocol feature vector library;

[0014] A parsing module, configured to identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags;

[0015] An analysis module, configured to analyze the semantic relevance of different protocol data in the original data object set by using a protocol semantic adaptive learning algorithm, and automatically learn and establish a dynamically updated protocol semantic mapping matrix;

[0016] A recording module, configured to convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record semantic matching values ​​to obtain a target data pool;

[0017] The output module is used to extract the fused data from the target data pool according to the target device protocol specification, and re-encode the output protocol data frame according to the target protocol data frame format.

[0018] In the third aspect, a multi-protocol integrated IoT device intelligent gateway data conversion device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the multi-protocol integrated IoT device intelligent gateway data conversion device executes the above-mentioned multi-protocol integrated IoT device intelligent gateway data conversion method.

[0019] In a fourth aspect, a computer-readable storage medium is provided, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the above-mentioned multi-protocol fusion IoT device intelligent gateway data conversion method.

[0020] The technical solution provided by this application solves the problem of insufficient accuracy of traditional protocol identification methods by adopting the technical features of the protocol feature vector library. The protocol feature vector library is based on multi-dimensional feature extraction of protocol header features, data structure patterns and timing features. Compared with the single header field matching method of the existing technology, it significantly improves the accuracy and scalability of device identification in a multi-protocol environment. The construction of the original data object set introduces protocol semantic tags, so that the data parsing process not only stays at the format level, but also goes deep into the semantic level, laying the foundation for subsequent intelligent processing. The protocol semantic adaptive learning algorithm, as the core technical feature of this application, breaks through the limitations of traditional static mapping rules. By dynamically analyzing the semantic relevance of different protocol data, it automatically establishes and updates the semantic mapping relationship between protocols. This adaptive learning mechanism enables the system to continuously optimize the conversion strategy as the application scenario changes, and can adapt to the access of new protocols and new devices without manual intervention. The technical feature of the dynamically updated protocol semantic mapping matrix ensures the timeliness and accuracy of the mapping relationship. When a new semantic association pattern is detected, the matrix structure can be automatically adjusted. This dynamic feature solves the fundamental problem of poor protocol scalability in the existing technology.

[0021] The combination of standard data format conversion and semantic matching record not only achieves the unification of data formats between heterogeneous protocols, but more importantly, establishes a quantitative evaluation mechanism for conversion quality. The recording of semantic matching values ​​enables each conversion operation to have traceable quality indicators. This feature fills the gap in the existing technology in terms of conversion quality assessment. The construction of the target data pool provides a high-quality data foundation for subsequent protocol adaptation by integrating multi-device data and quality assessment information. Compared with the direct conversion mode of traditional methods, this layered processing architecture improves the reliability and maintainability of the overall conversion process. The re-encoding and output technical features of the protocol data frame ensure that the converted data can be correctly identified and processed by the target device through precise target protocol format adaptation. This end-to-end conversion capability solves the problem of insufficient protocol compatibility in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a schematic diagram of an embodiment of a multi-protocol integrated IoT device intelligent gateway data conversion method in an embodiment of the present application;

[0024] Figure 2 This is a schematic diagram of an embodiment of a multi-protocol integrated IoT device intelligent gateway data conversion system in an embodiment of the present application;

[0025] Figure 3 This is a schematic block diagram of the structure of a multi-protocol integrated IoT device intelligent gateway data conversion device in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The embodiments of the present application provide a method and system for data conversion of an intelligent gateway for IoT devices with multi-protocol integration. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the multi-protocol integrated IoT device intelligent gateway data conversion method includes:

[0028] Step S101: Collect data packets from the AGV controller, intelligent three-dimensional warehouse equipment, and robot scheduling system, extract protocol header features, data structure patterns, and timing features, and construct a protocol feature vector library;

[0029] Step S102: Identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags;

[0030] Step S103: Analyze the semantic relevance of different protocol data in the original data object set through a protocol semantic adaptive learning algorithm, automatically learn and establish a dynamically updated protocol semantic mapping matrix;

[0031] Step S104: Convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record the semantic matching degree values ​​to obtain a target data pool;

[0032] Step S105 : extracting fused data from the target data pool according to the target device protocol specification, and re-encoding the output protocol data frame according to the target protocol data frame format.

[0033] It is understandable that the execution subject of this application can be a multi-protocol integrated IoT device intelligent gateway data conversion system, or a terminal or server, which is not limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0034] Specifically, a multi-protocol identification foundation is established through a protocol feature vector library. The intelligent gateway listens to the communication ports of the AGV controller, intelligent vertical library equipment, and robot scheduling system, captures the data packets sent by each device, and performs cache processing to form a raw data packet set. In the protocol header feature extraction process, the gateway extracts key information from the first 16 bytes of each data packet. The AGV controller uses Modbus protocol data packets containing function code fields, the intelligent vertical library equipment uses OPC-UA protocol containing node identifiers, and the robot scheduling system uses MQTT protocol containing topic names. These features constitute the protocol header feature set. Vectorization encoding processing converts the extracted feature values into digital vector representations according to the data length field position, checksum calculation method, and frame structure arrangement order. Each protocol type corresponds to a unique feature vector stored in the protocol feature vector library.

[0035] In the device protocol type identification stage, the protocol header features of the data packets to be identified are matched with the standard feature vectors stored in the protocol feature vector library to calculate the similarity. The protocol type is determined by calculating the Euclidean distance between the feature vectors. The data packet field segmentation and analysis are performed according to the identified protocol type. The Modbus protocol data packet is segmented according to the address domain, function code domain, data domain, and verification domain. The OPC-UA protocol is segmented according to the message header, security header, sequence header, and message body. The MQTT protocol is segmented according to the fixed header, variable header, and message payload. The parsed field information forms a parsed field set containing device address, function instruction, and data content. The semantic classification and labeling process analyzes the function instruction type and data content attribute. The read operation corresponds to data query function, the write operation corresponds to data modification function, and the status query corresponds to device monitoring function. Each data field obtains the corresponding semantic label according to its function attribute. The object encapsulation process groups the data fields with semantic labels according to the device source and arranges them according to the time sequence to form a raw data object set containing protocol semantic labels.

