A method and system for constructing a coal mine underground roadway surveying control network using UWB technology

Abstract

The application discloses a kind of methods for constructing coal mine underground roadway survey control network using UWB technology, belong to coal mine underground survey and mine intelligent technology field.The method includes: multiple UWB reference stations are arranged in underground roadway and initial value of coordinate is established;Distance observation value is obtained by two-way ranging of UWB mobile station and reference station;UWB ranging error compensation model is established to compensate the error of ranging value;Variational Bayes self-adaptive Kalman filtering or weighted least squares method is used to solve the three-dimensional coordinates of mobile station based on the corrected ranging value;According to the coordinate sequence of mobile station, combined with the direction constraint of roadway, the tunneling footage increment is calculated;Dynamic adjustment and self-calibration of survey control network are carried out, including abnormal observation identification and elimination, base station integrity score and automatic switching.The application realizes centimeter-level ranging and positioning in underground roadway, improves the automation level of footage measurement, and is suitable for GNSS unavailable underground environment of mine.

Description

Technical Field

[0001] This invention discloses a method and system for constructing a measurement and control network for underground coal mine roadways using UWB technology, belonging to the fields of underground coal mine measurement, intelligent mining, wireless positioning, geological mapping, and automated measurement of mining roadway advance. Terminology Explanation

[0002] Before understanding the technical solution of this invention, the key terms involved in this invention are defined and explained as follows: UWB: Ultra-Wideband, is a wireless technology that uses nanosecond-level narrow pulses or wideband signals to achieve high-precision ranging, positioning, and communication. Underground roadway measurement control network: A control network deployed in underground roadways of a coal mine to provide spatial reference, coordinate constraints, and measurement transfer relationships; Base station: A fixed UWB measurement node located within the tunnel with known or adjusted coordinates; Rover station: A mobile UWB measurement node, which can be a handheld terminal, a vehicle-mounted terminal, or a mobile terminal installed on tunneling equipment, used to collect distance measurement data between the station and the base station and calculate spatial coordinates; DS-TWR: Double-Sided Two-Way Ranging, a two-way, two-sided ranging method that reduces the impact of device clock asynchrony on ranging results through multiple round-trip communications; TOF: Time of Flight, refers to the time it takes for a wireless signal to travel from the transmitter to the receiver. TOA: Time of Arrival, refers to the time information when the signal arrives at the receiving end; LOS: Line of Sight, refers to the unobstructed direct propagation between the transmitter and receiver. NLOS: Non-Line of Sight, refers to a state in which the signal propagation path is not direct due to obstruction, diffraction, or reflection between the transmitter and receiver. Multipath effect: The phenomenon that wireless signals, after being reflected and diffracted by coal walls, roofs, support components, equipment surfaces, etc., form multiple propagation paths and superimpose, affecting the ranging accuracy; VBAKF: Variational Bayesian Adaptive Kalman Filter, is an algorithm that adaptively adjusts filter parameters to improve the accuracy of localization and state estimation when noise statistics are unknown or changing. Dynamic adjustment: The process of real-time or near-real-time correction of node coordinates, observation weights and residuals in a measurement control network based on newly added observation data, changes in node status or changes in control network topology; GDOP: Geometric Dilution of Precision, used to evaluate the impact of the spatial configuration of measurement nodes on positioning accuracy; Footage measurement: The operation of measuring the actual advance distance of a coal mine tunneling face or mining roadway. Background Technology

[0003] With the continuous advancement of smart mine construction in my country, underground coal mine production has placed higher demands on high-precision, automated, minimally manned, and real-time measurement technologies. The measurement and control network of underground mining roadways is a crucial spatial foundation for geological mapping, tunneling footage statistics, mining engineering acceptance, production scheduling, equipment positioning, disaster early warning, and intelligent mining control. Traditional underground measurement mainly relies on total stations, traverse surveying, manual distance measurement with tape measures, or manual recording of tunneling footage. While these methods offer high single-measurement accuracy, they suffer from drawbacks such as long operation times, high labor intensity, low automation, inability to perform continuous measurements, and difficulty in reflecting real-time dynamic changes in tunneling.

[0004] Underground mine roadways are typical GNSS-denied environments, where satellite navigation signals cannot directly reach the interior. The confined space, long roadways, and limited cross-sections of coal mines mean that coal walls, roofs, floors, metal supports, transport equipment, pipelines, cables, and personnel and vehicles all affect wireless signal propagation. Furthermore, variations in underground temperature and humidity, dust, roadway bends, equipment obstructions, and non-line-of-sight propagation cause ranging errors to exhibit nonlinear and non-stationary characteristics as the environment changes. Therefore, simply applying ground-based wireless positioning methods is insufficient to meet the requirements for constructing centimeter-level measurement control networks and automating tunnel advance measurements in underground coal mine roadways.

[0005] Ultra-wideband (UWB) technology boasts advantages such as large bandwidth, narrow pulse width, high temporal resolution, strong anti-interference capability, low power consumption, strong multipath resolution, high ranging accuracy, and moderate equipment cost. UWB technology has been widely applied in indoor positioning, personnel positioning, equipment positioning, warehousing and logistics, and robot navigation. Compared to wireless positioning technologies such as Bluetooth, Wi-Fi, and RFID, UWB, based on signal time-of-flight, theoretically offers higher ranging accuracy, making it suitable for centimeter- to decimeter-level positioning in short to medium distances. Coal mine tunnels exhibit significant long-distance linear spatial characteristics, and under certain conditions, UWB signals may also exhibit tunnel waveguide effects, giving them the potential to achieve long-distance coverage within tunnels.

