A construction personnel positioning method and positioning system in a power tunnel

Abstract

The present invention provides a method and a system for positioning construction personnel in an electric power tunnel, belonging to the field of positioning technology, and comprising obtaining measurement data; constructing a TDOA model, an AOA model and an RSSI model based on TDOA data, AOA data and RSSI values, respectively, and in combination with the position of a UWB receiver, and optimizing each model respectively by taking magnetic field induction data as a position constraint condition; quantifying the difference between the predicted value and the actual measured value of each optimized model to obtain an error equation set; converting each error equation in the error equation set into an error term, and integrating all error terms to construct an objective function; optimizing and solving the least squares objective function by a gradient descent method to obtain an optimal estimate of the construction personnel's position. The present invention effectively improves the positioning accuracy and real-time performance of construction personnel by fusing a variety of measurement data, constructing a plurality of positioning models, and optimizing and solving by a gradient descent method, and provides a strong guarantee for the safe construction of electric power tunnels.

Description

Technical Field

[0001] The present invention relates to the field of positioning technology, and in particular to a method and a system for positioning construction personnel in a power tunnel. Background Art

[0002] During the construction and maintenance of power tunnels, ensuring the safety of construction workers is of vital importance. Due to the complex environment of power tunnels, there are many potential safety hazards, such as insufficient light, narrow space, and numerous equipment. Once an accident occurs, it will pose a huge threat to the life safety of construction workers. Therefore, achieving accurate positioning of construction workers and timely grasping their location information is of great significance to improving construction safety and emergency response speed.

[0003] Traditional personnel positioning methods mostly rely on a single positioning technology, such as RFID, Bluetooth, WiFi, etc., but these technologies have many limitations in complex environments. For example, although RFID technology is low-cost and easy to install, its signal transmission range is limited and it is easily affected by multipath effects; although Bluetooth positioning technology is low-cost and low-energy, it also has problems with signal interference and multipath effects; although WiFi positioning technology has high accuracy, it requires the deployment of a large number of base stations in the tunnel, which is costly and complex to deploy.

[0004] Therefore, it is necessary to provide a method and system for positioning construction personnel in a power tunnel to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method and a system for positioning construction personnel in a power tunnel. Through key technical means such as multi-source data fusion, multi-model construction and optimization, and gradient descent optimization solution, the present invention effectively overcomes the shortcomings of existing positioning technology in the tunnel environment, realizes high-precision and real-time positioning of the construction personnel, and provides strong support for the construction safety and management of power tunnels.

[0006] The present invention provides a method for positioning construction personnel in a power tunnel, the positioning method comprising the following steps:

[0007] S1: acquiring measurement data in real time and performing preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data;

[0008] S2: Based on the TDOA data, AOA data and RSSI value, and in combination with the position of the UWB receiver, a TDOA model, an AOA model and an RSSI model are respectively constructed, and the magnetic field sensing data is used as a position constraint to optimize each model;

[0009] S3: quantify the difference between the predicted value of each optimized model and the actual measured value to obtain a set of error equations;

[0010] S4: converting each error equation in the error equation set into an error term, and integrating all error terms to construct an objective function, wherein the objective function adopts a least squares criterion;

[0011] S5: The least squares objective function is optimized and solved by the gradient descent method to obtain the optimal estimate of the construction personnel's position.

[0012] Specifically, step S1 specifically includes the following steps:

[0013] S101: capturing TDOA data and AOA data in real time through a UWB receiver, and capturing RSSI value and magnetic field sensing data in real time through a composite tag, wherein the composite tag includes a magnetic field sensor and an RSSI module;

[0014] S102: screening and synchronizing the measurement data to obtain screened measurement data;

[0015] S103: performing standardization processing on the screened measurement data to obtain standardized measurement data.

