Working area personnel positioning method and system based on data communication technology

Abstract

The invention discloses a working area personnel positioning method and system based on a data communication technology, and relates to the technical field of personnel positioning, and the method comprises the steps: estimating the position of a worker in real time through the fusion of an RFID tag and a multi-band signal, and the weighted fusion of a triangulation positioning algorithm and a dynamic signal; according to the method, extended Kalman filtering, multi-target tracking and particle filtering technologies are combined, signal interference is compensated, positioning precision is optimized, a fuzzy control algorithm is adopted for safety early warning, edge calculation and cloud calculation are combined, historical data are analyzed, and the moving trend of workers is predicted. According to the invention, by combining RFID tags, multi-band signals, dynamic signal fusion, multi-target tracking, particle filtering and fuzzy control technologies, high-precision and real-time personnel positioning and safety management are realized, the position of a worker can be accurately estimated, signal interference, attenuation and reflection effects in an environment can be effectively compensated, an early warning strategy is dynamically adjusted, and the working efficiency is improved. And the safety of the working area is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of worker positioning, and in particular to a method and system for positioning workers in a work area based on data communication technology. Background Art

[0002] In modern industrial and commercial environments, worker safety and efficient workspace management have become crucial issues that cannot be ignored. With companies placing increasing emphasis on safety management, particularly in hazardous work areas, warehouses, production lines, and other crowded environments, the ability to accurately track worker locations in real time and conduct dynamic safety management has become a crucial means of improving work efficiency and ensuring personnel safety.

[0003] Traditional personnel positioning methods generally use wireless radio frequency identification (RFID) technology for data communication, identifying the identity and location of workers through RFID tags and readers. Although RFID technology has certain advantages in some specific application scenarios, it also has problems such as limited positioning accuracy, environmental noise interference, and difficulty in positioning multiple people. Especially in complex working environments, the positioning error of a single signal source is large. Therefore, we propose a personnel positioning method and system for work area based on data communication technology. Summary of the Invention

[0004] The object of the present invention is to provide a method and system for locating personnel in a work area based on data communication technology, so as to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for locating personnel in a work area based on data communication technology, comprising the following steps:

[0006] Step 1: Communicate data between the RFID tag and the RFID reader to obtain the staff member's identity information, record the RFID tag signal of each staff member in the work area, and use multi-band signal collection technology to collect multi-band radio signals;

[0007] Step 2: Based on the strength of the RFID tag signal and the radio signal, a triangulation positioning algorithm is used to preliminarily estimate the position of the worker, and a dynamic signal fusion algorithm based on weighted average is used to perform weighted fusion on the position data of multiple signal sources;

[0008] Step 3: Based on multi-target tracking technology, the positions of multiple workers in the work area are tracked in real time using the extended Kalman filter and joint probabilistic data association algorithm. The motion trajectories of the workers are identified using the dynamic time warping algorithm to correct nonlinear errors in position and motion state.

[0009] Step 4: Introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. Combined with adaptive noise filtering technology, dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments.

[0010] Step 5: Use fuzzy control algorithms to determine whether a worker has entered a dangerous area and adjust the warning strategy based on the worker's status and trajectory. When a worker approaches a dangerous area, the system dynamically adjusts the alarm threshold in real time through a multi-level condition judgment mechanism, triggering different levels of safety alerts.

[0011] Step 6: Combine edge computing with cloud computing to distribute and process the location information of all staff members. Use machine learning algorithms to analyze historical data, predict staff movement trends, and provide intelligent decision-making support for safety management.

[0012] Preferably, the specific steps of step 1 are:

[0013] Step 1.1: Set up multiple RFID readers in the work area to receive RFID tags worn by staff through wireless signals and read the staff's identity information;

[0014] Step 1.2, associating the staff member's identity information with the unique identification code of their RFID tag, and transmitting the associated information to the data processing unit for subsequent processing;

[0015] Step 1.3: Filter the RFID signal to remove abnormal data caused by environmental noise or signal interference;

[0016] Step 1.4: Using multi-band radio signal collection technology, configure multiple sensor nodes in the area to collect radio signals of different frequency bands.

[0017] Preferably, the triangulation positioning algorithm in step 2 calculates the position of the target staff member based on the known positions of multiple base stations, through the signal arrival time and the difference in signal strength. The calculation process is: if the positions of multiple signal sources and the signal strength and time difference received by the staff from these signal sources are known, the coordinates of the staff member can be solved by the principle of geometric triangles.

