Anti-interference method of UWB and inertial navigation fusion positioning for complex environment of coal mine

Abstract

The present application relates to the technical field of coal mine underground personnel and equipment safety protection, especially to a UWB and inertial navigation fusion positioning anti-interference method for complex coal mine environment. The UWB and inertial navigation fusion positioning anti-interference method for complex coal mine environment comprises the following steps: S1: collecting UWB signal data of personnel, and obtaining UWB signal data, dust detection data, inertial navigation data and rotary encoder data of a heading machine in the process of the heading machine working on a coal mine heading face, and simultaneously obtaining UWB ranging data between the personnel and the heading machine; S2: constructing an inertial navigation credible data sequence of the heading machine based on the inertial navigation data; S3: processing the UWB signal data of the heading machine, and constructing a UWB credible data sequence of the heading machine in combination with the dust detection data. The present application can adaptively reduce the fusion weight of unreliable data, and improve the stability and accuracy of real-time position solution of the heading machine.

Description

Technical Field

[0001] This invention relates to the field of safety protection technology for personnel and equipment in underground coal mines, and in particular to a positioning anti-interference method that integrates UWB and inertial navigation for complex coal mine environments. Background Technology

[0002] During underground coal mine tunneling operations, the narrow tunnel spaces, dense electromechanical equipment, and complex electromagnetic environment, coupled with harsh working conditions such as high concentrations of dust, strong mechanical vibrations, and frequent slippage of the tunneling machine tracks, severely interfere with the real-time and accurate perception of the relative positions of personnel and the tunneling machine. Current intelligent coal mine construction places high-precision and high-reliability positioning requirements on personnel approach protection and equipment collaborative control. Especially during tunneling machine cutting and movement, if the dynamic relative positional relationship between personnel and the tunneling machine cannot be accurately obtained, collision accidents due to blind spots or misjudged distances can easily occur. In existing technologies, positioning methods based on ultra-wideband (UWB) signals are widely used for personnel positioning in underground mines due to their low power consumption and high ranging accuracy. However, UWB signals are easily affected by coal dust scattering and absorption, multipath effects, and reflections from metal support structures in coal mine tunneling faces, leading to signal power attenuation, initial diameter detection deviation, and ranging jumps. While relying solely on inertial navigation for tunneling machine attitude estimation is unaffected by dust and electromagnetic interference, the integral error accumulates over time, and the heading drift is significant under severe vibration and slippage conditions of the tunneling machine. Therefore, there is an urgent need for a positioning anti-interference method that integrates UWB and inertial navigation for the complex environment of coal mines, in order to improve the accuracy of real-time position calculation for personnel and tunneling machines and the dynamic safety early warning capability under complex working conditions. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies in sensing the location of surrounding personnel during tunneling machine operation, this invention provides a positioning anti-interference method that integrates UWB and inertial navigation for complex coal mine environments.

[0004] The technical implementation scheme of the present invention is as follows: a positioning anti-interference method for the fusion of UWB and inertial navigation in complex coal mine environments, comprising the following steps:

[0005] S1: Collect UWB signal data of personnel, and acquire UWB signal data, dust detection data, inertial navigation data and rotary encoder data of the tunneling machine during its operation in the coal mine tunneling face, and acquire UWB distance measurement data between personnel and tunneling machine.

[0006] S2: Construct a reliable inertial navigation data sequence for the tunnel boring machine based on the inertial navigation data;

[0007] S3: Process the UWB signal data of the tunneling machine and construct a reliable UWB data sequence of the tunneling machine by combining it with the dust detection data;

[0008] S4: Based on the inertial navigation reliable data sequence and the tunneling machine's UWB reliable data sequence, solve for the tunneling machine's real-time position and the tunneling machine's position covariance matrix;

[0009] S5: Calculate the slippage state factor based on the rotary encoder data and the real-time position of the tunneling machine, use the tunneling machine as a mobile base station, and calculate the reliability of the mobile base station based on the slippage state factor, the real-time position of the tunneling machine, and the position covariance matrix of the tunneling machine.

[0010] S6: Based on the reliability of the mobile base station, the UWB signal data and UWB ranging data of the personnel, calculate the real-time location of the personnel;

[0011] S7: Calculate dynamic safety risks based on the real-time location of the personnel and the real-time location of the tunneling machine.

[0012] Preferably, the process of collecting UWB signal data from personnel and acquiring UWB signal data, dust detection data, inertial navigation data, and rotary encoder data of the tunneling machine during its operation at the coal mine tunneling face, while simultaneously acquiring UWB ranging data between personnel and the tunneling machine, includes: acquiring the location of UWB base stations; collecting UWB signal data from personnel, UWB signal data from the tunneling machine, and UWB signal data of the tunneling machine's movement at each UWB base station; acquiring dust detection data, inertial navigation data, and rotary encoder data of the tunneling machine during its operation at the coal mine tunneling face through sensors deployed in the environment; and acquiring UWB ranging data between personnel and the tunneling machine based on the tunneling machine's UWB base stations.

[0013] Preferably, the step of constructing a reliable inertial navigation data sequence for the tunnel boring machine based on the inertial navigation data includes: the inertial navigation data including angular velocity, correction interval duration, and vibration intensity; obtaining the rotational cumulative threshold and inertial navigation cumulative error reference time calibrated by the inertial navigation experiment; calculating the angular velocity modulus integral based on the angular velocity; obtaining the high-frequency vibration intensity and vibration intensity threshold based on the vibration intensity through Gaussian filtering; calculating the reliability of the inertial navigation data within a preset time window; and associating the reliability of the inertial navigation data with the inertial navigation data to construct a reliable inertial navigation data sequence.

[0014] Preferably, the calculation of the reliability of inertial navigation data within a preset time window includes: Inertial navigation reliability calculation formula:

[0015] ;

[0016] In the formula, To improve the reliability of inertial navigation, To correct the interval duration, The reference time for the cumulative error of the inertial navigation system. For the integral of the magnitude of the angular velocity, For rotation accumulation threshold, For high-frequency vibration intensity, This is the vibration intensity threshold.

[0017] Preferably, the step of processing the UWB signal data of the tunneling machine and constructing a reliable UWB data sequence for the tunneling machine in conjunction with the dust detection data includes: the UWB signal data includes the initial diameter power and the total power of the base stations; the dust concentration of the UWB base stations is obtained based on the dust detection data; an initial topology map of the UWB base stations is constructed according to the location of the UWB base stations; UWB base stations with an initial diameter power greater than a preset power threshold in the initial topology map are selected as UWB base stations participating in positioning; a positioning topology map is constructed; the number of connected components in the positioning topology map is analyzed and obtained; the linear factor and quadratic factor of the dust error are calibrated through a dust error experiment; the horizontal geometric precision factor, the optimal horizontal geometric precision factor, and the horizontal geometric precision attenuation reference value are obtained through geometric matrix operations based on the positioning topology map; the reliability of the UWB signal data within a preset time window is calculated; and the reliability of the UWB signal data is correlated with the inertial navigation data to construct a reliable UWB data sequence.