[0036] The protocol semantic adaptive learning algorithm establishes a semantic correlation model by analyzing the original data object set, the protocol semantic sample library collects the corresponding relationship between AGV position data and stereoscopic warehouse location data, and records the association mode of robot task instructions and device control parameters. The semantic similarity calculation is based on two dimensions of semantic tag matching degree and data content correlation degree, the semantic tag matching degree is calculated by comparing the functional attributes of different protocol data objects, the data content correlation degree is calculated by analyzing the numerical range and trend of data values, and the comprehensive score of the two dimensions constitutes the semantic correlation score matrix. In the process of establishing the protocol semantic mapping matrix, the X-axis value of the AGV controller position coordinates corresponds to the row number of the stereoscopic warehouse equipment location identification, the Y-axis value corresponds to the column number, the task type field of the robot scheduling system corresponds to the function instruction field of each execution device, forming the mapping relationship from the source protocol field to the target protocol field. The dynamic updating mechanism adjusts the mapping relationship weight value according to the newly received original data object, and when a new mode of AGV and shelf interaction is detected, the corresponding mapping relationship is automatically added in the protocol semantic mapping matrix.

[0037] The data format conversion stage performs data conversion between heterogeneous protocols according to the protocol semantic mapping matrix, the field information of the source protocol data object is matched with the source protocol field in the mapping matrix to determine the corresponding target protocol field mapping relationship. The conversion of AGV controller position coordinate data into the location code format recognized by stereoscopic warehouse equipment involves coordinate system conversion, which converts the absolute coordinates of AGV into relative position codes of shelves, and the conversion of task instructions of robot scheduling system into control command formats accepted by each execution device requires instruction format adaptation. Semantic matching degree calculation compares the semantic consistency of field between standard data object and original protocol data object, the calculation process includes field name matching, data type consistency and numerical range rationality test, consistent values are recorded in the conversion log as semantic matching degree. The target data pool stores the converted standard data objects and corresponding semantic matching degree values, and manages multi-device fusion data in time stamp order.

[0038] The protocol data frame output process first queries the matching fusion data from the target data pool according to the target device protocol type, and selects the corresponding standard data object according to the device address and functional requirements. Protocol adaptation data generation converts the data content in standard format into the data type required by the target protocol, including data precision adjustment, encoding format conversion and byte sequence adjustment. Data frame assembly adds frame header information according to the frame structure requirements of the target protocol, arranges the data fields according to the byte sequence specified by the protocol, calculates and adds the check code, and constructs the standard protocol data frame. The communication interface sends the assembled protocol data frame to the target device through the corresponding physical interface, and records the sending time stamp and transmission state information at the same time.

[0039] In a specific embodiment, the process of performing step S101 can specifically include the following steps:

[0040] Listen to the communication port of the AGV controller, intelligent three-dimensional warehouse equipment and robot scheduling system, cache the received data packets to obtain the original data packet set;

[0041] Based on the first 16 bytes of each data packet in the original data packet set, the protocol header identification extraction process is performed to obtain the protocol header feature set containing the function code, node identifier and topic name;

[0042] The feature values in the protocol header feature set are vectorized and encoded according to the data length field position, check code calculation method and frame structure arrangement order to obtain the protocol feature vector;

[0043] According to the protocol feature vector, a feature index table is established for storage management processing to obtain a protocol feature vector library recording the feature vectors corresponding to each protocol type.

[0044] Specifically, during the communication port listening process, the intelligent gateway simultaneously listens to the RS485 serial port of the AGV controller, the Ethernet port of the intelligent three-dimensional warehouse equipment and the WiFi interface of the robot scheduling system through socket programming technology, and adopts a multi-thread concurrent listening mechanism to ensure real-time capture of data packets from each device. The cache processing mechanism stores the received data packets in the memory buffer according to the arrival time sequence, and each data packet record contains the source device identifier, the receiving timestamp, the data packet length and the original byte stream content, forming an original data packet set. The buffer uses a ring queue structure to manage data packet storage, and automatically covers the earliest data packet when the buffer is full, ensuring the continuity and real-time nature of the data stream. The data packet deduplication processing identifies duplicate data packets by calculating the hash value of each data packet, avoiding data redundancy caused by network retransmission.

[0045] The protocol header identification extraction process analyzes the first 16 bytes based on the frame structure characteristics of different protocols. The first byte of the Modbus protocol is the device address, the second byte is the function code, the first four bytes of the OPC-UA protocol are the message type identifier, and the first byte of the MQTT protocol contains the message type and flag information. The function code extraction identifies the read-hold register, write-single register, read-input register and other operation types for the Modbus protocol, the node identifier extraction parses the unique identifier information of the server node for the OPC-UA protocol, and the topic name extraction parses the topic string published and subscribed for the MQTT protocol. The protocol header feature set construction process stores the extracted function code, node identifier and topic name according to the protocol type, and establishes a correspondence table between the protocol type and the header feature. The feature value verification checks whether the extracted header feature meets the protocol specification requirements, and filters out the data packets with format errors or damage.

[0046] The vectorized encoding process converts the character strings and numerical features in the protocol header feature set into numerical vector representations. The data length field position encoding records the byte offset of the key data field in the data packet. The checksum calculation method encoding identifies the type of checksum algorithm used by different protocols. The frame structure arrangement order encoding describes the arrangement pattern of the data fields. The encoding process uses the one-hot encoding method to convert discrete protocol features into binary vectors of fixed length. Continuous numerical features are mapped to a numerical range of zero to one using normalization. The dimension of the protocol feature vector is designed to be 64 bits, of which the first 16 bits represent the protocol type, the middle 32 bits represent the header feature information, and the last 16 bits represent the structural features. The vector similarity calculation uses the Euclidean distance method to measure the degree of difference between different protocol feature vectors. The similarity threshold is set for accurate identification of the protocol type.

