Construction tool positioning management method and system based on RFID technology

Abstract

This invention discloses a method and system for managing the positioning of construction tools based on RFID technology, relating to the field of tool positioning technology. The method includes: constructing a spatially accessible topology map, establishing a reader deployment parameter table, and building a positioning model containing five signal hiding states; collecting raw positioning data, inputting observed features into the positioning model, determining the current signal hiding state through a forward recursion algorithm, and classifying and labeling the positioning data; mapping the positioning data to the spatially accessible topology map to obtain candidate spatial units for positioning, identifying abnormal data through spatial recursion constraint verification, and generating a standardized positioning dataset through spatial consistency verification; based on the standardized positioning dataset, determining the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the unit, generating and outputting a standardized positioning result data packet. This effectively solves the problems of large RFID positioning deviations and discontinuous trajectories caused by factors such as metal obstruction and multipath reflection in complex construction environments.

Description

Technical Field

[0001] This invention relates to the field of tool positioning technology, specifically to a method and system for managing the positioning of construction tools based on RFID technology. Background Technology

[0002] Construction sites have a wide variety of tools that are highly mobile. Traditional management methods that rely on manual inspection and recording are inefficient and prone to errors, making it difficult to meet the real-time and accuracy requirements of modern construction management. Radio Frequency Identification (RFID) technology, as a non-contact automatic identification method, has advantages such as fast reading, multi-target identification, and strong environmental adaptability. It has been gradually introduced into the field of construction tool management to achieve automated tool positioning and tracking. Currently, common RFID positioning solutions mainly involve deploying a fixed reader network to collect information such as tag signal strength RSSI and estimate the tool position based on signal strength attenuation models or geometric positioning algorithms, thereby providing location data support for tool management on construction sites.

[0003] However, in complex construction environments, conventional RFID positioning methods face significant limitations. First, construction sites typically have numerous metal components, concrete walls, and other obstructions that can easily shield and reflect RFID signals, leading to abnormal signal attenuation and prominent multipath effects. Positioning results relying solely on RSSI are prone to significant errors. Second, factors such as densely packed tools, mechanical vibrations, and frequent personnel movement can cause unstable tag readings, intermittent signals, or even interruptions. This makes traditional positioning methods based on instantaneous signal strength prone to problems such as location jumps and discontinuous trajectories, affecting the continuity and reliability of positioning. Furthermore, existing RFID... Most positioning technologies are simply a combination of general wireless positioning algorithms and RFID hardware, failing to fully consider the spatial topology of construction sites and the state changes of tools during actual operation. For example, there are physical barriers and access path constraints between different construction areas, and tool movement should meet spatial accessibility conditions. However, existing methods lack modeling and verification of such spatial topological relationships. At the same time, conventional methods usually do not distinguish and process the types of signal interference in a fine-grained manner, resulting in insufficient positioning robustness under different interference scenarios such as metal obstruction, multipath reflection, and signal interruption, making it difficult to adapt to the complex and dynamic actual working conditions of construction sites. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a method and system for managing the positioning of construction tools based on RFID technology. By introducing a hidden Markov model for signal state determination and combining spatial topological constraints with a multi-source data fusion correction mechanism, it effectively solves the problems of large RFID positioning deviations and discontinuous trajectories caused by factors such as metal obstruction and multipath reflection in complex construction environments, thereby improving the accuracy and robustness of tool positioning.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a construction tool positioning management method based on RFID technology, comprising: The construction site is divided into multiple spatial units, a spatial reachability topology map is constructed, RFID fixed readers are deployed in each spatial unit, a reader deployment parameter table is established, and a positioning model containing at least five signal hiding states is built. Raw positioning data is collected and observation features are extracted. The observation features are input into the positioning model and combined with the historical state judgment results. The current signal hiding state is determined by the forward recursion algorithm, and the positioning data is classified and labeled. The classified and labeled location data is mapped to a spatially reachable topology to obtain candidate spatial units for location. Abnormal data is identified through spatial recursive constraint verification, and a standardized location dataset is generated through spatial consistency verification. Based on the standardized positioning dataset, the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit are determined, and a standardized positioning result data packet is generated and output.

