Underground engineering construction machinery positioning method, system, electronic device, computer storage medium and program product

Abstract

This invention discloses a method for locating construction machinery in underground engineering, comprising the following steps: extracting topological key points from the BIM model of the construction area to construct a three-point-one-line chain topological model; when the construction machinery enters the construction area, determining the initial topological point position, collecting the displacement increment output by the odometer and the angle increment output by the inertial measurement unit in real time, and calculating the increment along the current topological edge starting from the currently confirmed topological point. Upon confirmation of the construction machinery reaching the next topological point through distance, angle, and visual confirmation conditions, a topological point refresh operation is performed. The absolute coordinates of the construction machinery are obtained by superimposing the coordinates of the currently confirmed topological point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit. This invention also discloses a corresponding positioning system, electronic device, storage medium, and program product. This invention can reduce computational load, increase calculation speed, reduce cumulative errors, and improve reliability.

Description

Technical Field

[0001] This invention relates to the field of navigation technology, and in particular to a method for positioning underground construction machinery, electronic equipment, computer storage media, and program products. Background Technology

[0002] The construction environment of underground engineering projects (such as tunnels, mines, and underground pipe corridors) is characterized by the absence of satellite signals, limited space, insufficient lighting, and dust and humidity, making traditional surface positioning technologies, represented by satellite positioning (GPS / BDS / RTK), unsuitable.

[0003] The existing technical approach for fusing BIM (Building Information Modeling) models with real-time sensor data to achieve construction machinery positioning mainly involves: acquiring 3D point cloud data around the construction machinery using environmental perception sensors such as LiDAR; registering the current frame point cloud with the global BIM point cloud model (e.g., Iterative Nearest Point (ICP) registration or Normal Distribution Transform (NDT) registration) to obtain the real-time pose of the construction machinery in the BIM coordinate system; simultaneously acquiring displacement and angle data from odometers or inertial measurement units (IMUs) and using these as predicted values ​​in multi-source sensor fusion (e.g., Extended Kalman Filter (ESKF)); and using the fused pose as the positioning output. This approach uses BIM as a passive correction benchmark, resulting in consistently high computational resource consumption and excessive reliance on the integrity and real-time access of the BIM model. Furthermore, wheeled odometers used in tunnel construction inherently suffer from slippage / slippage issues. On muddy or gravel surfaces, the slippage rate can reach 3%–5%. Relying solely on odometer calculations results in a cumulative error of 0.6–1.0 meters every 20 meters. Without periodic correction, the cumulative positioning error in long tunnels (e.g., several kilometers in length) will continue to increase to tens or even hundreds of meters, causing construction machinery to completely lose its ability to determine its position within the topological segment. Simultaneously, the angular drift of the MEMS IMU accumulates over time without external correction, further exacerbating the calculation errors. Summary of the Invention

[0004] To address the aforementioned shortcomings of existing technologies, the present invention aims to provide a method for locating underground construction machinery. Existing technologies rely on global BIM matching, resulting in high computational load, poor real-time performance, and excessive dependence on real-time access to the BIM model under the frame-by-frame continuous correction paradigm. The present invention also aims to provide a system, electronic equipment, computer storage medium, and program for implementing the method for locating underground construction machinery.

[0005] The technical solution of the present invention is as follows: A method for positioning underground engineering construction machinery, comprising the following steps:

[0006] S1. Extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model; the chain topological model consists of sequentially connected topological points and topological edges, each topological point records its absolute coordinates in the BIM coordinate system, and each topological edge records the distance between adjacent topological points; the chain model is arranged according to the construction progress order, and the topological points are sequentially the previous point, the current point, and the next point.

[0007] S2. When construction machinery enters the construction area, determine the location of the initial topology point and use the initial topology point as the starting reference for chain calculation.

[0008] S3. Real-time acquisition of displacement increments from the odometer and angle increments from the inertial measurement unit, including heading angle and pitch angle. Based on the displacement increments and angle increments, starting from the currently confirmed topology point, incremental calculations are performed along the current topology edge to calculate the cumulative offset distance and cumulative angle change of the construction machinery on the current topology edge in real time.

[0009] S4. Confirm whether the construction machinery has reached the next topological point by detecting distance conditions, angle conditions and visual confirmation conditions. The distance condition is that the current cumulative offset distance is greater than the length of the current topological edge. The angle condition is that the current cumulative angle change exceeds the corner angle of the current topological edge. The visual confirmation condition is that the visual sensor recognizes the preset feature mark of the next topological point.

