A kind of laser radar-based heading machine tunnel relative positioning method and system

Abstract

The application discloses a kind of based on laser radar's heading machine roadway relative positioning method and system, belong to coal mine heading machine positioning technical field, comprising: through the laser radar of heading machine carrying initial stationary state roadway three-dimensional point cloud acquisition, extract roof, left and right effective point cloud and fit virtual plane;Based on the position of radar in body coordinate system obtains the normal distance from center to each plane, combined with roadway design parameter to establish plane constraint. Utilize the initial pose parameter of inertial measurement unit and normal distance to determine the fixed coordinate of roadway in absolute coordinate system. Real-time acquisition of radar center to each virtual plane normal distance and normal vector angle under motion state, substitute into fixed coordinate and plane constraint equation, solve the spatial coordinates and pose of heading machine in absolute coordinate system. The method does not need auxiliary sign, and has strong anti-interference, can realize continuous stable real-time relative positioning.

Description

Technical Field

[0001] This invention belongs to the field of positioning technology for underground tunneling equipment in coal mines, specifically relating to a method for relative positioning of tunneling machine roadways based on lidar point clouds. Background Technology

[0002] Currently, there are three main types of positioning technologies for underground tunneling machines: inertial navigation positioning technology, which mainly uses a three-axis accelerometer and gyroscope to calculate the position of the tunneling machine body; however, long-term operation can cause zero-point drift of the gyroscope and accelerometer, and with the accumulation of noise, errors can accumulate. Long-term operation requires manual calibration after shutdown, affecting production efficiency and failing to meet the needs of continuous, long-distance operations. Optical positioning technology primarily involves deploying total stations and backsight prisms at roadway benchmarks, then installing prisms on the machine body. The total station uses laser and angle measurements to collect the position coordinates of multiple prisms to obtain the machine body coordinates. Its drawback is the dusty and poor lighting conditions in the underground environment, which compromises measurement accuracy. Ultra-wideband (UWB) wireless positioning technology uses UWB base stations as positioning benchmarks. UWB tags moving with the machine body, based on the TOF / TDOA ranging principle, communicate with the base station via ultra-wideband pulse bidirectional communication. Combining the coordinates of multiple base stations and ranging data, triangulation is used to calculate the tag's three-dimensional coordinates. Finally, the data processing terminal calculates the machine body's posture parameters based on the fixed pose of the tag and through spatial geometric transformation, transmitting the data in real-time to the main control system. However, the numerous electromechanical devices involved in underground roadway operations can interfere with the pulses, leading to distorted positioning results and significant errors. In summary, existing positioning solutions generally suffer from error accumulation, susceptibility to environmental interference, reliance on external markers, the need for manual maintenance, and the inability to operate stably for extended periods. Summary of the Invention

[0003] To address the problems of error accumulation, susceptibility to environmental interference, reliance on external markers, need for manual calibration, and inability to operate continuously and stably in existing underground tunneling machine (TBM) positioning technologies, this invention provides a TBM roadway relative positioning method based on lidar. This method aims to achieve high-precision, real-time, and robust positioning of the TBM relative to the roadway under harsh underground working conditions, providing reliable position and posture perception support for automated cutting, autonomous guiding, and unmanned operation of the TBM.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A method for relative positioning of a tunnel boring machine in a roadway based on lidar, comprising: The LiDAR mounted on the tunneling machine collects three-dimensional point cloud data of the tunnel in its initial static state, extracts the effective point cloud of the tunnel roof, left side and right side, and performs fitting of each virtual plane of the tunnel. Based on the position of the lidar in the fuselage coordinate system, the normal distance from the lidar center to each virtual plane is obtained, and the planar constraint conditions are constructed in combination with the design geometric parameters of the tunnel. Based on the pose parameters of the tunneling machine in its initial state obtained by the inertial measurement unit mounted on the tunneling machine, combined with the normal distance from the center of the lidar to each virtual plane, the fixed coordinates of the roadway in the absolute coordinate system are determined. The normal distance from the center of the lidar to each virtual plane in the tunnel and the angle between the plane normal vectors are obtained under the motion state. Based on the fixed coordinates of the tunnel in the absolute coordinate system, the real-time spatial coordinates and pose data of the tunneling machine in the absolute coordinate system are obtained through planar constraints.