[0006] Furthermore, by collecting CAD construction drawings and measured boundary data from the construction site, and combining the physical boundaries of walls, fences, construction area partitions, work process areas, and fixed access paths for personnel and equipment, the construction site is divided into non-overlapping spatial units that cannot be directly physically crossed. The boundary coordinates of each spatial unit are marked using a Cartesian coordinate system, each spatial unit is assigned a unique string code, the physical reachability relationship between any two spatial units is verified and recorded, unit associations without direct access conditions are deleted, and a construction-specific spatial reachability topology map is constructed.

[0007] Furthermore, at least one fixed RFID reader is deployed within each spatial unit to acquire the reader's two-dimensional planar coordinates, establish a unique mapping relationship between the reader and the spatial unit, and form a reader deployment parameter table. RFID tags are bound to construction tools, tag and tool information is entered, and initial positioning parameters are set for the tools. Based on the spatial reachability topology map and the reader deployment parameter table, a hidden Markov positioning model is built, classifying five signal hiding states: normal reception, metal obstruction, multipath reflection, trajectory drift, and signal interruption. The observation parameters corresponding to each state are calibrated using sample data from interference scenarios to complete model initialization.

[0008] Furthermore, all fixed RFID readers are triggered according to a preset clock synchronization mechanism to synchronously scan tags at a fixed collection cycle, collecting tag UID codes, RSSI values, signal reception duration, number of valid reads per cycle, reader number, and collection timestamp to generate a single-cycle raw positioning dataset. The raw positioning dataset is preprocessed and cleaned to remove invalid data due to communication failures, empty scans, or incomplete formats, retaining valid data that matches the tool tag UID codes. From each valid data, three types of observation features are extracted: signal strength fluctuation amplitude, continuous reading stability score, and consistency of repeated readings across readers, forming a standardized observation feature sequence.

[0009] Furthermore, the standardized observation feature sequence is input into the localization model, and the single-frame observation probability of each observation feature corresponding to the five signal hidden states is calculated through the observation probability matrix. The hidden state determination result of the previous cycle tool is retrieved, and the state transition probability of each hidden state in the current cycle is obtained by combining it with the state transition probability matrix. The single-frame observation probability and the state transition probability are weighted and fused using a forward recursive algorithm to obtain the global posterior probability of each hidden state, and the hidden state corresponding to the maximum value is selected as the current final state determination result. Based on the final determination result, the data is classified and labeled: the normal reception state is compliant data, the metal blockage, multipath reflection, and trajectory drift state is data to be corrected, and the signal interruption state is missing data.

[0010] Furthermore, the compliant data and data to be corrected after classification and labeling are mapped to a construction-specific spatial reachability topology map. The spatial unit code is matched with the reader number to determine the candidate spatial unit for the positioning of the construction tool. The historical positioning compliant spatial unit code sequence of the tool in the last N frames is retrieved to form a historical and current spatial unit comparison dataset. Based on the physical reachability relationship in the topology map, spatial recursive constraint verification is performed on the comparison dataset to verify whether there is a direct physical reachability relationship between the current candidate spatial unit and the compliant spatial units in the last N consecutive frames. If the relationship is satisfied, it is marked as compliant positioning data; otherwise, it is marked as abnormal positioning data and collected according to the tag UID code.

[0011] Furthermore, for abnormal data in the metal obstruction state, the compliant spatial unit of the tool in the previous cycle is used, and the positioning range within the compliant spatial unit is reduced to 50% of the compliant coordinate error range of the previous cycle; for abnormal data in the multipath reflection state, the candidate spatial unit corresponding to the reflected signal is removed, and the physical reachable spatial unit in the spatial reachable topology map that is closest to the original candidate spatial unit is selected as the new candidate spatial unit; for abnormal data in the trajectory drift state, all physical reachable spatial units between the historical compliant spatial unit and the current candidate spatial unit are locked to form a limited reachable unit interval; for abnormal data in the signal interruption state, the current cycle's completed coordinates are calculated using linear interpolation based on the tool's historical N consecutive frames of compliant coordinates, and matched to the spatial unit corresponding to the topology map.