[0016] Specifically, step S2 specifically includes the following steps:

[0017] S201: Preliminarily estimate the potential position of the construction personnel through TDOA data, AOA data and RSSI value to obtain a preliminary position estimation result;

[0018] S202: constructing a model based on the preliminary position estimation result and the known position of the UWB receiver to obtain a TDOA model;

[0019] S203: constructing a model according to the preliminary position estimation result and the azimuth information of the UWB receiver to obtain an AOA model;

[0020] S204: constructing a model based on the preliminary location estimation result and the RSSI value collected by the composite tag to obtain an RSSI model;

[0021] S205: Using the magnetic field sensing data as a position constraint condition, and optimizing the TDOA model, the AOA model, and the RSSI model.

[0022] Specifically, step S205 specifically includes the following steps:

[0023] S205a: Analyze the magnetic field induction data based on the magnetic field distribution characteristics of the power tunnel where the construction workers are located, and obtain a spatial distribution model of the magnetic field intensity in the tunnel;

[0024] S205b: limiting the location range of the construction personnel according to the spatial distribution model of the magnetic field strength to form a location constraint condition;

[0025] S205c: Apply the position constraint condition to the TDOA model, the AOA model and the RSSI model, and adjust the parameters of each model until the error between the predicted position of each model and the expected position under the magnetic field constraint condition is minimized.

[0026] Specifically, step S3 specifically includes the following steps:

[0027] S301: By comparing the predicted value of each model with the actual measured value, the prediction deviation of each model is evaluated and quantified to obtain the error margin of each model, and the error margin is expressed as: ;

[0028] in, For the The measurement point The error margin of the model is It is the index of TDOA model, AOA model and RSSI model. For the The actual measured value of the measuring point, For Model In the The predicted value at each measurement point;

[0029] S302: According to the error amplitude of each model, an error equation set is constructed by an error quantification method. The error equation set is expressed as: ;

[0030] in, is the set of error equations, is the number of measurement points, and , For the TDOA model The error margin at each measurement point is For the AOA model The error margin at each measurement point is For RSSI model The error margin at each measurement point.

[0031] Specifically, step S4 specifically includes the following steps:

[0032] S401: transform each error equation in the error equation set to obtain an error term representing the prediction error of each model, wherein the mathematical expression of the error term is: ;

[0033] S402: weighting the error terms based on the credibility of the prediction errors of each model to obtain a set of weighted error terms. The mathematical expression of the weighted error terms is: ;

[0034] S403: constructing an objective function using the least squares criterion based on the weighted error term set, the mathematical expression of the objective function is: ;

[0035] in, is the objective function.

[0036] Specifically, step S5 specifically includes the following steps:

[0037] S501: Initialize by setting initial parameters and hyperparameters of the gradient descent method to obtain a starting point for optimization solution;

[0038] S502: applying a gradient descent method to the objective function and iteratively updating parameters to obtain a parameter sequence that gradually approaches an optimal solution;

[0039] S503: The parameter updating process is terminated according to a preset convergence criterion to obtain an optimal estimate of the construction personnel's position. The mathematical expression is: .

[0040] The present invention also provides a construction personnel positioning system in a power tunnel, which is used to execute the construction personnel positioning method in a power tunnel, and the positioning system comprises:

[0041] A data processing module, used for acquiring measurement data in real time and performing preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data;

[0042] A model building module, for building a TDOA model, an AOA model and an RSSI model based on the TDOA data, the AOA data and the RSSI value and in combination with the position of the UWB receiver, and optimizing each model by taking the magnetic field sensing data as a position constraint;

[0043] The error quantification module is used to quantify the difference between the predicted value and the actual measured value of each optimized model to obtain a set of error equations;

[0044] An objective function construction module, used for converting each error equation in the error equation set into an error term, and integrating all error terms to construct an objective function, wherein the objective function adopts a least squares criterion;

[0045] The positioning optimization solution module is used to optimize and solve the least squares objective function through the gradient descent method to obtain the optimal estimate of the construction personnel's position.