[0005] Preferably, the extraction of effective point clouds from the tunnel roof, left flank, and right flank includes the following steps: The tunnel environment is scanned in three dimensions at a set frequency to obtain real-time raw three-dimensional point cloud data including the tunnel roof, left side, right side, floor and tunneling face area; Extract the 3D coordinates, reflection intensity, laser line number, and hardware timestamp of each point from the original 3D point cloud data, and filter out out-of-limit points and erroneous points by conditional filtering. Based on the calibration and installation parameters of the lidar at the top of the central axis of the main fuselage platform, the three-dimensional self-occlusion area is determined, including the lower fuselage platform and the front cutting arm. The three-dimensional point cloud data in this area is filtered out to obtain the filtered point cloud data.

[0006] Preferably, the fitting of the virtual plane includes the following steps: Using the 3D ROI positioning method, with the LiDAR as the coordinate origin, the tunnel area is located. The 3D point cloud data of non-tunnel areas is removed, and the semantic point clouds of the tunnel roof, left side and right side are separated. The virtual plane is then fitted by the least squares method.

[0007] Preferably, the step of obtaining the normal distance from the center of the lidar to each virtual plane based on the lidar position in the fuselage coordinate system, and constructing planar constraint conditions in conjunction with the design geometric parameters of the tunnel, includes the following steps: Determine the installation data of the lidar and the coordinate system rotation matrices for the lidar coordinate system, fuselage coordinate system, and absolute coordinate system; Based on the coordinate system rotation matrix, the coordinates of the lidar center in the fuselage coordinate system are obtained. Based on the virtual planes of the tunnel roof, left side and right side, the vertical distance from the lidar center to the virtual plane is determined. The initial absolute attitude of the tunneling machine is obtained by inertial measurement unit and the normal distance from the radar to the roadway roof, left side and right side obtained by lidar scanning. The initial spatial attitude of the tunneling machine in absolute coordinate system is calculated and the fixed coordinates of the roadway in absolute coordinate system are determined. Based on the fixed coordinates of the tunnel in the absolute coordinate system, planar constraints are constructed as follows: Under the constraints of the top plate plane, the equation of the top plate plane in the absolute coordinate system is: Combined with the distance of the radar to the top plate ,have to: ; in, Design height for alleyways, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the top plate plane Quantity; On the left-side plane constraint, the equation of the left-side plane in the absolute coordinate system is determined as follows: Combined with the distance of the radar to the left side ,have to: ; in, Design the width of the alleyway, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the left side plane Quantity; On the left side plane constraint, determine the equation of the right side plane in the absolute coordinate system as follows: Combined with the distance of the radar to the right side ,have to: ; in, The normal vector of the right side plane Quantity.

[0008] Preferably, the acquisition of the real-time spatial coordinates includes the following steps: Obtain the normal distances from the center of the lidar to each virtual plane in the tunnel as measured by the lidar under motion conditions, and calculate the three-dimensional position of the fuselage in the tunnel using planar constraints: Wherein, the Z-coordinate is: ; coordinate: Joint left and right side planar constraints: ; X-coordinate: calculated recursively from the planar constraints of the point cloud in consecutive frames from the lidar.

[0009] Preferably, the acquisition of the pose data includes the following steps: Real-time acquisition of the normal vectors of the laser radar to each virtual plane; The yaw angle is calculated based on the difference in the Y component of the left and right side normal vectors, the pitch angle is calculated based on the Z component of the top plate normal vector, and the roll angle is calculated based on the difference in the X component of the left and right side normal vectors. The position and orientation of the tunnel boring machine in the roadway are determined by the yaw angle, pitch angle, and roll angle.