[0006] However, UWB ranging and positioning in underground roadway environments still face challenges such as multipath effects, non-line-of-sight propagation, node geometry degradation, base station failure, time-varying ranging noise, temperature and humidity influences, and increased measurement errors due to dynamic movement. Traditional positioning methods based on a single UWB base station or a small number of nodes are insufficient to meet the requirements of stability, redundant observation, coordinate transfer accuracy, and dynamic self-calibration capabilities for underground mining roadway control network construction. Therefore, it is necessary to propose a UWB measurement and control network construction method for underground coal mine roadway environments. This method achieves high-precision, long-distance, continuous, and automated operation of the underground roadway measurement and control network through multi-reference station joint networking, reasonable spatial topology layout, ranging error modeling, adaptive filtering, dynamic adjustment, and anomaly observation removal. Summary of the Invention

[0007] (a) Purpose of the invention To address the problems of low efficiency in manual measurement, insufficient accuracy in wireless positioning, weak geometry of underground roadways, significant multipath and non-line-of-sight errors, and lack of self-calibration capability in existing technologies, this invention provides a method for constructing a measurement and control network for underground roadways in coal mines using UWB technology. This method enables centimeter-level distance measurement and high-precision positioning in underground roadways, improves the automation level of mining roadway advance measurement, and enhances the stability and robustness of the measurement and control network.

[0008] (II) Technical Solution To address the problems of low efficiency in manual measurement, insufficient accuracy in wireless positioning, weak geometry of underground roadways, significant multipath and non-line-of-sight errors, and lack of self-calibration capability in existing technologies, this invention provides a method for constructing a measurement and control network for underground roadways in coal mines using UWB technology. This method enables centimeter-level distance measurement and high-precision positioning in underground roadways, improves the automation level of mining roadway advance measurement, and enhances the stability and robustness of the measurement and control network.

[0009] This invention proposes a method for constructing a measurement and control network for underground coal mine roadways using UWB technology. The method uses multiple fixed UWB reference stations as the underground spatial control framework and one or more UWB rover stations as measurement nodes. Distance observations between reference stations and rover stations, and between reference stations themselves, are obtained through bilateral two-way ranging. Through roadway topology constraints, multi-base station redundant observations, ranging error compensation, variational Bayesian adaptive Kalman filtering, and dynamic adjustment algorithms, the method achieves three-dimensional coordinate calculation of the underground roadway, dynamic calibration of the control network, and automated measurement of tunneling footage.

[0010] The overall technical approach of this invention includes the following steps: S1: Deploy multiple UWB reference stations in the underground roadway and establish the initial coordinate values ​​of each UWB reference station; S2: By performing bilateral and bidirectional distance measurement between the UWB rover and the multiple UWB reference stations, the distance observation values ​​between the UWB rover and each visible UWB reference station are obtained. S3: Establish a UWB ranging error compensation model for underground roadways, perform error compensation on the distance observation values, and obtain the corrected ranging values; S4: Based on the corrected ranging value, the three-dimensional coordinates of the UWB rover station are calculated using variational Bayesian adaptive Kalman filtering or weighted least squares method. S5: Based on the three-dimensional coordinate sequence of the UWB rover station and combined with the roadway direction constraint, calculate the incremental tunneling footage along the roadway direction; S6: Perform dynamic adjustment and self-calibration on the measurement control network consisting of the multiple UWB reference stations and the UWB rover stations. The dynamic adjustment and self-calibration includes: performing statistical tests on the observation residuals between reference stations and between reference stations and rover stations to identify abnormal observations, downweighting or removing abnormal observations, and realizing automatic switching of base stations based on the integrity score of the reference stations.

[0011] The steps described above are explained below: Step S1, further, involves deploying multiple UWB reference stations within the underground roadway and establishing initial coordinate values ​​for each UWB reference station. The multiple UWB reference stations are deployed using an asymmetric spatial gradient configuration, with adjacent reference stations positioned at different heights on either side of the roadway or on the same side, forming an interlaced geometric configuration in both horizontal and vertical directions. The antennas of each reference station are at least 30 cm away from the coal face, metal supports, and large equipment. Each rover measurement location simultaneously observes no fewer than four reference stations. The coordinates of the UWB reference stations are established and initialized using existing underground traverse point constraints or distance adjustment between base stations.

[0012] Step S2, further, involves performing bilateral two-way ranging (DS-TWR) between the UWB rover and the multiple UWB reference stations to obtain distance observations between the UWB rover and each visible UWB reference station. The DS-TWR method calculates the time of flight (TOF) of the wireless signal through multiple message exchanges (POLL-RESP-FINAL) between the rover and the reference stations, and then converts the distance observations based on the electromagnetic wave propagation speed, thereby reducing ranging errors caused by node clock asynchrony.

[0013] Step S3, further, establishes an underground roadway UWB ranging error compensation model to compensate for the errors in the distance observation values, obtaining corrected ranging values. The ranging error compensation model compensates for at least one of systematic errors, multipath errors, non-line-of-sight errors, and random noise. The error compensation model utilizes calibrated distances or redundant observations from the control network, taking parameters such as temperature, humidity, signal strength, first-path power, received power, distance, and signal-to-noise ratio as inputs, and outputs corrected ranging values.

[0014] Step S4, further, based on the corrected ranging values, the three-dimensional coordinates of the UWB rover are calculated using variational Bayesian adaptive Kalman filtering (VBAKF) or weighted least squares method. The weighted least squares method iteratively calculates the rover coordinates until the coordinate corrections meet a preset convergence threshold. The VBAKF adaptively estimates the process noise covariance and observation noise covariance using variational Bayesian inference, and introduces prior elevation constraints derived from the tunnel design slope or traverse point elevation during the filtering process to improve the stability of three-dimensional positioning.

[0015] Step S5, further, based on the three-dimensional coordinate sequence of the UWB rover station and combined with the roadway direction constraint, calculate the incremental advance along the roadway direction. The incremental advance is calculated by projecting the position difference vector between the rover station at the current moment and the initial moment onto the roadway centerline or the unit direction vector of the advance direction. The rover station coordinates at multiple consecutive moments are filtered, and abnormal positions that significantly deviate from the centerline are eliminated using the roadway direction constraint.

[0016] Step S6, further, performs dynamic adjustment and self-calibration on the measurement control network composed of the multiple UWB reference stations and the UWB rover. The dynamic adjustment and self-calibration includes: statistically testing the observation residuals between reference stations and between reference stations and rover to identify abnormal observations; downweighting or removing abnormal observations; and automatically switching base stations based on the reference station integrity score. Distance observations between reference stations and between reference stations and rover are periodically collected, and the overall control network is adjusted using an iterative least squares method. Statistical testing is performed on the residuals, and abnormal observations are identified and removed or downweighted according to the 3σ criterion. The reference station integrity score is determined by a weighted average of at least one of observation success rate, residual stability, communication quality, and power status. When some reference stations fail or their scores fall below a preset threshold, the system automatically switches to an adjacent available reference station combination through redundant observations to maintain continuous operation of the control network.