[0018] Preferably, the dynamic signal fusion algorithm steps of the weighted average value in step 2 are:

[0019] Weight allocation: assigning different weights to each signal source. The weight allocation factors include signal quality, signal source stability, and signal-to-noise ratio of the signal source;

[0020] Dynamic adjustment: dynamically adjust the weight of each signal source;

[0021] Weighted average: Based on the weight of each signal source, the position information of all signal sources is integrated through the weighted average method to obtain the final position of the staff member. The specific calculation formula is:

[0022]

[0023] Among them, P final is the final position, P i is the preliminary positioning result of the i-th signal source, w i is the weight of the i-th signal source, and n is the number of signal sources.

[0024] Preferably, the specific steps of step 3 are:

[0025] Step 3.1. Assign an initial state to each worker, including position, velocity, and covariance matrix, and use the state vector representation of the Kalman filter:

[0026]

[0027] Among them, x k and y k The current position of the staff member, v x,k and v y,k The speed component of the worker;

[0028] Step 3.2: Use the extended Kalman filter to predict and correct the position of the worker. First, predict the position:

[0029] x k - =Fx k-1 +Bu k ;

[0030] Then make corrections based on the received measurement data:

[0031] K k =P k - H T (HP k - H T +R) -1 ;

[0032] x k =x k - +K k (z k -Hx k -);

[0033] Among them, K k is the Kalman gain, P k- is the predicted covariance, H is the measurement matrix, and R is the measurement noise matrix.

[0034] Preferably, the specific steps of step 3 further include:

[0035] Step 3.3: Associate the measurement data of multiple workers and use the joint probability data association algorithm to assign observations and tracking objects. By calculating the joint probability of each possible association scheme, the best association result is selected. The calculation formula is:

[0036]

[0037] Among them, m i represents the i-th hypothesis, P(m i ) is the prior probability of the hypothesis, P(z k |x k ,m i ) is the likelihood of observing a given location;

[0038] Step 3.4: Use the dynamic time warping algorithm to optimize the worker's motion trajectory. The dynamic time warping algorithm reduces nonlinear errors by aligning and correcting the time series of the trajectory. Its calculation formula is:

[0039] D(i,j)=dist(x i ,y j )+min(D(i-1,j),D(i,j-1),D(i-1,j-1));

[0040] Among them, dist(x i ,y j ) is the position point x i with y j The distance between them.

[0041] Preferably, the specific steps of step 4 include:

[0042] Step 4.1: Use the ion filter algorithm to compensate for multipath effects, signal attenuation, and reflection effects. The specific process of particle filtering is as follows:

[0043] Particle Generation:

[0044]

[0045] Particle weighting:

[0046]

[0047] Particle resampling:

[0048]

[0049] Step 4.2: Combine the adaptive noise filtering technology and adjust the filtering parameters according to the signal quality and environmental changes.

[0050] Preferably, the specific steps of step 5 are:

[0051] Step 5.1: Use the fuzzy control algorithm to determine whether the staff has entered the dangerous area. The control rules of the fuzzy control algorithm are:

[0052]

[0053] Among them, T threshold is the dynamic safety threshold, x danger are the coordinates of the center of the danger zone, is the location coordinates of the staff;

[0054] Step 5.2: When a worker approaches a dangerous area, the alarm threshold is dynamically adjusted through a multi-level condition judgment mechanism, and the early warning strategy is adjusted in a timely manner according to the worker's movement trend and location.

[0055] Preferably, the specific steps of step 6 include:

[0056] Step 6.1: Use edge computing technology to locally process data from various positioning devices and sensors, reducing data transmission delays and improving real-time response capabilities.

[0057] Step 6.2: Upload all location information to the cloud, analyze and predict the trend of the staff's historical location data through machine learning algorithms, predict the staff's future movement trajectory, and provide decision support for safety managers.

[0058] The present invention also provides a work area personnel positioning system based on data communication technology, comprising:

[0059] The collection module is used to communicate data with the RFID reader through the RFID tag to obtain the identity information of the staff member and record the RFID tag signal of each staff member in the work area, while collecting multi-band radio signals using multi-band signal collection technology;

[0060] The initial estimation module is used to preliminarily estimate the location of the worker based on the strength of the RFID tag signal and the radio signal using a triangulation positioning algorithm, and to perform weighted fusion of the location data of multiple signal sources using a dynamic signal fusion algorithm based on a weighted average;

[0061] The optimization module is used to track the positions of multiple workers in the work area in real time using multi-target tracking technology, extended Kalman filtering and joint probabilistic data association algorithms, and to identify the workers' motion trajectories using a dynamic time warping algorithm to correct nonlinear errors in position and motion state.

[0062] The adjustment module is used to introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. It also combines adaptive noise filtering technology to dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments.

[0063] The early warning module uses a fuzzy control algorithm to determine whether a worker has entered a dangerous area and adjusts the early warning strategy based on the worker's status and trajectory. When a worker approaches a dangerous area, the system dynamically adjusts the alarm threshold in real time through a multi-level condition judgment mechanism, triggering different levels of safety alerts.