[0018] Preferably, the calculation of the reliability of UWB signal data within a preset time window includes:

[0019] UWB credibility calculation formula:

[0020] ;

[0021] In the formula, For UWB credibility, The number of UWB base stations participating in the positioning. For the first The first-path power of a UWB base station For the first The total power of each UWB base station It is a very small positive number. For the first Dust concentration of a UWB base station The linearity factor for dust error. The second factor is the dust error factor. To the maximum permissible dust error, Geometric topology factor;

[0022] ;

[0023] In the formula, Geometric topological factor, The horizontal geometric precision factor, The optimal horizontal geometric accuracy factor. This is a reference value for the attenuation of horizontal geometric accuracy. To determine the number of connected components in the topological graph.

[0024] Preferably, the step of solving for the real-time position and position covariance matrix of the tunneling machine based on the reliable inertial navigation data sequence and the reliable UWB data sequence of the tunneling machine includes: solving for the real-time position and position covariance matrix of the tunneling machine based on the DS evidence theory, according to the reliable inertial navigation data sequence and the reliable UWB data sequence of the tunneling machine.

[0025] Preferably, the step of calculating the slippage state factor based on the rotary encoder data and the real-time position of the tunneling machine, using the tunneling machine as a mobile base station, and calculating the mobile base station reliability based on the slippage state factor, the real-time position of the tunneling machine, and the position covariance matrix of the tunneling machine includes: obtaining the body length and the radius of the traveling wheels of the tunneling machine; calculating the theoretical displacement increment of the tracks based on the rotary encoder data and the radius of the traveling wheels; calculating the actual displacement increment based on the actual position of the tunneling machine; calculating the slippage state factor based on the theoretical displacement increment of the tracks and the actual displacement increment; and calculating the mobile base station reliability based on the slippage state factor, the position covariance matrix of the tunneling machine, and the inertial navigation reliability.

[0026] Mobile base station credibility formula:

[0027] ;

[0028] In the formula, To ensure the credibility of mobile base stations, For UWB credibility, To improve the reliability of inertial navigation, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, This is a reference value for the location covariance. This is the slippage state factor.

[0029] Preferably, the step of calculating the real-time location of a person based on the reliability of the mobile base station, the UWB signal data of the person, and the UWB ranging data includes: obtaining the ranging value between the person and the UWB base station, the ranging noise variance between the person and the UWB base station, and the basic ranging noise variance based on the UWB signal data of the person; obtaining the ranging value between the person and the mobile base station based on the UWB ranging data; and solving the real-time location of the person with the objective function value as the goal.

[0030] ;

[0031] In the formula, The objective function value, The number of UWB base stations participating in the positioning. For the first The variance of ranging noise of a UWB base station For real-time location of personnel, This indicates the real-time position of the tunneling machine. For the first Location of each UWB base station For personnel and the Ranging values ​​of a UWB base station, To ensure the credibility of mobile base stations, Based on the variance of the basic ranging noise, For Euclidean distance This refers to the distance measured between people and mobile base stations.

[0032] Preferably, the step of calculating dynamic safety risks based on the real-time positions of personnel and tunneling machines includes: calculating a comprehensive positioning error based on the slippage state factor, the tunneling machine position covariance matrix, the square of the personnel positioning error, and the tunneling machine body length; calculating the relative distance between personnel and tunneling machines based on the real-time positions of personnel and tunneling machines; obtaining personnel trajectories and tunneling machine trajectories based on the real-time positions of personnel and tunneling machines; calculating a personnel intention factor based on the angle between the personnel trajectory and the tunneling machine trajectory; and calculating dynamic safety risks based on the comprehensive positioning error and the relative distance between personnel and tunneling machines.

[0033] Comprehensive positioning error formula:

[0034] ;

[0035] In the formula, To account for the overall positioning error, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, Let this be the square of the personnel positioning error. For slippage state factor, This refers to the length of the tunneling machine's fuselage.

[0036] Dynamic security risk formula:

[0037]

[0038] In the formula, For dynamic security risks, As the reference distance, The relative distance between personnel and the tunneling machine. To account for the overall positioning error, For the maximum positioning error, For personnel intention factors.

[0039] The beneficial effects are:

[0040] This invention introduces an inertial navigation reliability calculation formula that includes parameters such as the integral of angular velocity modulus and the correction interval duration. It uses the angular velocity accumulation threshold and the high-frequency vibration intensity extracted by Gaussian filtering for joint constraints, so that when the tunneling machine is under severe cutting vibration or has not undergone external correction for a long time, the inertial navigation reliability can be adaptively attenuated. In the subsequent fusion process, it weakens the interference of angular velocity noise caused by vibration and heading divergence caused by time accumulation on position estimation.

[0041] Based on the linear and quadratic coefficients calibrated by dust error experiments, this invention uses dust concentration as a correction factor for UWB reliability calculation. Combined with the horizontal geometric accuracy factor and the number of connected components in the positioning topology, it achieves a quantitative assessment of the risks of UWB non-line-of-sight transmission and node isolation. When dust concentration increases sharply or the geometric configuration deteriorates due to base station layout obstruction, UWB reliability can reflect the availability of the current signal in real time, reducing the pollution of the positioning accuracy of the tunneling machine caused by the positive deviation error of ranging due to signal penetration of dust clusters.

[0042] This invention further reuses the tunneling machine as a mobile base station to assist personnel positioning. Addressing the issue that tunneling machine tracks are prone to slippage on wet or broken surfaces, a slippage state factor is constructed by comparing the difference between the theoretical and actual displacement increments of the rotary encoder. Combined with the trace of the tunneling machine's position covariance matrix and the aforementioned inertial navigation and UWB reliability, the reliability of the mobile base station is defined. When slippage causes an increase in the tunneling machine's position estimation covariance or distortion of the displacement increment, the personnel positioning algorithm can automatically cut off or reduce its dependence on the slipping tunneling machine as a mobile base station. This prevents the instantaneous positioning deviation caused by slippage from being mapped to errors in personnel position calculation, ensuring the smoothness and continuity of personnel trajectories under dynamic disturbances of the tunneling machine. Attached Figure Description

[0043] Figure 1 This is a flowchart of the positioning anti-interference method of the present invention, which integrates UWB and inertial navigation for complex coal mine environments.

