High-precision intelligent distance measuring and positioning device for heading machine

Abstract

The invention discloses a high-precision intelligent distance measurement positioning device for a heading machine, and belongs to the technical field of heading positioning. An elastic base station network comprises a plurality of groups of miniature base stations arranged on the side wall and the top plate of a roadway; the multi-source sensing module fuses multi-source data through the data acquisition and processing device to realize continuous dynamic positioning in the working process of the heading machine, the target card provides an absolute distance reference, the inertial navigation unit calculates the pose of the heading machine in real time through a built-in MEMS gyroscope and an accelerometer, the geological radar scans the stratum structure in the heading direction in real time, and the real-time positioning of the heading machine is realized. And the laser radar scans a roadway contour by emitting a laser beam to perform point cloud matching with a preset three-dimensional digital map. The probability of tunneling deviation caused by settlement or displacement is reduced by arranging the vertical crossing positioning network and dynamically correcting the positioning reference, the working environment and self state information of the tunneling machine can be comprehensively sensed from different dimensions, and the actual distance and position relation between the tunneling machine and objects around a roadway can be judged in an auxiliary mode.

Description

Technical Field

[0001] The present invention belongs to the technical field of tunneling positioning, and in particular relates to a high-precision intelligent distance measuring and positioning device for a tunneling machine. Background Art

[0002] As a key equipment for tunneling in coal mines and other mines, the integrated boring, anchoring and guarding machine operates in a complex and changeable environment, facing challenges such as high dust, high humidity and limited operating space. Traditional tunneling distance measurement methods rely on manual labor and simple mechanical equipment, and have problems such as low accuracy, low efficiency and safety hazards. Existing tunneling machines usually change the relative position between the base station and the target card due to settlement during operation, making the static model positioning invalid. To solve these problems, an intelligent distance measurement device is urgently needed to improve the accuracy and automation level of tunneling work. Summary of the Invention

[0003] The object of the present invention is to provide a high-precision intelligent distance measurement and positioning device for a roadheader to solve the problems raised in the above background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solutions: a high-precision intelligent ranging and positioning device for a roadheader, comprising a flexible base station network and a multi-source sensing module. The flexible base station network communicates with the multi-source sensing module via LoRa. The flexible base station network is used to build a positioning network covering the tunnel. The flexible base station network includes multiple groups of micro base stations arranged on the side walls and roof of the tunnel.

[0005] The multi-source sensing module uses a data acquisition and processing device to fuse multi-source data to achieve continuous dynamic positioning during the operation of the tunnel boring machine. The multi-source sensing module includes a target card, an inertial navigation unit, a geological radar, an auxiliary antenna and a lidar. The target card provides an absolute distance reference. The inertial navigation unit uses a built-in MEMS gyroscope and accelerometer to calculate the tunnel boring machine posture in real time. The geological radar scans the stratum structure in the tunneling direction in real time. The auxiliary antenna is used to enhance the detection capability of the geological radar. The lidar scans the tunnel contour by emitting a laser beam and performs point cloud matching with a preset three-dimensional digital map to provide absolute position correction.

[0006] In the implementation process, the elastic base station network is composed of multiple groups of micro base stations arranged on the sidewall and roof of the roadway, communicates with the multi-source sensing module through LoRa, constructs a positioning network covering the roadway, the micro base station as a positioning reference point, transmits signals to the multi-source sensing module on the roadheader, provides relative position information, realizes all-around positioning and tracking of the roadheader, in the roadway, each micro base station cooperates with each other to overcome the problems of limited positioning range of single base station and signal easy to be blocked, etc., to provide a macro position framework for the positioning of the roadheader, the target card provides an absolute distance reference based on UWB technology, suppresses multipath interference through frequency hopping spread spectrum technology, dynamically adjusts the power to adapt to the dust environment, the inertial navigation unit is built-in MEMS gyroscope and accelerometer, real-time calculates the pose of the roadheader, fills the positioning vacancy in the signal interruption period, the laser radar can accurately obtain the distance information of the surrounding objects by emitting laser beams and measuring the time of reflected light, and construct a high-precision three-dimensional environment map of the roadway, and combined with the laser emitter arranged on the top wall of the advance support, the working environment and the state information of the roadheader can be comprehensively perceived from different dimensions, which can assist in judging the actual distance and position relationship between the roadheader and the surrounding objects of the roadway, when the elastic base station network has positioning error due to settlement, the laser radar data can be used for calibration and correction, the roadheader pose information monitored by the inertial navigation unit in real time can also help the elastic base station network to more accurately understand the motion state of the roadheader, and after these data are transmitted to the control system of the roadheader, they can be used to realize automatic navigation, speed control, attitude adjustment and other functions, such as according to the positioning information of the laser radar and the target card, the roadheader can automatically plan and maintain the correct driving route in the roadway, and according to the attitude data of the inertial navigation unit, the body of the roadheader is adjusted in real time to ensure its stable operation.