[0046] Compared with the related art, the method and system for positioning construction personnel in a power tunnel provided by the present invention have the following beneficial effects:

[0047] The present invention obtains measurement data and performs preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field induction data; then, based on the TDOA data, AOA data and RSSI value, and in combination with the position of a UWB receiver, a TDOA model, an AOA model and an RSSI model are respectively constructed, and the magnetic field induction data is used as a position constraint to optimize each model; then, the difference between the predicted value and the actual measured value of each optimized model is quantified to obtain an error equation set; then, each error equation in the error equation set is converted into an error term, and all error terms are integrated to construct an objective function; finally, the least squares objective function is optimized and solved by a gradient descent method to obtain the optimal estimated value of the construction personnel's position. The present invention effectively improves the positioning accuracy and real-time performance of the construction personnel by fusing multiple measurement data, constructing multiple positioning models, and optimizing and solving them by a gradient descent method, thereby providing a strong guarantee for the safe construction of power tunnels. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flowchart of a method for locating construction personnel in a power tunnel provided by the present invention;

[0049] Figure 2 This is a module structure diagram of a construction personnel positioning system in a power tunnel provided by the present invention. DETAILED DESCRIPTION

[0050] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only the parts related to the present invention, rather than all structures, are shown in the accompanying drawings. In addition, the embodiments of the present invention and the features in the embodiments may be combined with each other without conflict.

[0051] It should also be noted that, for ease of description, only the parts related to the present invention, but not all of the contents, are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0052] In the method for locating construction workers in power tunnels proposed by the present invention, UWB receivers and composite tags play a vital role. UWB (Ultra Wide Band) receivers are advanced wireless communication technology devices known for their high-precision time and spatial resolution, and can capture ultra-wideband pulse signals from tags worn by construction workers. These signals carry TDOA (Time Difference Of Arrival) and AOA (Angle Of Arrival) information. Based on this information, UWB receivers can calculate the time difference and angle between the arrival of signals at different receivers, thereby providing key positioning clues for positioning algorithms.

[0053] The composite tag, as an integrated multifunctional intelligent device, is carried by construction workers. It not only has a UWB module for transmitting and receiving ultra-wideband signals, but also integrates a magnetic field sensor and an RSSI (Received Signal Strength Indicator) module. The magnetic field sensor can detect changes in the magnetic field of the surrounding environment. This information is particularly important in power tunnels under complex electromagnetic environments because it can provide additional environmental features to help the positioning system more accurately determine the location of construction workers. The RSSI module is responsible for measuring signal strength, which helps to estimate the relative distance between the tag and the receiver, especially in non-line-of-sight propagation environments. The RSSI value can be used as auxiliary information to enhance the robustness and accuracy of positioning.

[0054] Through the collaborative work of UWB receivers and composite tags, this positioning method can collect rich measurement data, including TDOA, AOA, RSSI values ​​and magnetic field sensing data. After preprocessing, these data are used to build and optimize the positioning model, and finally the optimal position estimate of the construction personnel is obtained through the least squares criterion and gradient descent method, achieving high-precision positioning in the special environment of power tunnels.

[0055] Embodiment 1

[0056] The present invention provides a method for positioning construction personnel in a power tunnel, referring to Figure 1 As shown, the positioning method includes the following steps:

[0057] S1: Acquire measurement data in real time and perform preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data.

[0058] In this embodiment, the UWB receiver captures TDOA (time difference of arrival) and AOA (angle of arrival) data in real time, while the composite tag collects RSSI (received signal strength) values ​​and magnetic field sensing data. The composite tag includes a magnetic field sensor and RSSI module, which can be carried and transmit data in real time. The preprocessing stage involves data screening, synchronization and standardization to ensure data quality, remove outliers, synchronize timestamps of different receivers, and format the data into a consistent standard format to provide reliable basic data for subsequent processing.

[0059] Preprocessing ensures that the positioning algorithm uses high-quality, synchronized and uniformly formatted measurement data, reduces noise and error sources, and improves positioning accuracy and stability.

[0060] S2: Based on the TDOA data, AOA data and RSSI value, and in combination with the position of the UWB receiver, a TDOA model, an AOA model and an RSSI model are respectively constructed, and at the same time, the magnetic field sensing data is used as a position constraint condition to optimize each model.