[0010] S5. When it is confirmed that the construction machinery has reached the next topology point, execute the topology point refresh operation: move the previous point, current point and subsequent point in the chain topology model forward in sequence, reset the cumulative values ​​of the odometer and inertial measurement unit based on the coordinates of the new current topology point, and continue to perform incremental calculations along the next topology edge.

[0011] S6. The absolute coordinates of the construction machinery are calculated as follows: The coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit are superimposed to obtain the real-time position of the construction machinery in the BIM coordinate system.

[0012] Furthermore, the topological key points include bifurcation points, turning points, passing tunnels, and mileage markers, wherein the mileage markers are auxiliary topological points that are uniformly interpolated at preset equal intervals along the tunnel advance direction.

[0013] Furthermore, the distance condition is that the current cumulative offset distance is greater than 95% of the length of the current topological edge, and the angle condition is that the current cumulative angle change exceeds the angle threshold, wherein the angle threshold is the corner angle of the current topological edge - 2°.

[0014] Furthermore, the preset feature markers of the topological points include at least one of geometric shape features, reflection identification features, or structural features.

[0015] Furthermore, it includes a topology consistency verification step, specifically periodically checking the compatibility between the absolute coordinates of the construction machinery and the chain topology model. The compatibility includes whether the vertical distance from the current absolute coordinate position to the current topology edge exceeds the preset geometric tolerance, whether the difference between the absolute value of the cumulative heading angle change and the preset heading angle of the current topology edge exceeds the preset angle tolerance, and whether the dwell time on the current topology edge exceeds the upper limit of the normal passage time.

[0016] Furthermore, the topology consistency verification step sets up multi-level responses based on the degree of deviation of the compatibility test results. The multi-level responses include shortening the verification cycle and recording a warning log when there is a slight deviation, pausing the output of customized data when there is a moderate deviation, and prompting the manual movement of construction machinery to the nearest known topology point and reconfirmation of the initial topology point when there is a severe deviation.

[0017] The second technical solution of the present invention is: a system for positioning underground engineering construction machinery, comprising:

[0018] The BIM processing module is used to extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model. The chain topological model consists of sequentially connected topological points and topological edges. Each topological point records its absolute coordinates in the BIM coordinate system, and each topological edge records the distance between adjacent topological points. The chain model is arranged according to the construction progress order, and the topological points are sequentially the previous point, the current point, and the next point.

[0019] The initialization module is used to determine the initial topology point position when construction machinery enters the construction area, and to use the initial topology point as the starting point for chain calculation.

[0020] The calculation module is used to collect the displacement increment output by the odometer and the angle increment output by the inertial measurement unit in real time. The angles include the heading angle and the pitch angle. Based on the displacement increment and the angle increment, the module calculates the incremental displacement along the current topological edge starting from the currently confirmed topological point and calculates the cumulative offset distance and cumulative angle change of the construction machinery on the current topological edge in real time.

[0021] The condition detection module is used to confirm whether the construction machinery has reached the next topological point by detecting distance conditions, angle conditions, and visual confirmation conditions. The distance condition is that the current cumulative offset distance is greater than the length of the current topological edge, the angle condition is that the current cumulative angle change exceeds the corner angle of the current topological edge, and the visual confirmation condition is that the visual sensor recognizes the preset feature mark of the next topological point.

[0022] The topology refresh module is used to perform a topology refresh operation when it is confirmed that the construction machinery has reached the next topology point: the previous point, current point and subsequent point in the chain topology model are moved forward in sequence, the cumulative values ​​of the odometer and inertial measurement unit are reset based on the coordinates of the new current topology point, and incremental calculation is continued along the next topology edge;

[0023] The coordinate calculation module calculates the absolute coordinates of the construction machinery in the following way: by superimposing the coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit, the real-time position of the construction machinery in the BIM coordinate system is obtained.

[0024] Furthermore, a verification module is included for topology consistency verification: periodically checking the compatibility between the absolute coordinates of the construction machinery and the chain topology model. The compatibility includes whether the vertical distance from the current absolute coordinate position to the current topology edge exceeds the preset geometric tolerance, whether the difference between the absolute value of the cumulative change in heading angle and the preset heading angle of the current topology edge exceeds the preset angle tolerance, and whether the dwell time on the current topology edge exceeds the upper limit of the normal passage time.