[0044] Figure 2 This is a flowchart illustrating the risk calculation process of the UWB and inertial navigation fusion for complex coal mine environments, as described in this invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] A positioning anti-interference method combining UWB and inertial navigation for complex coal mine environments, such as... Figure 1 and Figure 2 As shown, it includes the following steps:

[0047] S1: Collect UWB signal data of personnel, and acquire UWB signal data, dust detection data, inertial navigation data and rotary encoder data of the tunneling machine during its operation in the coal mine tunneling face, and acquire UWB distance measurement data between personnel and tunneling machine.

[0048] The system acquires the location of UWB base stations, collects UWB signal data of personnel, tunneling machines, and mobile UWB signal data of each UWB base station, acquires dust detection data, inertial navigation data, and rotary encoder data of the tunneling machine during its operation at the coal mine tunneling face through sensors deployed in the environment, and acquires UWB distance measurement data between personnel and the tunneling machine based on the tunneling machine's UWB base station.

[0049] It needs further explanation that the coordinates of multiple UWB base stations pre-deployed within the coal mine tunneling face are obtained to form a positioning reference network within the tunnel. Each pre-deployed UWB base station receives UWB pulse signals emitted by the positioning cards worn by each personnel, and analyzes this data to obtain raw measurement information including signal arrival timestamps, signal strength indicators, initial path signal power, and total received power. It also receives UWB pulse signals emitted by UWB positioning tags installed on the tunneling machine, obtaining the signal arrival time, initial path power, and total power information of the tunneling machine relative to each fixed UWB base station. Finally, dust concentration sensors in the environment continuously collect data according to a preset sampling period. The dust concentration in the air at the working face is recorded along with the corresponding sampling time. The three-axis angular velocity, three-axis acceleration, and the correction interval between two external corrections given by the internal algorithm are read from the inertial navigation measurement unit installed on the tunneling machine body. The vibration intensity information calculated by the high-frequency component of the accelerometer is extracted. The pulse count or angle increment signal output by the rotary encoder installed on the tunneling machine track mechanism is read to estimate the theoretical travel distance of the tunneling machine track. The signal is transmitted and received directly between the positioning card worn by the personnel and the UWB positioning tag installed on the tunneling machine. The real-time distance between the personnel and the tunneling machine is calculated using the two-way time-of-flight principle.

[0050] S2: Construct a reliable inertial navigation data sequence for the tunnel boring machine based on the inertial navigation data;

[0051] Inertial navigation data includes angular velocity, correction interval duration, and vibration intensity. The rotation accumulation threshold and inertial navigation cumulative error reference time calibrated by the inertial navigation experiment are obtained. The angular velocity modulus integral is calculated based on the angular velocity. Based on the vibration intensity, the high-frequency vibration intensity and vibration intensity threshold are obtained through Gaussian filtering. The reliability of the inertial navigation data within a preset time window is calculated. The reliability of the inertial navigation data is associated with the inertial navigation data to construct a reliable inertial navigation data sequence.

[0052] ;

[0053] In the formula, To improve the reliability of inertial navigation, To correct the interval duration, The reference time for the cumulative error of the inertial navigation system. For the integral of the magnitude of the angular velocity, For rotation accumulation threshold, For high-frequency vibration intensity, This is the vibration intensity threshold.

[0054] It should be further explained that the inertial navigation data specifically includes angular velocity data output by the inertial navigation measurement unit, correction interval data between two external correction events, and vibration intensity data extracted from the high-frequency components of the accelerometer. The angular velocity data is triaxial angular velocity, that is, the angular rate of rotation of the tunneling machine body around three mutually orthogonal coordinate axes, in radians per second. The correction interval refers to the time length between the moment when the inertial navigation measurement unit last received external reference information for error correction and the current moment, in seconds. The vibration intensity data reflects the mechanical vibration energy level of the tunneling machine body during coal and rock cutting or movement, and is usually characterized by the root mean square value of the acceleration amplitude. Two parameter values, pre-calibrated through inertial navigation experiments, are obtained: the rotation accumulation threshold and the inertial navigation accumulation error reference time. The rotation accumulation threshold is obtained through turntable experiments. The inertial measurement unit is fixed on a precision turntable, which is then rotated at different angular velocities. The angular velocity data output by the inertial measurement unit is recorded. For each set of experiments, the integral value of the angular velocity modulus within a fixed time window is calculated. Simultaneously, a high-precision optical positioning system is used to measure the actual rotation angle. By comparing the rotation angle calculated by the inertial navigation system with the actual rotation angle, the integral value of the angular velocity modulus at which the error begins to increase is identified and used as the rotation accumulation threshold. The inertial navigation accumulation error reference time represents the time taken when the inertial navigation system is in operation without external interference. Under no-correction conditions, the characteristic time of error growth over time, in seconds, is used to determine the reference time for inertial navigation cumulative error. This is achieved through a combination of static and dynamic experiments. The inertial measurement unit (INS) is placed on a stable platform and held still for a period of time, such as ten minutes, and its attitude angular drift is recorded and the drift rate is calculated. Then, it is run at a constant speed on a motion track. The position calculated by the INS is compared with the actual position on the track, and the curve of position error growth over time is analyzed. This curve is fitted with an exponential function, and the time corresponding to the error growing to e times the initial error (approximately 2.718 times the natural constant) is taken as the reference time for INS cumulative error. The angular velocity modulus integral is calculated based on the angular velocity data, which includes three axes. The components are denoted as angular velocities around the X-axis, Y-axis, and Z-axis, respectively. At each sampling moment, the sum of squares of the three axial angular velocity components at that moment is calculated, and the square root of the sum is obtained to obtain the angular velocity modulus at the current moment. The angular velocity modulus reflects the instantaneous intensity of the tunneling machine's rotation at the current moment. Within a preset time window, such as 10 seconds, the angular velocity modulus is integrated over time. The product of the angular velocity modulus at each moment in the window and the sampling time interval is accumulated and summed to obtain the angular velocity modulus integral within the time window. The physical meaning of the angular velocity modulus integral is the total rotation angle experienced by the tunneling machine within the time window, in radians.High-frequency vibration intensity and vibration intensity threshold are obtained based on vibration intensity data obtained through Gaussian filtering. The vibration intensity data is extracted from the accelerometer output signal of the inertial navigation measurement unit. It includes both low-frequency components caused by the normal movement and attitude changes of the tunneling machine, and high-frequency components caused by cutting impacts and mechanical collisions. The high-frequency vibration intensity is obtained by Gaussian filtering the original vibration intensity time series. The high-frequency vibration intensity reflects the degree of instantaneous impact or severe vibration experienced by the tunneling machine in a short period. The vibration intensity threshold is determined as follows: Under normal operating conditions where the tunneling machine is moving normally and without slippage, accelerometers are installed at multiple locations on the machine body to continuously collect vibration data. The collected vibration signals are Gaussian filtered to extract the high-frequency components. The average value and standard deviation of the high-frequency vibration intensity are calculated. The average value plus three times the standard deviation is used as the upper limit of the vibration intensity under normal operating conditions to obtain the vibration intensity threshold. If the vibration characteristics of the tunneling machine differ when moving on a soft coal floor or a rock floor, the maximum value under different operating conditions is taken as the final threshold. After obtaining the above intermediate variables, the reliability of the inertial navigation data within a preset time window is calculated. The reliability calculation formula comprehensively evaluates the quality of inertial navigation data from three dimensions: time cumulative error, rotational cumulative error, and vibration interference. The exponential part considers the smoothing degradation effect caused by time growth and rotational accumulation, while the max function part sets a hard cutoff condition triggered by vibration exceeding limits. The calculated inertial navigation reliability can be dynamically adjusted in real time according to changes in the tunnel boring machine's operating conditions. A reliable inertial navigation data sequence is constructed by associating the reliability of the inertial navigation data with the inertial navigation data itself. Specifically, within each preset time window, the calculated inertial navigation reliability value is bound to the corresponding inertial navigation dataset within that window. The inertial navigation dataset includes the timestamp of the window's start time, statistics of angular velocity data within the window, correction interval duration, vibration intensity statistics, and incremental information on the tunnel boring machine's position and attitude obtained through dead reckoning within the window. The inertial navigation reliability is marked as an additional attribute field in this data record. The inertial navigation datasets corresponding to multiple consecutive time windows and their respective inertial navigation reliability values ​​are arranged in chronological order to form a reliable inertial navigation data sequence.