[0061] In this embodiment, TDOA, AOA and RSSI models are constructed based on the preprocessed TDOA data, AOA data and RSSI values. These models use the location information and measurement data of the UWB receiver to estimate the potential location of the construction personnel. The magnetic field sensing data is used as a constraint to optimize the model to ensure that the location predicted by the model conforms to the magnetic field distribution characteristics of the actual environment and improve the accuracy of positioning.

[0062] Through model optimization, especially after introducing magnetic field constraints, the positioning results are closer to the real environment, reducing positioning errors and improving the practical value of the positioning system.

[0063] S3: Quantify the differences between the predicted values ​​of each optimized model and the actual measured values ​​to obtain a set of error equations.

[0064] In this embodiment, the difference between each model prediction value and the actual measurement value is quantified to form a set of error equations. Each error equation represents the degree of deviation between the model prediction and the actual measurement at a certain point, which provides a mathematical basis for the subsequent construction of the objective function and is the basis of least squares optimization in the positioning algorithm.

[0065] S4: Convert each error equation in the error equation set into an error term, and integrate all error terms to construct an objective function, wherein the objective function adopts a least squares criterion.

[0066] In this embodiment, the error equation is converted into error terms, each of which reflects the magnitude of the model prediction error. All error terms are integrated into an objective function, using the least squares criterion, that is, the objective function of minimizing the sum of squares of all error terms.

[0067] The construction of the objective function provides a clear mathematical form for the subsequent optimization solution, ensuring that the positioning algorithm can be optimized in the direction of reducing the overall error.

[0068] S5: The least squares objective function is optimized and solved by the gradient descent method to obtain the optimal estimate of the construction personnel's position.

[0069] In this embodiment, the constructed objective function is optimized using the gradient descent method, and the objective function value is gradually reduced by iteratively updating the parameters until the convergence condition is reached, thereby obtaining the optimal estimated value of the construction personnel's position.

[0070] Through optimization, the positioning algorithm can find the parameter combination that minimizes the error, thereby obtaining the most accurate estimate of the construction worker's location.

[0071] Specifically, step S1 specifically includes the following steps:

[0072] S101: capturing TDOA data and AOA data in real time through a UWB receiver, and capturing RSSI value and magnetic field sensing data in real time through a composite tag, wherein the composite tag includes a magnetic field sensor and an RSSI module.

[0073] In this embodiment, the UWB receiver and the composite tag collect measurement data in real time. The UWB receiver captures TDOA (time difference of arrival) and AOA (angle of arrival) data, while the composite tag collects RSSI (received signal strength) values ​​and magnetic field sensing data. The composite tag integrates a magnetic field sensor and RSSI module, which can be carried with you and transmit data to the receiver in real time.

[0074] Real-time data capture ensures the responsiveness and dynamic tracking capabilities of the positioning system, providing instant positioning information even in fast-moving scenarios. At the same time, the integrated design of the composite tag reduces hardware requirements, simplifies field deployment, and improves system portability and efficiency.

[0075] S102: Screening and synchronizing the measurement data to obtain screened measurement data.

[0076] In this embodiment, outliers and noise are removed from the collected data through a data screening process, and only valid and reliable measurement data are retained. In addition, the data from multiple receivers are time-stamped and synchronized to ensure that all data are consistent in time, which is particularly important for TDOA and AOA data because they rely on precise time synchronization.

[0077] Data screening and synchronization improve the data quality of the positioning system, reduce positioning errors caused by outliers and asynchrony, and enhance the stability and reliability of positioning results.

[0078] S103: performing standardization processing on the screened measurement data to obtain standardized measurement data.

[0079] In this embodiment, the screened measurement data is standardized, including but not limited to the unification of data format, consistency correction of dimensions and normalization of data range. Standardization ensures that all measurement data are comparable and compatible, which facilitates subsequent data analysis and model construction.

[0080] Data standardization eliminates the differences between different data sources, improves the processability of data and the generalization ability of the model, and reduces errors caused by inconsistent data formats and dimensions.

[0081] Specifically, step S2 specifically includes the following steps:

[0082] S201: Preliminary estimation of the potential location of the construction personnel is performed through TDOA data, AOA data and RSSI value to obtain a preliminary location estimation result.