[0047] The feature index table establishment process categorizes and stores protocol feature vectors by protocol type. Each protocol type corresponds to a unique index number. The index table records information such as the protocol type number, protocol name, feature vector, and identification threshold. Storage management uses a hash table structure to quickly retrieve protocol feature vectors. The hash function calculates the hash value based on the first 16 bits of the protocol feature vector. Conflict resolution uses a chain address method to handle situations where different protocols are mapped to the same hash value. The protocol feature vector library supports dynamic expansion, automatically adding corresponding feature vector records when a new protocol type is detected. A version management mechanism records the update history of protocol feature vectors and supports feature vector updates during protocol version upgrades.

[0048] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0049] The protocol header features of the data packet to be identified are matched and calculated with the feature vectors stored in the protocol feature vector library to obtain the protocol type identification result;

[0050] Based on the protocol type identification result, the data packet is segmented and parsed according to the corresponding protocol specification to obtain a parsed field set containing the device address, function instruction and data content;

[0051] Perform semantic classification and tagging based on the functional instruction type and data content attributes in the parsed field set to obtain data fields with read operation semantic labels, write operation semantic labels, and status query semantic labels;

[0052] The data fields are encapsulated into objects according to the device source and time sequence to obtain a set of original data objects containing protocol semantic tags.

[0053] Specifically, a similarity algorithm is used to compare the protocol header features of the data packet to be identified with the standard feature vectors stored in the protocol feature vector library. The Euclidean distance calculation method is used to measure the degree of difference between the feature vectors. The calculation process compares the 64-bit feature vector extracted from the data packet to be identified with the standard vector of each protocol type in the library bit by bit. The square sum of the differences in each bit is accumulated and squared to obtain the distance value. The protocol type corresponding to the standard vector with the smallest distance value is the identification result. When the distance value between the feature vector of the data packet sent by the AGV controller and the Modbus standard vector is 0.12, the distance value between the feature vector of the OPC-UA standard vector is 0.78, and the distance value between the feature vector of the MQTT standard vector is 0.85, the identification result is determined to be the Modbus protocol. The confidence assessment mechanism sets the identification threshold of 0.30. When the minimum distance value exceeds the threshold, it is marked as an unknown protocol type to avoid subsequent processing errors caused by misidentification.

[0054] The field segmentation and parsing process calls the corresponding protocol parsing rules based on the protocol type identification results. The Modbus protocol is segmented according to the format of device address occupying 1 byte, function code occupying 1 byte, data length occupying 1 byte, data content occupying N bytes, and CRC check occupying 2 bytes. The OPC-UA protocol is segmented according to the format of message type occupying 4 bytes, security policy occupying 4 bytes, message length occupying 4 bytes, and message body occupying M bytes. The MQTT protocol is segmented according to the format of fixed header occupying 1-4 bytes, topic length occupying 2 bytes, topic name occupying L bytes, and message payload occupying P bytes. The parsing process locates the start and end position of each field by byte offset, extracts the corresponding field content from the original byte stream, and converts it to the corresponding data type. The parsed field set records the name, data type, byte length, and specific value of each field to form a structured data representation.

[0055] Semantic classification and tagging analyzes the semantic meaning of data fields based on the function instruction type and data content attributes within the parsed field set. Read operation semantic tags correspond to Modbus protocol read instructions such as function codes 01, 02, 03, and 04; write operation semantic tags correspond to write instructions such as function codes 05, 06, 15, and 16; and status query semantic tags correspond to function codes related to device status monitoring and diagnostics. Semantic classification for the OPC-UA protocol is based on the service type field: the Read service is tagged with read operation semantics, the Write service is tagged with write operation semantics, and the Browse service is tagged with status query semantics. Semantic classification for the MQTT protocol is based on content analysis of topic names: topics containing the control keyword are tagged with write operation semantics, topics containing the status keyword are tagged with status query semantics, and topics containing the data keyword are tagged with read operation semantics. The semantic tag generation process converts the numerical encoding of the function instruction into a textual tag, while also recording the tag's confidence level and applicable scope.

[0056] Object encapsulation groups semantically tagged data fields according to device origin. Data fields originating from AGV controllers are encapsulated as AGV data objects, data fields originating from intelligent stereoscopic warehouse devices are encapsulated as stereoscopic warehouse data objects, and data fields originating from the robot scheduling system are encapsulated as scheduling data objects. Time series arrangement sorts data packets based on their reception timestamps, ensuring that data objects are organized in chronological order. The data object structure includes attributes such as device identification, protocol type, semantic label, data field set, timestamp, and data integrity check code. The encapsulation process uses JSON format to store the structured information of data objects, facilitating subsequent data processing and protocol conversion operations. The original data object collection manages each data object through a linked list structure, supporting dynamic addition and deletion operations.

[0057] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0058] Extract the semantic labels and data content of each protocol data object from the original data object set, and establish a protocol semantic sample library. The protocol semantic sample library contains samples of the semantic correspondence between AGV position data and three-dimensional warehouse storage location data, and robot task instructions and equipment control parameters.

[0059] The semantic similarity values ​​between different protocol data objects are calculated based on the protocol semantic sample library, and the semantic relevance scoring matrix is ​​obtained by calculating the semantic label matching degree and data content relevance.