[0012] Furthermore, a spatial consistency review is performed on the compliant positioning data and the corrected positioning data. This review verifies whether the spatial unit codes and coordinate values ​​corresponding to all positioning data conform to the physical boundary coordinates of the spatial reachability topology map, and whether the coordinate values ​​are within the effective signal coverage radius recorded in the reader deployment parameter table. Any residual abnormal data that fails any review is removed. All the verified positioning data are then categorized and collected according to the tool tag UID code to generate a standardized positioning dataset containing the tag UID code, spatial unit code, coordinates within the spatial unit, positioning status label, data acquisition timestamp, and corresponding reader number.

[0013] Furthermore, the reader signal stability score and spatial unit positioning reliability benchmark value are extracted from the standardized positioning dataset. Weighting coefficients are assigned according to the signal stability score (60%) and the positioning reliability benchmark value (40%), and the weighted sum is used to obtain the comprehensive weight value of a single positioning data point. Based on the comprehensive weight value, a weighted fusion calculation is performed on the multi-source positioning data of the same construction tool to obtain the fused positioning coordinates within the spatial unit. The 3σ criterion is used to remove residual outliers with excessive dispersion, and the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit are determined. The tag UID code is bound one by one with the target spatial unit code, two-dimensional plane coordinates, positioning status label, and positioning result confidence score to generate a standardized positioning result data package and output it.

[0014] A construction tool positioning management system based on RFID technology includes: The initialization module divides the construction site into multiple spatial units, constructs a spatial reachability topology map, deploys RFID fixed readers for each spatial unit, establishes a reader deployment parameter table, and builds a positioning model containing at least five signal hiding states. The state determination module collects raw positioning data and extracts observation features. It inputs the observation features into the positioning model, combines them with historical state determination results, and uses a forward recursion algorithm to determine the current signal hiding state and classify and label the positioning data. The correction module maps the classified and labeled location data to a spatially reachable topology to obtain candidate spatial units for location. It identifies abnormal data through spatial recursive constraint verification and generates a standardized location dataset through spatial consistency verification. The positioning output module, based on a standardized positioning dataset, determines the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit, and generates and outputs a standardized positioning result data packet.

[0015] (III) Beneficial Effects This invention provides a method and system for positioning and managing construction tools based on RFID technology, which has the following beneficial effects: (1) By dividing the construction site into discrete spatial units and constructing a spatial reachable topology, a structured management basis is provided for subsequent RFID signal identification. RFID readers are deployed and parameter tables are established. By combining the hidden Markov model to refine the signal hiding state, the refined modeling of the construction environment and the modeling of the interference state are realized, which improves the spatial adaptability and interference robustness of the positioning system and lays a key foundation for subsequent data classification, correction and accurate positioning.

[0016] (2) By introducing a hidden Markov model, the RFID signal is judged in a fine-grained manner, and the signal is classified into five types: normal, obstruction, reflection, drift and interruption. This enables accurate identification and classification of complex construction interference scenarios, providing a key basis for subsequent differential data correction. It not only effectively suppresses the positioning jump and trajectory discontinuity caused by abnormal signals, but also significantly improves the robustness of the positioning system to dynamic interference and its state perception capability.

[0017] (3) By mapping the classified data to the spatially accessible topology map and performing spatial recursive constraint verification based on historical positioning data, abnormal positioning caused by signal interference or physical inaccessibility can be effectively identified. Combined with different signal states, such as occlusion, reflection, drift, and interruption, differential data correction is performed to significantly suppress position jumps and trajectory discontinuities. Spatial consistency verification is performed to remove residual abnormal data and generate a standard dataset. The continuity and spatial rationality of the positioning results are systematically improved, and the robustness of the system in complex construction environments is enhanced.