[0007] In a specific embodiment, the laser radar adopts an IP68 dustproof and waterproof shell, the inside of which is filled with nitrogen, a protective cover is sleeved on the outer surface of the laser radar, one end of the protective cover is in an arc structure, the protective cover is made of high-transmittance polycarbonate, and a flow guide groove with a depth of 1-2 mm is arranged on the outer surface of the protective cover, the inclination angle of the flow guide groove is 30°-45°, and the flow guide groove guides the coal dust to slide along the groove by using the natural airflow in the roadway.

[0008] In the implementation process, the protective cover is sleeved in front of the lens of the laser radar, which is convenient for quick disassembly and cleaning, the distance between the protective cover and the lens is 5-10 mm to avoid contact, one end of the protective cover is in an arc structure, and a flow guide groove is designed on the surface of the protective cover, which guides the coal dust to slide along the groove by using the natural airflow in the roadway, such as the wind pressure generated by the movement of the roadheader during tunneling, to reduce the planar deposition and the probability of affecting the detection accuracy of the laser radar due to the accumulation of coal dust, and improve the measurement accuracy.

[0009] In a specific embodiment, the data acquisition and processing device interacts with the intelligent tunneling system of the tunnel boring machine. The data acquisition and processing device can use a fusion algorithm to perform edge computing on the data fused by the multi-source sensor modules, wherein the edge computing includes running an adaptive Kalman filter, sliding window optimization, and building an LSTM compensation model;

[0010] Adaptive Kalman filter: input UWB ranging value, lidar point cloud matching result and inertial navigation unit posture integral, output roadway straight section, sharp bend section and coal dust peak value> 300mg / m 3 Dynamic weights of UWB, LiDAR and INS;

[0011] Sliding window optimization: Fusion of 5 seconds of historical data minimizes pose estimation residuals and eliminates multipath hopping and vibration noise;

[0012] Build an LSTM compensation model: the input includes real-time dust concentration, humidity, and UWB signal strength, and the output is a ranging correction coefficient.

[0013] In the above implementation process, by adaptively adjusting sensor weights, eliminating noise interference, and compensating for environmental influences, multi-source sensor data is effectively integrated, the error of a single sensor is reduced, and the overall positioning accuracy is improved. Sliding window optimization can eliminate multipath hopping and vibration noise, making the positioning results more stable in complex tunnel environments and ensuring the reliability of positioning information during the operation of the tunnel boring machine. The LSTM compensation model and adaptive Kalman filter can dynamically adjust the processing strategy according to environmental changes such as coal dust and humidity, so that the device can work normally under different tunnel working conditions and improve its adaptability to complex and changing environments.

[0014] In a specific implementation scheme, the micro base stations are installed every 15 m along the side walls of the tunnel and additionally deployed every 30 m on the roof to form three-dimensional cross coverage.

[0015] In the above implementation process, micro base stations are installed every 15 meters on the side walls of the tunnel to ensure a denser distribution of signal sources along the longitudinal direction of the tunnel, reduce signal blind spots, and ensure that the tunnel boring machine can always receive stable signals during its movement in the tunnel. Additional base stations are deployed every 30 meters on the top plate to increase the signal coverage dimension in the vertical direction, forming a three-dimensional cross-coverage with the side wall base stations, avoiding signal obstruction or poor reception due to changes in the tunnel shape, tunnel boring machine position and posture, thereby providing the tunnel boring machine with a stable and reliable positioning signal.

[0016] In a specific embodiment, the laser radar is fixed on the top of the tunneling machine by a mounting bracket at an angle of 20°-30°, the target card and the inertial navigation unit are both mounted on the center of gravity of the tunneling machine through silica gel damping base, and the inertial navigation unit is located between the target card and the laser radar, the geological radar is installed 1-2 m behind the cutter head of the tunneling machine, and the antenna is directed towards the tunneling face, and the auxiliary antenna is arranged on both sides of the tunneling machine close to the body.

[0017] In the above implementation process, the laser radar can accurately capture the geometric shape of the ground surface such as the profile of the tunnel and the position of the tunneling equipment, obtain the point cloud data of the tunnel, and complete the construction of the three-dimensional environment of the tunnel, and the geological radar can detect the underground rock structure, aquifer or cavity, etc., and combine the ground surface point cloud with the underground profile to construct the tunnel surface-underground integrated model, when the laser radar detects a surface crack, the geological radar is triggered to perform a deep scan below the crack to determine whether there is a collapse risk, if there is a collapse risk, a warning is automatically generated, and a detour path is planned to improve the safety of tunneling. The surface features such as rock cracks and terrain undulations detected by the laser radar are used to calibrate the electromagnetic wave propagation path of the geological radar, or the underground cavity or vein information detected by the geological radar is used to correct the errors caused by surface shielding in the modeling of the laser radar, so that deviations in the tunneling process can be discovered and corrected in time, and the accuracy and efficiency of the tunneling operation are improved.