[0060] Based on the score values ​​in the semantic relevance scoring matrix, a semantic mapping relationship between the data fields of the protocols is established. The location coordinate field of the AGV controller is mapped to the shelf identification field of the three-dimensional warehouse equipment. The task type field of the robot scheduling system is mapped to the function instruction field of each execution device to form a protocol semantic mapping matrix.

[0061] The mapping relationship weight values ​​in the protocol semantic mapping matrix are updated based on the newly received original data objects. When a new semantic association pattern is detected, the matrix structure is automatically adjusted to generate a dynamically updated protocol semantic mapping matrix.

[0062] Specifically, by traversing the original data object set, the semantic label and data content of each data object are extracted, a correspondence is established between the position coordinate data of the AGV controller and the shelf location coding data of the intelligent three-dimensional warehouse equipment, and the association pattern between the task instructions of the robot scheduling system and the control parameters of each execution device is recorded. The sample extraction process analyzes the X coordinate, Y coordinate, and orientation angle fields in the AGV data object, extracts the shelf number, layer number, and location number fields in the three-dimensional warehouse data object, and collects the task type, priority, and execution time fields in the robot scheduling data object. The semantic correspondence sample record format includes attributes such as source protocol type, source data field, target protocol type, target data field, and association strength. The association strength is calculated by counting the co-occurrence frequency of data fields in the same time window. The sample library is stored in a key-value pair structure, where the key is the semantic identifier of the data field and the value is a set of data records containing all associated information of the field.

[0063] Semantic similarity calculation analyzes the degree of association between data objects from different protocols based on data records in the protocol semantic sample library. Semantic label matching is calculated by comparing the semantic label types of data objects. The matching degree between read operation labels is 1.0, the matching degree between read operation labels and write operation labels is 0.3, and the matching degree between read operation labels and status query labels is 0.6. Data content relevance calculation analyzes the numerical range, changing trend, and temporal correlation of data fields. The degree of overlap between the X-value range of the AGV position coordinates and the row number range of the three-dimensional warehouse determines the correlation value. The negative correlation between robot task priority and device response time reflects the relevance of task scheduling. The semantic relevance scoring matrix construction process weights the semantic label matching and data content relevance according to the weight coefficient. The weight coefficient is dynamically adjusted based on the protocol type and application scenario to form a scoring matrix that reflects the strength of semantic association between data objects from different protocols.

[0064] The process of establishing the protocol semantic mapping relationship selects data object combinations with scores exceeding the threshold based on the semantic relevance scoring matrix. The mapping relationship between the AGV controller's location coordinate field and the stereoscopic warehouse equipment's shelf identification field is established based on the semantic similarity of spatial location. The mapping relationship between the robot scheduling system's task type field and the execution equipment function instruction field is established based on the semantic consistency of the operation function. The mapping relationship definition includes parameters such as the source field identifier, target field identifier, conversion rules, and mapping weights. The conversion rules describe the method for converting source field values ​​to target field values, and the mapping weights reflect the reliability of the mapping relationship. The protocol semantic mapping matrix uses a sparse matrix to store mapping relationships. The matrix row index corresponds to the source protocol data field, the column index corresponds to the target protocol data field, and the matrix element value represents the mapping weight. The matrix structure supports many-to-many mapping relationships. The same source field is mapped to multiple target fields, and the same target field accepts mappings from multiple source fields.

[0065] The dynamic update mechanism continuously adjusts the structure and weight values ​​of the protocol semantic mapping matrix based on newly received raw data objects. The update process analyzes the consistency of new data objects with existing mapping relationships. When the semantic association pattern of a new data object matches an existing mapping relationship, the corresponding mapping weight is increased. When a new semantic association pattern is detected, a new mapping relationship record is added to the matrix. The weight value update uses an exponential smoothing algorithm, combining the influence coefficient of the new data object with the attenuation coefficient of historical data to calculate the updated weight value. Matrix structure adjustments include adding new row and column indexes, deleting invalid mapping relationships, and merging similar mapping patterns. The adjustment process uses threshold control to avoid increased computational complexity caused by excessive expansion of the matrix. The adaptive learning mechanism records the frequency of use and conversion accuracy of the mapping relationship and adjusts the priority and confidence of the mapping relationship based on statistical data.

[0066] In a specific embodiment, the process of executing the step of establishing the semantic mapping relationship between the data fields of the protocols according to the score values ​​in the semantic relevance score matrix may specifically include the following steps:

[0067] The protocol data object combinations with score values ​​greater than a preset threshold are screened from the semantic relevance score matrix to establish a high-relevance protocol object library. The high-relevance protocol object library records the correlation between AGV position data and stereoscopic warehouse storage location data, and the matching degree between robot task instructions and equipment control parameters.

[0068] A data field mapping rule table is established based on the correlation and matching values ​​in the high-correlation protocol object library. The mapping rule table defines the mapping relationship between the X-axis value of the AGV controller position coordinate and the row number of the three-dimensional warehouse equipment shelf identification, and the Y-axis value and the column number.

[0069] The field correspondence in the mapping rule table is converted into a matrix form for storage. The row index of the matrix represents the source protocol data field, the column index represents the target protocol data field, and the matrix element value represents the mapping weight. The initial protocol semantic mapping matrix is ​​constructed.

[0070] The accuracy of the initial protocol semantic mapping matrix is ​​verified based on the actual data conversion results, the mapping weight values ​​in the matrix are adjusted, and the newly discovered field mapping relationships are supplemented to complete the construction of the protocol semantic mapping matrix.