[0018] (4) By dynamically weighting and fusing multi-source positioning data and removing discrete points, high-precision and high-reliability final positioning is achieved. The weight of each data is determined based on signal quality and spatial compliance. Multiple data from the same tool are fused to optimize positioning coordinates. Anomalies are removed based on the 3σ criterion to ensure robust results. Finally, a standardized positioning result data package is generated, which includes confidence assessment and status marking. This provides the management system with accurate, consistent and interpretable positioning information, directly supporting tool search and management decisions. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the steps of the construction tool positioning and management method based on RFID technology according to the present invention; Figure 2 This is a schematic diagram of the construction tool positioning management method based on RFID technology of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figures 1-2 This invention provides a method for positioning and managing construction tools based on RFID technology, comprising the following steps: Step 1: Divide the construction site into multiple spatial units, construct a spatial reachability topology map, deploy RFID fixed readers for each spatial unit, establish a reader deployment parameter table, and build a positioning model containing at least five signal hiding states; Step one includes the following: Step 101: Collect CAD construction drawings and on-site measured boundary data from the construction site. Combine these with physical boundaries such as walls, fences, construction area partitions, work process areas, and fixed access paths for personnel and equipment to divide the entire construction site into multiple non-overlapping spatial units that cannot be directly physically crossed. Use a Cartesian coordinate system to mark the boundary coordinates of each spatial unit, assign a unique and non-repeating string code to each spatial unit, verify and record the physical reachability relationship between any two spatial units, delete unit associations that have no direct access conditions or cannot be crossed, and construct a fixed construction-specific spatial reachability topology map. Step 102: According to the spatial unit distribution of the construction-specific space reachable topology map, at least one RFID fixed reader with clock synchronization function shall be deployed in each spatial unit, with the reader antenna facing the unobstructed area inside the corresponding spatial unit; obtain the two-dimensional plane coordinates of each RFID fixed reader by on-site total station measurement or engineering benchmark layout, and use the point-by-point signal testing method to determine and record the effective signal coverage radius and signal boundary attenuation threshold of each reader, establish a unique mapping relationship between a single RFID fixed reader and a single spatial unit, and compile a reader deployment parameter table; Step 103: Select anti-metal RFID electronic tags and uniquely bind each RFID electronic tag to a non-wearing, unobstructed part of the construction tool to be managed. Enter the tag UID code, corresponding tool number, and tool type into the management terminal database. Set initial positioning parameters for each construction tool with a bound tag, including initial home spatial unit code, initial positioning confidence benchmark value, and hidden Markov state decision critical score. Complete the tool tag binding and tool initial parameter configuration. Step 104: Based on the construction-specific space reachability topology map and reader deployment parameter table, build a dedicated hidden Markov model for construction tool positioning; divide the signal state during the positioning process into five hidden states: normal reception, metal obstruction, multipath reflection, trajectory drift, and signal interruption; through sample data of various interference scenarios pre-collected at the construction site, calibrate the RFID signal strength range, reception continuity, and trajectory offset observation parameters corresponding to each hidden state, and complete the model initialization and parameter solidification.

[0022] When using this method, refer to steps 101 to 104: By dividing the construction site into discrete spatial units and constructing a spatially accessible topology map, a structured management foundation is provided for subsequent RFID signal identification. RFID readers are deployed and parameter tables are established. By combining hidden Markov models to refine the signal hiding state, refined modeling of the construction environment and interference state modeling are achieved, improving the spatial adaptability and interference robustness of the positioning system. This lays a key foundation for subsequent data classification, correction, and accurate positioning.