[0018] In a specific embodiment, the laser radar is a 16-line scan, the wavelength is 1550 nm, the vertical field of view is ±15°, and the point cloud density is >100 points / m 2 .

[0019] In the above implementation process, the 16-line scan can obtain more distance information in the vertical direction, forming relatively dense point cloud data, the 1550 nm wavelength is relatively less susceptible to interference during transmission, which can ensure the stability of signal transmission, the vertical field of view of ±15° combined with the inclined installation angle further expands the perception range of the tunnel space, and the high point cloud density makes the generated tunnel profile information more accurate, which is beneficial to subsequent matching with the preset map and construction of the full-dimensional tunnel map.

[0020] In a specific embodiment, the geological radar and the laser radar are combined to output a full-dimensional tunnel map containing a three-dimensional surface model, an underground structure profile and a risk annotation.

[0021] In the above implementation process, the full-dimensional tunnel map provides an accurate reference for the path planning and navigation of the tunneling machine, and in combination with the positioning system of the tunneling machine, the position of the tunneling machine can be accurately determined on the map, and the optimal tunneling route can be planned according to the map information to avoid collision with obstacles and entering dangerous areas, thereby improving the efficiency and quality of tunneling, and also helping to realize the automation and intelligentization of tunneling operations.

[0022] In a specific embodiment, the dynamic weight distribution is as follows:

[0023] Straight line segment, coal dust <100mg / m 3 : The weights of UWB, LiDAR and Inertial Navigation Unit are 70%, 20% and 10% respectively;

[0024] Sharp bend section, curvature radius <10m: the weights of UWB, lidar and inertial navigation unit are 30%, 60% and 10% respectively;

[0025] Coal dust peak > 300 mg / m 3 : The weights of UWB, lidar and inertial navigation unit are 85%, 5% and 10% respectively.

[0026] In the above implementation, UWB is weighted 70% for straight sections with low coal dust concentrations, as it offers high ranging accuracy in open, low-interference straight sections, providing accurate distance information. LiDAR is weighted 20% to assist in correcting and verifying position information and monitoring the roadway contour. The Inertial Navigation Unit (INU) is weighted 10% to primarily maintain attitude and assist in positioning in straight sections, but its weight is relatively low due to its cumulative error characteristics. LiDAR is weighted 60% for sharp bends, where it can accurately sense the position and orientation of the TBM relative to the roadway by scanning the roadway contour. UWB is weighted 30%, but its weight is reduced due to signal obstruction caused by the roadway's curvature. The INU is weighted 10% to assist in monitoring the TBM's attitude changes during turns. When coal dust peaks are high, UWB is weighted 85% to minimize the impact of UWB signals in high dust environments. LiDAR is weighted 5%, as high dust levels significantly impact LiDAR scanning accuracy. The INU is weighted 10% to continuously monitor attitude.

[0027] In a specific implementation scheme, the structure of the LSTM environment compensation model is a 2-layer LSTM network.

[0028] In the above implementation process, the first layer first performs preliminary feature extraction and processing on input data such as real-time dust concentration, humidity, and UWB signal strength, capturing the basic patterns and trends in the data. The second layer, the LSTM network, builds on the first layer to further explore more complex and in-depth feature relationships. For example, the first layer can identify the simple change trend of dust concentration over time. The second layer can combine other factors such as humidity to analyze the comprehensive impact of dust concentration on UWB ranging under different humidity conditions, thereby more accurately establishing the connection between environmental parameters and ranging errors. This enables the model to more effectively learn and utilize historical environmental data information and accurately compensate for the impact of environmental factors on positioning.

[0029] In a specific embodiment, the internal part of each group of the micro base station is equipped with a MEMS tilt sensor and a strain gauge, the target card includes a UWB module and a directional antenna, wherein the UWB module suppresses multipath interference through frequency hopping spread spectrum technology, and the directional antenna is used to enhance the directivity and anti-multipath ability of the signal, the dynamic power adjustment range of the UWB module is -30dBm to -10dBm, the power adjustment step is 0.5dBm, and the power is increased by 0.5dBm every time the coal dust concentration increases by 100mg / m 3 .