[0064] The integrated module combines edge computing with cloud computing to perform distributed storage and processing of the location information of all staff members. It also analyzes historical data through machine learning algorithms, predicts the movement trends of staff members, and provides intelligent decision-making support for safety management.

[0065] Technical effects and advantages of the present invention:

[0066] The present invention achieves high-precision, real-time personnel positioning and safety management by combining RFID tags, multi-band signals, dynamic signal fusion, multi-target tracking, particle filtering and fuzzy control technology. It can not only accurately estimate the position of staff members and track their movement trajectory in real time, but also effectively compensate for signal interference, attenuation and reflection effects in the environment, dynamically adjust early warning strategies, and improve the safety of the work area. At the same time, combined with edge computing and cloud computing technologies, it realizes distributed data processing and intelligent decision support, greatly improving the response speed and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 Schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0068] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0069] The present invention provides Figure 1A method for locating personnel in a work area based on data communication technology is shown, comprising the following steps:

[0070] Step 1: Communicate data between the RFID tag and the RFID reader to obtain the staff member's identity information, record the RFID tag signal of each staff member in the work area, and use multi-band signal collection technology to collect multi-band radio signals;

[0071] The specific steps of step 1 are:

[0072] Step 1.1: Set up multiple RFID readers in the work area to receive wireless signals from RFID tags worn by workers and read their identity information. The deployment of multiple RFID readers ensures wider signal coverage in the work area and can receive RFID tag signals from more workers, thereby improving system reliability. The deployment of multiple RFID readers ensures wider signal coverage in the work area and can receive RFID tag signals from more workers, thereby improving system reliability and being able to obtain the location and identity information of workers in real time, providing information support for subsequent multi-target tracking.

[0073] Step 1.2: Associate the worker's identity information with the unique identification code of their RFID tag and transmit this associated information to the data processing unit for subsequent processing. By associating the worker's identity information with the unique identification code of their RFID tag, each worker's location information, identity information, and work tasks can be effectively managed. This ensures that the information remains consistent when transmitted from each reader to the data processing unit, facilitating subsequent calculation, analysis, and processing. After the information is transmitted to the data processing unit, deeper analysis such as dynamic signal fusion and trajectory optimization can be performed.

[0074] Step 1.3: Filter the RFID signal to remove abnormal data caused by environmental noise or signal interference. Filtering can effectively remove noise, interference signals, and abnormal data in the environment, ensuring that the RFID signal acquired by the system is more stable and accurate. By removing unnecessary noise, the quality of the RFID signal is improved, making the subsequent positioning and tracking process more accurate. Filtering helps reduce errors caused by signal instability and enhances system reliability.

[0075] Step 1.4: Utilize multi-band radio signal collection technology to configure multiple sensor nodes within the area to collect radio signals from different frequency bands. Multi-band signal collection enables simultaneous reception of signals from multiple radio frequency bands, such as Wi-Fi and Zigbee, enhancing the system's adaptability to diverse environmental conditions. Signals from different frequency bands complement each other, providing more information sources, optimizing worker location estimates, and reducing errors caused by signal attenuation or interference. Multi-band technology can effectively address various signal interferences in the environment (such as multipath effects) and improve the system's stability and robustness in complex environments.

[0076] Step 2: Based on the strength of the RFID tag signal and the radio signal, a triangulation positioning algorithm is used to preliminarily estimate the position of the worker, and a dynamic signal fusion algorithm based on weighted average is used to perform weighted fusion on the position data of multiple signal sources;

[0077] The triangulation algorithm in step 2 calculates the location of the target worker based on the known locations of multiple base stations and the differences in signal arrival times and signal strengths. The calculation process is as follows: if the locations of multiple signal sources and the signal strengths and time differences received by the worker from these signal sources are known, the coordinates of the worker can be solved using the principle of geometric triangles.

[0078] Triangulation based on signal strength: Assuming there are multiple signal sources, the coordinates of each signal source are known, and the target location is estimated by the difference in signal strength. Signal strength is usually related to distance, and the decrease in strength is inversely proportional to the increase in distance;

[0079] Assume there are three signal sources at known locations:

[0080] A(x1,y1),

[0081] B(x2,y2),

[0082] C(x3,y3),

[0083] The target position is (x, y), and the received signal strength of the signal source is P A 、P B 、P C , the relationship between signal strength and distance is expressed as:

[0084]

[0085] Among them, P0 is the signal strength of the reference position, d is the distance between the target and the signal source, and n is the path loss exponent. In order to calculate the position of the target based on the signal strength, it is necessary to calculate the received signal strength P A 、P B 、P C Convert to distance dA d B d C :

[0086]

[0087] These distance information are then used to construct a triangulation positioning equation, which is solved by the least squares method to obtain the target position (x, y).