[0010] The present invention also provides a relative positioning system for tunnel boring machines based on lidar, comprising: The lidar unit is used to collect three-dimensional point cloud data of the tunnel. An inertial measurement unit is used to acquire the orientation parameters of the tunneling machine in its initial state. The data preprocessing unit is used to extract effective point clouds from the tunnel roof, left sidewall, and right sidewall. The pose calculation unit is used to fit each virtual plane of the tunnel based on the effective point cloud of the tunnel roof, left side and right side; it is also used to obtain the normal distance from the center of the lidar to each virtual plane based on the lidar position in the fuselage coordinate system, and to construct planar constraint conditions in combination with the design geometric parameters of the tunnel. The reference positioning unit is used to determine the fixed coordinates of the tunnel in the absolute coordinate system based on the pose parameters of the tunneling machine in its initial state and the normal distance from the center of the lidar to each virtual plane. It is also used to obtain the real-time spatial coordinates and pose data of the tunneling machine in the absolute coordinate system based on the normal distance from the center of the lidar to each virtual plane of the tunnel and the angle between the plane normal vectors measured by the lidar in the motion state, and based on the fixed coordinates of the tunnel in the absolute coordinate system and through plane constraints.

[0011] Preferably, it further includes: The communication unit is used for internal data interaction and interference-resistant transmission with external devices. The visualization unit, including an intrinsically safe display screen for mining or a remote monitoring terminal, is used for real-time display of spatial coordinates and pose data.

[0012] Preferably, the lidar is mounted behind the cutting mechanism, at the top center of the main platform of the fuselage; wherein the scanning center axis of the lidar is collinear and coincides with the longitudinal center axis of the fuselage.

[0013] The relative positioning method for tunnel boring machines based on lidar provided by this invention has the following advantages: This invention determines the spatial relationship between the tunnel, the aircraft body, and the lidar by acquiring tunnel structure parameters, fuselage parameters, and lidar sensing parameters under a unified spatial coordinate system, avoiding errors and data conflicts caused by the inconsistency of coordinate systems in traditional multi-sensor positioning. This method constructs virtual plane constraints by extracting stable structures from the tunnel roof, left side, and right side, eliminating the need for external markers such as total stations, base stations, and prisms. This solves the problems of traditional positioning methods requiring manual deployment, frequent calibration, and high maintenance costs.

[0014] This invention establishes a positioning benchmark and solidifies the tunnel coordinates using initial inertial navigation data in a single step. After the tunneling machine moves, it relies solely on lidar for pure geometric constraints to calculate its pose, completely avoiding inertial navigation drift and multi-source data interference. This achieves high-precision positioning with no cumulative error and continuous stability. When the tunneling machine experiences body offset or attitude change, the system can calculate the pose in real time and directly output correction commands, avoiding problems such as cutting deviation and tunnel skew caused by traditional positioning lag. Automatic guidance and cutting of the tunneling machine can be achieved without manual intervention, truly supporting unmanned and intelligent operation of underground tunneling equipment. Attached Figure Description

[0015] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating the implementation of relative positioning using the lidar in this invention. Figure 2 A schematic diagram of a tunneling machine relative positioning method based on lidar provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of lidar installation provided in an embodiment of the present invention; Figure 4 This is an example diagram of the point cloud preprocessing process provided in an embodiment of the present invention; Figure 5 Example diagram of the transformation relationship between the tunnel, fuselage, and radar coordinate systems, and the fuselage position calculation parameters provided for embodiments of the present invention; Figure 6 The bidirectional positioning process provided in the embodiments of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0018] Example 1 Taking the relative positioning method for tunnel boring machines in roadways based on lidar provided by this invention as an example, the relative positioning method for tunnel boring machines in roadways provided by this invention is further discussed in detail as follows: Figure 1-2 As shown, it includes the following steps: Step 11: Rigidly install the mining explosion-proof lidar on the top centerline of the tunneling machine body, as shown in the installation position. Figure 3 As shown, the system is divided into two stages: initial calibration and dynamic positioning. In the initial static state, IMU data is collected to establish the initial attitude reference of the machine body. After the tunneling machine moves, IMU and odometer data are no longer used, and the pose calculation is achieved solely by the 3D point cloud of the lidar. The 3D point cloud data of the tunnel is collected in real time to complete the data time synchronization of the initial stage.