[0030] In the above implementation process, the MEMS tilt sensor equipped in the internal part of the micro base station can monitor the inclination angle change of the roadway in real time, and the strain gauge can sense the deformation stress condition of the roadway, through these sensors, the deformation of the roadway caused by settlement, geological movement, etc. can be found in time, data support is provided for the safety evaluation and maintenance of the roadway, and the stability of the positioning system is ensured, the micro base station periodically collects its own attitude and settlement data, exchanges position information with adjacent base stations through a cooperative positioning algorithm, constructs a dynamic reference system, and if the settlement is greater than 3mm or the inclination is greater than 0.5°, a correction instruction is automatically sent to the central controller, and if the deformation of the roadway causes the position or attitude of the base station to change, the positioning model can be adjusted according to the monitoring data, and the accuracy of the positioning of the roadheader is ensured not to be affected by the deformation of the roadway.

[0031] The UWB module in the target card uses frequency hopping spread spectrum technology to transmit signals on multiple frequencies, effectively suppresses multipath interference, and the directional antenna enhances the directivity and anti-multipath ability of the signal, so that the signal is more concentratedly transmitted and received, reduces the interference caused by signal scattering and reflection, further improves the positioning accuracy, and at the same time, the UWB module has a dynamic power adjustment function, in the high-concentration coal dust environment of the roadway, the automatic power increase can enhance the signal strength, ensure that the signal can effectively penetrate the coal dust, maintain stable communication and ranging, and ensure that the target card works normally in harsh environments.

[0032] Compared with the prior art, the beneficial effects of the present application are:

[0033] 1. The application reduces the probability of deviation caused by settlement or displacement by laying out a three-dimensional intersection positioning network and dynamically correcting the positioning reference. The laser radar scans the tunnel profile by emitting a laser beam and matches the point cloud with the preset three-dimensional digital map to accurately obtain the point cloud data of the tunnel. It can comprehensively perceive the tunneling machine working environment and its own state information from different dimensions, and can assist in judging the actual distance and position relationship between the tunneling machine and the surrounding objects. When the elastic base station network has positioning error due to settlement, the laser radar data can be used for calibration and correction. The tunneling machine attitude information monitored by the inertial navigation unit in real time can also help the elastic base station network to more accurately understand the motion state of the tunneling machine. After these data are transmitted to the control system of the tunneling machine, the tunneling machine can automatically plan and maintain the correct driving route in the tunnel, timely discover and correct the deviation in the tunneling process, and improve the accuracy and efficiency of the tunneling operation.

[0034] 2. The application sets a protective cover in front of the lens of the laser radar, which is not only convenient to disassemble and clean, but also can guide the coal dust to slide along the groove by using the wind pressure generated by the movement of the tunneling machine during tunneling, reduce the deposition on the plane, and reduce the deposition of coal ash on the lens of the laser radar without using air flow sweeping. This can not only reduce the probability of affecting the detection accuracy of the laser radar due to coal dust accumulation, but also reduce the subsequent maintenance cost.

[0035] 3. The application switches the weight mode according to the real-time scene, such as the increase of the curvature radius of the tunnel or the increase of the dust concentration, to reduce the influence of the narrow and curved tunnel environment on the positioning accuracy, reduce signal superposition and interference, and improve the tunneling accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 It is a schematic diagram of the layout position of the application;

[0037] Figure 2 It is a schematic diagram of the cooperation of the application and the laser emitter;

[0038] Figure 3 It is a schematic diagram of the plane structure of the laser radar of the application;

[0039] Figure 4 It is a schematic diagram of the plane assembly structure of the laser radar and the protective cover of the application.

[0040] In the figure: 1, micro base station; 2, target card; 3, inertial navigation unit; 4, geological radar; 5, auxiliary antenna; 6, laser radar; 7, protective cover. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] See also Figures 1-4 The present invention provides a high-precision intelligent ranging and positioning device for a tunnel boring machine, including a flexible base station network and a multi-source sensing module. The flexible base station network communicates with the multi-source sensing module through LoRa. The flexible base station network is used to build a positioning network covering the tunnel to overcome the problems of limited positioning range of a single base station and easy signal obstruction, so as to realize all-round and uninterrupted positioning and tracking of the tunnel boring machine. The flexible base station network includes multiple groups of micro base stations 1 arranged on the side walls and top plates of the tunnel. The multi-source sensing module integrates multi-source data through a data acquisition and processing device to realize continuous dynamic positioning of the tunnel boring machine during operation. The multi-source sensing module includes a target card 2, an inertial navigation unit 3, a geological radar 4, an auxiliary antenna 5 and a laser radar 6. The target card 2 provides an absolute distance reference, and the inertial navigation unit 3 uses a built-in MEMS gyroscope and an accelerometer to calculate the tunnel boring machine position in real time. The geological radar 4 scans the stratum structure in the direction of excavation in real time, and the auxiliary antenna 5 is used to enhance the detection capability of the geological radar 4. The lidar 6 scans the tunnel contour by emitting a laser beam and matches the point cloud with the preset three-dimensional digital map to provide absolute position correction. The micro base station 1 is installed every 15m along the side wall of the tunnel and supplemented every 30m on the roof to form a three-dimensional cross coverage. Multiple groups of micro base stations 1 are equipped with MEMS tilt sensors and strain gauges. The target card 2 includes a UWB module and a directional antenna. The UWB module suppresses multipath interference through frequency hopping spread spectrum technology. The directional antenna is used to enhance the signal's directionality and anti-multipath capability. The dynamic power adjustment range of the UWB module is -30dBm to -10dBm, with a power adjustment step of 0.5dBm. For every 100mg / m³ increase in coal dust concentration, the power is increased by 0.5dBm.