[0055] S3: Process the UWB signal data of the tunneling machine and construct a reliable UWB data sequence of the tunneling machine by combining it with the dust detection data;

[0056] UWB signal data includes initial path power and total base station power. Dust concentration of UWB base stations is obtained based on dust detection data. An initial topology map of UWB base stations is constructed according to their locations. UWB base stations with initial path power greater than a preset power threshold in the initial topology map are selected as participating base stations for positioning. A positioning topology map is constructed, and the number of connected components in the positioning topology map is analyzed and obtained. The linear and quadratic factors of dust error are calibrated through dust error experiments. Based on the positioning topology map, the horizontal geometric precision factor, the optimal horizontal geometric precision factor, and the horizontal geometric precision attenuation reference value are obtained through geometric matrix operations. The reliability of UWB signal data within a preset time window is calculated. The reliability of UWB signal data is correlated with inertial navigation data to construct a reliable UWB data sequence.

[0057] UWB credibility calculation formula:

[0058] ;

[0059] In the formula, For UWB credibility, The number of UWB base stations participating in the positioning. For the first The first-path power of a UWB base station For the first The total power of each UWB base station It is a very small positive number. For the first Dust concentration of a UWB base station The linearity factor for dust error. The second factor is the dust error factor. To the maximum permissible dust error, Geometric topology factor;

[0060] ;

[0061] In the formula, Geometric topological factor, The horizontal geometric precision factor, The optimal horizontal geometric accuracy factor. This is a reference value for the attenuation of horizontal geometric accuracy. To determine the number of connected components in the topological graph.

[0062] It should be further explained that the UWB signal data of the tunneling machine includes two key parameters for the positioning tag signal received by each fixed UWB base station: first-path power and total base station power. First-path power refers to the power value of the direct path signal component detected by the UWB receiver, that is, the energy carried by the first path of arrival of the signal from the transmitter to the receiver through line-of-sight propagation. Total base station power refers to the sum of the power of all multipath signal components received by the UWB receiver within a set integration time window. This includes both the direct path component and the power superposition of the reflection path components generated by reflections from the tunnel wall and equipment surface. The ratio of first-path power to total base station power reflects the proportion of energy occupied by the direct path in the current signal propagation channel. The higher the ratio, the more ideal the line-of-sight propagation conditions and the less severe the multipath interference. Using the dust concentration data collected by the dust concentration sensors deployed in the coal mine tunneling face environment in step S1, combined with the spatial location information of each fixed UWB base station, a dust concentration value at the current moment is assigned to each fixed UWB base station participating in the positioning through spatial interpolation or proximity matching. Since the number of dust concentration sensors is usually less than the number of UWB base stations, and the distribution of dust in the tunnel space is somewhat uneven, the measured value of the dust concentration sensor closest to a certain UWB base station is taken as the dust concentration value corresponding to that UWB base station. All the fixed UWB base stations pre-deployed in the tunnel are taken as nodes in the figure. Edges between nodes are constructed based on the spatial distance and signal connectivity between each UWB base station. The specific construction rule is as follows: if two UWB base stations can receive each other's UWB signals during actual operation, then these two nodes are connected by an edge in the graph. After traversing all UWB base station node pairs according to this rule, an undirected graph structure is formed. This undirected graph is the initial topology graph of the UWB base stations. The initial topology graph reflects the spatial interconnection of the entire UWB base station network. UWB base stations with a first-path power greater than a preset power threshold in the initial topology graph are selected as UWB base stations participating in positioning. Based on the initial topology graph, further selection is performed according to the first-path power of the UWB signal received by each UWB base station from the tunnel boring machine. A power threshold is preset, determined by experimentally measuring the minimum initial path power that ensures UWB ranging error remains within acceptable limits under different dust concentrations and distances. For each UWB base station node in the initial topology, the initial path power of the received positioning tag signal from the tunnel boring machine is compared with the preset power threshold. If the initial path power of a UWB base station is less than or equal to the preset power threshold, the signal received by that UWB base station is considered too weak, indicating severe obstruction or attenuation, and the reliability of the ranging result is insufficient. UWB base station nodes with initial path power greater than the preset power threshold are retained, and new edge connections are established between these retained nodes. The resulting new undirected graph structure is the positioning topology.The obtained positioning topology is analyzed using graph theory. A connected component is a maximal connected subgraph in the positioning topology, where there is at least one path connecting any two nodes, and no edges connect different connected components. The number of connected components reflects the degree of topological fragmentation of the UWB base station network used for tunneling machine positioning. If the number of connected components is 1, it means that all UWB base stations involved in positioning are in the same connected network, and the positioning algorithm can use the mutual constraints between these base stations for robust solution. If the number of connected components is greater than 1, it means that the UWB base stations involved in positioning are divided into multiple isolated subgroups, which is caused by severe signal obstruction in some areas or signal interruption of some UWB base stations due to dust interference. In this case, the reliability of the positioning results will be affected. Before the formal deployment of the system or during periodic calibration, the quantitative impact of dust concentration on UWB ranging error is determined through specialized experimental methods. In the experiment, a pair of UWB transceivers were deployed in a controlled dust environment, keeping the actual distance between them constant. Different concentrations of coal dust or rock dust were artificially released, and the distance measurements output by the UWB ranging system under each dust concentration condition were recorded. The ranging error between these measurements and the actual distance was calculated. Using dust concentration as the independent variable and ranging error as the dependent variable, a curve showing the relationship between ranging error and dust concentration was obtained through least-squares polynomial fitting. The coefficient of the first-order term of the fitted polynomial is the linear factor 'a' of the dust error, and the coefficient of the second-order term is the quadratic factor 'b' of the dust error. The units of 'a' and 'b' are meters per milligram per cubic meter (m / m³) and meters per milligram per cubic meter (m / m³) squared, respectively. These two factors reflect the attenuation and distortion characteristics of the UWB signal in a specific coal mine dust environment; a larger factor indicates a more significant impact of the environmental dust on the accuracy of UWB ranging.