[0017] Furthermore, the multiple UWB reference stations mentioned in step S1 are arranged in an asymmetric spatial gradient manner, with adjacent reference stations being arranged at different heights on both sides of the roadway or on the same side, so as to form an interlaced geometric configuration in the plane and elevation directions.

[0018] Further, the signal flight time of the bilateral two-way ranging in step S2 is calculated using the following formula:

[0019] Among them, T round1 and T round2 T represents the round-trip measurement time. reply1 and T reply2 T represents the internal response time of the two devices respectively. of Indicates the signal's flight time.

[0020] Furthermore, the ranging error compensation model described in step S3 compensates for at least one of the system error, multipath error, non-line-of-sight error, and random noise; the corrected ranging value is obtained by subtracting the distance-related error correction term, the system bias correction term, and the non-line-of-sight error correction term from the original ranging value.

[0021] Furthermore, the variational Bayesian adaptive Kalman filter described in step S4 adaptively estimates the process noise covariance and observation noise covariance through variational Bayesian inference, and introduces prior elevation constraints derived from the tunnel design slope or traverse point elevation during the filtering process.

[0022] Furthermore, the advance increment mentioned in step S5 is calculated using the following formula:

[0023] Among them, P o and P t These are the initial and current positions of the UWB mobile station, respectively, where u is the unit direction vector of the tunnel centerline or the tunneling direction.

[0024] Furthermore, in step S6, abnormal observations are identified based on residuals exceeding a preset threshold. The integrity score of the base station is determined by weighting at least one of the following: observation success rate, residual stability, communication quality, and power status. The objective function of the control network adjustment is to minimize the weighted sum of squared observation residuals.

[0025] Furthermore, the UWB base station is fixedly installed on the sidewall or top anchor bolt of the tunnel, and the UWB mobile station is handheld, vehicle-mounted, robot-mounted, or mounted on tunneling equipment.

[0026] Furthermore, an automatic measurement method for tunneling footage in underground coal mine roadways based on a UWB measurement and control network is provided. The UWB measurement and control network constructed using the above method continuously acquires the three-dimensional coordinates of UWB rover stations located near the tunneling face or on the tunneling equipment, and performs projection calculations in conjunction with roadway direction constraints to automatically obtain the tunneling footage increment.

[0027] Furthermore, a UWB measurement and control network system for underground coal mines is provided, comprising: multiple fixed UWB reference stations deployed in underground roadways to provide spatial coordinate references; at least one UWB rover station for bilateral and bidirectional ranging with the multiple fixed UWB reference stations; and a background data processing system for receiving ranging data and performing error compensation, coordinate calculation, footage calculation, dynamic adjustment, and self-calibration steps in the above method.

[0028] In view of the problems of low efficiency of manual measurement, insufficient accuracy of wireless positioning, weak geometry of underground roadways, significant multipath and non-line-of-sight errors, and lack of self-calibration capability of control network in the existing technology, the present invention has the following beneficial effects.

[0029] First, it achieves centimeter-level ranging and high-precision positioning in underground roadways. Through DS-TWR ranging, error compensation, adaptive filtering, and dynamic adjustment, this invention enables centimeter-level ranging in underground roadway environments. Examples show that the static ranging RMSE can reach approximately 1.67 cm, the dynamic ranging RMSE can reach approximately 2.79 cm, the static planar positioning RMSE can reach approximately 1.97 cm, and the dynamic planar positioning RMSE can reach approximately 4.69 cm.

[0030] Second, it improves the automation level of advance measurement in mining roadways. This invention utilizes a UWB rover to acquire position coordinates in real time and automatically calculates advance based on roadway direction constraints. This significantly reduces the time and labor intensity of manual advance measurement, and improves the real-time performance, accuracy, and automation level of advance statistics.

[0031] Third, it adapts to the underground coal mine environment where GNSS is unavailable. This invention does not rely on GNSS signals, but utilizes UWB reference stations deployed underground to construct an independent space measurement and control network, making it suitable for underground roadways, tunnels, mines, and other environments where satellite signals are inaccessible.

[0032] Fourth, it enhances the stability and robustness of the downhole measurement and control network. Through redundant observations from multiple reference stations, dynamic adjustment, outlier observation removal, and automatic base station switching, this invention can maintain the operation of the measurement and control network when some observations are interfered with or some base stations are in abnormal condition, thereby improving system reliability.

[0033] Fifth, it reduces the cost and safety risks of downhole measurement operations. This invention can reduce the number of times surveyors need to enter the vicinity of the tunneling face for manual measurements, thereby reducing the risks of downhole operations and improving the efficiency of measurement data acquisition.

[0034] Sixth, it possesses long-distance continuous measurement capabilities. The UWB system described in this invention has an effective measurement distance exceeding 300m in underground roadways and can automatically switch between different base stations, making it suitable for continuous measurement in long-distance mining roadways.

[0035] Seventh, it has strong scalability. This invention can be integrated with inertial navigation, temperature and humidity sensors, LoRa communication, underground robots, tunneling machine posture measurement systems, and mine production scheduling systems to form a high-precision underground spatial positioning service platform for smart mines. Attached Figure Description

[0037] Figure 1 This is a flowchart of the overall process of the present invention; Figure 2 A schematic diagram of the structure of a UWB fixed reference station; Figure 3 This is a schematic diagram of the UWB rover structure. Figure 4 A schematic diagram of the zigzag, staggered layout of UWB reference stations in underground roadways; Figure 5 This is a timing diagram of the UWB bilateral two-way ranging DS-TWR. Figure 6 Flowchart of UWB ranging error compensation and dynamic adjustment; Figure 7 A schematic diagram of tunnel advance calculation based on UWB rover coordinates; Figure 8 A schematic diagram of the on-site test layout of the mortar conveyor roadway at the 2106 working face of Yingpanhao Coal Mine.