[0019] The steps for the lidar to initially collect point cloud information and measurement data from other sensors are as follows: Step 111: Determine the installation reference plane of the tunneling machine body. Select the position at the center of the top of the main platform of the machine body behind the cutting mechanism as the installation point of the lidar, so that the radar scanning center axis is collinear and coincident with the longitudinal center axis of the machine body, and avoid the machine body platform, cutting arm and other structures from obstructing the scanning field of view.

[0020] Step 112: Securely install the mine explosion-proof lidar at the above-mentioned location using a rigid fixing method, complete the radar installation attitude calibration, and record the installation offset and installation angle of the radar relative to the center of the fuselage as fixed parameters for subsequent coordinate system transformation.

[0021] Step 113: Activate the lidar and perform a three-dimensional scan of the tunnel environment at a set frequency to acquire real-time raw three-dimensional point cloud data including the tunnel roof, left side, right side, floor, and tunneling face area.

[0022] Step 114: Initial calibration stage, acquire IMU angular velocity and acceleration data, and calculate the initial absolute attitude (yaw) of the tunnel boring machine. , up and down , roll The IMU data serves as the sole initial reference for establishing the global coordinate system. After the motion, the IMU measurement values ​​drift, so the IMU data is no longer used in subsequent calculations.

[0023] Step 115: Complete the time synchronization and caching of the initial point cloud and IMU reference data, which will serve as the input for solidifying the tunnel coordinates.

[0024] Step 12: Based on the collected real-time data, preprocess the 3D point cloud of the tunnel, removing noise points, self-occluding points, and invalid point clouds, and extracting the valid point clouds of the tunnel roof, left side, and right side. The set of corner coordinates of the cutting surface can be determined through the following steps: Step 121: Analyze the LiDAR's proprietary protocol, automatically identify the line count mode, and extract the 3D coordinates of each point. x, y, z The criteria include: reflection intensity, laser line number, and hardware timestamp. Hard criteria are used for filtering, such as eliminating points exceeding limits, dust or water mist points with low reflection intensity, and mismeasured points on high-reflection surfaces, based on distance range or reflection intensity range.

[0025] Step 122: Based on the calibration and installation parameters of the radar at the top of the central axis of the main fuselage platform, a three-dimensional self-masking region is predefined, including the lower fuselage platform, the forward cutting arm, etc., and the point cloud within this region is directly filtered out. If the cutting arm is in an active state, the masking region can be dynamically updated according to its real-time angle. A radius filtering method is used to first remove outliers and dust noise, achieving preliminary data preprocessing.

[0026] Step 123: Identify the point cloud set of the two sidewalls and one roof in the tunnel, use the 3D ROI positioning method, with the LiDAR as the coordinate origin, to locate the tunnel area and set... x,y,z The spatial range requires x In the axial direction, the focus should be on both sides of the fuselage and the established stable roadways. y The axial direction covers the typical coverage area of ​​a coal mine. z In the axial direction, the area below the base plate needs to be directly removed, and then all point clouds not within the above 3D frame need to be removed, resulting in a sharp reduction in data volume.

[0027] The initial point cloud screening by cropping the Region of Interest (ROI) can be determined by the following steps: Step 125: Using semantic recognition of point clouds, the position ranges of the left and right sides and the top plate are initially divided. The candidate point cloud of the input ROI to be clipped is output as a point cloud module with semantic labels {left side, right side, top plate, other}.

[0028] Step 126: Calculate the top 30% quantiles of the Z values ​​for all points within the ROI as Zi. t1， Let Z≥Z t1 The points are marked as "candidate areas for the top plate"; the median Z-value quantile is calculated as Z0. t2 , will Z t2 ≤Z≤Z t1 The points are marked as "candidate areas for both sides"; Z is... <Z t2Points marked as "other" are removed. Then, by constraining the normal direction, the geometric features of each semantic region are accurately confirmed. An algorithm accelerated by integral images is used, with a neighborhood search radius of 0.3m, to calculate the unit normal vector of each point. n x ,n y ,n z ), by filtering in the "top plate candidate area", those that meet the requirements n z Points with a value less than -0.85 are marked as "top plate" and satisfy the following conditions in the "right side candidate area". n y Points with a value less than -0.85 are marked as "right side"; in the "left side candidate region", points satisfying... n y Points with a value greater than 0.85 are marked as "left side"; points that do not meet the above conditions are marked as "other" and removed.