[0071] Specifically, the semantic relevance scoring matrix is ​​filtered by setting a preset threshold of 0.75. Each score value in the matrix is ​​traversed and compared with the threshold. When the semantic relevance score of the AGV position data and the three-dimensional warehouse storage location data is 0.89, which is greater than the threshold, the protocol data object combination is added to the high-relevance protocol object library. The screening process also checks the matching degree between the robot task instructions and the device control parameters. When the matching score is 0.82, which exceeds the preset threshold, the association information of the combination is recorded. The high-relevance protocol object library uses a dual-key index structure to store data. The first key is the source protocol data object identifier, and the second key is the target protocol data object identifier. The corresponding values ​​include attributes such as relevance value, matching value, co-occurrence frequency, and temporal correlation. The relevance value in the library reflects the spatial semantic correlation between the AGV position data and the three-dimensional warehouse storage location data, and the matching value indicates the degree of functional semantic consistency between the robot task instructions and the device control parameters. The data record also contains the valid time range and confidence assessment of the protocol object combination to ensure the timeliness and reliability of the mapping relationship.

[0072] The data field mapping rule table is created by formulating specific field conversion rules based on the association and matching values ​​recorded in the highly relevant protocol object library. The X-axis values ​​of the AGV controller's position coordinates are mapped to the row numbers of the rack storage equipment's shelf locations through a linear transformation. This transformation formula maps the X-coordinate range of the AGV's work area to the row number range of the rack storage shelf. The mapping relationship between the Y-axis values ​​and column numbers is determined based on the geometric correspondence between the AGV's operating path and the rack storage layout, taking into account parameters such as the coordinate system origin, scale factor, and offset. The mapping rule table record format includes information such as the source field identifier, source data type, target field identifier, target data type, conversion function, and inverse conversion function. The conversion function describes the specific method for calculating the target field value from the source field value, while the inverse conversion function allows the source field value to be deduced from the target field value. The rule table also contains mapping constraints. The mapping relationship becomes invalid when the AGV position coordinates are outside the valid operating range, and the mapping operation is suspended when the rack storage location is under maintenance. The mapping rule priority is set based on the combined score of association and matching, with higher-scoring mapping rules receiving higher execution priority.

[0073] The matrix conversion process converts the field correspondences in the mapping rule table into a numerical matrix representation. The matrix row index uses a hash function to map the string identifiers of the source protocol data fields to integer indices, and the column index uses the same method to process the target protocol data field identifiers. The matrix element values ​​store the mapping weights. The weight calculation comprehensively considers factors such as relevance, matching, and historical conversion success rates, and a weighted average method is used to obtain the final weight value. The initial protocol semantic mapping matrix uses a sparse matrix storage format, storing only matrix elements with non-zero weights, saving storage space and accelerating matrix operations. The matrix structure contains a row index array, a column index array, and a weight value array, supporting fast row and column lookups and weight updates. The matrix maintenance process includes weight threshold filtering. When the mapping weight falls below the minimum threshold, the corresponding element is deleted from the matrix to prevent low-quality mapping relationships from interfering with the data conversion process.

[0074] The verification and adjustment process verifies the accuracy of the initial protocol semantic mapping matrix through actual data conversion operations. Typical protocol conversion cases from historical data are selected as test samples. Data conversion is performed according to the mapping relationships defined in the matrix, and the conversion results are compared with the expected results. Accuracy is assessed using two metrics: conversion success rate and semantic preservation. Conversion success rate measures the proportion of correctly formatted data after conversion, while semantic preservation measures the consistency of the semantic meaning of the data before and after conversion. The weight adjustment mechanism corrects the mapping weights in the matrix based on the verification results. When the conversion success rate of a mapping relationship is lower than expected, the corresponding weight value is reduced. When a new valid mapping pattern is discovered, the corresponding element is added to the matrix. The supplementation process identifies missing mapping relationships discovered during verification, analyzes the data field combinations in conversion failure cases, establishes new mapping rules, and updates the matrix structure. The dynamic adjustment algorithm uses an exponential decay mechanism to address the impact of historical data. Recent verification results have a greater impact on weight adjustment than earlier verification results, ensuring that the matrix adapts to protocol evolution and changes in application scenarios.

[0075] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0076] Selecting a protocol data object to be converted from the original data object set, matching the field information of the data object with the source protocol field in the protocol semantic mapping matrix, and determining the corresponding target protocol field mapping relationship;

[0077] Based on the mapping relationship, the position coordinate data of the AGV controller is converted into the cargo location coding format recognized by the three-dimensional warehouse equipment, and the task instructions of the robot scheduling system are converted into the control command format accepted by each execution device to generate a standard data object;

[0078] Calculate the semantic consistency value of the fields between the standard data object and the original protocol data object, record the consistency value as the semantic matching degree in the conversion log, and establish a conversion quality tracking record;

[0079] The standard data objects and the corresponding semantic matching values ​​are stored in the cache area in timestamp order to form a target data pool containing multi-device fusion data and quality assessment information.

[0080] Specifically, the protocol data object selection process selects the data objects to be converted from the original data object set according to the priority rules. The priority is based on a comprehensive assessment of the data object's timestamp age, semantic label importance, and conversion urgency. The field information matching process compares each data field in the selected protocol data object with the source protocol field in the protocol semantic mapping matrix one by one. The comparison methods include field name exact matching, data type compatibility check, and semantic label consistency verification. The matching algorithm uses a combination of string similarity calculation and semantic vector distance measurement. When the X-coordinate field of the AGV controller data object successfully matches the source field X_coordinate in the mapping matrix, the target field mapping relationship corresponding to the matrix is ​​searched. The target protocol field mapping relationship determination process obtains the target field identifier and mapping weight corresponding to the source field through a matrix search operation. The mapping relationship record contains the conversion rule function, data type conversion method, and validity verification conditions.