[0083] In this embodiment, TDOA (Time Difference of Arrival) data processing: The signal arrival time difference captured by multiple UWB receivers is used to solve the signal source position through the least square method.

[0084] AOA (Angle of Arrival) data processing: Using the multi-antenna array in the UWB receiver, the angle of incidence of the signal is calculated and determined by the cross-correlation method.

[0085] RSSI (Received Signal Strength Indicator) value analysis: Evaluates the relationship between signal strength and distance, using fingerprint database matching to estimate distance.

[0086] S202: Build a model based on the preliminary position estimation result and the known position of the UWB receiver to obtain a TDOA model.

[0087] In this embodiment, the nonlinear equation system is solved by taking the preliminary position estimate as the initial point and minimizing the residual sum of squares between the TDOA measurement and the theoretical TDOA through a Gauss-Newton method iterative algorithm.

[0088] Multipath effect correction: Taking into account the possible multipath signal propagation in the tunnel, a correction factor is introduced or a multipath model is used to adjust the calculation.

[0089] S203: construct a model according to the preliminary position estimation result and the orientation information of the UWB receiver to obtain an AOA model.

[0090] In this embodiment, the antenna array layout is optimized: a uniform circular array is used to determine the best antenna layout to maximize the positioning accuracy.

[0091] Signal incident angle calculation: The signal incident angle is determined by vector synthesis using the phase difference and wavelength of the antenna array.

[0092] S204: construct a model based on the preliminary location estimation result and the RSSI values ​​collected by the composite tags to obtain an RSSI model.

[0093] In this embodiment, environmental feature modeling: an environmental model is established, including the reflection coefficients of walls, devices and other objects, which are used to correct RSSI measurements.

[0094] Signal attenuation compensation: Applies a signal attenuation formula that takes into account frequency, distance, and environmental factors to obtain more accurate distance estimates.

[0095] S205: Using the magnetic field sensing data as a position constraint condition, and optimizing the TDOA model, the AOA model, and the RSSI model.

[0096] In this embodiment, magnetic field sensing data is integrated: the data of magnetometers worn by construction workers are used in combination with the magnetic field distribution map in the tunnel to limit the possible location range.

[0097] Joint optimization: Use algorithms such as Kalman filtering or particle filtering to fuse the results of TDOA, AOA, and RSSI models to obtain the optimal position estimate.

[0098] Application of constraints: Introduce prior knowledge, such as the impossibility of construction workers passing through walls or being outside underground facilities, to filter out unreasonable solutions.

[0099] Specifically, step S205 specifically includes the following steps:

[0100] S205a: Analyze the magnetic field induction data based on the magnetic field distribution characteristics of the power tunnel where the construction workers are located to obtain a spatial distribution model of the magnetic field intensity in the tunnel.

[0101] In this embodiment, the magnetic field data of the power tunnel where the construction workers are located are collected, and these data are provided by the magnetic field sensor in the composite tag. These data are analyzed to identify the change pattern and trend of the magnetic field intensity.

[0102] Statistical analysis is used to construct a spatial distribution model of magnetic field intensity. The model should be able to reflect the variation of magnetic field intensity in different areas of the tunnel. Considering that there may be wires, cables and other metal structures in the power tunnel, which will affect the magnetic field, the model should include the influence of these environmental characteristics.

[0103] S205b: Limiting the location range of the construction personnel according to the spatial distribution model of the magnetic field strength to form a location constraint condition.

[0104] In this embodiment, the magnetic field strength thresholds of different areas are determined based on the magnetic field distribution model. These thresholds can be used to limit the possible location range of construction personnel. The magnetic field sensing data actually received by the construction personnel is compared with the magnetic field distribution model to determine which areas with matching magnetic field strengths the construction personnel may be located in. The above-mentioned limited location range is converted into mathematical constraints, which will be applied in the subsequent model optimization process to ensure that the predicted position conforms to the actual magnetic field distribution.