[0029] Step 127: Using spatial and normal similarity constraints, aggregate semantic points into connected clusters, eliminating isolated noise. From the point sets labeled "Top Plate," "Right Side," and "Left Side," select 10-20 points with the smallest curvature in their normal neighborhood as initial seed points. After generating seed points, set growth criteria—spatial distance < 0.2m and normal vector angle < 15°. Starting from the seed points, grow outwards, incorporating neighboring points that meet the criteria into the same semantic cluster. Utilize the continuity of the planar structure to expand the semantic region coverage, such as... Figure 4 This indicates the point cloud preprocessing process.

[0030] Step 13: Based on the point cloud preprocessing results, perform virtual plane fitting processing on the extracted effective point cloud of the roadway structure; calculate the normal distance from the LiDAR center to each virtual plane based on the LiDAR position and the position of each virtual plane, and construct planar constraint conditions in conjunction with the roadway design geometric parameters; in the initial stage, establish an absolute attitude reference using inertial measurement data to complete the calibration and solidification of the roadway's absolute coordinates; after the tunneling machine moves, use only the solidified roadway coordinates as the reference, and directly calculate the spatial pose of the tunneling machine in the roadway by the angle between the normal distance of the LiDAR point cloud and the plane normal vector; the specific steps of the data calculation unit in constructing the virtual plane and calculating the pose are as follows: Step 131: Perform virtual plane fitting and relative distance calculation. Fit the semantically clear point cloud into a standard mathematical virtual plane and calculate the perpendicular distance of the laser radar to the plane. For each semantic subset, use the least squares method to fit the optimal standard mathematical plane. The plane equation is in the form of: Ax + By + Cz + D = 0 in( A,B,C () is the unit normal vector of the plane. D This is a plane constant term.

[0031] Step 132: Let the coordinates of the lidar center in the fuselage coordinate system be (x0, y0, z0). Then, the vertical distance from the lidar center to the virtual plane is... d The calculation formula is: ; Finally, the distance to the left gang was obtained. d l Distance to the right side d r Distance to the top slab d f .

[0032] Step 133, as follows Figure 5 The absolute coordinate system relationships shown in this scheme mainly involve three coordinate systems: the radar coordinate system, the radar coordinate system, and the radar coordinate system. o r With the radar installation center as the origin, the red line represents the radar coordinate system and the coordinates of the two sides and one top scanned below it; fuselage coordinate system o b With the center of the fuselage as the origin, the blue line represents the fuselage coordinate system, defined as follows: x b The shaft is forward along the tunneling axis. y b The axis runs horizontally to the right along the tunnel. z b Axis elevation pointing upwards; absolute coordinate system o w The origin is a fixed reference point in the tunnel, and the axis is aligned with the tunnel space. x w ,y w ,z w The transformation relationship, expressed by the homogeneous transformation matrix, is as follows:

[0033] ; in, The rotation matrix from the fuselage coordinate system to the absolute coordinate system (including) ; Translation vector (including) ; The rotation matrix from the radar coordinate system to the fuselage coordinate system (derived from the installation angle) calculate); Translation vector (by installation offset) constitute).

[0034] Step 134: Initial calibration stage. Combine the initial absolute attitude of the tunneling machine obtained by the inertial measurement unit with the normal distance from the radar to the two sides and the roof obtained by the lidar scan. Calculate the initial spatial attitude of the tunneling machine in the absolute coordinate system and use it to locate the fixed coordinates of the roadway in the absolute coordinate system, thus completing the one-time solidification of the absolute coordinates of the roadway.

[0035] Once the absolute coordinates of the tunnel are fixed, they will serve as the sole fixed reference for all subsequent positioning processes. This will also serve as the reference coordinate system for the relative positioning of the tunnel boring machine's body posture within the tunnel.