[0081] The data conversion process converts source protocol data into the target protocol format based on a defined mapping relationship. The conversion of AGV controller position coordinate data into the rack location coding format of the high-bay warehouse equipment involves coordinate system transformation and encoding rule adaptation. Position coordinate conversion utilizes a linear transformation method, converting the X and Y values ​​of the AGV working coordinate system into the row and column numbers of the high-bay warehouse shelf using a scale factor and offset. This conversion process accounts for differences in coordinate system origins and unit conversions. The conversion of robot scheduling system task instructions into the execution device control command format requires instruction semantic parsing and command format reconstruction. Task types are converted from enumerated values ​​in the scheduling system into control codes identified by the execution device, and task parameters are converted from the scheduling format into a device-specific parameter format. The standard data object generation process organizes the converted data fields into a unified data structure, including standard fields such as device identification, data type, value content, timestamp, and quality indicators. Data objects are stored using a key-value pair structure, with the key being the standardized field name and the value being the converted data content, while retaining traceability information for the original data.

[0082] The semantic consistency calculation process evaluates the conversion quality by comparing the semantic correspondence of the standard data object and the original protocol data object. The calculation method includes field mapping integrity check, data value rationality verification, and semantic correlation degree evaluation. The field mapping integrity is calculated by the ratio of the number of successfully converted fields to the total number of fields. The data value rationality is verified by checking whether the converted value is within the expected range. The semantic correlation degree is calculated by analyzing the semantic similarity of the data before and after conversion. The consistency value calculation uses a weighted average method to combine the integrity, rationality, and correlation degree according to the weight coefficients to obtain the final semantic matching degree value. The conversion log record includes conversion time, source data object identifier, target data object identifier, conversion rule identifier, semantic matching degree value, and abnormal information. The log is stored in time sequence format for easy tracking and analysis. The conversion quality tracking record evaluates the performance of the conversion algorithm by statistical analysis of the semantic matching degree distribution, conversion success rate trend, and abnormal pattern recognition in the conversion log.

[0083] The target data pool construction process arranges and stores the standard data objects and corresponding semantic matching degree values in chronological order with millisecond-level timestamp precision to ensure data sorting accuracy. The cache area uses a circular buffer structure to manage storage space, automatically discarding the oldest data records when the buffer capacity is insufficient, maintaining the real-time and effectiveness of the data pool. The multi-device fusion data organization method establishes an index structure according to device type and data type, with separate index tables for AGV data, stereoscopic warehouse data, and robot scheduling data, supporting fast data retrieval and association query. Quality evaluation information includes semantic matching degree, conversion delay, data integrity, and reliability rating for each data record, and is stored with the data record to form a complete data quality archive. The data pool management mechanism includes data deduplication, abnormal data marking, and expired data cleaning functions to ensure data quality and storage efficiency in the data pool.

[0084] In a specific embodiment, the process of performing step S105 can specifically include the following steps:

[0085] According to the protocol type of the target device, query the matching fusion data from the target data pool, filter out the corresponding standard data object and semantic matching degree value according to the device address and functional requirements, and establish a target device data set;

[0086] Based on the standard data object content in the target device data set and the field definition of the target protocol specification, convert the data content in standard format to the data type and encoding format required by the target protocol, and generate protocol adaptation data;

[0087] According to the data frame structure requirements of the target protocol, the protocol adaptation data is subjected to frame header addition, data field sorting, and checksum calculation processing, and the various parts of the data frame are assembled according to the byte order specified by the protocol to construct a standard protocol data frame;

[0088] The standard protocol data frame is sent to the target device through the corresponding communication interface, and the sending timestamp and transmission status information are recorded at the same time to complete the output processing of the protocol data frame.

[0089] Specifically, the fusion data query process performs index lookup operations in the target data pool based on the protocol type of the target device. When the target device is an intelligent stereoscopic library device using the OPC-UA protocol, the query algorithm traverses the data records in the data pool marked as related to the stereoscopic library. Device address filtering is based on the network address and device identifier of the target device for precise matching. Functional requirement filtering is based on the data type and operation type requested by the target device. The filtering conditions include data timeliness verification and semantic matching threshold checks. The target device data set establishment process organizes standard data objects that meet the filtering conditions and the corresponding semantic matching values ​​into a structured data set. The data set contains attributes such as data record identification, data content, quality rating, and timestamp. The data set is managed using a linked list structure, supports dynamic addition and deletion operations, and maintains the association and dependency relationships between data records. The query results are sorted based on a comprehensive score based on the semantic matching value and timeliness, with high-quality and recent data records ranked first.

[0090] The protocol adaptation data generation process performs data type conversion and encoding format adjustments based on the standard data object content in the target device dataset and the field definitions of the target protocol specification. Data type conversions include integer to floating-point, string to enumeration value, and Boolean to place flag conversions. Encoding format adjustments involve byte order conversion, data precision adjustment, and unit conversion. When the AGV position coordinates in the standard data object are floating-point values, they are converted to the double-precision floating-point format required by the OPC-UA protocol. When the robot task priority is an enumeration value, it is converted to the integer encoding required by the Modbus protocol. Field name mapping converts the generic field names of the standard data object to the specific field names of the target protocol specification. Field order adjustment rearranges the data fields according to the data structure requirements of the target protocol. Protocol adaptation data is stored in an intermediate format, which contains the converted data content, target protocol identification, and format verification information. The intermediate format facilitates subsequent data frame assembly operations.

[0091] The data frame construction process encapsulates the protocol adaptation data in a frame format according to the target protocol's frame structure specifications. The frame header addition process adds header fields such as the protocol identifier, version information, and message type, based on the target protocol type. The data field sorting process reorganizes the protocol adaptation data according to the field order specified by the target protocol. The Modbus protocol arranges the data fields in the order of device address, function code, data length, and data content, while the OPC-UA protocol arranges them in the order of message header, security header, sequence header, and message body. Checksum calculation uses the checksum algorithm specified by the target protocol. The Modbus protocol uses the CRC16 checksum algorithm for data frame checksum calculation, while the MQTT protocol uses the message digest algorithm to calculate the message integrity checksum. The byte order assembly process concatenates the frame header, data field, and checksum according to the protocol's byte order, generating a standard protocol data frame that conforms to the target protocol specification. The data frame verification process verifies that the generated data frame conforms to the protocol specification, including frame length verification, field format verification, and checksum correctness verification.