[0023] Step 2: Collect raw positioning data and extract observation features. Input the observation features into the positioning model, combine them with the historical state judgment results, and use the forward recursion algorithm to determine the current signal hiding state and classify and label the positioning data. Step two includes the following: Step 201: According to the preset clock synchronization mechanism, all RFID fixed readers are triggered to start tag scanning synchronously at the preset fixed collection cycle, and non-contact identification is performed on the anti-metal RFID electronic tags of construction tools within their respective signal coverage areas; the raw positioning data fed back by the tags in a single cycle is collected in real time, including: tag UID code, signal reception strength RSSI value, signal continuous reception duration, number of valid tag reads in a single cycle, corresponding reader number and collection timestamp, generating a single-cycle raw positioning dataset and uploading it to the background management terminal in real time; Step 202: After receiving the single-cycle raw positioning dataset, the background management terminal performs preprocessing and cleaning operations: invalid data entries with reader communication failures, no tag feedback during empty scanning, or incomplete data formats are removed, and valid identification data corresponding one-to-one with the tool tag UID code are retained; for each valid data entry, three types of observation features are extracted: signal strength fluctuation amplitude, continuous reading stability score, and consistency of repeated readings across readers, forming a standardized observation feature sequence that matches the Hidden Markov Model; Step 203: Input the standardized observation feature sequence into the construction tool positioning hidden Markov model. Based on the observation probability matrix that has been pre-trained and calibrated through samples, for the current single observation feature, directly look up the table to calculate the observation probability of the five hidden states corresponding to the observation feature: normal reception, metal blockage, multipath reflection, trajectory drift, and signal interruption. Output the single-frame observation probability value of a single construction tool corresponding to the five hidden states in the current period, and normalize the single-frame observation probability value to ensure that the sum of the probability values ​​of all states is 1. Step 204: Retrieve the final hidden state determination result of the tool in the previous period, and combine it with the state transition probability matrix of the Hidden Markov Model pre-training to obtain the state transition probability of each hidden state in the current period; use the forward recursion algorithm to perform weighted fusion calculation on the single frame observation probability value and the corresponding state transition probability to obtain the global posterior probability of each hidden state, and select the hidden state corresponding to the maximum global posterior probability as the final state determination result of the tool in the current period; Step 205: Based on the final status determination result, classify and label the single-cycle valid positioning data: label the data corresponding to the normal reception status as compliant data, label the data corresponding to the metal blockage, multipath reflection, and trajectory drift status as data to be corrected, and label the data corresponding to the signal interruption status as missing data; simultaneously remove invalid data determined to be strong electromagnetic interference or data distortion, and complete the classification, filtering and status labeling of the positioning data.

[0024] When using this method, refer to steps 201 to 205: By introducing a Hidden Markov Model, fine-grained state determination of RFID signals is performed, classifying signals into five types: normal, obstructed, reflected, drifting, and interrupted. This enables accurate identification and classification of complex construction interference scenarios, providing a key basis for subsequent differential data correction. It not only effectively suppresses the problems of positioning jumps and trajectory discontinuities caused by signal anomalies, but also significantly improves the robustness of the positioning system to dynamic interference and its state perception capability.

[0025] Step 3: Map the classified and labeled location data to a spatially reachable topology to obtain candidate spatial units for location. Identify abnormal data through spatial recursive constraint verification and generate a standardized location dataset through spatial consistency verification. Step three includes the following: Step 301: Map the compliant data and data to be corrected after classification, filtering and status labeling to the construction-specific space reachable topology map. Match the spatial unit code corresponding to the reader deployment parameter table with the reader number associated with the data to determine the candidate spatial unit for the location of a single construction tool within a single cycle. At the same time, retrieve the compliant spatial unit code sequence corresponding to the tool's historical location of the last N frames from the backend management terminal. The spatial unit corresponding to the compliant location data is the compliant spatial unit, forming a historical and current spatial unit comparison dataset containing the spatial unit codes of the past N frames, the current candidate spatial unit codes, and the tool tag UID code, where N≥3. Step 302: Based on the physical reachability relationships of spatial units pre-recorded in the construction-specific spatial reachability topology map, perform spatial recursive constraint verification on the historical and current spatial unit comparison dataset: verify one by one whether there is a direct physical reachability relationship between the current candidate spatial unit and the compliant spatial units of the previous N consecutive frames. If the reachability relationship is satisfied, mark the corresponding positioning data as compliant positioning data and retain it. If the relationship is not satisfied, mark the corresponding positioning data as abnormal positioning data and collect it separately according to the tool label UID code. Step 303: For the collected abnormal positioning data, combined with the hidden state type marked in Step 205, execute a differentiated data correction strategy: In the metal occlusion state, use the compliant spatial unit of the tool in the previous cycle, and reduce the positioning range within the compliant spatial unit to 50% of the error range of the compliant positioning coordinates in the previous cycle of the tool; in the multipath reflection state, directly remove the candidate spatial unit corresponding to the reflected signal, and select the physical reachable spatial unit in the construction-specific space reachable topology map that is closest to the original candidate spatial unit as the new candidate spatial unit; in the trajectory drift state, lock all physical reachable spatial units between the historical compliant spatial unit and the current candidate spatial unit to form a limited reachable unit interval; in the signal interruption state, based on the compliant positioning coordinates of the tool in the past N consecutive frames, use linear interpolation to calculate the completed coordinates of the current cycle, and match the completed coordinates to the corresponding spatial unit in the construction-specific space reachable topology map; Step 304: Perform a unified spatial consistency review on the compliant positioning data and the corrected positioning data: verify whether the spatial unit codes and coordinate values ​​corresponding to all data conform to the physical boundary coordinate limits of the construction-specific space reachable topology map, and at the same time verify whether the coordinate values ​​are within the effective signal coverage radius recorded in the corresponding reader deployment parameter table, and remove any residual abnormal data that fails either of the two reviews; Step 305: Integrate the compliant positioning data that has passed the spatial consistency verification with the corrected positioning data, classify and collect them according to the tool label UID code, generate a standardized positioning dataset containing the label UID code, spatial unit code, coordinates within the spatial unit, positioning status label, data collection timestamp, and corresponding reader number, and upload it synchronously to the backend management terminal.