[0036] Based on the solidified absolute coordinates of the roadway, planar constraints are constructed: Regarding the roof plane constraints, the roof plane equation in the absolute coordinate system is... Combined with the distance of the radar to the top plate ,have to: ; in, Design height for alleyways, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the top plate plane Components. The process of establishing the left-side plane constraint is as follows: the equation of the left-side plane in the absolute coordinate system is... Combined with the distance of the radar to the left side ,have to:

[0037] ; in, Design the width of the alleyway, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the left side plane Components. The process of establishing the right-side plane constraint is as follows: the equation of the right-side plane in the absolute coordinate system is... Combined with the distance of the radar to the right side ,have to:

[0038] ; in, The normal vector of the right side plane Quantity.

[0039] Step 135: After the tunneling machine moves, using the fixed absolute coordinates of the tunnel as a reference, the three-dimensional position of the machine body in the tunnel is directly calculated using only the normal distance between the lidar and the tunnel plane. ; Coordinates combined with left and right plane constraints, then eliminated ,have to: ; Solving Then, the consistency of the distance between the left and right sides can be verified.

[0040] The X-coordinate is recursively calculated from the planar constraints of the point cloud in consecutive frames from the lidar.

[0041] Step 136: The attitude angle of the tunneling machine after its movement is calculated only by the angle between the radar measured normal vector and the standard normal vector of the roadway, without relying on inertial data.

[0042] The tunnel roof, left and right walls are fixed structures, and their normal vectors maintain a standard vertical direction in the absolute coordinate system. When the fuselage shifts or deflects, the normal vectors acquired by the radar will tilt, and the attitude can be calculated from the tilt angle. Rotation matrix Yaw angle Pitch angle Roll angle The function (assuming the rotation sequence is yaw → pitch → roll): ; Yaw angle It is calculated by the difference in the Y components of the left and right side normal vectors, such as , Lianlide: ; Pitch angle It is determined by the top plate normal vector. Z Component solution, such as ,have to: ; Roll angle It is calculated by the difference in the X components of the left and right side normal vectors, such as , ,have to: ; This calculation yields the orientation of the tunnel boring machine (TBM) in the tunnel. However, due to potential errors in plane fitting caused by underground dust and vibration noise, least squares fitting is necessary. This transforms the constraint equations into a nonlinear least squares problem, iteratively solving for the orientation parameters and minimizing the plane equation residuals. Subsequently, the TBM's orientation data is generated. The accuracy of the orientation parameters is then verified; if they exceed the marked range, they are flagged as abnormal. This completes the final steps.

[0043] Step 14: After the tunneling machine moves further, repeat steps 135 and 136. Based on the fixed absolute coordinate reference of the tunnel, continuously use the lidar point cloud to calculate the three-dimensional position and attitude angle of the machine body in the tunnel, and output the spatial pose of the tunneling machine relative to the tunnel in real time to achieve continuous, stable and drift-free relative positioning.

[0044] Specifically, the data transmission unit transforms the abstract pose into an intuitive graphical interface, allowing the driver to clearly determine the relative relationship between the machine body and the tunnel. The pose output by the system can be directly connected to the main control of the tunneling machine, enabling automatic correction, automatic guidance, and automatic cutting under unmanned working conditions without manual intervention, significantly improving the tunnel forming quality and tunneling efficiency.

[0045] Example 2 The present invention also provides a relative positioning system for tunnel boring machines based on lidar, comprising: The lidar unit is used to collect three-dimensional point cloud data of the tunnel. An inertial measurement unit is used to acquire the orientation parameters of the tunneling machine in its initial state. The data preprocessing unit is used to extract effective point clouds from the tunnel roof, left sidewall, and right sidewall. The pose calculation unit is used to fit each virtual plane of the tunnel based on the effective point cloud of the tunnel roof, left side and right side; it is also used to obtain the normal distance from the center of the lidar to each virtual plane based on the lidar position in the fuselage coordinate system, and to construct planar constraint conditions in combination with the design geometric parameters of the tunnel. The reference positioning unit is used to determine the fixed coordinates of the tunnel in the absolute coordinate system based on the pose parameters of the tunneling machine in its initial state and the normal distance from the center of the lidar to each virtual plane. It is also used to obtain the real-time spatial coordinates and pose data of the tunneling machine in the absolute coordinate system based on the normal distance from the center of the lidar to each virtual plane of the tunnel and the angle between the plane normal vectors measured by the lidar in the motion state, and based on the fixed coordinates of the tunnel in the absolute coordinate system and through plane constraints.