[0025] Another technical solution of the present invention is: an electronic device, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor;

[0026] The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the aforementioned underground engineering construction machinery positioning method.

[0027] Another technical solution of the present invention is: a computer-readable storage medium storing a computer program thereon, wherein when the computer program is executed by a processor, the aforementioned method for positioning underground construction machinery is implemented.

[0028] Another technical solution of the present invention is: a computer program product, including a computer program, wherein when the computer program product is executed by a processor, it implements the aforementioned method for positioning underground construction machinery.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] (1) The model is extremely simple, and the computational complexity is greatly reduced. The complex global BIM topology map is simplified into a three-point chain model, which only requires maintaining the coordinates of three topology points (previous point, current point, and next point) and the parameters (distance and angle) of two topology edges, which greatly reduces the storage and computation requirements. Existing BIM fusion positioning technology usually requires maintaining the feature parameters of the global BIM point cloud (such as NDT normal distribution grid or ICP matching point set). Taking a 10Hz laser frame rate as an example, global BIM registration needs to be performed once every 0.1 seconds, and global matching needs to be performed 10 times per second. The present invention adopts a low-frequency periodic calculation strategy (such as performing incremental calculation and verification once every 5 seconds). Taking a 20m segment length and a 1m / s travel speed as an example, only about 4 positioning calculations need to be performed in a 20-second single segment travel process, while the existing method needs to perform 200 global BIM registrations in the same period (10Hz×20s). The BIM matching frequency is reduced by about 50 times, and the computational resource occupation is reduced from continuous high intensity to low frequency sparse mode.

[0031] (2) Fast calculation and significantly improved real-time performance. Odometer and IMU incremental calculation are the main positioning methods. The odometer displacement increment is directly obtained by the pulse counting of the photoelectric encoder (50-100Hz), and the IMU angle increment is obtained by the angular velocity integration of the gyroscope (100-200Hz). Both are microsecond-level operations, which are much faster than ICP point cloud registration (millisecond to tens of milliseconds) and NDT probability density calculation. The positioning data refresh delay is reduced from tens of milliseconds in existing methods to microseconds, ensuring that construction machinery can obtain the position calculation results at any time without positioning interruption or waiting.

[0032] (3) Reliability confirmed, with three conditions working together and complementing each other. The odometer, IMU, and vision sensor have complementary failure modes: the odometer falsely reports displacement when slipping / idling, but the IMU can detect angle mismatch (the heading does not change or changes erratically when slipping); in long straight tunnels, the IMU angle changes slightly, but the vision sensor can actively identify feature markers (highly reflective targets or preset geometric patterns); when a single sensor misleads, it can be intercepted by the other two types of linkage. The triple AND linkage of distance, angle, and vision constitutes a safety redundancy at the sensor level, ensuring that topology point refresh is triggered only when the next topology point is actually reached, avoiding incorrect segment switching caused by misjudgment by a single sensor.

[0033] (4) Controllable error, event-driven discrete reset. Unlike the "time-driven" continuous correction method of frame-by-frame re-registration in existing BIM fusion positioning, this invention adopts an "event-driven" discrete reset strategy. Coordinate anchoring and sensor reset are only performed when the construction machinery passes through the topology point and the three conditions are met simultaneously. In a 20m long travel segment, the odometer slip rate is estimated at 3% to 5%. The cumulative distance error before reaching the end of the travel segment is about 0.6 to 1.0 meters. After refreshing, the error is zero and is not inherited to the next travel segment. This mechanism has low requirements for the real-time accessibility of the BIM model. It only needs to read the pre-stored coordinates of the next topology point at the moment of topology point refresh, instead of querying the BIM database every frame. Moreover, in long straight tunnel sections, there is no need to waste computing power on meaningless re-matching.

[0034] (5) Highly adaptable and suitable for complex tunnel environments. Visually confirmed feature markers (highly reflective targets, preset geometric patterns, structural feature corners, etc.) are robust to typical tunnel conditions such as dust, humidity, and low light; the odometer and IMU are standard sensors for engineering machinery, requiring no additional hardware investment; the synergistic complementarity of the three conditions enables the system to maintain reliability even when the performance of a single sensor deteriorates.

[0035] (6) Fully autonomous positioning without the need for external facilities. The BIM topology chain model is constructed in one go during the offline preprocessing stage. During the online positioning stage, the construction machinery can independently complete all calculations without the need for external reference stations, beacon networks or real-time queries of BIM databases, thus achieving truly fully autonomous positioning. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the underground engineering construction machinery positioning method of the present invention.