[0038] Figure 9 Experimental diagram of the 2106 working face at Yingpanhao Coal Mine The labels for the various figures are explained as follows: Figure 1 In the middle section: 101—Underground site reconnaissance, 102—UWB measurement equipment configuration, 103—Measurement control network layout, 104—UWB ranging data acquisition, 105—Range error compensation, 106—Coordinate calculation and dynamic adjustment, 107—Result output, 108—Tunnel design data, 109—Known traverse points, 110—Installation location conditions, 111—Anomaly observation removal module, 112—Base station automatic switching module, 113—Control network self-calibration module.

[0039] Figure 2 In the middle: 200—UWB fixed reference station, 201—antenna, 202—RF front end, 203—UWB RF transceiver module, 204—main controller, 205—data storage unit, 206—communication unit, 207—environmental sensor, 208—power management module, 209—downhole gateway or host computer, 210—mounting base.

[0040] Figure 3In the middle: 300—UWB rover, 301—antenna, 302—UWB transceiver module, 303—main control unit, 304—coordinate calculation unit, 305—communication unit, 306—display unit, 307—operation buttons, 308—inertial measurement unit, 309—battery unit, 310—handheld housing, 311—fixed reference station, 312—back-end system or downhole gateway.

[0041] Figure 4 In the middle: 400—underground mining roadway, 401—upper side or high-level installation zone, 402—lower side or low-level installation zone, 403—high-level reference station, 404—low-level reference station, 405—UWB mobile station, 406—roadway extension and excavation direction, 407—span observation link.

[0042] Figure 5 In the middle: 500—UWB rover, 501—UWB base station, 502—POLL message, 503—RESP message, 504—FINAL message, 505—time of flight calculation unit.

[0043] Figure 6 In the diagram: 600—Original UWB ranging value, 601—Signal quality parameters, 602—Environmental parameters, 603—LOS / NLOS identification module, 604—Range error comprehensive compensation module, 605—Corrected ranging value, 606—Dynamic adjustment and coordinate update module, 607—Weight determination module, 608—Abnormal observation removal module, 609—Control network self-calibration module.

[0044] Figure 7 In the diagram: 700—underground roadway, 701—unit vector of roadway direction, 702—initial measurement point P0, 703—current measurement point P t 704—Increment in footage, 705—Formula for calculating footage.

[0045] Figure 8 In the middle: 800—2106 working face grout transport test section, 801—densified deployment area, 802—conventional deployment area, 803—handheld UWB mobile terminal, 804—long distance ranging link. Detailed Implementation

[0046] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are for illustrative purposes only and should not be considered as limiting the scope of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0047] This embodiment uses the 800 section of the 2106 working face haulage roadway in Yingpanhao Coal Mine as the test site to detail the configuration of the UWB measurement equipment and the layout of the reference stations. The roadway has an average coal thickness of approximately 7.15m and an average dip angle of approximately 3°. Support methods include roof anchor mesh, dovetail support, T-shaped steel strips, and sidewall anchor bolts. The total length of the test section is approximately 320m, and the on-site test setup is as follows... Figure 8 As shown.

[0048] The measurement equipment required for this solution includes a UWB fixed reference station, a UWB rover, a communication and data processing unit, a power supply unit, and optional environmental sensor units. Depending on the application scenario, the UWB rover can be designed as a handheld terminal, a vehicle-mounted terminal, a robot-mounted terminal, or a tunneling equipment-mounted terminal.

[0049] like Figure 2 As shown, the UWB fixed reference station 200 is a fundamental node in the underground roadway measurement and control network, and its hardware modules mainly include: The UWB radio frequency transceiver module 203 is used to transmit and receive UWB pulse signals to complete ranging communication with other base stations or rover stations. In this embodiment, a DW1000 type UWB radio frequency transceiver module that supports high-precision timestamp recording is preferably used, or a domestically produced UWB chip module with equivalent ranging capabilities.

[0050] The RF front-end 202 and antenna 201 module includes an RF front-end chip, a power amplifier, a low-noise amplifier, a filter circuit, and an SMA antenna. The antenna should have good omnidirectional radiation characteristics and adaptability to the downhole environment. During installation, the distance between the antenna and the coal face, metal support, or large equipment should meet the requirements for reducing multipath effects, preferably greater than 30cm.

[0051] The main controller 204 is used to control the UWB module's transmission and reception, record timestamps, execute the ranging protocol, buffer observation data, and communicate with the host computer or gateway. This embodiment uses an STM32F103CBT6 microcontroller, but GD32 series chips or other embedded microcontrollers with SPI, UART, and low-power control capabilities can also be used.

[0052] The communication unit 206 is used for data transmission between the base station and the host computer, gateway, or other nodes. It can employ wired RS485, CAN, Ethernet, LoRa, Wi-Fi, or a wireless communication method permitted in underground coal mines. For long-distance tunnels, low-power, long-range, and interference-resistant LoRa or wired communication methods are preferred. This embodiment uses LoRa wireless communication.

[0053] The power management module 208 is used to provide a stable power supply for the base station, and can use an intrinsically safe underground power supply, a mining DC power supply, a rechargeable battery, or an external power supply. The power supply module should meet the intrinsically safe design requirements for underground coal mines.

[0054] The environmental sensor module 207, optionally configured with temperature, humidity, air pressure, dust, or attitude sensors, is used to collect environmental parameters affecting UWB ranging errors and participate in the construction of the ranging error compensation model. This embodiment is configured with temperature and humidity sensors.

[0055] The protective enclosure of the reference station should be made of flame-retardant, anti-static, and corrosion-resistant materials, and should meet the corresponding protection level requirements according to the underground environment. A permanent explosion-proof mark, MA mark, and nameplate should be prominently displayed on the enclosure. The reference station is fixedly installed to the sidewall or top anchor bolts of the roadway using mounting bracket 210.

[0056] The base station transmits data with the downhole gateway or host computer 209 through the communication unit 206.

[0057] like Figure 3 As shown, the UWB rover 300 is used for mobile measurement or footage measurement in underground roadways. Its hardware modules mainly include: The UWB transceiver module 302 is used for bilateral, bidirectional ranging with multiple fixed reference stations. This embodiment uses the same DW1000 module as the reference station.

[0058] The main control unit 303 is used to control the ranging process, store data, and perform preliminary coordinate calculations.

[0059] The communication unit 305 is used to send ranging results, coordinate results or equipment status to the background system or the downhole gateway 312.