[0088] Time difference-based triangulation relies on the time difference between multiple signal sources to estimate the target's position, assuming that the signal emitted by the target is received by at least three signal sources and the positions of the signal sources are known;

[0089] For three known signal sources:

[0090] A(x1,y1),

[0091] B(x2,y2),

[0092] C(x3,y3),

[0093] The target position is (x, y). Assuming that the signal propagation time from the target to each signal source is t1, t2, and t3 respectively, the corresponding distance is:

[0094] d1=c·t1,

[0095] d2=c·t2,

[0096] d3=c·t3,

[0097] Where c is the speed of light;

[0098] By establishing three equations about the target position and using the data of known signal sources and time differences, the following set of equations is constructed:

[0099]

[0100] These equations are solved by the least squares method to obtain the target position.

[0101] The steps of the dynamic signal fusion algorithm of the weighted average in step 2 are:

[0102] Weight allocation: assigning different weights to each signal source. The weight allocation factors include signal quality, signal source stability, and signal-to-noise ratio. Assigning different weights to each signal source can determine its contribution to the final position based on its signal quality, stability, and signal-to-noise ratio, ensuring that signal sources with higher accuracy have greater influence. By reasonably allocating weights, the impact of low-quality signal sources on the final result can be reduced, avoiding errors caused by unstable signal sources.

[0103] Dynamic adjustment: Dynamically adjust the weight of each signal source. As the environment changes, signal strength may fluctuate. The system will dynamically adjust the weight of each signal source to maintain the accuracy of position estimation. For example, if the signal strength of a signal source decreases, the system will automatically reduce the weight of that signal source and increase the weight of other signal sources to avoid errors caused by the instability of a single signal source. By dynamically adjusting the weight, it avoids position errors caused by relying on unstable signal sources in harsh environments and enhances the system's adaptability to environmental changes.

[0104] Weighted average: Based on the weight of each signal source, the position information of all signal sources is integrated through the weighted average method to obtain the final position of the staff member. The specific calculation formula is:

[0105]

[0106] Among them, P final is the final position, P i is the preliminary positioning result of the i-th signal source, w i is the weight of the i-th signal source, and n is the number of signal sources. The position information of multiple signal sources is fused by the weighted average method, which combines the advantages of multiple signal sources. The final position information is more accurate and reliable. The system can filter out errors caused by unstable signal sources or environmental influences, and obtain a more accurate staff location. Combining data from multiple signal sources can improve the accuracy and flexibility of the system under different conditions, especially in the case of multipath effects or signal loss.

[0107] Step 3: Based on multi-target tracking technology, the positions of multiple workers in the work area are tracked in real time using the extended Kalman filter and joint probabilistic data association algorithm. The motion trajectories of the workers are identified using the dynamic time warping algorithm to correct nonlinear errors in position and motion state.

[0108] The specific steps of step 3 are:

[0109] Step 3.1. Assign an initial state to each worker, including position, velocity, and covariance matrix, and use the state vector representation of the Kalman filter:

[0110]

[0111] Among them, x k and y k The current position of the staff member, v x,k and v y,kThe velocity component of the staff is set at the time of the operation. By setting the initial position, velocity and covariance matrix for each staff member, basic information is provided for the subsequent tracking process. The state vector contains the current position and velocity components of the staff member, providing a complete dynamic system model. The covariance matrix reflects the initial estimate of uncertainty in the system and provides the necessary mathematical basis for subsequent Kalman filter prediction and correction. The choice of initial state directly affects the effect of Kalman filter. Reasonable initial state setting can make the subsequent prediction and correction process more accurate.

[0112] Step 3.2: Use the extended Kalman filter to predict and correct the position of the worker. First, predict the position:

[0113] x k - =Fx k-1 +Bu k ;

[0114] Then make corrections based on the received measurement data:

[0115] K k =P k - H T (HP k - H T +R) -1 ;

[0116] x k =x k - +K k (z k -Hx k - );

[0117] Among them, K k is the Kalman gain, P k - is the predicted covariance, H is the measurement matrix, and R is the measurement noise matrix. The extended Kalman filter first predicts the position of the staff member through the known system model. The prediction step uses the state equation of the Kalman filter to estimate the position and speed of the staff member, so that the system can know the approximate position of the target in advance. Through the received measurement data (such as RFID signals and other wireless signals), the extended Kalman filter uses the Kalman gain to correct the prediction results, further reducing the error and improving the estimation accuracy. The Kalman gain reflects the weight relationship between the predicted value and the measured value, and automatically optimizes the update process. The extended Kalman filter updates the state and covariance through continuous iteration, gradually reducing the error accumulation and improving the real-time tracking accuracy of the system. This step can effectively eliminate high-frequency noise or signal errors in the system;