[0105] S205c: Apply the position constraint condition to the TDOA model, the AOA model and the RSSI model, and adjust the parameters of each model until the error between the predicted position of each model and the expected position under the magnetic field constraint condition is minimized.

[0106] In this embodiment, the parameters of the TDOA model, the AOA model, and the RSSI model are adjusted so that the position predicted by the model is as close as possible to the expected position under the magnetic field constraint condition.

[0107] An optimization algorithm is used to minimize the error between the predicted position and the expected position under magnetic field constraints. The magnetic field constraints are incorporated into the loss function of the model to ensure that the optimization process takes into account the errors of both the positioning model and the magnetic field distribution model, thereby obtaining a more accurate position estimate.

[0108] Specifically, step S3 specifically includes the following steps:

[0109] S301: By comparing the predicted value of each model with the actual measured value, the prediction deviation of each model is evaluated and quantified to obtain the error margin of each model, and the error margin is expressed as: ;

[0110] in, For the The measurement point The error margin of the model is It is the index of TDOA model, AOA model and RSSI model. For the The actual measured value of the measuring point, For Model In the The predicted value at each measurement point.

[0111] In this embodiment, the error magnitudes of all models at all measurement points are calculated to form an error magnitude list or matrix, in preparation for the next step of constructing an error equation set.

[0112] S302: According to the error amplitude of each model, an error equation set is constructed by an error quantification method. The error equation set is expressed as: ;

[0113] in, is the set of error equations, is the number of measurement points, and , For the TDOA model The error margin at each measurement point is For the AOA model The error margin at each measurement point is For RSSI model The error margin at each measurement point.

[0114] In this embodiment, the error equation set is sorted to ensure that all equations are arranged in the order of model and measurement points, which is convenient for subsequent objective function construction.

[0115] Specifically, step S4 specifically includes the following steps:

[0116] S401: transform each error equation in the error equation set to obtain an error term representing the prediction error of each model, wherein the mathematical expression of the error term is: .

[0117] In this embodiment, each error equation in the error equation set is converted into an error term. The error term is usually in the form of a square error. The conversion into an error term is to facilitate the construction of the objective function. By minimizing the sum of all error terms, the model parameters can be optimized to make the model prediction closer to the actual measurement value.

[0118] S402: weighting the error terms based on the credibility of the prediction errors of each model to obtain a set of weighted error terms. The mathematical expression of the weighted error terms is: .

[0119] In this embodiment, weighting is performed to take into account the relative importance of different models and measurement points when constructing the objective function, so as to ensure that errors from different sources can be more reasonably balanced when optimizing the objective function.

[0120] S403: constructing an objective function using the least squares criterion based on the weighted error term set, the mathematical expression of the objective function is: ;

[0121] in, is the objective function.

[0122] In this embodiment, the objective function is constructed to find a set of model parameters in the subsequent optimization process so that the sum of squares of all weighted error terms is minimized, thereby obtaining the optimal model parameter estimation.

[0123] Specifically, step S5 specifically includes the following steps:

[0124] S501: Initialize by setting initial parameters and hyperparameters of the gradient descent method to obtain a starting point for the optimization solution.

[0125] In this embodiment, initialization parameters: a set of initial parameters is selected as the starting point of optimization, and these parameters can be randomly generated or based on prior knowledge or preliminary estimation.

[0126] Set hyperparameters: Set the hyperparameters of the gradient descent method, including the learning rate, maximum number of iterations, and convergence threshold. The learning rate controls the step size of each iteration, the maximum number of iterations sets the maximum running time of the algorithm, and the convergence threshold is used to determine when to stop iterating.

[0127] S502: Applying a gradient descent method to the objective function and iteratively updating parameters to obtain a parameter sequence that gradually approaches an optimal solution.

[0128] In this embodiment, the gradient of the objective function is calculated at the current parameter. The gradient points to the direction in which the function value grows fastest. Therefore, the parameters need to be updated in the reverse direction to reduce the objective function value. The parameters are updated in the opposite direction of the gradient. The gradient calculation and parameter update steps are repeated until the maximum number of iterations is reached or the modulus of the gradient is less than the convergence threshold.