[0060] The display unit 306 and operation buttons 307 can be equipped with a display screen, buttons or indicator lights for on-site viewing of ranging distance, current base station, coordinates and equipment status.

[0061] Battery unit 309 can be a power bank, intrinsically safe battery or other mining safety power source.

[0062] An optional inertial measurement unit 308, including an accelerometer, gyroscope, and magnetometer, is used to fuse with UWB ranging results to improve dynamic measurement stability.

[0063] Optional temperature and humidity sensors can be used to help compensate for ranging errors caused by environmental changes.

[0064] The aforementioned components are integrated into the handheld housing 310. The rover can be used handheld or mounted on a tunneling machine, transfer machine, transport vehicle, inspection robot, or roadway surveying trolley. When the rover is mounted on mobile equipment, the antenna installation position should be kept relatively fixed, and the spatial offset relationship between the antenna phase center and the equipment reference point should be established. In this embodiment, the rover is used handheld and carried by the surveyor to move along the roadway for measurement. The rover transmits data with the backend system or underground gateway 312 through the communication unit 305 and obtains distance observation values ​​through the ranging link with the fixed reference station 311.

[0065] The data processing and backend system is deployed on the surface server, receiving UWB ranging observations, environmental sensor data, equipment status data, and control network coordinate data. It performs error compensation, coordinate calculation, dynamic adjustment, anomaly detection, and tunneling footage calculation. The system's main functions include: UWB ranging data parsing and time synchronization; base station coordinate management and control network topology management; ranging error modeling and real-time correction; rover station 3D coordinate calculation; adaptive Kalman filtering and VBAKF filtering; control network dynamic adjustment; anomaly observation removal and base station integrity diagnosis; and automatic tunneling footage calculation and report generation.

[0066] This invention addresses the characteristics of underground coal mine roadways, such as their narrow and elongated spaces, limited cross-sections, complex support structures, and significant obstruction affecting distance measurement. Especially for long, straight, or nearly straight roadways, it preferentially employs an asymmetric spatial gradient (alternating high and low elevations) layout method. Within the roadway cross-section, three-dimensional geometric constraints are formed through differentiated combinations of sidewall and elevation positions. The main setup method is as follows: Multiple UWB base stations are sequentially deployed along the tunnel direction, numbered 1, 2, 3...n; Adjacent reference stations are respectively deployed at different heights on both sides of the roadway or on the same side, so that they form an interlaced geometric configuration in both the horizontal and vertical directions; Odd-numbered reference stations can be placed at the bottom of the coal face or near low-position anchor bolts, while even-numbered reference stations can be placed near higher-position anchor bolts or in the opposite direction. The distance between the base station antenna and the coal face, metal support and large equipment should preferably be greater than 30cm to reduce multipath effects; The distance between adjacent base stations is determined based on the roadway conditions, the effective UWB coverage distance, and the measurement accuracy requirements. For densely populated measurement areas, the adjacent spacing can be 5m to 20m; for conventional long-distance roadways, the adjacent spacing can be 30m to 80m; under the condition of meeting the ranging quality requirements, the effective coverage distance of a single base station can exceed 300m. For each rover station measurement location, it is preferable to observe no fewer than 3 reference stations simultaneously; if high-precision three-dimensional coordinate calculation is to be performed, it is preferable to observe no fewer than 4 reference stations simultaneously.

[0067] Specifically, such as Figure 4 As shown, in the underground mining roadway 400, an asymmetric spatial gradient layout method is adopted to sequentially deploy multiple UWB reference stations along the roadway direction. High-level reference station 403 and low-level reference station 404 are deployed on the upper side high-level installation zone 401 and the lower side low-level installation zone 402, respectively, so that adjacent reference stations form an interlaced geometric configuration in the plane and elevation directions.

[0068] like Figure 8 As shown, a total of 8 UWB reference stations were deployed in the test section: devices 1, 3, 5, and 7 were located near the second anchor bolt at the bottom of the coal face (corresponding to low-level reference station 404), and devices 2, 4, 6, and 8 were located near the fourth anchor bolt (corresponding to high-level reference station 403), thus creating a differentiated distribution at the sidewall and elevation positions. In the dense deployment area 801, the spacing between devices 1 to 4 was approximately 6m, used to form a localized dense deployment area; in the conventional deployment area 802, the spacing between devices 4 to 8 was approximately 60m, used to verify the coverage capability of long-distance roadways. During installation, the antenna was ensured to be more than 30cm away from the coal face to reduce the impact of multipath effects. A span-of-sight observation link 407 was formed between adjacent base stations and between base stations and rover stations, and redundant ranging links were established between non-adjacent base stations.

[0069] In other specific implementations, for long mining roadways with relatively regular orientations, a linear enhancement deployment method can be adopted. Its main features are as follows: base stations are deployed along the roadway orientation to form the main survey line; span connections are introduced in key sections to enable distance constraints between non-adjacent base stations; localized densification of the deployment is implemented in areas with roadway bends, intersections, slope changes, dense equipment, and severe obstruction; and the control network constraint strength is improved through redundant distance observations between base stations, avoiding error accumulation caused by a simple chain structure.

[0070] For connecting roadways, intersections, and areas where the working face roadway intersects with the cut-off point, a branch-constrained layout method can be adopted. Its main characteristics are as follows: key UWB reference stations are set up at roadway intersections or branch points; UWB reference stations are deployed along different branch roadways; multi-directional distance constraints are formed between branch nodes and reference stations in each direction; and the consistency of coordinate transfer between different roadway branches is enhanced through redundant observations in the intersection area.

[0071] For mining areas or working faces requiring long-term stable operation, a multi-level composite deployment method can be adopted. Its main characteristics are as follows: A composite network consisting of a main control network, a densification network, and a mobile measurement network is constructed; the main control network comprises a small number of stable reference stations, connecting known traverse points or permanent control points underground; the densification network extends along the roadway to improve local measurement density and the observability of mobile stations; the mobile measurement network is composed of mobile stations and visible reference stations in real time, used for locating current measurement points and calculating footage; when some reference stations fail, the system automatically switches to adjacent available reference station combinations through redundant observations to maintain continuous operation of the control network.