[0118] Step 3.3: Associate the measurement data of multiple workers and use the joint probability data association algorithm to assign observations and tracking objects. By calculating the joint probability of each possible association scheme, the best association result is selected. The calculation formula is:

[0119]

[0120] Among them, m i represents the i-th hypothesis, P(m i ) is the prior probability of the hypothesis, P(z k |x k ,m i ) is the likelihood of observing a given position. The joint probabilistic data association algorithm can associate data between multiple workers and signal sources by calculating the association probability between each observation data and the worker. The joint probabilistic data association takes into account the interference and mutual influence between multiple targets, ensuring that each measurement data is matched with the most appropriate worker. Through joint probability calculation, the joint probabilistic data association can effectively track the movement of multiple workers in a dynamic environment. The system is no longer affected by a single target, but considers the correlation between multiple workers, avoiding the occurrence of mismatching. The algorithm can select the best association result by calculating the joint probability of each possible association scheme, further optimizing the positioning accuracy of the workers and solving the complex interference problem between multiple targets.

[0121] Step 3.4: Use the dynamic time warping algorithm to optimize the worker's motion trajectory. The dynamic time warping algorithm reduces nonlinear errors by aligning and correcting the time series of the trajectory. Its calculation formula is:

[0122] D(i,j)=dist(x i ,y j )+min(D(i-1,j),D(i,j-1),D(i-1,j-1));

[0123] Among them, dist(x i ,y j ) is the position point x i with y jThe dynamic time warping algorithm can eliminate nonlinear errors in trajectories by aligning time series. For example, due to speed differences between different workers or changes in the environment, the time series of trajectories may be offset on the time axis. The dynamic time warping algorithm can align these trajectories to reduce the error accumulation caused by nonlinear errors. The dynamic time warping algorithm can optimize the worker's motion trajectory to reduce errors caused by environmental noise, measurement errors, or rapidly changing motion patterns, making the final trajectory smoother and more consistent with actual motion laws. Through the dynamic time warping algorithm, the system can accurately identify the worker's motion trajectory and ensure that the position estimate matches the actual trajectory, thereby improving the ability to predict and manage worker behavior.

[0124] Step 4: Introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. Combined with adaptive noise filtering technology, dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments.

[0125] The specific steps of step 4 include:

[0126] Step 4.1: Use the ion filter algorithm to compensate for multipath effects, signal attenuation, and reflection effects. The specific process of particle filtering is as follows:

[0127] Particle Generation:

[0128]

[0129] Particle filtering represents the state space of the system by generating multiple particles (samples). Each particle represents a possible state (such as the position of a worker). These particles are usually evenly distributed within the assumed initial state range and are used to estimate the uncertainty of the current position. The generated particle distribution helps the system detect all potential possible positions in complex environments. The particle distribution determines the error range of the system in the initial stage. Reasonable particle initialization can effectively reduce the error accumulation in the subsequent tracking process.

[0130] Particle weighting:

[0131]

[0132] Particle weighting is performed by evaluating the "weight" of each particle based on measurement data (such as signal strength, time delay, etc.). This weight represents the possibility of the state represented by the particle in the current environment. The difference between the actual information received and the predicted value is used to calculate the weight, so that the particles are more concentrated in the possible state area. Through weighting, the particle filter can automatically identify and enhance the real positioning information, thereby improving positioning accuracy. Especially in complex environments (such as multipath effects or signal reflection areas), particle weighting can effectively filter out irrelevant hypotheses;

[0133] Particle resampling:

[0134]

[0135] Particle resampling regenerates a new particle set by discarding low-weight particles and retaining high-weight particles, making the particle set more concentrated in the most likely state area. Resampling can avoid positioning errors caused by excessive dispersion of particles, allowing the system to focus more accurately on the most likely target position. Resampling effectively solves the "particle depletion" problem in particle filtering, prevents the situation where a large number of particles have too low weights, and thus improves tracking accuracy.

[0136] Step 4.2: Combined with adaptive noise filtering technology, the filtering parameters are adjusted according to the signal quality and environmental changes. As the environment and signal conditions change (such as signal quality fluctuations, multipath effects, environmental noise, etc.), the system can dynamically adjust the filtering parameters (such as noise covariance matrix, particle weight calculation, etc.). Adaptive noise filtering adjusts the filtering parameters in real time according to the quality of real-time measurement data to ensure optimal signal processing performance. By dynamically adjusting the filter parameters, the system can adapt to signal fluctuations and instabilities in various working environments, especially in complex or unknown environments, to avoid signal loss or inaccurate positioning. Adaptive noise filtering can effectively reduce errors caused by signal noise, reflection effects, attenuation and multipath effects. By automatically adjusting the noise parameters, the system can adapt to different noise environments, thereby improving signal accuracy and enhancing positioning stability. This adaptive adjustment mechanism can work stably in complex environments, avoid system performance fluctuations due to environmental changes, and enhance the robustness and reliability of the system in dynamic environments.