[0063] Based on the positioning topology map, the horizontal geometric precision factor, the optimal horizontal geometric precision factor, and the horizontal geometric precision attenuation reference value are obtained through geometric matrix operations. The specific process of geometric matrix operations is as follows: Based on the coordinates of all UWB base stations participating in positioning and the approximate position of the tunneling machine in the positioning topology map, an observation matrix is ​​constructed. Then, the pseudo-inverse of the observation matrix is ​​calculated, and the horizontal position covariance is extracted from it and converted into the horizontal geometric precision factor. The optimal horizontal geometric precision factor is the minimum value that can be theoretically achieved under the current geometric distribution of base stations. The horizontal geometric precision attenuation reference value is an empirical constant, indicating that when the horizontal geometric precision factor increases to this value, the decrease in positioning accuracy begins to become significant. The horizontal geometric precision attenuation reference value is obtained through geometric simulation. Under the typical base station layout of a coal mine tunneling face, the distribution of the horizontal geometric precision factor at different positions is calculated, and the average and standard deviation of the horizontal geometric precision factor at all positions are statistically analyzed. The average plus one standard deviation is used as the reference value.

[0064] The reliability of UWB signal data within a preset time window is calculated. In the UWB reliability calculation formula, the first product factor reflects the average reliability of each UWB base station involved in the positioning in terms of signal quality. The second product factor is used to evaluate the impact of dust concentration on the ranging accuracy of the i-th UWB base station. The third product factor is used to evaluate the degree of degradation of the current UWB base station geometric distribution configuration relative to the optimal configuration. Even if the signal quality of each UWB base station is good, if these UWB base stations are all clustered within a narrow azimuth angle range relative to the tunneling machine, the horizontal positioning accuracy will be affected by the geometric configuration. If the model is poor and deteriorates, the reliability should be lowered. The fourth product factor is used to evaluate the impact of the connectivity of the positioning topology on the positioning robustness. In a split network topology, there is a lack of mutual constraints between UWB base stations in different connected components. The positioning algorithm is prone to getting trapped in local optima or producing multiple ambiguities, so the reliability should be reduced. The reliability of UWB signal data is associated with inertial navigation data to construct a UWB reliable data sequence. In a preset time window, the calculated UWB reliability is aligned and bound with the UWB signal data of the tunneling machine and the corresponding inertial navigation data in the time window on the timestamp.

[0065] S4: Based on the inertial navigation reliable data sequence and the tunneling machine's UWB reliable data sequence, solve for the tunneling machine's real-time position and the tunneling machine's position covariance matrix;

[0066] Based on the DS evidence theory, the real-time position of the tunneling machine and the position covariance matrix of the tunneling machine are obtained by solving the problem based on the reliable inertial navigation data sequence and the reliable UWB data sequence of the tunneling machine.

[0067] It should be further explained that the inertial navigation trusted data sequence and the tunnel boring machine's UWB trusted data sequence output by step S3 are obtained. The inertial navigation trusted data sequence contains the tunnel boring machine's position data calculated by the inertial navigation system at each sampling time and the corresponding trusted value. The tunnel boring machine's UWB trusted data sequence contains the tunnel boring machine's position data calculated by the UWB positioning system and the corresponding trusted value. The two trusted data sequences are synchronized in time.

[0068] Based on the DS evidence theory, the true position of the tunnel boring machine (TBM) is considered as the target proposition to be identified. An identification framework is constructed, which includes all the position points of the TBM in the tunnel coordinate system. For each sampling time, the TBM position data and its credibility in the inertial navigation credible data sequence at that time are assigned as one evidence body, and the TBM position data and its credibility in the UWB credible data sequence at that time are assigned as another evidence body. Each evidence body is assigned a basic probability value to different position assumptions in the identification framework according to its credibility. The basic probability value provided by the inertial navigation credible data sequence reflects the degree of inference and trust of the inertial navigation system to the current position of the TBM, and the basic probability value provided by the UWB credible data sequence reflects the degree of inference and trust of the UWB positioning system to the current position of the TBM.

[0069] The Dempster combination rule in DS evidence theory is applied to fuse the basic probability assignments of the two pieces of evidence. During the fusion calculation, for each position hypothesis in the identification framework, the basic probability assignment of the inertial navigation reliable data sequence to that hypothesis is multiplied by the basic probability assignment of the tunnel boring machine's UWB reliable data sequence to that hypothesis, and then divided by a normalization constant. This normalization constant is obtained by subtracting the degree of conflict between the two pieces of evidence from one. The degree of conflict is calculated as the sum of the products of the basic probability assignments of all conflicting position hypothesis pairs. Through this fusion calculation, a joint basic probability assignment is obtained. Based on the fused joint basic probability assignment, the real-time position of the tunnel boring machine is solved, and the position hypothesis with the highest basic probability assignment among the joint basic probability assignments is selected. Let the real-time position of the tunneling machine be denoted as . Solve for the tunneling machine position covariance matrix. When solving for the tunneling machine position covariance matrix, the joint basic probability is assigned as a weight. Calculate the second-order central moments of all position hypotheses in the identification framework relative to the selected real-time position of the tunneling machine. For each position hypothesis, calculate the deviation vector between the position hypothesis and the real-time position of the tunneling machine. Multiply the product of this deviation vector and its transpose by the joint basic probability assigned to the position hypothesis. Summate all the product results to obtain the tunneling machine position covariance matrix. This tunneling machine position covariance matrix quantifies the uncertainty of the fused real-time position estimate of the tunneling machine. Its diagonal elements reflect the position variance in each direction, and the off-diagonal elements reflect the position covariance between different directions.