[0046] The communication unit is used for internal data interaction and interference-resistant transmission with external devices. The visualization unit, including an intrinsically safe display screen for mining or a remote monitoring terminal, is used for real-time display of spatial coordinates and pose data.

[0047] The relative positioning system for tunneling machines based on lidar described in this embodiment can fully realize the positioning method in the aforementioned embodiments through the coordinated linkage of the seven functional units. It does not require external targets or base stations, and achieves passive and drift-free positioning by relying on the geometry of the tunnel itself. It is suitable for complex underground working conditions and can be directly connected to the automatic cutting and autonomous walking system of the tunneling machine. It provides high-precision and stable positioning support for the intelligent operation of the tunneling machine, and solves the technical pain points of traditional positioning systems, such as high dependence on manual labor, low accuracy, and poor stability.

[0048] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the present invention patent. No reference numerals in the claims should be construed as limiting the scope of the claims. Any simple variations or equivalent substitutions of technical solutions that can be readily obtained by those skilled in the art within the scope of the technology disclosed in the present invention are within the protection scope of the present invention.

Claims

1. A method for relative positioning of a tunnel boring machine in a roadway based on lidar, characterized in that, include: The LiDAR mounted on the tunneling machine collects three-dimensional point cloud data of the tunnel in its initial static state, extracts the effective point cloud of the tunnel roof, left side and right side, and performs fitting of each virtual plane of the tunnel. Based on the position of the lidar in the fuselage coordinate system, the normal distance from the lidar center to each virtual plane is obtained, and the planar constraint conditions are constructed in combination with the design geometric parameters of the tunnel. Based on the pose parameters of the tunneling machine in its initial state obtained by the inertial measurement unit mounted on the tunneling machine, combined with the normal distance from the center of the lidar to each virtual plane, the fixed coordinates of the roadway in the absolute coordinate system are determined. The normal distance from the center of the lidar to each virtual plane in the tunnel and the angle between the plane normal vectors are obtained under the motion state. Based on the fixed coordinates of the tunnel in the absolute coordinate system, the real-time spatial coordinates and pose data of the tunneling machine in the absolute coordinate system are obtained through planar constraints.

2. The method according to claim 1, characterized in that, The extraction of effective point clouds from the tunnel roof, left flank, and right flank includes the following steps: The tunnel environment is scanned in three dimensions at a set frequency to obtain real-time raw three-dimensional point cloud data including the tunnel roof, left side, right side, floor and tunneling face area; Extract the 3D coordinates, reflection intensity, laser line number, and hardware timestamp of each point from the original 3D point cloud data, and filter out out-of-limit points and erroneous points by conditional filtering. Based on the calibration and installation parameters of the lidar at the top of the central axis of the main fuselage platform, the three-dimensional self-occlusion area is determined, including the lower fuselage platform and the front cutting arm. The three-dimensional point cloud data in this area is filtered out to obtain the filtered point cloud data.

3. The method according to claim 3, characterized in that, The fitting of the virtual plane includes the following steps: Using the 3D ROI positioning method, with the LiDAR as the coordinate origin, the tunnel area is located. The 3D point cloud data of non-tunnel areas is removed, and the semantic point clouds of the tunnel roof, left side and right side are separated. The virtual plane is then fitted by the least squares method.