[0137] Step 5: Use fuzzy control algorithms to determine whether a worker has entered a dangerous area and adjust the warning strategy based on the worker's status and trajectory. When a worker approaches a dangerous area, the system dynamically adjusts the alarm threshold in real time through a multi-level condition judgment mechanism, triggering different levels of safety alerts.

[0138] The specific steps of step 5 are:

[0139] Step 5.1: Use the fuzzy control algorithm to determine whether the staff has entered the dangerous area. The control rules of the fuzzy control algorithm are:

[0140]

[0141] Among them, T threshold is the dynamic safety threshold, x danger are the coordinates of the center of the danger zone, is the position coordinate of the worker; the fuzzy control algorithm converts the distance between the worker and the dangerous area into a fuzzy set (such as "far", "middle", and "near"), and judges whether the worker has entered the dangerous area based on these fuzzy values. This fuzzy judgment method can handle the uncertainty and ambiguity in the environment, and is more flexible and adaptable than traditional hard judgments (such as distance threshold judgments). The fuzzy control system can adapt to changing situations according to changes in distance, the complexity of the working environment, and different safety standards, and avoid wrong judgments due to errors. The safety threshold is dynamically adjusted according to the current status of the worker (such as position, speed) and environmental factors. As the worker moves, the dynamic threshold is updated in real time. For example, when the worker moves quickly, the system may need to lower the safety threshold of the dangerous area, thereby triggering an early warning. By dynamically adjusting the safety threshold, the system can make personalized responses based on the actual situation of the worker (such as direction of travel, speed, etc.) to ensure that potential safety threats are discovered in a timely manner;

[0142] Step 5.2, when the staff approaches the danger zone, the alarm threshold is dynamically adjusted through the multi-level condition judgment mechanism, and the early warning strategy is adjusted in time according to the movement trend and position of the staff; the multi-level condition judgment mechanism responds to different danger levels by setting multiple alarm thresholds. For example, when the staff enters the periphery of the danger zone, the first level early warning is triggered; when the staff approaches the danger zone, the second level early warning is triggered; when the staff enters the danger zone, the third level early warning is triggered. Each early warning level has different response measures (such as different alarm intensity, different response time, etc.). This multi-level mechanism can increase the early warning intensity step by step as the staff gradually approaches the danger zone, ensuring that safety management can be carried out in a hierarchical manner, avoiding unnecessary emergency responses or failure to respond to risks in time due to early or late early warnings. According to the movement trend and position changes of the staff, the system It can dynamically adjust the alarm threshold in real time. For example, if the staff moves faster, the system will automatically lower the distance threshold for triggering the alarm and issue an early warning; if the staff moves slower, the alarm threshold can be appropriately relaxed to avoid excessive alarms. By analyzing the speed, direction and behavior patterns of the staff in real time and dynamically adjusting the alarm threshold, the system can be made more intelligent and avoid false alarms or missed alarms caused by traditional static threshold settings. According to the real-time location and trend of the staff, the system can adjust the early warning strategy, such as using different alarm methods (sound, light signals, vibration, etc.) to remind the staff to enter the dangerous area, or through manual intervention for emergency treatment. By flexibly adjusting the early warning strategy, the system can dynamically respond according to the actual behavior of the staff, reduce false alarms caused by untimely or too frequent early warnings, and improve the accuracy of safety management.

[0143] Step 6: Combine edge computing with cloud computing to distribute and process the location information of all staff members. Use machine learning algorithms to analyze historical data, predict staff movement trends, and provide intelligent decision-making support for safety management.

[0144] The specific steps of step 6 include:

[0145] Step 6.1. Use edge computing technology to process data from various positioning devices and sensors locally, reducing data transmission delays and improving real-time response capabilities. Edge computing processes data close to the data source (such as on-site sensor nodes), avoiding the delay of transmitting large amounts of data to the remote cloud for processing. This enables positioning data to be processed more quickly, thereby improving real-time response capabilities. Local processing can immediately respond to environmental changes and changes in staff locations, improving the system's response speed to emergencies. Edge computing reduces the amount of data transmitted and transfers data processing and analysis to the local area, thereby alleviating pressure on network bandwidth. For systems that need to process large amounts of sensor data, especially in high-density device deployments and wireless network environments, edge computing can greatly improve system stability and processing capabilities.