[0129] S503: The parameter updating process is terminated according to a preset convergence criterion to obtain an optimal estimate of the construction personnel's position. The mathematical expression is: .

[0130] In this embodiment, when the modulus of the gradient is less than the preset convergence threshold, or the number of iterations reaches the maximum value, the parameter update process is considered to have converged, and the final parameter is the optimal estimate, which makes the objective function reach the minimum value. Once the optimal parameters are found, these parameters can be used in the positioning model to predict the position of the construction workers and obtain the optimal estimate.

[0131] The working principle of the method for positioning construction personnel in a power tunnel provided by the present invention is as follows: by acquiring TDOA (time difference of arrival), AOA (angle of arrival), RSSI (received signal strength indication) and magnetic field sensing data in real time, and building multiple positioning models (TDOA model, AOA model, RSSI model) based on these data, and optimizing each model by using magnetic field sensing data as position constraint conditions. This method can make full use of the advantages of different positioning technologies, effectively make up for the shortcomings of a single technology, and improve the accuracy and reliability of positioning.

[0132] In the positioning process, the gradient descent method is also used to optimize the least squares objective function to obtain the optimal estimate of the construction personnel's position. As a commonly used optimization method, the gradient descent method has the characteristics of simple implementation and high computational efficiency. Through this method, complex objective functions can be quickly solved, further improving positioning accuracy and real-time performance.

[0133] In summary, the method for positioning construction personnel in power tunnels proposed in this application effectively improves the positioning accuracy and real-time performance of construction personnel by integrating multiple measurement data, constructing multiple positioning models, and using the gradient descent method for optimization and solution, thereby providing a strong guarantee for the safe construction of power tunnels.

[0134] Embodiment 2

[0135] The present invention also provides a construction personnel positioning system in a power tunnel, which is used to execute the construction personnel positioning method in a power tunnel, referring to Figure 2 As shown, the positioning system includes:

[0136] The data processing module 100 is used to acquire measurement data in real time and perform preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data.

[0137] The model building module 200 is used to build a TDOA model, an AOA model and an RSSI model based on the TDOA data, the AOA data and the RSSI value and in combination with the position of the UWB receiver, and optimize each model using the magnetic field sensing data as a position constraint condition.

[0138] The error quantification module 300 is used to quantify the difference between the predicted value of each optimized model and the actual measured value to obtain an error equation set.

[0139] The objective function construction module 400 is used to convert each error equation in the error equation set into an error term and integrate all error terms to construct an objective function, wherein the objective function adopts the least squares criterion.

[0140] The positioning optimization solution module 500 is used to optimize and solve the least squares objective function through the gradient descent method to obtain the optimal estimate of the construction personnel's position.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, the storage medium including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically-erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0143] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A method for positioning construction personnel in a power tunnel, characterized in that: The positioning method comprises the following steps: S1: acquiring measurement data in real time and performing preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data; S2: Based on the TDOA data, AOA data and RSSI value, and in combination with the position of the UWB receiver, a TDOA model, an AOA model and an RSSI model are respectively constructed, and the magnetic field sensing data is used as a position constraint to optimize each model; S3: quantify the difference between the predicted value of each optimized model and the actual measured value to obtain a set of error equations; S4: converting each error equation in the error equation set into an error term, and integrating all error terms to construct an objective function, wherein the objective function adopts a least squares criterion; S5: The least squares objective function is optimized and solved by the gradient descent method to obtain the optimal estimate of the construction personnel's position.

2. A method for locating construction personnel in a power tunnel according to claim 1, characterized in that: Step S1 specifically includes the following steps: S101: capturing TDOA data and AOA data in real time through a UWB receiver, and capturing RSSI value and magnetic field sensing data in real time through a composite tag, wherein the composite tag includes a magnetic field sensor and an RSSI module; S102: screening and synchronizing the measurement data to obtain screened measurement data; S103: performing standardization processing on the screened measurement data to obtain standardized measurement data.