[0072] In this invention, the coordinates of the UWB reference station can be established in the following ways: Using existing traverse points, total station measurement results, or mine measurement control points underground, determine the coordinates of several primary UWB reference stations; Using UWB distance measurements and known point constraints between reference stations, the coordinates of other reference stations are determined through free network adjustment or constrained adjustment. During the extension of the tunnel, new UWB reference stations will be added to the existing control network, and the coordinates of the control network will be updated through incremental adjustment. For newly installed reference stations, temporary initial coordinate values ​​can be used first, and then coordinate iterative correction can be performed using distance measurement data from multiple time periods.

[0073] In this embodiment, using existing traverse points underground, the coordinates of primary reference stations No. 1, 3, 5, and 7 were determined using a total station. Using UWB distance measurements between the reference stations and known point constraints, the coordinates of reference stations No. 2, 4, 6, and 8 were determined through constrained adjustment. The coordinate initialization accuracy was better than 5cm, meeting the requirements for control network adjustment.

[0074] Let the coordinate parameter vector of the base station be:

[0075] The distance observation equation between base stations is:

[0076] Among them, L ij V represents the distance measurement observation between the i-th and j-th reference stations. ij To observe the residuals.

[0077] The objective function for control network adjustment is:

[0078] Among them, P ij For observation weights. Weights can be determined based on ranging distance, signal quality, LOS / NLOS status, and environmental parameters:

[0079] This invention preferably uses the bilateral two-way ranging DS-TWR method to obtain the distance between nodes, and its ranging timing is as follows: Figure 5 As shown, this method calculates the time-of-flight of the wireless signal through multiple message exchanges between the tag and the base station, thus mitigating ranging errors caused by node clock asynchrony.

[0080] In a typical DS-TWR ranging process, the UWB rover 500 sends a POLL message 502 to the UWB base station 501. Upon receiving this message, the UWB base station 501 returns a RESP message 503, and the UWB rover 500 then sends a FINAL message 504. Multiple round-trip times and response times can be obtained based on the transmission and reception timestamps. The time-of-flight calculation unit 505 calculates the signal time of flight (TOF) based on the transmission and reception timestamps, and then converts this into the distance observation value.

[0081] Flight time can be calculated using the following formula:

[0082] Among them, T round1 and T round2 T represents the round-trip measurement time. reply1 and T reply2 T represents the internal response time of the two devices respectively. of Indicates the signal's flight time.

[0083] The original distance between nodes is calculated using the following formula:

[0084] Where, d m Here, is the original ranging value, and c is the propagation speed of electromagnetic waves in air. Considering variations in downhole air temperature and humidity, as well as hardware delays, corrections can be made to the propagation speed and system delay.

[0085] The station and the visible reference station sequentially complete the DS-TWR ranging, and each ranging cycle acquires distance observations from more than 4 reference stations.

[0086] In the UWB observation model, the unknown three-dimensional coordinates of the rover at a certain moment are:

[0087] The known coordinates of the i-th UWB base station are:

[0088] The theoretical distance from the i-th base station to the rover is:

[0089] The UWB ranging observation equation is:

[0090] Where, d i Let ℇ be the distance measurement observation between the i-th base station and the rover. i The ranging error includes systematic error, multipath error, non-line-of-sight error, and random noise.

[0091] In underground tunnel environments, the total distance measurement error can be expressed as:

[0092] Among them, ℇ sys This is a system error, mainly caused by hardware delay and clock frequency offset; ℇ mp This is a multipath error, mainly caused by reflections from the coal face, roof, floor, and metal supports; ℇ NLOS This is a non-line-of-sight error, mainly caused by obstruction, diffraction, and reflection paths replacing the direct path; ℇ noise It is random noise.

[0093] To improve the accuracy of UWB ranging in underground roadways, this invention establishes a ranging error compensation model suitable for underground roadway environments. The error compensation process is as follows: Figure 6 As shown. The relationship between the original distance measurement value and the true distance can be expressed as: The relationship between the original distance measurement value and the true distance can be expressed as:

[0094] Where, d m The original UWB ranging value, d t Let α be the true distance, β be the distance-related error coefficient, and β be the distance offset corresponding to the system's fixed delay. η This is the random error term.

[0095] The variance of random error can be modeled as a function that varies with distance:

[0096] in, denoted as the initial ranging noise variance, and k is the distance correlation variance growth coefficient.

[0097] When there are mixed LOS and NLOS states, a mixed error model can be constructed:

[0098] Among them, P LOS and P NLOS Let be the probabilities of the line-of-sight state and the non-line-of-sight state, respectively, and satisfy the following:

[0099] like Figure 6 As shown, the system acquires the original UWB ranging value 600, signal quality parameters 601 (including first path power, received power, signal-to-noise ratio and ranging success rate) and environmental parameters 602 (including temperature and humidity). The LOS / NLOS identification module 603 determines the line-of-sight and non-line-of-sight propagation status, and the ranging error comprehensive compensation module 604 corrects system errors, multipath errors, NLOS errors and distance-related errors.

[0100] The distance correction value can be obtained by the following formula:

[0101] Where, d c This is the corrected distance measurement value. The distance correlation error coefficient is the estimated value. For the estimated system bias, This is the estimated value for non-line-of-sight error.

[0102] In other specific implementations, a multivariate mathematical regression model can be established using calibration distance, total station calibration distance, or redundant observations from the control network. Parameters such as temperature, humidity, signal strength, initial diameter power, received power, distance, and signal-to-noise ratio can be used as input variables to establish a ranging correction function.

[0103] Where T is temperature, H is humidity, and P is temperature. fp For the initial diameter power, P rx Here, SNR is the received power, and SNR is the signal-to-noise ratio. The corrected ranging value is:

[0104] Among them, ℇ sys This is a system error, mainly caused by hardware delay and clock frequency offset; ℇ mp This is a multipath error, mainly caused by reflections from the coal face, roof, floor, and metal supports; ℇ NLOS This is a non-line-of-sight error, mainly caused by obstruction, diffraction, and reflection paths replacing the direct path; ℇ noise It is random noise.

[0105] The error coefficients are predetermined by calibration measurements at multiple known distance points in the tunnel before the test, and are adjusted in real time according to environmental parameters during the measurement process.