[0146] Step 6.2: Upload all location information to the cloud. Machine learning algorithms are used to analyze and predict trends in the historical location data of workers, predict future movement trajectories, and provide decision support for safety managers. After uploading all worker location information to the cloud, the cloud platform can perform large-scale data analysis and processing. Combining historical data with location changes, machine learning algorithms (such as time series analysis, regression analysis, and cluster analysis) are used to predict future movement trends. These predictions can help safety managers identify potential risks in advance and make corresponding adjustments. Through in-depth analysis of historical data using machine learning algorithms, the system can identify worker behavior patterns, regular routes, and potential behavioral anomalies, thereby predicting in advance the tendency of workers to enter dangerous areas. Based on the predictions, safety managers can adjust the distribution of security personnel or set appropriate safety zones in advance, reducing unnecessary resource waste. The cloud's computing power enables real-time processing of large-scale data and cross-regional sharing, allowing the location information of workers in multiple work areas to be monitored and analyzed simultaneously. Safety managers can view the safety status of different work areas in real time from anywhere in the world, supporting cross-regional safety management and collaborative decision-making. Through cloud-based data sharing and analysis, multiple management teams can share the same data set and work together, improving the efficiency and coordination of cross-regional management.

[0147] The present invention also provides a work area personnel positioning system based on data communication technology, comprising:

[0148] The collection module is used to communicate data with the RFID reader through the RFID tag to obtain the identity information of the staff member and record the RFID tag signal of each staff member in the work area, while collecting multi-band radio signals using multi-band signal collection technology;

[0149] The initial estimation module is used to preliminarily estimate the location of the worker based on the strength of the RFID tag signal and the radio signal using a triangulation positioning algorithm, and to perform weighted fusion of the location data of multiple signal sources using a dynamic signal fusion algorithm based on a weighted average;

[0150] The optimization module is used to track the positions of multiple workers in the work area in real time using multi-target tracking technology, extended Kalman filtering and joint probabilistic data association algorithms, and to identify the workers' motion trajectories using a dynamic time warping algorithm to correct nonlinear errors in position and motion state.

[0151] The adjustment module is used to introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. It also combines adaptive noise filtering technology to dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments.

[0152] The early warning module uses a fuzzy control algorithm to determine whether a worker has entered a dangerous area and adjusts the early warning strategy based on the worker's status and trajectory. When a worker approaches a dangerous area, the system dynamically adjusts the alarm threshold in real time through a multi-level condition judgment mechanism, triggering different levels of safety alerts.

[0153] The integrated module combines edge computing with cloud computing to perform distributed storage and processing of the location information of all staff members. It also analyzes historical data through machine learning algorithms, predicts the movement trends of staff members, and provides intelligent decision-making support for safety management.

[0154] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for locating personnel in a work area based on data communication technology, comprising the following steps: Step 1: Communicate data between the RFID tag and the RFID reader to obtain the staff member's identity information, record the RFID tag signal of each staff member in the work area, and use multi-band signal collection technology to collect multi-band radio signals; Step 2: Based on the strength of the RFID tag signal and the radio signal, a triangulation positioning algorithm is used to preliminarily estimate the position of the worker, and a dynamic signal fusion algorithm based on weighted average is used to perform weighted fusion on the position data of multiple signal sources; Step 3: Based on multi-target tracking technology, the positions of multiple workers in the work area are tracked in real time using the extended Kalman filter and joint probabilistic data association algorithm. The motion trajectories of the workers are identified using the dynamic time warping algorithm to correct nonlinear errors in position and motion state. Step 4: Introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. Combined with adaptive noise filtering technology, dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments. Step 5: Use the fuzzy control algorithm to determine whether the staff has entered the danger zone, and adjust the warning strategy according to the status and trajectory of the staff. The alarm threshold is adjusted dynamically in real time through a multi-level condition judgment mechanism; Step 6: Combine edge computing with cloud computing to perform distributed storage and processing of staff location information, analyze historical data through machine learning algorithms, and predict staff movement trends.

2. A method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 1 are: Step 1.1: Set up multiple RFID readers in the work area to receive RFID tags worn by staff through wireless signals and read the staff's identity information; Step 1.2, associating the staff member's identity information with the unique identification code of their RFID tag, and transmitting the associated information to the data processing unit for subsequent processing; Step 1.3: Filter the RFID signal to remove abnormal data caused by environmental noise or signal interference; Step 1.4: Using multi-band radio signal collection technology, configure multiple sensor nodes in the area to collect radio signals of different frequency bands.

3. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The triangulation positioning algorithm in step 2 calculates the position of the target staff member based on the known positions of multiple base stations, the signal arrival time and the difference in signal strength. The calculation process is as follows: if the positions of multiple signal sources and the signal strength and time difference received by the staff member from these signal sources are known, the coordinates of the staff member can be solved by the principle of geometric triangles.

4. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The dynamic signal fusion algorithm steps of the weighted average value in step 2 are as follows: Weight allocation: assigning different weights to each signal source. The weight allocation factors include signal quality, signal source stability, and signal-to-noise ratio of the signal source; Dynamic adjustment: dynamically adjust the weight of each signal source; Weighted average: Based on the weight of each signal source, the position information of all signal sources is integrated through the weighted average method to obtain the final position of the staff member. The specific calculation formula is: Among them, P final is the final position, P i is the preliminary positioning result of the i-th signal source, w i is the weight of the i-th signal source, and n is the number of signal sources.

5. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 3 are: Step 3.

1. Assign an initial state to each worker, including position, velocity, and covariance matrix, and use the state vector representation of the Kalman filter: Among them, x k and y k The current position of the staff member, v x,k and v y,k The speed component of the worker; Step 3.2: Use the extended Kalman filter to predict and correct the position of the worker. First, predict the position: x k - =Fx k-1 +Bu k 4 Then make corrections based on the received measurement data: K k =P k - H T (HP k - H T +R) -1 ; x k =x k - +K k (z k -Hx k -); Among them, K k is the Kalman gain, P k - is the predicted covariance, H is the measurement matrix, and R is the measurement noise matrix.

6. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 3 also include: Step 3.3: Associate the measurement data of multiple workers and use the joint probabilistic data association algorithm to assign observations and tracking objects. By calculating the joint probability of each possible association scheme, the best association result is selected. The calculation formula is: Among them, m i represents the i-th hypothesis, P(m i ) is the prior probability of the hypothesis, P(z k |x k ,m i ) is the likelihood of observing a given location; Step 3.4: Use the dynamic time warping algorithm to optimize the worker's motion trajectory. The dynamic time warping algorithm reduces nonlinear errors by aligning and correcting the time series of the trajectory. Its calculation formula is: D(i,j)=dist(x i ,y j )+min(D(i-1,j),D(i,j-1),D(i-1,j-1)); Among them, dist(x i ,y j ) is the position point x i with y j The distance between them.

7. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 4 include: Step 4.1: Use the ion filter algorithm to compensate for multipath effects, signal attenuation, and reflection effects. The specific process of particle filtering is as follows: Particle Generation: Particle Weighting: Particle resampling: Step 4.2: Combine the adaptive noise filtering technology and adjust the filtering parameters according to the signal quality and environmental changes.

8. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 5 are: Step 5.1: Use the fuzzy control algorithm to determine whether the staff has entered the dangerous area. The control rules of the fuzzy control algorithm are: Among them, T threshold is the dynamic safety threshold, x danger are the coordinates of the center of the danger zone, is the location coordinates of the staff; Step 5.2: When a worker approaches a dangerous area, the alarm threshold is dynamically adjusted through a multi-level condition judgment mechanism, and the early warning strategy is adjusted in a timely manner according to the worker's movement trend and location.

9. The method for locating personnel in a work area based on data communication technology according to claim 1, characterized in that: The specific steps of step 6 include: Step 6.1: Use edge computing technology to locally process data from various positioning devices and sensors, reducing data transmission delays and improving real-time response capabilities. Step 6.2: Upload all location information to the cloud, analyze and predict the trend of the staff's historical location data through machine learning algorithms, predict the staff's future movement trajectory, and provide decision support for safety managers.

10. A work area personnel positioning system based on data communication technology, characterized in that: include: The collection module is used to communicate data with the RFID reader through the RFID tag to obtain the identity information of the staff member and record the RFID tag signal of each staff member in the work area, while collecting multi-band radio signals using multi-band signal collection technology; The initial estimation module is used to preliminarily estimate the location of the worker based on the strength of the RFID tag signal and the radio signal using a triangulation positioning algorithm, and to perform weighted fusion of the location data of multiple signal sources using a dynamic signal fusion algorithm based on a weighted average; The optimization module is used to track the positions of multiple workers in the work area in real time using multi-target tracking technology, extended Kalman filtering and joint probabilistic data association algorithms, and to identify the workers' motion trajectories using a dynamic time warping algorithm to correct nonlinear errors in position and motion state. The adjustment module is used to introduce a particle filter algorithm to model and compensate for signal interference, signal attenuation, and reflection effects in the environment. It also combines adaptive noise filtering technology to dynamically adjust the filter parameters to address multipath effects and signal loss issues in different working environments. The early warning module uses a fuzzy control algorithm to determine whether a worker has entered a dangerous area and adjusts the early warning strategy based on the worker's status and trajectory. When a worker approaches a dangerous area, the system dynamically adjusts the alarm threshold in real time through a multi-level condition judgment mechanism, triggering different levels of safety alerts. The integrated module combines edge computing with cloud computing to perform distributed storage and processing of the location information of all staff members. It also analyzes historical data through machine learning algorithms, predicts the movement trends of staff members, and provides intelligent decision-making support for safety management.

Images