[0026] When using this method, refer to steps 301 to 305: By mapping classified data to a spatially accessible topology and performing spatial recursive constraint verification based on historical positioning data, abnormal positioning caused by signal interference or physical inaccessibility can be effectively identified. Combined with different signal states, such as occlusion, reflection, drift, and interruption, differentiated data correction is performed to significantly suppress position jumps and trajectory discontinuities. Spatial consistency is checked, residual abnormal data is removed, and a standard dataset is generated. This systematically improves the continuity and spatial rationality of positioning results and enhances the robustness of the system in complex construction environments.

[0027] Step 4: Based on the standardized positioning dataset, determine the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit, and generate and output the standardized positioning result data packet.

[0028] Step four includes the following: Step 401: Retrieve the standardized positioning dataset from the backend management terminal, and simultaneously retrieve the reader deployment parameter table and tool initial parameter configuration data. Extract the reader signal stability score and spatial unit positioning reliability benchmark value corresponding to each positioning data in the dataset. Step 402: Based on the principle of positive correlation between signal quality and spatial compliance, assign dynamic weighting coefficients to each positioning data: reader signal stability score accounts for 60%, spatial unit positioning reliability benchmark value accounts for 40%, and the comprehensive weight value of a single positioning data is obtained by weighted summation, with the comprehensive weight value ranging from 0 to 1; Step 403: Based on the comprehensive weight value of each positioning data, perform weighted fusion calculation on the multi-source positioning data of the same construction tool to calculate the fused positioning coordinates of the tool in the corresponding spatial unit. Using the fused positioning coordinates as the benchmark, use the 3σ criterion to remove residual outliers with excessive dispersion during the fusion calculation process, and determine the final target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit. Step 404: Bind the unique label UID code of the construction tool, the target spatial unit code, the two-dimensional plane coordinates within the unit, the current positioning status label, and the positioning result confidence score one by one to generate a standardized positioning result data package containing the above core information; Step 405: The standardized positioning result data packet is synchronously output to the back-end management terminal in real time to complete the entire process of single-cycle construction tool positioning management. At the same time, the data packet is archived to the management terminal database according to the timestamp to realize the traceability and query of positioning data.

[0029] When using this method, refer to steps 401 to 405: By dynamically weighting and fusing multi-source positioning data and removing discrete points, high-precision and high-reliability final positioning is achieved. The weight of each data point is determined based on signal quality and spatial compliance. Multiple data points from the same tool are fused to optimize positioning coordinates. Outliers are removed based on the 3σ criterion to ensure robust results. Finally, a standardized positioning result data package is generated, which includes confidence assessment and status marking. This provides the management system with accurate, consistent, and interpretable positioning information, directly supporting tool search and management decisions.