4. The method according to claim 1, characterized in that, The process of obtaining the normal distance from the center of the lidar to each virtual plane based on the lidar position in the fuselage coordinate system, and constructing planar constraint conditions in conjunction with the design geometric parameters of the tunnel, includes the following steps: Determine the installation data of the lidar and the coordinate system rotation matrices for the lidar coordinate system, fuselage coordinate system, and absolute coordinate system; Based on the coordinate system rotation matrix, the coordinates of the lidar center in the fuselage coordinate system are obtained. Based on the virtual planes of the tunnel roof, left side and right side, the vertical distance from the lidar center to the virtual plane is determined. The initial absolute attitude of the tunneling machine is obtained by inertial measurement unit and the normal distance from the radar to the roadway roof, left side and right side obtained by lidar scanning. The initial spatial attitude of the tunneling machine in absolute coordinate system is calculated and the fixed coordinates of the roadway in absolute coordinate system are determined. Based on the fixed coordinates of the tunnel in the absolute coordinate system, planar constraints are constructed as follows: Under the constraints of the top plate plane, the equation of the top plate plane in the absolute coordinate system is: Combined with the distance of the radar to the top plate ,have to: ； in, Design height for alleyways, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the top plate plane Quantity; On the left-side plane constraint, the equation of the left-side plane in the absolute coordinate system is determined as follows: Combined with the distance of the radar to the left side ,have to: ； in, Design the width of the alleyway, For the fuselage in the absolute coordinate system coordinate, For the radar in the fuselage coordinate system coordinate, The normal vector of the left side plane Quantity; On the left side plane constraint, determine the equation of the right side plane in the absolute coordinate system as follows: Combined with the distance of the radar to the right side ,have to: ； in, The normal vector of the right side plane Quantity.

5. The method according to claim 3, characterized in that, The acquisition of the real-time spatial coordinates includes the following steps: Obtain the normal distances from the center of the lidar to each virtual plane in the tunnel as measured by the lidar under motion conditions, and calculate the three-dimensional position of the fuselage in the tunnel using planar constraints: Wherein, the Z-coordinate is: ； coordinate: Joint left and right side planar constraints: ； X-coordinate: calculated recursively from the planar constraints of the point cloud in consecutive frames from the lidar.

6. The method according to claim 4, characterized in that, The acquisition of the pose data includes the following steps: Real-time acquisition of the normal vectors of the laser radar to each virtual plane; The yaw angle is calculated based on the difference in the Y component of the left and right side normal vectors, the pitch angle is calculated based on the Z component of the top plate normal vector, and the roll angle is calculated based on the difference in the X component of the left and right side normal vectors. The position and orientation of the tunnel boring machine in the roadway are determined by the yaw angle, pitch angle, and roll angle.

7. A relative positioning system for tunnel boring machines based on lidar, characterized in that, include: The lidar unit is used to collect three-dimensional point cloud data of the tunnel. An inertial measurement unit is used to acquire the orientation parameters of the tunneling machine in its initial state. The data preprocessing unit is used to extract effective point clouds from the tunnel roof, left sidewall, and right sidewall. The pose calculation unit is used to fit each virtual plane of the tunnel based on the effective point cloud of the tunnel roof, left side and right side; it is also used to obtain the normal distance from the center of the lidar to each virtual plane based on the lidar position in the fuselage coordinate system, and to construct planar constraint conditions in combination with the design geometric parameters of the tunnel. The reference positioning unit is used to determine the fixed coordinates of the tunnel in the absolute coordinate system based on the pose parameters of the tunneling machine in its initial state and the normal distance from the center of the lidar to each virtual plane. It is also used to obtain the real-time spatial coordinates and pose data of the tunneling machine in the absolute coordinate system based on the normal distance from the center of the lidar to each virtual plane of the tunnel and the angle between the plane normal vectors measured by the lidar in the motion state, and based on the fixed coordinates of the tunnel in the absolute coordinate system and through plane constraints.

8. The system according to claim 7, characterized in that, Also includes: The communication unit is used for internal data interaction and interference-resistant transmission with external devices. The visualization unit, including an intrinsically safe display screen for mining or a remote monitoring terminal, is used for real-time display of spatial coordinates and pose data.

9. The system according to claim 7, characterized in that, The lidar is mounted behind the cutting mechanism, at the top center of the main platform of the fuselage; wherein the scanning center axis of the lidar is collinear and coincides with the longitudinal center axis of the fuselage.

Images