[0092] The protocol data frame output processing process selects the physical communication interface corresponding to the target device through the communication interface management module. The interface types include serial port, Ethernet, wireless network and other different communication methods. The sending operation adopts asynchronous transmission mode to write the standard protocol data frame into the sending buffer of the communication interface. The transmission control module is responsible for managing the fragmented transmission and retransmission processing of the data packet. The sending timestamp records the exact time when the data frame is sent. The transmission status information includes the confirmation of successful transmission, network delay measurement and error code record. The status monitoring mechanism tracks the transmission process of the data frame in real time. When a transmission failure is detected, the retransmission mechanism is triggered. When the confirmation response of the target device is received, the transmission status is updated to success. The output log record contains information such as the target device identification, protocol type, data frame content, sending timestamp, transmission delay and transmission results. The log data is used for transmission performance analysis and fault diagnosis.

[0093] The above describes the data conversion method of the multi-protocol integrated IoT device intelligent gateway in the embodiment of the present application. The following describes the data conversion system of the multi-protocol integrated IoT device intelligent gateway in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the multi-protocol integrated IoT device intelligent gateway data conversion system includes:

[0094] A construction module is used to collect data packets from AGV controllers, intelligent stereoscopic warehouse equipment, and robot scheduling systems, extract protocol header features, data structure patterns, and timing features, and build a protocol feature vector library;

[0095] A parsing module, configured to identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags;

[0096] An analysis module, configured to analyze the semantic relevance of different protocol data in the original data object set by using a protocol semantic adaptive learning algorithm, and automatically learn and establish a dynamically updated protocol semantic mapping matrix;

[0097] A recording module, configured to convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record semantic matching values ​​to obtain a target data pool;

[0098] The output module is used to extract the fused data from the target data pool according to the target device protocol specification, and re-encode the output protocol data frame according to the target protocol data frame format.

[0099] above Figure 2 From the perspective of modular functional entities, the multi-protocol integrated IoT device intelligent gateway data conversion system in the embodiment of the present invention is described in detail. The multi-protocol integrated IoT device intelligent gateway data conversion device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0100] Reference Figure 3 In the embodiment of the present invention, a multi-protocol fusion IoT device intelligent gateway data conversion device is also provided. The multi-protocol fusion IoT device intelligent gateway data conversion device can be a server, and its internal structure can be as follows Figure 3 As shown. The multi-protocol fusion IoT device intelligent gateway data conversion device includes a processor, memory, display screen, input device, network interface and database connected through a system bus. Among them, the computer-designed processor is used to provide computing and control capabilities. The memory of the multi-protocol fusion IoT device intelligent gateway data conversion device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the multi-protocol fusion IoT device intelligent gateway data conversion device is used to store the corresponding data in this embodiment. The network interface of the multi-protocol fusion IoT device intelligent gateway data conversion device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0101] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the multi-protocol integrated IoT device intelligent gateway data conversion device to which the solution of the present invention is applied.

[0102] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions. When the instructions are run on a computer, the computer executes the steps of the multi-protocol integrated IoT device intelligent gateway data conversion method.

[0103] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0104] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a multi-protocol integrated IoT device intelligent gateway data conversion device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0105] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-protocol integrated IoT device intelligent gateway data conversion method, characterized in that: The method comprises: Collect data packets from AGV controllers, intelligent warehouse equipment, and robot scheduling systems, extract protocol header features, data structure patterns, and timing features, and build a protocol feature vector library; Identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags; The method comprises the following steps: analyzing the semantic relevance of different protocol data in the original data object set by a protocol semantic adaptive learning algorithm, automatically learning and establishing a dynamically updated protocol semantic mapping matrix, including: extracting the semantic label and data content of each protocol data object from the original data object set, establishing a protocol semantic sample library, wherein the protocol semantic sample library contains semantic correspondence samples of AGV position data and three-dimensional warehouse cargo location data, robot task instructions and equipment control parameters; calculating the semantic similarity values ​​between different protocol data objects based on the protocol semantic sample library, and obtaining a semantic relevance scoring matrix by calculating the semantic label matching degree and data content correlation degree; establishing a semantic mapping relationship between data fields between protocols according to the scoring values ​​in the semantic relevance scoring matrix, mapping the position coordinate field of the AGV controller to the cargo location identification field of the three-dimensional warehouse device, and mapping the task type field of the robot scheduling system to the function instruction field of each execution device, thereby forming a protocol semantic mapping matrix; updating the mapping relationship weight value in the protocol semantic mapping matrix based on the newly received original data object, automatically adjusting the matrix structure when a new semantic association pattern is detected, and generating a dynamically updated protocol semantic mapping matrix; Convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record the semantic matching value to obtain a target data pool; The fused data is extracted from the target data pool according to the target device protocol specification, and the output protocol data frame is re-encoded according to the target protocol data frame format.

2. The multi-protocol integrated IoT device intelligent gateway data conversion method according to claim 1 is characterized in that: The data packets of the AGV controller, intelligent stereoscopic warehouse equipment and robot scheduling system are collected, protocol header features, data structure patterns and timing features are extracted, and a protocol feature vector library is constructed, including: Monitor the communication ports of the AGV controller, intelligent three-dimensional warehouse equipment, and robot scheduling system, cache the received data packets, and obtain the original data packet set; Performing protocol header identification extraction processing based on the first 16 bytes of each data packet in the original data packet set to obtain a protocol header feature set including a function code, a node identifier, and a subject name; Performing vectorized encoding processing on the feature values ​​in the protocol header feature set according to the data length field position, the check code calculation method and the frame structure arrangement order to obtain a protocol feature vector; A feature index table is established according to the protocol feature vectors for storage management processing, and a protocol feature vector library that records feature vectors corresponding to each protocol type is obtained.