[0070] S5: Calculate the slippage state factor based on the rotary encoder data and the real-time position of the tunneling machine, use the tunneling machine as a mobile base station, and calculate the reliability of the mobile base station based on the slippage state factor, the real-time position of the tunneling machine, and the position covariance matrix of the tunneling machine.

[0071] The machine body length and travel wheel radius of the tunneling machine are obtained. The theoretical track displacement increment is calculated based on the rotary encoder data and the travel wheel radius. The actual displacement increment is calculated based on the actual position of the tunneling machine. The slippage state factor is calculated based on the theoretical track displacement increment and the actual displacement increment. The mobile base station reliability is calculated based on the slippage state factor, the tunneling machine position covariance matrix, and the inertial navigation reliability.

[0072] Mobile base station credibility formula:

[0073] ;

[0074] In the formula, To ensure the credibility of mobile base stations, For UWB credibility, To improve the reliability of inertial navigation, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, This is a reference value for the location covariance. This is the slippage state factor.

[0075] It needs further explanation that the two inherent geometric parameters of the tunneling machine are its body length and the radius of its traveling wheels. The body length refers to the maximum external dimension of the tunneling machine along its longitudinal axis in the horizontal plane, measured in meters. The traveling wheel radius refers to the pitch circle radius of the drive wheel in the track drive system that meshes with the track plates and transmits power, also measured in meters. The traveling wheel radius determines the theoretical travel distance of the tunneling machine corresponding to each pulse recorded by the rotary encoder or each unit angle of rotation. The specific method for calculating the theoretical displacement increment of the track based on the rotary encoder data and the traveling wheel radius is as follows: The rotary encoder continuously outputs pulse signals or angle increment signals during the tunneling machine's movement. Within a preset time window, the change in the number of pulses output by the rotary encoder is accumulated. The accumulated pulse count is divided by the standard pulse count output by the rotary encoder per revolution to obtain the rotation of the drive wheel or driven wheel within that time window. The total number of rotations is multiplied by the radius of the traveling wheel and then by 2π. The product is the theoretical distance the tunneling machine's tracks should travel forward within this time window, which is the theoretical track displacement increment. Using the coordinate change of the tunneling machine's real-time position between two different moments obtained in step S4, the actual displacement of the tunneling machine in the actual space is calculated. The real-time position coordinates of the tunneling machine at the beginning of the current time window are recorded as the first position coordinates. The real-time position coordinates of the tunneling machine at the end of the current time window are recorded as the second position coordinates. The Euclidean distance between the second position coordinates and the first position coordinates is calculated. This distance value is the actual displacement increment of the tunneling machine within this time window. The difference between the theoretical track displacement increment and the actual displacement increment is calculated, and then the difference is divided by the theoretical track displacement increment to obtain a ratio between 0 and 1. This ratio is the slippage state factor. When the track does not slip at all, the difference is 0 and the slippage state factor is 0; when the track slips completely, that is, when the tunneling machine spins in place, the actual displacement increment approaches 0 and the slippage state factor approaches 1.

[0076] Using the tunnel boring machine (TBM) as a mobile base station, the reliability of the mobile base station is calculated based on the slippage state factor, the real-time position of the TBM, the position covariance matrix of the TBM, and other relevant reliability parameters. The UWB positioning tag installed on the TBM then functions as a mobile base station, sending or receiving UWB signals to or from personnel positioning cards to provide distance measurement information between personnel and the TBM. Since the position of the TBM itself is obtained through fusion estimation in step S4, the position estimation result has a certain degree of uncertainty. Furthermore, the positioning accuracy of the TBM itself and its reliability as a reference base station will decrease in the slippage state. Therefore, the reliability of the TBM as a mobile base station is quantitatively evaluated. The reliability of the mobile base station is equal to the product of the UWB reliability, the inertial navigation reliability, the position covariance attenuation term, and the slippage penalty term. The position covariance attenuation term is calculated in the form of a natural exponent, where the exponent is the trace of the TBM's position covariance matrix divided by the negative value of the reference value. The larger the trace, the more obvious the attenuation. The slippage penalty term is one minus the slippage state factor, and the more severe the slippage, the smaller this term becomes.

[0077] S6: Based on the reliability of the mobile base station, the UWB signal data and UWB ranging data of the personnel, calculate the real-time location of the personnel;

[0078] The ranging values ​​between the person and the UWB base station, the ranging noise variance between the person and the UWB base station, and the basic ranging noise variance are obtained based on the UWB ranging data of the person. The ranging values ​​between the person and the mobile base station are obtained based on the UWB ranging data. The real-time position of the person is solved with the objective function value minimized.

[0079] ;

[0080] In the formula, The objective function value, The number of UWB base stations participating in the positioning. For the first The variance of ranging noise of a UWB base station For real-time location of personnel, This indicates the real-time position of the tunneling machine. For the first Location of each UWB base station For personnel and the Ranging values ​​of a UWB base station, To ensure the credibility of mobile base stations, Based on the variance of the basic ranging noise, For Euclidean distance This refers to the distance measured between people and mobile base stations.

[0081] It needs further explanation that, based on the personnel's UWB signal data, the ranging values ​​between the personnel and UWB base stations, the ranging noise variance between the personnel and UWB base stations, and the basic ranging noise variance are obtained. For each fixed UWB base station participating in personnel positioning, the straight-line distance between the personnel positioning card and the fixed UWB base station is calculated using the signal arrival timestamp and the signal time-of-flight ranging principle. The total number of fixed UWB base stations participating in positioning depends on the number of fixed UWB base stations in the tunnel that can effectively receive the personnel positioning card signal and whose signal quality meets the preset requirements at the current moment. The ranging noise variance characterizes the uncertainty of the ranging value of the UWB base station. The calculation process involves obtaining the initial... The power ratio reference limit is preset for the initial diameter signal power value and the total received power value. In a coal mine roadway, a straight roadway with low dust concentration and no large equipment obstruction is selected. The average value of the ratio of the initial diameter power to the total power is measured at multiple locations under line-of-sight conditions. The average value is rounded down to 80% and taken as the power ratio reference limit. The proportion of the initial diameter power value read in the current cycle to the total received power value is calculated and compared with the power ratio reference limit. If the proportion reaches or exceeds the reference limit, the first adjustment factor is maintained at one. If the proportion is lower than the reference limit, the first adjustment factor is determined as the power ratio reference limit divided by the above proportion. The lower the proportion, the larger the value of the first adjustment factor. The signal strength indication values ​​of the fixed UWB base station are read for the current period and the four immediately preceding periods, totaling five consecutive values. During a stable period when the tunneling machine is stationary and no personnel are moving in the tunnel, the base station signal strength indication values ​​are continuously collected, and their standard deviation is calculated. Three times the obtained standard deviation is taken as the upper limit of the strength fluctuation tolerance. The standard deviation of the above five signal strength indication values ​​is calculated and compared with the upper limit of the strength fluctuation tolerance. If the standard deviation is less than or equal to the upper limit of the tolerance, the second adjustment factor is maintained at one. If the standard deviation is greater than the upper limit of the tolerance, the second adjustment factor is determined as the result of the standard deviation divided by the upper limit of the tolerance. The more severe the fluctuation, the larger the value of the second adjustment factor.