[0043] Furthermore, multiple groups of micro base stations 1 are deployed on the side walls and roof of the tunnel to form three-dimensional cross-coverage, and a positioning network covering the tunnel is constructed through LoRa communication. This overcomes the problem of limited signal range and easy obstruction of a single base station. Each micro base station 1 supports collaborative calibration and dynamically corrects the positioning reference, reducing the probability of tunneling deviation due to settlement or displacement. In addition, the relatively dense base station layout uses principles such as triangulation positioning to more accurately calculate the position of the tunnel boring machine in the tunnel. For example, when the tunnel boring machine is located between two side wall base stations, it can be accurately located by measuring the distance using the signals of the two micro base stations 1. The micro base station 1 on the roof can also assist in providing vertical position information, further improving positioning accuracy and meeting the high-precision positioning requirements of the tunnel boring machine. At the same time, when some micro base stations 1 experience signal anomalies due to tunnel deformation or settlement, other base stations can still operate normally, ensuring the reliability of the positioning system. For example, if a micro base station 1 on a side wall is displaced due to tunnel settlement, the adjacent micro base stations 1 on the side wall and the micro base station 1 on the roof can continue to provide effective signals to maintain the positioning function for the tunnel boring machine and reduce the impact of environmental changes on positioning.

[0044] The target card 2 provides an absolute distance reference. By exchanging signals with the micro base station 1 in the flexible base station network, it uses technologies such as UWB to accurately measure the distance to the micro base station 1, providing key distance information for the positioning of the roadheader. The inertial navigation unit 3, with its built-in MEMS gyroscope and accelerometer, monitors the acceleration and angular velocity of the roadheader in real time. It infers the roadheader's position in real time through integration operations and other methods. It can independently provide information on the roadheader's motion status in a short period of time, playing a particularly important role in the event of temporary signal loss or interference. The geological radar 4 scans the stratum structure in the direction of excavation in real time, detecting the lithology and geological structure of the stratum ahead, providing early warning of potential geological risks, and assisting the roadheader in adjusting excavation parameters. Its data also provides certain environmental reference information for positioning. The auxiliary antenna 5 is used to enhance the detection capability of the geological radar 4, enabling it to obtain more accurate and comprehensive stratum information. The lidar 6 scans the roadway contour by emitting a laser beam, matching the point cloud with a preset three-dimensional digital map, and accurately obtains the roadway's point cloud data, providing absolute position correction for the roadheader and improving positioning accuracy.

[0045] The laser radar 6 adopts an IP68 dustproof and waterproof shell, which is filled with nitrogen. The outer surface of the laser radar 6 is covered with a protective cover 7, and one end of the protective cover 7 is an arc-shaped structure. The protective cover 7 is made of high-transmittance polycarbonate. The outer surface of the protective cover 7 is provided with a guide groove with a depth of 1-2mm. The inclination angle of the guide groove is 30°~45°, and the guide groove uses the natural airflow in the tunnel to guide the coal dust to slide along the groove.

[0046] Furthermore, the IP68 dustproof and waterproof casing can prevent dust from entering the interior of the laser radar 6, avoiding damage to internal components caused by coal dust accumulation, affecting optical path transmission and measurement accuracy, and can also protect internal circuits and optical components from damage for a long time in a humid tunnel environment or when encountering water, ensuring the normal operation of the equipment. The internal nitrogen filling can prevent the internal metal parts of the laser radar 6 from contacting with oxygen, inhibiting oxidation reactions, preventing rust and corrosion, and extending the service life of the equipment. At the same time, it can also inhibit condensation of water vapor inside the laser radar 6 when the humidity in the tunnel is high, and avoid lens fogging. The protective cover 7 made of high-transmittance polycarbonate has good light transmittance and has little effect on the transmission and reception of laser signals by the laser radar 6. At the same time, it can prevent coal dust from directly contacting the lens. The design of the guide groove utilizes the natural airflow in the tunnel to guide coal dust to slide along the groove, reducing the adhesion amount by more than 70%, reducing the probability of coal dust adhering to the surface of the protective cover 7, and thereby reducing the obstruction and interference of the laser signal, ensuring measurement accuracy, and at the same time reducing the frequency of manual cleaning, saving maintenance time and cost, and improving the operating efficiency and continuity of the tunnel boring machine.