[0106] Rover coordinate calculation involves measuring distances between the rover and multiple UWB reference stations at a given measurement moment to obtain the corrected distance observation value d. c,i The coordinates of the rover station are calculated using nonlinear least squares or weighted least squares methods.

[0107] The observation equation is:

[0108] Linearizing the observation equation, it can be written as:

[0109] Where L is the observation correction vector, A is the design matrix, ΔX is the coordinate correction vector to be determined, and V is the residual vector.

[0110] The weighted least squares solution is:

[0111] Iterative update of rover coordinates:

[0112] Iteration stops when the coordinate corrections meet the following condition:

[0113] in, The preset convergence threshold can be set according to the measurement accuracy requirements, for example, from 0.001m to 0.01m.

[0114] To improve the stability and robustness of the rover during dynamic measurement, this invention employs a variational Bayesian adaptive Kalman filter method to dynamically filter the UWB positioning results.

[0115] The state vector of the rover station can be represented as:

[0116] Where, p k Let v be the position vector. k Let a be the velocity vector. k These are acceleration or motion model parameters. For implementations that integrate inertial sensors, the state vector may also include attitude error, accelerometer bias, and gyroscope bias.

[0117] The state prediction equation is:

[0118] The observation equation is:

[0119] Among them, F k Let G be the state transition matrix. k H is the process noise driving matrix. k Let w be the observation matrix. k For process noise, e k To observe noise.

[0120] Traditional Kalman filtering requires pre-defined process noise covariance and observation noise covariance. However, in underground roadway environments, ranging noise varies with distance, obstruction, environment, and movement status. Therefore, this invention introduces variational Bayesian inference to adaptively estimate the statistical characteristics of noise during the filtering process.

[0121] By setting the prediction error covariance to follow an inverse Wissaud distribution, variational inference is used to minimize the KL divergence between the approximate posterior distribution and the true posterior distribution.

[0122] State X is estimated synchronously through fixed-point iteration. k and P k Covariance is used to adaptively adjust the filter gain, thereby improving the system's ability to adapt to non-stationary noise and abnormal ranging values.

[0123] In three-dimensional positioning of underground roadways, the accuracy in the elevation direction may be weaker than in the planar direction due to the limited height distribution of nodes within the roadway. This invention can introduce an elevation constraint model:

[0124] Alternatively, elevation prior constraints can be established based on the tunnel design slope, floor elevation, traverse point elevation, and equipment installation height to improve the stability of three-dimensional positioning.

[0125] This embodiment describes in detail the complete process of automatic calculation of tunneling footage and dynamic adjustment and self-calibration of the control network.

[0126] Let the unit vector of the design centerline or measured centerline of the tunnel be:

[0127] The position of the rover at the initial or reference time is:

[0128] The current location of the mobile station is:

[0129] The advance increment along the tunnel direction is:

[0130] When considering only planar advance, a unit vector in the planar direction can be used:

[0131] The corresponding planar advance is:

[0132] To reduce the impact of single measurement errors on the advance calculation, this invention can perform wavelet threshold denoising, moving average or Kalman filtering on the rover coordinates at multiple consecutive times, and eliminate abnormal positions that are significantly deviated from the centerline by using roadway direction constraints.

[0133] During operation, the accuracy of underground UWB measurement and control networks can be affected by factors such as the addition of new reference stations, reference station failure, equipment relocation, changes in the roadway environment, and variations in ranging quality. This invention proposes a dynamic adjustment and self-calibration method for the control network, comprising the following steps: Periodically collect distance observations between reference stations and between reference stations and rover stations; The observation weights are determined based on the observation distance, signal quality, environmental parameters, and historical residuals. The overall adjustment of the control network is performed using the iterative least squares method. Perform statistical tests on the residuals to identify outlier observations; Deweight, remove, or issue alarms for abnormal base stations or abnormal ranging links; When a new reference station is added, the control network coordinates are updated using the incremental adjustment method; When a base station fails, the system automatically switches to a combination of available base stations to maintain the continuity of rover measurements.

[0134] like Figure 6 As shown, after the corrected ranging values ​​are weighted according to the observation quality by the weight determination module 607, they enter the dynamic adjustment and coordinate update module 606 for control network updates and rover positioning. The anomaly removal module 608 performs statistical verification on the residuals, and the control network self-calibration module 609 realizes automatic base station switching based on the integrity score of the reference station to maintain the long-term operational accuracy of the control network.

[0135] Anomalies can be identified according to the following criteria:

[0136] Where, r i For the i-th observation residual, This corresponds to the standard deviation of the observations. When an observation meets the above conditions, it is determined to be an outlier and is either removed or downweighted.

[0137] The integrity evaluation index of the base station can be comprehensively determined by observation success rate, average residual, residual standard deviation, signal quality, and power supply status:

[0138] in, For the integrity score of the i-th base station, S i To observe the success rate metric, R i C is the residual stability index.i E is a communication quality indicator. i These are power status indicators, with w1, w2, w3, and w4 being weighting coefficients.

[0139] A field test was conducted at the 2106 working face of Yingpanhao Coal Mine, and the following measurement results were obtained.

[0140] In the static ranging test, multiple sets of ranging data within the range of 5.7m to 76.3m were selected for analysis. The experimental results show that the overall UWB ranging error under static conditions is relatively small, with a mean error of approximately 0.56cm, a median of approximately 0.82cm, a standard deviation of approximately 1.63cm, a root mean square error of approximately 1.67cm, a maximum positive error of approximately 2.37cm, and a maximum negative error of approximately −2.86cm. All ranging errors are controlled within ±3cm.

[0141] In the dynamic ranging test, the handheld mobile device was used to perform measurements while in motion. The results showed that the average dynamic ranging error was approximately 1.31 cm, the standard deviation was approximately 2.59 cm, the root mean square error was approximately 2.79 cm, the maximum positive error was approximately 4.94 cm, and the maximum negative error was approximately −4.07 cm. All ranging errors were controlled within ±5 cm.

[0142] In the static coordinate positioning test, using the total station measurement results as a reference, the UWB system's planar position measurement error ranged from 1.32cm to 2.24cm. The mean static planar position error was approximately 1.93cm, the standard deviation was approximately 0.35cm, and the root mean square error was approximately 1.97cm. The mean three-dimensional positioning error was approximately 2.70cm, the standard deviation was approximately 0.89cm, and the root mean square error was approximately 2.85cm.