3. A method for locating construction personnel in a power tunnel according to claim 2, characterized in that: Step S2 specifically includes the following steps: S201: Preliminarily estimate the potential position of the construction personnel through TDOA data, AOA data and RSSI value to obtain a preliminary position estimation result; S202: constructing a model based on the preliminary position estimation result and the known position of the UWB receiver to obtain a TDOA model; S203: constructing a model according to the preliminary position estimation result and the azimuth information of the UWB receiver to obtain an AOA model; S204: constructing a model based on the preliminary location estimation result and the RSSI value collected by the composite tag to obtain an RSSI model; S205: Using the magnetic field sensing data as a position constraint condition, and optimizing the TDOA model, the AOA model, and the RSSI model.

4. A method for locating construction personnel in a power tunnel according to claim 3, characterized in that: Step S205 specifically includes the following steps: S205a: Analyze the magnetic field induction data based on the magnetic field distribution characteristics of the power tunnel where the construction workers are located, and obtain a spatial distribution model of the magnetic field intensity in the tunnel; S205b: limiting the location range of the construction personnel according to the spatial distribution model of the magnetic field strength to form a location constraint condition; S205c: Apply the position constraint condition to the TDOA model, the AOA model and the RSSI model, and adjust the parameters of each model until the error between the predicted position of each model and the expected position under the magnetic field constraint condition is minimized.

5. A method for locating construction personnel in a power tunnel according to claim 4, characterized in that: Step S3 specifically includes the following steps: S301: By comparing the predicted value of each model with the actual measured value, the prediction deviation of each model is evaluated and quantified to obtain the error margin of each model, and the error margin is expressed as: ; in, For the The measurement point The error margin of the model is It is the index of TDOA model, AOA model and RSSI model. For the The actual measured value of the measuring point, For Model In the The predicted value at each measurement point; S302: According to the error amplitude of each model, an error equation set is constructed by an error quantification method. The error equation set is expressed as: ; in, is the set of error equations, is the number of measurement points, and , For the TDOA model The error margin at each measurement point is For the AOA model The error margin at each measurement point is For RSSI model The error margin at each measurement point.

6. A method for locating construction personnel in a power tunnel according to claim 5, characterized in that: Step S4 specifically includes the following steps: S401: transform each error equation in the error equation set to obtain an error term representing the prediction error of each model, wherein the mathematical expression of the error term is: ; S402: weighting the error terms based on the credibility of the prediction errors of each model to obtain a set of weighted error terms. The mathematical expression of the weighted error terms is: ; S403: constructing an objective function using the least squares criterion based on the weighted error term set, the mathematical expression of the objective function is: ; in, is the objective function.

7. A method for locating construction personnel in a power tunnel according to claim 6, characterized in that: Step S5 specifically includes the following steps: S501: Initialize by setting initial parameters and hyperparameters of the gradient descent method to obtain a starting point for optimization solution; S502: applying a gradient descent method to the objective function and iteratively updating parameters to obtain a parameter sequence that gradually approaches an optimal solution; S503: The parameter updating process is terminated according to a preset convergence criterion to obtain an optimal estimate of the construction personnel's position. The mathematical expression is: 。 8. A construction personnel positioning system in a power tunnel, used to execute a construction personnel positioning method in a power tunnel as claimed in any one of claims 1 to 7, characterized in that: The positioning system comprises: A data processing module, used for acquiring measurement data in real time and performing preprocessing, wherein the measurement data includes TDOA data, AOA data, RSSI value and magnetic field sensing data; A model building module, for building a TDOA model, an AOA model and an RSSI model based on the TDOA data, the AOA data and the RSSI value and in combination with the position of the UWB receiver, and optimizing each model by taking the magnetic field sensing data as a position constraint; The error quantification module is used to quantify the difference between the predicted value and the actual measured value of each optimized model to obtain a set of error equations; An objective function construction module, used for converting each error equation in the error equation set into an error term, and integrating all error terms to construct an objective function, wherein the objective function adopts a least squares criterion; The positioning optimization solution module is used to optimize and solve the least squares objective function through the gradient descent method to obtain the optimal estimate of the construction personnel's position.

Images