[0047] The data acquisition and processing device interacts with the roadheader's intelligent tunneling system. It uses a fusion algorithm to perform edge computing on the data fused by the multi-source sensor modules. This edge computing includes running an adaptive Kalman filter, sliding window optimization, and building an LSTM compensation model.

[0048] Adaptive Kalman filter: input UWB ranging value, LiDAR 6 point cloud matching results and inertial navigation unit 3 posture integration, output roadway straight section, sharp bend section and coal dust peak value> 300mg / m 3 Dynamic weights of UWB, LiDAR 6 and Inertial Navigation Unit 3;

[0049] Sliding window optimization: Fusion of 5 seconds of historical data minimizes pose estimation residuals and eliminates multipath hopping and vibration noise;

[0050] Build an LSTM compensation model: the input includes real-time dust concentration, humidity, and UWB signal strength, and the output is the ranging correction coefficient. The dynamic weight distribution is as follows:

[0051] Straight line segment, coal dust <100mg / m 3 : The weights of UWB, LiDAR 6, and Inertial Navigation Unit 3 are 70%, 20%, and 10%, respectively;

[0052] Sharp bend, with a curvature radius of less than 10 m: the weights of UWB, LiDAR 6, and Inertial Navigation Unit 3 are 30%, 60%, and 10%, respectively;

[0053] Coal dust peak > 300 mg / m 3: The weights of UWB, lidar 6 and inertial navigation unit 3 are 85%, 5% and 10% respectively, and the structure of the LSTM environmental compensation model is a 2-layer LSTM network.

[0054] Furthermore, the data acquisition and processing device obtains multi-source data such as UWB ranging values, point cloud matching results, and pose integrals from the multi-source sensor module, and uses a fusion algorithm to perform edge computing, that is, processing the data directly on a device close to the data source, eliminating the need to transmit large amounts of data to a remote server. This reduces data transmission delays and network bandwidth usage, and can not only improve the convenience of ranging, but also improve ranging efficiency.

[0055] The adaptive Kalman filter takes in multi-source sensor data based on different roadway conditions, such as straight sections, sharp bends, and high-dust environments. Leveraging the adaptive nature of the Kalman filter algorithm, it dynamically adjusts the weights of the UWB, LiDAR 6, and Inertial Navigation Unit 3 in positioning. In straight sections with low dust concentrations, the UWB is more accurate and therefore given a higher weight. In sharp bends, the LiDAR 6 perceives the environmental contours more accurately, so its weight is increased. In high-dust environments, the UWB is less affected and therefore has an increased weight, thereby improving positioning accuracy in various scenarios. The sliding window optimization uses 5 seconds of historical data as a sliding window. Through calculation and optimization, it minimizes the pose estimation residual. This process identifies and eliminates jumps caused by multipath effects and noise generated by equipment vibration, resulting in smoother and more stable positioning results. The LSTM compensation model takes as input environmental parameters such as real-time dust concentration, humidity, and UWB signal strength. A two-layer LSTM network is used to construct the model. The model learns the relationship between environmental parameters and ranging error and outputs ranging correction coefficients to dynamically compensate for UWB ranging data and other data, reducing the impact of environmental factors on positioning.

[0056] The laser radar 6 is fixed on the top of the tunnel boring machine at an angle of 20° to 30° through a mounting bracket. The target card 2 and the inertial navigation unit 3 are both mounted at the center of gravity of the tunnel boring machine through a silicone damping base for vibration reduction, and the inertial navigation unit 3 is located between the target card 2 and the laser radar 6. The geological radar 4 is installed 1 to 2 meters behind the cutter head of the tunnel boring machine, with the antenna facing the tunneling surface. The auxiliary antenna 5 is deployed on both sides of the tunnel boring machine near the fuselage. The laser radar 6 has a 16-line scan, a wavelength of 1550nm, a vertical field of view of ±15°, and a point cloud density of >100 points / m 2 The geological radar 4 is combined with the lidar 6 to output a full-dimensional tunnel map including a three-dimensional surface model, underground structure profile and risk annotations.

[0057] Furthermore, the geological radar 4 and the lidar 6 combine to produce a full-dimensional roadway map, integrating both above-ground and underground information. The geological radar 4 provides underground structural profile information, such as the stratification and geological structure characteristics of the strata. The lidar 6 acquires a three-dimensional surface model, including the shape, size, and internal obstacle distribution of the roadway. The fusion of the two creates a comprehensive map of the roadway, providing operators with a comprehensive and intuitive view of the roadway and helping them better understand the excavation environment.