3. The multi-protocol fusion IoT device intelligent gateway data conversion method according to claim 1 is characterized in that: The method of identifying the protocol type of each device based on the protocol feature vector library and parsing the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags includes: Matching and calculating the protocol header features of the data packet to be identified with the feature vectors stored in the protocol feature vector library to obtain a protocol type identification result; According to the protocol type identification result, the data packet is subjected to field segmentation and parsing according to the corresponding protocol specification to obtain a parsed field set including the device address, function instruction and data content; Perform semantic classification and tagging based on the functional instruction type and data content attributes in the parsed field set to obtain data fields with read operation semantic tags, write operation semantic tags, and status query semantic tags; The data fields are encapsulated into objects according to the device source and time sequence to obtain a set of original data objects containing protocol semantic tags.

4. The multi-protocol fusion IoT device intelligent gateway data conversion method according to claim 1 is characterized in that: The semantic mapping relationship between the data fields of the protocols is established according to the scoring values ​​in the semantic relevance scoring matrix, the position coordinate field of the AGV controller is mapped to the shelf identification field of the three-dimensional warehouse equipment, and the task type field of the robot scheduling system is mapped to the function instruction field of each execution device, forming a protocol semantic mapping matrix, including: Filtering protocol data object combinations whose score values ​​are greater than a preset threshold from the semantic relevance scoring matrix to establish a high-relevance protocol object library, wherein the high-relevance protocol object library records the correlation between AGV position data and stereoscopic warehouse storage location data, and the matching degree between robot task instructions and equipment control parameters; A data field mapping rule table is established based on the correlation and matching values ​​in the high-correlation protocol object library, wherein the mapping rule table defines a mapping relationship between the X-axis value of the AGV controller position coordinate and the row number of the three-dimensional warehouse equipment storage location identification, and the Y-axis value and the column number; The field correspondence in the mapping rule table is converted into a matrix form for storage, wherein the row index of the matrix represents the source protocol data field, the column index represents the target protocol data field, and the matrix element value represents the mapping weight, thereby constructing an initial protocol semantic mapping matrix; The accuracy of the initial protocol semantic mapping matrix is ​​verified based on the actual data conversion results, the mapping weight values ​​in the matrix are adjusted, and the newly discovered field mapping relationships are supplemented to complete the construction of the protocol semantic mapping matrix.

5. The multi-protocol fusion IoT device intelligent gateway data conversion method according to claim 1 is characterized in that: The converting different protocol data objects into a standard data format according to the protocol semantic mapping matrix and recording semantic matching values ​​to obtain a target data pool includes: Selecting a protocol data object to be converted from the original data object set, matching the field information of the data object with the source protocol field in the protocol semantic mapping matrix, and determining a corresponding target protocol field mapping relationship; Based on the mapping relationship, the position coordinate data of the AGV controller is converted into a cargo location coding format recognized by the three-dimensional warehouse equipment, and the task instructions of the robot scheduling system are converted into a control command format accepted by each execution device to generate a standard data object; Calculate the field semantic consistency value between the standard data object and the original protocol data object, record the consistency value as the semantic matching degree in the conversion log, and establish a conversion quality tracking record; The standard data objects and corresponding semantic matching degree values ​​are stored in a cache area in the order of timestamps to form a target data pool containing multi-device fusion data and quality assessment information.

6. The multi-protocol fusion IoT device intelligent gateway data conversion method according to claim 1 is characterized in that: The step of extracting fused data from the target data pool according to the target device protocol specification and re-encoding the output protocol data frame according to the target protocol data frame format includes: Querying matching fusion data from the target data pool according to the protocol type of the target device, screening corresponding standard data objects and semantic matching values ​​according to the device address and functional requirements, and establishing a target device data set; Based on the standard data object content in the target device data set and the field definition of the target protocol specification, convert the data content in the standard format into the data type and encoding format required by the target protocol to generate protocol adaptation data; According to the data frame structure requirements of the target protocol, the protocol adaptation data is subjected to frame header addition, data field sorting and check code calculation processing, and the various parts of the data frame are assembled according to the byte order specified by the protocol to construct a standard protocol data frame; The standard protocol data frame is sent to the target device through the corresponding communication interface, and the sending timestamp and transmission status information are recorded at the same time to complete the output processing of the protocol data frame.

7. A multi-protocol integrated IoT device intelligent gateway data conversion system, characterized in that: A method for converting data from a multi-protocol-integrated IoT device intelligent gateway according to any one of claims 1 to 6, wherein the multi-protocol-integrated IoT device intelligent gateway data conversion system comprises: A construction module is used to collect data packets from AGV controllers, intelligent stereoscopic warehouse equipment, and robot scheduling systems, extract protocol header features, data structure patterns, and timing features, and build a protocol feature vector library; A parsing module, configured to identify the protocol type of each device based on the protocol feature vector library and parse the device address, function instruction and data content in the data packet to construct a set of original data objects containing protocol semantic tags; An analysis module, configured to analyze the semantic relevance of different protocol data in the original data object set by using a protocol semantic adaptive learning algorithm, and automatically learn and establish a dynamically updated protocol semantic mapping matrix; A recording module, configured to convert different protocol data objects into a standard data format according to the protocol semantic mapping matrix, and record semantic matching values ​​to obtain a target data pool; The output module is used to extract the fused data from the target data pool according to the target device protocol specification, and re-encode the output protocol data frame according to the target protocol data frame format.

8. A multi-protocol integrated IoT device intelligent gateway data conversion device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the method for converting data of a multi-protocol-fused IoT device intelligent gateway according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor executes the multi-protocol integrated IoT device intelligent gateway data conversion method according to any one of claims 1 to 6.

Images