[0030] This invention also provides a construction tool positioning management system based on RFID technology, comprising: The initialization module divides the construction site into discrete spatial units, constructs a spatial reachability topology map, deploys RFID fixed readers for each unit, establishes a reader deployment parameter table, and builds a positioning model containing five signal hiding states. The state determination module collects raw positioning data and extracts observation features. It inputs the observation features into the positioning model, combines them with historical state determination results, and uses a forward recursion algorithm to determine the current signal hiding state and classify and label the positioning data. The correction module maps the classified and labeled positioning data to a spatially reachable topology to obtain positioning candidate spatial units, identifies abnormal data through spatial recursive constraint verification, and generates a standardized positioning dataset through spatial consistency verification. The positioning output module, based on a standardized positioning dataset, determines the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the unit, and generates and outputs a standardized positioning result data packet.

[0031] In the application, the various formulas mentioned are all calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of data to obtain the most recent real-world results. The coefficients in the formulas are set by those skilled in the art according to the actual situation.

[0032] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, and combinations thereof. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0033] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0034] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for positioning and managing construction tools based on RFID technology, characterized in that: include: The construction site is divided into multiple spatial units, a spatial reachability topology map is constructed, RFID fixed readers are deployed in each spatial unit, a reader deployment parameter table is established, and a positioning model containing at least five signal hiding states is built. Raw positioning data is collected and observation features are extracted. The observation features are input into the positioning model and combined with the historical state judgment results. The current signal hiding state is determined by the forward recursion algorithm, and the positioning data is classified and labeled. The classified and labeled location data is mapped to a spatially reachable topology to obtain candidate spatial units for location. Abnormal data is identified through spatial recursive constraint verification, and a standardized location dataset is generated through spatial consistency verification. Based on the standardized positioning dataset, the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit are determined, and a standardized positioning result data packet is generated and output.

2. The method for positioning and managing construction tools based on RFID technology according to claim 1, characterized in that: By collecting CAD construction drawings and on-site measured boundary data, and combining the physical boundaries of walls, fences, construction area partitions, work process areas, and fixed access paths for personnel and equipment, the construction site is divided into non-overlapping spatial units that cannot be directly physically crossed. The boundary coordinates of each spatial unit are marked using a Cartesian coordinate system, each spatial unit is assigned a unique string code, the physical reachability relationship between any two spatial units is verified and recorded, unit associations without direct access conditions are deleted, and a construction-specific spatial reachability topology map is constructed.

3. The method for positioning and managing construction tools based on RFID technology according to claim 2, characterized in that: At least one fixed RFID reader is deployed inside each spatial unit. The two-dimensional plane coordinates of the reader are obtained, a unique mapping relationship between the reader and the spatial unit is established, and a reader deployment parameter table is generated. Bind RFID tags to construction tools, enter tag and tool information, and set initial positioning parameters for the tools; A hidden Markov localization model was built based on a spatially accessible topology map and a reader deployment parameter table. Five signal hiding states were defined: normal reception, metal obstruction, multipath reflection, trajectory drift, and signal interruption. The observation parameters corresponding to each state were calibrated using sample data from interference scenarios to complete the model initialization.

4. The method for positioning and managing construction tools based on RFID technology according to claim 1, characterized in that: All RFID fixed readers are triggered according to a preset clock synchronization mechanism to synchronously scan tags at a fixed collection cycle, collecting tag UID codes, RSSI values, signal reception duration, number of valid reads per cycle, reader number, and collection timestamp to generate a single-cycle raw positioning dataset. The raw positioning dataset is preprocessed and cleaned to remove invalid data due to communication failures, empty scans, or incomplete formats, retaining valid data that matches the tool tag UID codes. From each valid data, three types of observation features are extracted: signal strength fluctuation amplitude, continuous reading stability score, and consistency of repeated readings across readers, forming a standardized observation feature sequence.

5. The method for positioning and managing construction tools based on RFID technology according to claim 4, characterized in that: The standardized observation feature sequence is input into the localization model. The single-frame observation probability of each observation feature corresponding to the five signal hidden states is calculated through the observation probability matrix. The hidden state determination result of the previous cycle tool is retrieved. The state transition probability of each hidden state in the current cycle is obtained by combining the state transition probability matrix. The single-frame observation probability and the state transition probability are weighted and fused using the forward recursion algorithm to obtain the global posterior probability of each hidden state. The hidden state corresponding to the maximum value is selected as the current final state determination result. Based on the final judgment results, the data is classified and labeled. Data in normal reception state is compliant data, data in the state of metal obstruction, multipath reflection, and trajectory drift is data to be corrected, and data in the state of signal interruption is missing data.