[0058] Risk marking on the full-dimensional tunnel map clearly identifies potential geological risks detected by the geological radar 4, such as faults and water-bearing areas, and dangerous conditions in the tunnel identified by the lidar 6, such as over-break, under-break areas, and obstacles. Operators can take measures in advance based on these markings, such as adjusting the excavation speed, changing the support method, or performing geological treatment, effectively reducing risks in the excavation operation and ensuring construction safety.

[0059] The full-dimensional tunnel map provides an accurate environmental reference for the positioning and navigation of the roadheader. Combined with the positioning information of the target card 2 and the inertial navigation unit 3, the roadheader can accurately determine its own position on the map and plan the optimal excavation path based on the map. The accuracy and comprehensiveness of the map help to realize the automated and intelligent navigation of the roadheader, improving the efficiency and quality of excavation.

[0060] The working principle and usage process of the present invention are as follows: First, multiple groups of micro base stations 1 are deployed on the side walls and roof of the tunnel. The micro base stations 1 are installed every 15 meters along the side walls of the tunnel and additionally deployed every 30 meters on the roof to form a three-dimensional cross coverage. Then, the parameters of each sensor are initialized, and the MEMS tilt sensor is calibrated to a horizontal error of less than 0.1°. The micro base stations 1 periodically collect their own posture and settlement data, and exchange position information with adjacent base stations through a collaborative positioning algorithm to establish a dynamic reference system. If settlement > 3mm or tilt > 0.5° is detected, correction instructions are automatically sent to the central controller.

[0061] The UWB module in the target card 2 then periodically sends frequency-hopping signals to communicate with the deployed micro base station 1. By calculating the flight time of the signal from the UWB module to the micro base station 1, and combining it with the speed of light, the distance between the target card 2 and the micro base station 1 is accurately calculated, thereby determining the position of the roadheader. The target card then interacts with the laser transmitter installed on the top wall of the advance support in the tunnel to identify and determine the posture of the roadheader and its position in the tunnel.

[0062] The laser radar 6 scans the tunnel contour every 2 seconds to generate three-dimensional point cloud data. During the process of the laser radar 6 scanning the tunnel contour, the auxiliary antenna 5 transmits or receives electromagnetic waves at multiple angles to generate an underground dielectric constant distribution map to identify abnormal structures. Then, the data detected by the geological radar 4 and the laser radar 6 are unified into the same time-space coordinate system through the timestamp and posture data of the inertial navigation unit 3. The surface features output by the laser radar 6 and the underground features output by the geological radar 4 are extracted to construct a joint feature vector. The full-dimensional map containing the surface three-dimensional model, underground structure profile and risk annotation is output, and matched with the preset map through the ICP algorithm to correct the absolute position. When the laser radar 6 detects signs of collapse on the surface, it triggers the geological radar 4 to perform a high-density scan of the collapsed area to verify the stability of the underground structure. The laser radar 6 can accurately capture the surface geometry. Such as the tunnel contour and the position of the tunneling equipment, etc., to obtain the tunnel point cloud data and complete the construction of the tunnel three-dimensional environment. The geological radar 4 can detect the underground rock structure, aquifers or cavities, etc., and combine the surface point cloud with the underground profile to construct the tunnel surface-underground integrated model. When the laser radar 6 detects a surface crack, it triggers the geological radar 4 to perform a deep scan below the crack to determine whether there is a collapse risk. If there is a collapse risk, an early warning is automatically generated and an avoidance path is planned to improve the tunneling safety. The electromagnetic wave propagation path of the geological radar 4 is calibrated by the surface features detected by the laser radar 6, such as rock cracks and terrain undulations, or the underground cavities or mineral veins detected by the geological radar 4 are used to correct the errors caused by surface obstruction in the modeling of the laser radar 6. Deviations in the tunneling process can be discovered and corrected in a timely manner, thereby improving the accuracy and efficiency of the tunneling operation.

[0063] The inertial navigation unit 3 collects the angular velocity and acceleration data of the roadheader in real time, calculates its position and posture through kinematic model integration, and inputs the detected data into the data acquisition and processing device for edge computing together with the data collected by the UWB module and the lidar 6. The output result is then linked with the DAS deployed in the tunnel to monitor the strain of the entire tunnel, adjust the positioning algorithm parameters in advance, dynamically adjust the base station coordinates, and transmit the feedback corrected position and coordinate information to the roadheader control system. According to the preset tunneling route and accuracy requirements, the roadheader's travel direction and cutting parameters are adjusted in real time to achieve precise tunneling and continuous high-precision positioning;

[0064] At the same time, during the excavation process, the guide trough uses the wind pressure generated by the movement of the tunnel boring machine to guide the coal dust to slide along the trough, reducing the probability of coal dust adhering to the surface of the protective cover 7, reducing the obstruction and interference to the laser signal, and ensuring measurement accuracy.