[0082] The determination of the basic ranging noise variance involves conducting static and dynamic ranging experiments on the same type of UWB transceiver equipment in a dust-free environment with a wide line of sight. Using the total station ranging value as the reference true value, the statistical variance of a large number of ranging sample errors is calculated. This variance is used as the noise variance of the UWB system under ideal conditions. The ranging value between personnel and the mobile base station is obtained based on the UWB ranging data. This UWB ranging data is the UWB ranging data between personnel and the tunneling machine obtained in step S1. The real-time distance value is obtained by directly measuring the two-way time-of-flight distance between the personnel positioning card and the UWB positioning tag installed on the tunneling machine. This UWB ranging data between personnel and the tunneling machine is the ranging value between personnel and the mobile base station. This ranging value reflects the current distance between the personnel and the tunneling machine. The actual geometric distance between the tunnel boring machine and the mobile base station is determined. After acquiring the above data, an objective function is constructed to solve for the real-time location of the personnel. The objective function aims to minimize the real-time location of the personnel. The objective function is used to solve for the real-time location of the personnel, and its value is the weighted sum of squared residuals of two terms. The first term is the sum of the squared residuals corresponding to each fixed base station. The residual is the difference between the calculated distance from the personnel to the base station and the measured distance. The weight is the reciprocal of the ranging noise variance of that base station. The second term is the squared residual corresponding to the mobile base station. The residual is the difference between the calculated distance from the personnel to the tunnel boring machine and the measured distance. The weight is the mobile base station reliability divided by the basic ranging noise variance. By minimizing this objective function, the contribution of ranging information from each base station is adaptively balanced, and the interference of unreliable ranging values ​​on the positioning results is suppressed.

[0083] S7: Calculate dynamic safety risks based on the real-time location of the personnel and the real-time location of the tunneling machine.

[0084] The comprehensive positioning error is calculated based on the slippage state factor, the tunneling machine position covariance matrix, the square of the personnel positioning error, and the tunneling machine body length. The relative distance between the personnel and the tunneling machine is calculated based on the real-time positions of the personnel and the tunneling machine. The personnel trajectory and the tunneling machine trajectory are obtained based on the real-time positions of the personnel and the tunneling machine. The personnel intention factor is calculated based on the angle between the personnel trajectory and the tunneling machine trajectory. The dynamic safety risk is calculated based on the comprehensive positioning error and the relative distance between the personnel and the tunneling machine.

[0085] Comprehensive positioning error formula:

[0086] ;

[0087] In the formula, To account for the overall positioning error, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, Let this be the square of the personnel positioning error. For slippage state factor, This refers to the length of the tunneling machine's fuselage.

[0088] Dynamic security risk formula:

[0089]

[0090] In the formula, For dynamic security risks, As the reference distance, The relative distance between personnel and the tunneling machine. To account for the overall positioning error, For the maximum positioning error, For personnel intention factors.

[0091] It should be further explained that the comprehensive positioning error is calculated based on the slippage state factor, the tunneling machine position covariance matrix, the square of the personnel positioning error, and the tunneling machine body length. The comprehensive positioning error is used to quantify the overall uncertainty of the relative position estimation between the personnel and the tunneling machine. The comprehensive positioning error consists of two parts: the first part is the square root of the sum of the tunneling machine position estimation variance and the square of the personnel positioning error, which represents the uncertainty brought about by the pure algorithm statistical error; the second part is the product of the slippage state factor and the tunneling machine body length, which represents the additional position uncertainty boundary introduced by the tunneling machine slippage. The sum of the two parts is the conservatively estimated comprehensive positioning error.

[0092] After obtaining the comprehensive positioning error and the relative distance between personnel and the tunneling machine, such as Figure 2 As shown, dynamic safety risk is calculated. The dynamic safety risk is used to calculate real-time safety risk indicators. The risk value is obtained by multiplying three factors. The first factor is the baseline distance divided by the relative distance between the personnel and the tunneling machine plus a small positive number. The closer the distance, the larger this factor, reflecting basic geometric risk. The second factor is the ratio of the comprehensive positioning error to the maximum positioning error. The greater the positioning uncertainty, the larger this factor, amplifying the basic risk. The third factor is the personnel intention factor, which is the cosine of the angle between the personnel trajectory and the tunneling machine trajectory, reflecting the consistency of the trajectory directions of the personnel and the tunneling machine. Initially, the personnel trajectory and the tunneling machine trajectory have only one point, and the personnel intention factor is zero. Through this formula, the risk level can be automatically increased when the positioning quality deteriorates, achieving a confidence-based safety early warning.

[0093] The final calculation result of dynamic safety risk is a dimensionless value. The larger the value, the higher the potential risk of a safety accident between personnel and tunneling machine at the current moment. In practical applications, several risk levels are divided according to the value range, and different audible and visual alarm prompts or automatic shutdown control strategies are adopted for different risk levels.

[0094] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A positioning anti-interference method integrating UWB and inertial navigation for complex coal mine environments, characterized in that, Includes the following steps: S1: Collect UWB signal data of personnel, and acquire UWB signal data, dust detection data, inertial navigation data and rotary encoder data of the tunneling machine during its operation in the coal mine tunneling face, and acquire UWB distance measurement data between personnel and tunneling machine. S2: Construct a reliable inertial navigation data sequence for the tunnel boring machine based on the inertial navigation data; S3: Process the UWB signal data of the tunneling machine and construct a reliable UWB data sequence of the tunneling machine by combining it with the dust detection data; S4: Based on the inertial navigation reliable data sequence and the tunneling machine's UWB reliable data sequence, solve for the tunneling machine's real-time position and the tunneling machine's position covariance matrix; S5: Calculate the slippage state factor based on the rotary encoder data and the real-time position of the tunneling machine, use the tunneling machine as a mobile base station, and calculate the reliability of the mobile base station based on the slippage state factor, the real-time position of the tunneling machine, and the position covariance matrix of the tunneling machine. S6: Based on the reliability of the mobile base station, the UWB signal data and UWB ranging data of the personnel, calculate the real-time location of the personnel; S7: Calculate dynamic safety risks based on the real-time location of the personnel and the real-time location of the tunneling machine.

2. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments as described in claim 1, characterized in that, The process involves collecting UWB signal data from personnel and acquiring UWB signal data, dust detection data, inertial navigation data, and rotary encoder data of the tunneling machine during its operation at the coal mine tunneling face. Simultaneously, it acquires UWB ranging data between personnel and the tunneling machine. This includes: obtaining the location of UWB base stations; collecting UWB signal data from personnel, the tunneling machine, and the tunneling machine's movement UWB signal data from each UWB base station; acquiring dust detection data, inertial navigation data, and rotary encoder data of the tunneling machine during its operation at the coal mine tunneling face using sensors deployed in the environment; and acquiring UWB ranging data between personnel and the tunneling machine based on the tunneling machine's UWB base stations.

3. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The step of constructing a reliable inertial navigation data sequence for the tunnel boring machine based on the inertial navigation data includes: the inertial navigation data including angular velocity, correction interval duration, and vibration intensity; obtaining the rotational cumulative threshold and inertial navigation cumulative error reference time calibrated by the inertial navigation experiment; calculating the angular velocity modulus integral based on the angular velocity; obtaining the high-frequency vibration intensity and vibration intensity threshold based on the vibration intensity through Gaussian filtering; calculating the reliability of the inertial navigation data within a preset time window; and associating the reliability of the inertial navigation data with the inertial navigation data to construct a reliable inertial navigation data sequence.

4. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 3, characterized in that, The calculation of the reliability of inertial navigation data within a preset time window includes: Inertial navigation reliability calculation formula: ； In the formula, To improve the reliability of inertial navigation, To correct the interval duration, The reference time for the cumulative error of the inertial navigation system. For the integral of the magnitude of the angular velocity, For rotation accumulation threshold, For high-frequency vibration intensity, This is the vibration intensity threshold.

5. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The process of processing the UWB signal data of the tunneling machine and constructing a reliable UWB data sequence for the tunneling machine by combining it with the dust detection data includes: the UWB signal data including the initial diameter power and the total power of the base stations; obtaining the dust concentration of the UWB base stations based on the dust detection data; constructing an initial topology map of the UWB base stations according to their locations; selecting UWB base stations with an initial diameter power greater than a preset power threshold in the initial topology map as UWB base stations participating in positioning; constructing a positioning topology map; analyzing and obtaining the number of connected components in the positioning topology map; calibrating the linear factor and quadratic factor of the dust error through a dust error experiment; obtaining the horizontal geometric precision factor, the optimal horizontal geometric precision factor, and the horizontal geometric precision attenuation reference value through geometric matrix operations based on the positioning topology map; calculating the reliability of the UWB signal data within a preset time window; and correlating the reliability of the UWB signal data with the inertial navigation data to construct a reliable UWB data sequence.

6. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 5, characterized in that, The calculation of the reliability of UWB signal data within a preset time window includes: UWB credibility calculation formula: ; In the formula, For UWB credibility, The number of UWB base stations participating in the positioning. For the first The first-path power of a UWB base station For the first The total power of each UWB base station It is a very small positive number. For the first Dust concentration of a UWB base station The linearity factor for dust error. The second factor is the dust error factor. To the maximum permissible dust error, Geometric topological factor; ； In the formula, Geometric topological factor, The horizontal geometric precision factor, The optimal horizontal geometric accuracy factor. This is a reference value for the attenuation of horizontal geometric accuracy. To determine the number of connected components in the topological graph.

7. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The step of solving for the real-time position and position covariance matrix of the tunneling machine based on the reliable inertial navigation data sequence and the reliable UWB data sequence of the tunneling machine includes: solving for the real-time position and position covariance matrix of the tunneling machine based on the DS evidence theory.

8. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The step of calculating a slippage state factor based on the rotary encoder data and the real-time position of the tunneling machine, using the tunneling machine as a mobile base station, and calculating the reliability of the mobile base station based on the slippage state factor, the real-time position of the tunneling machine, and the position covariance matrix of the tunneling machine includes: obtaining the body length and the radius of the traveling wheels of the tunneling machine; calculating the theoretical displacement increment of the tracks based on the rotary encoder data and the radius of the traveling wheels; calculating the actual displacement increment based on the actual position of the tunneling machine; calculating the slippage state factor based on the theoretical displacement increment of the tracks and the actual displacement increment; and calculating the reliability of the mobile base station based on the slippage state factor, the position covariance matrix of the tunneling machine, and the reliability of the inertial navigation. Mobile base station credibility formula: ； In the formula, To ensure the credibility of mobile base stations, For UWB credibility, To improve the reliability of inertial navigation, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, This is a reference value for the location covariance. This is the slippage state factor.

9. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The step of calculating the real-time location of a person based on the reliability of the mobile base station, the UWB signal data of the person, and the UWB ranging data includes: obtaining the ranging value between the person and the UWB base station, the ranging noise variance between the person and the UWB base station, and the basic ranging noise variance based on the UWB signal data of the person; obtaining the ranging value between the person and the mobile base station based on the UWB ranging data; and solving the real-time location of the person with the objective function value as the goal. ； In the formula, The objective function value, The number of UWB base stations participating in the positioning. For the first The variance of ranging noise of a UWB base station For real-time location of personnel, This indicates the real-time position of the tunneling machine. For the first Location of each UWB base station For personnel and the Ranging values ​​of a UWB base station, To ensure the credibility of mobile base stations, Based on the variance of the basic ranging noise, For Euclidean distance This refers to the distance measured between people and mobile base stations.

10. The positioning anti-interference method based on UWB and inertial navigation fusion for complex coal mine environments according to claim 1, characterized in that, The calculation of dynamic safety risks based on the real-time positions of personnel and tunneling machines includes: calculating a comprehensive positioning error based on the slippage state factor, the tunneling machine position covariance matrix, the square of the personnel positioning error, and the tunneling machine body length; calculating the relative distance between personnel and tunneling machines based on the real-time positions of personnel and tunneling machines; obtaining the personnel trajectory and tunneling machine trajectory based on the real-time positions of personnel and tunneling machines; calculating the personnel intention factor based on the angle between the personnel trajectory and the tunneling machine trajectory; and calculating dynamic safety risks based on the comprehensive positioning error and the relative distance between personnel and tunneling machines. ； In the formula, To account for the overall positioning error, Let the position covariance matrix of the tunneling machine be... To represent finding the trace of a matrix, Let this be the square of the personnel positioning error. For slippage state factor, This refers to the length of the tunneling machine's fuselage. Dynamic security risk formula: In the formula, For dynamic security risks, As the reference distance, The relative distance between personnel and the tunneling machine. To account for the overall positioning error, For the maximum positioning error, For personnel intention factors.

Images