6. The method for positioning and managing construction tools based on RFID technology according to claim 1, characterized in that: The compliant data and data to be corrected after classification and labeling are mapped to a construction-specific spatial reachability topology map. The spatial unit code is matched with the reader number to determine the candidate spatial unit for the positioning of the construction tool. The historical positioning compliant spatial unit code sequence of the tool in the last N frames is retrieved to form a historical and current spatial unit comparison dataset. Based on the physical reachability relationship in the topology map, spatial recursion constraint verification is performed on the comparison dataset to verify whether there is a direct physical reachability relationship between the current candidate spatial unit and the compliant spatial units in the last N consecutive frames. If the relationship is satisfied, it is marked as compliant positioning data; otherwise, it is marked as abnormal positioning data and collected according to the tag UID code.

7. The method for positioning and managing construction tools based on RFID technology according to claim 6, characterized in that: For abnormal data in the metal obstruction state, the compliant spatial unit of the tool in the previous cycle is used, and the positioning range within the compliant spatial unit is reduced to 50% of the compliant coordinate error range of the previous cycle; for abnormal data in the multipath reflection state, the candidate spatial unit corresponding to the reflection signal is removed, and the physical reachable spatial unit that is closest to the original candidate spatial unit in the spatial reachable topology map is selected as the new candidate spatial unit; for abnormal data in the trajectory drift state, all physical reachable spatial units between the historical compliant spatial unit and the current candidate spatial unit are locked to form a limited reachable unit interval. Abnormal data in signal interruption state is calculated using linear interpolation based on the compliant coordinates of N consecutive frames in the tool's history, and then matched to the spatial unit corresponding to the topology map.

8. The method for positioning and managing construction tools based on RFID technology according to claim 7, characterized in that: Perform spatial consistency verification on compliant and corrected positioning data, verifying whether the spatial cell codes and coordinate values ​​corresponding to all positioning data conform to the physical boundary coordinates of the spatial reachable topology map, and whether the coordinate values ​​are within the effective signal coverage radius recorded in the reader deployment parameter table. Remove any residual abnormal data that fails any verification. Classify and collect all verified positioning data according to the tool label UID code to generate a standardized positioning dataset containing the label UID code, spatial cell code, coordinates within the spatial cell, positioning status label, data collection timestamp, and corresponding reader number.

9. The method for positioning and managing construction tools based on RFID technology according to claim 1, characterized in that: The reader signal stability score and spatial unit positioning reliability benchmark value are extracted from the standardized positioning dataset. Weighting coefficients are assigned according to the signal stability score (60%) and the positioning reliability benchmark value (40%), and the weighted sum is used to obtain the comprehensive weight value of a single positioning data point. Based on the comprehensive weight value, a weighted fusion calculation is performed on the multi-source positioning data of the same construction tool to obtain the fused positioning coordinates within the spatial unit. The 3σ criterion is used to remove residual outliers with excessive dispersion to determine the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit. The tag UID code is bound one by one with the target spatial unit code, two-dimensional plane coordinates, positioning status label, and positioning result confidence score to generate a standardized positioning result data package and output it.

10. A construction tool positioning management system based on RFID technology, used to implement the method described in any one of claims 1 to 9, characterized in that: include: The initialization module divides the construction site into multiple spatial units, constructs a spatial reachability topology map, deploys RFID fixed readers for each spatial unit, establishes a reader deployment parameter table, and builds a positioning model containing at least five signal hiding states. The state determination module collects raw positioning data and extracts observation features. It inputs the observation features into the positioning model, combines them with historical state determination results, and uses a forward recursion algorithm to determine the current signal hiding state and classify and label the positioning data. The correction module maps the classified and labeled location data to a spatially reachable topology to obtain candidate spatial units for location. It identifies abnormal data through spatial recursive constraint verification and generates a standardized location dataset through spatial consistency verification. The positioning output module, based on a standardized positioning dataset, determines the target spatial unit corresponding to the construction tool and the two-dimensional plane coordinates within the target spatial unit, and generates and outputs a standardized positioning result data packet.

Images