[0065] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A high-precision intelligent ranging and positioning device for a roadheader, comprising a flexible base station network and a multi-source sensing module, characterized in that: The elastic base station network communicates with the multi-source sensor module via LoRa, and the elastic base station network is used to build a positioning network covering the lane, and the elastic base station network includes multiple groups of micro base stations (1) arranged on the side walls and roof of the lane; The multi-source sensing module realizes continuous dynamic positioning of the tunnel boring machine during operation by fusing multi-source data through a data acquisition and processing device. The multi-source sensing module comprises a target card (2), an inertial navigation unit (3), a geological radar (4), an auxiliary antenna (5) and a laser radar (6), wherein the target card (2) provides an absolute distance reference, the inertial navigation unit (3) calculates the tunnel boring machine posture in real time through a built-in MEMS gyroscope and an accelerometer, the geological radar (4) scans the stratum structure in the tunneling direction in real time, the auxiliary antenna (5) is used to enhance the detection capability of the geological radar (4), and the laser radar (6) scans the tunnel contour by emitting a laser beam and performs point cloud matching with a preset three-dimensional digital map to provide absolute position correction.

2. A high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The laser radar (6) adopts an IP68 dustproof and waterproof shell, and is filled with nitrogen. The outer surface of the laser radar (6) is sleeved with a protective cover (7), and one end of the protective cover (7) is in an arc-shaped structure. The protective cover (7) is made of high-transmittance polycarbonate. The outer surface of the protective cover (7) is provided with a guide groove with a depth of 1-2 mm, and the guide groove has an inclination angle of 30° to 45°. The guide groove uses the natural airflow in the tunnel to guide the coal dust to slide along the groove.

3. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The data acquisition and processing device interacts with the intelligent tunneling system of the tunnel boring machine. The data acquisition and processing device can use a fusion algorithm to perform edge computing on the data fused by the multi-source sensor modules, where the edge computing includes running an adaptive Kalman filter, sliding window optimization, and building an LSTM compensation model. Adaptive Kalman filter: input UWB ranging value, laser radar (6) point cloud matching result and inertial navigation unit (3) posture integral, output roadway straight section, sharp bend section and coal dust peak value> 300mg / m 3 Dynamic weights of UWB, LiDAR (6) and Inertial Navigation Unit (3); Sliding window optimization: Fusion of 5 seconds of historical data minimizes pose estimation residuals and eliminates multipath hopping and vibration noise; Build an LSTM compensation model: the input includes real-time dust concentration, humidity, and UWB signal strength, and the output is a ranging correction coefficient.

4. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The micro base stations (1) are installed every 15 m along the side walls of the lanes and deployed every 30 m on the roof, forming a three-dimensional cross coverage.

5. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The laser radar (6) is fixed on the top of the tunnel boring machine at an angle of 20° to 30° via a mounting bracket. The target card (2) and the inertial navigation unit (3) are both mounted at the center of gravity of the tunnel boring machine via a silicone damping base for vibration reduction, and the inertial navigation unit (3) is located between the target card (2) and the laser radar (6). The geological radar (4) is mounted 1 to 2 meters behind the cutter head of the tunnel boring machine, with the antenna facing the tunneling surface. The auxiliary antenna (5) is deployed on both sides of the tunnel boring machine near the fuselage.

6. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The laser radar (6) has 16-line scanning, a wavelength of 1550nm, a vertical field of view of ±15°, and a point cloud density of >100 points / m 2 .

7. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The geological radar (4) is combined with the laser radar (6) to output a full-dimensional tunnel map including a three-dimensional surface model, an underground structure profile and risk annotations.

8. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 3, characterized in that: The dynamic weight distribution is as follows: Straight line segment, coal dust <100mg / m 3 : The weights of UWB, LiDAR (6) and Inertial Navigation Unit (3) are 70%, 20% and 10% respectively; Sharp bend section, curvature radius <10m: the weights of UWB, LiDAR (6) and inertial navigation unit (3) are 30%, 60% and 10% respectively; Coal dust peak > 300 mg / m 3 : The weights of UWB, LiDAR (6) and Inertial Navigation Unit (3) are 85%, 5% and 10% respectively.

9. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 3, characterized in that: The structure of the LSTM environmental compensation model is a 2-layer LSTM network.

10. The high-precision intelligent distance measurement and positioning device for a roadheader according to claim 1, characterized in that: The plurality of micro base stations (1) are equipped with MEMS tilt sensors and strain gauges. The target card (2) includes a UWB module and a directional antenna. The UWB module suppresses multipath interference through frequency hopping spread spectrum technology. The directional antenna is used to enhance the directionality and anti-multipath capability of the signal. The dynamic power adjustment range of the UWB module is from -30dBm to -10dBm, with a power adjustment step of 0.5dBm. The power adjustment step is ... 3 , power increased by 0.5dBm.

Images