[0143] In the dynamic coordinate positioning test, the average error of the UWB system's planar position measurement was approximately 4.45 cm, with a standard deviation of approximately 1.46 cm and a root mean square error of approximately 4.69 cm; the average error of the three-dimensional dynamic position measurement was approximately 6.56 cm, with a standard deviation of approximately 1.45 cm and a root mean square error of approximately 6.72 cm. These results demonstrate that the system can still meet the requirements for high-precision downhole positioning and footage measurement in dynamic environments.

[0144] In effective distance measurement tests, such as Figure 8 As shown, the handheld mobile terminal can measure a distance of approximately 314.58m between base station 1 and base station 8 via the long-distance ranging link 804. In the section between base stations 6 and 7, the system can automatically switch between available base stations, and the ranging data is smooth without significant jumps. The results indicate that the UWB measurement and control network described in this invention can achieve an effective measurement distance of over 300m in underground roadway environments and possesses long-distance continuous measurement capabilities.

[0145] In the verification of advance measurement, the advance measurement results measured by a manual total station were used as a reference. The advance increment error automatically calculated by the UWB system was within 5cm, which verified the effectiveness of the present invention in automatic advance measurement.

[0146] Alternative solutions The present invention is not limited to the above embodiments. The following alternative or extended solutions may be adopted without departing from the core idea of ​​the present invention.

[0147] UWB chip alternatives: The above embodiments use DW1000-class UWB RF transceiver modules. In practical applications, other UWB chips or modules with high-precision timestamp recording, bidirectional ranging, and low-power communication capabilities can also be used, including domestically produced UWB chips, DW3000 series modules, or UWB RF modules conforming to relevant standards.

[0148] Main controller alternatives: The main control module can be an STM32, GD32, ESP32, ARM Cortex series processor, RISC-V microcontroller, single-board computer, or other control unit with SPI, UART, low-power control and edge computing capabilities.

[0149] Alternative communication methods: Data transmission between the base station and the rover can utilize LoRa, RS485, CAN, Ethernet, Wi-Fi, 5G mining communication, leaky communication systems, or existing underground communication networks in coal mines. Changes in communication methods do not affect the core method of constructing a measurement and control network based on UWB ranging in this invention.

[0150] Alternative solutions for mobile station mounting: In the above embodiments, the mobile station is handheld. In practical applications, it can also be vehicle-mounted, robot-mounted, or mounted on tunneling equipment, as long as the spatial offset relationship between the antenna phase center and the equipment reference point is known.

[0151] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for constructing a measurement and control network for underground roadways in coal mines using UWB technology, characterized in that, Includes the following steps: S1: Deploy multiple UWB reference stations in the underground roadway and establish the initial coordinate values ​​of each UWB reference station; S2: By performing bilateral and bidirectional distance measurement between the UWB rover and the multiple UWB reference stations, the distance observation values ​​between the UWB rover and each visible UWB reference station are obtained. S3: Establish a UWB ranging error compensation model for underground roadways, perform error compensation on the distance observation values, and obtain the corrected ranging values; S4: Based on the corrected ranging value, the three-dimensional coordinates of the UWB rover station are calculated using variational Bayesian adaptive Kalman filtering or weighted least squares method. S5: Based on the three-dimensional coordinate sequence of the UWB rover station and combined with the roadway direction constraint, calculate the incremental tunneling footage along the roadway direction; S6: Perform dynamic adjustment and self-calibration on the measurement control network consisting of the multiple UWB reference stations and the UWB rover stations. The dynamic adjustment and self-calibration includes: performing statistical tests on the observation residuals between reference stations and between reference stations and rover stations to identify abnormal observations, downweighting or removing abnormal observations, and realizing automatic switching of base stations based on the integrity score of the reference stations.

2. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, The multiple UWB reference stations mentioned in step S1 are arranged in an asymmetric spatial gradient manner, with adjacent reference stations being arranged at different heights on both sides of the roadway or on the same side, so as to form an interlaced geometric configuration in the plane and elevation directions.

3. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, The signal flight time for the bilateral two-way ranging in step S2 is calculated using the following formula: Among them, T round1 and T round2 T represents the round-trip measurement time. reply1 and T reply2 T represents the internal response time of the two devices respectively. of Indicates the signal's flight time.

4. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, The ranging error compensation model described in step S3 compensates for at least one of systematic error, multipath error, non-line-of-sight error, and random noise; the corrected ranging value is obtained by subtracting the distance-related error correction term, the systematic bias correction term, and the non-line-of-sight error correction term from the original ranging value.

5. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, The variational Bayesian adaptive Kalman filter described in step S4 uses variational Bayesian inference to adaptively estimate the process noise covariance and observation noise covariance, and introduces prior elevation constraints from the tunnel design slope or traverse point elevation during the filtering process.

6. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, The advance increment mentioned in step S5 is calculated using the following formula: Among them, P o and P t These are the initial and current positions of the UWB mobile station, respectively, where u is the unit direction vector of the tunnel centerline or the tunneling direction.

7. The method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to claim 1, characterized in that, In step S6, abnormal observations are identified based on residuals exceeding a preset threshold. The integrity score of the reference station is determined by weighting at least one of the following: observation success rate, residual stability, communication quality, and power status. The objective function of the control network adjustment is to minimize the weighted sum of squared observation residuals.

8. A method for constructing a measurement and control network for underground coal mine roadways using UWB technology according to any one of claims 1 to 7, characterized in that, The UWB base station is fixedly installed on the side or top of the roadway anchor bolt, and the UWB mobile station can be handheld, vehicle-mounted, robot-mounted, or mounted on tunneling equipment.

9. A UWB measurement and control network system for underground coal mines, characterized in that, include: Multiple UWB fixed reference stations are deployed in underground roadways to provide spatial coordinate references; At least one UWB rover station is used for bilateral two-way ranging with the plurality of UWB fixed reference stations; A background data processing system is used to receive ranging data and execute the steps in the method for constructing a coal mine underground roadway measurement and control network using UWB technology as described in any one of claims 1 to 8.

Images