[0037] Figure 2 This is a schematic diagram of a chain topology model in a bifurcation tunnel scenario embodiment of the present invention.

[0038] Figure 3 This is a flowchart illustrating the topology consistency verification step in an embodiment of the present invention. Detailed Implementation

[0039] The specific technical solution of the present invention will be further described in detail below with reference to specific examples.

[0040] Example 1

[0041] Please combine Figure 1As shown, this embodiment uses a three-arm drilling rig in tunnel construction as an example to illustrate the positioning method for underground engineering construction machinery. The drilling rig operates near the tunnel face, with a typical travel speed of 0.5–2.0 m / s. It needs to obtain its absolute position in the BIM tunnel coordinate system in real time to guide borehole positioning and construction progress management. The specific positioning method includes the following steps:

[0042] S1. Extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model.

[0043] This process is an offline preprocessing step. Specifically, during the construction preparation phase, topological key points are automatically extracted using the BIM model from the tunnel design phase (such as a 3D tunnel model built with Revit or Civil 3D). The extraction algorithm performs a first-order neighborhood scan along the tunnel centerline to identify the following four types of topological key points:

[0044] (a) Bifurcation: Nodes with ≥3 neighboring connections along the tunnel centerline, such as the junction of the main tunnel and the cross passage, or the point where the left and right lines separate;

[0045] (b) Inflection point: The location where the rate of change of curvature of the tunnel centerline exceeds a preset threshold (e.g., 0.05 rad / m), corresponding to a horizontal turning point or a vertical curve slope change point;

[0046] (c) Passing tunnel: The location of pedestrian / vehicle cross tunnel openings identified by the boundary detection of tunnel sidewall openings;

[0047] (d) Mileage markers: Auxiliary topology points that are uniformly interpolated along the tunnel advance direction at preset equal intervals (such as 10m or 20m) (excluding geometric mutations, only used for densification and refresh of anchor points).

[0048] The extracted topological key points are sorted according to the construction advance direction. Three consecutive points are selected to construct a chain window: the previous point (the topological point that has been confirmed to have been passed), the current point (the next topological point that the construction machinery is currently heading to), and the next point (the next topological point after the current point). Each topological point records its three-dimensional absolute coordinates (E_i, N_i, H_i) in the BIM coordinate system, and each topological edge records the Euclidean distance L_i, horizontal heading angle θ_i, and pitch angle φ_i between adjacent topological points.

[0049] Taking the simplified tunnel in this embodiment as an example, four topological key points are extracted: point A (tunnel entrance reference point, mileage DK0+000), point B (bifurcation / turning point, mileage DK0+020, heading 15° to the right), point C (passing hole location, mileage DK0+035), and point D (mileage marker point, mileage DK0+060). A chain model is constructed according to the construction progress: A→B→C→D. The parameters of each topological edge are recorded as follows:

[0050] Section A→B: Euclidean distance L_AB = 20m, heading angle θ_AB = 0°, pitch angle φ_AB = 0° (straight line, flat slope);

[0051] Section B→C: Euclidean distance L_BC = 15m, heading angle θ_BC = 15° (right yaw), pitch angle φ_BC = +2° (uphill);

[0052] Section C→D: Euclidean distance L_CD = 25m, heading angle θ_CD = 15°, pitch angle φ_CD = -1° (downhill).

[0053] Please combine Figure 2 As shown in the example of a bifurcated tunnel scenario, point B (the bifurcation point) has two passable branches: the left branch B→C1 (main tunnel extension) and the right branch B→C2 (cross passage). The chain model stores the topological edge parameters of the two branches separately; that is, the chain model records all passable branches at point B and their topological edge parameters: Branch 1 is the B→C1 segment: L_BC1=15m, θ_BC1=15° rightward deviation, φ_BC1=+2° (main tunnel); Branch 2 is the B→C2 segment: L_BC2=12m, θ_BC2=-10° leftward deviation, φ_BC2=0° (cross passage). During the online positioning phase, the corresponding branch is selected based on the information issued by the construction task.

[0054] S2. When construction machinery enters the construction area, determine the location of the initial topology point and use the initial topology point as the starting reference for chain calculation.

[0055] After the drilling rig enters the construction area, the operator confirms or inputs the initial topology point on the onboard human-machine interface. Typically, the known reference point closest to the rig's current position is selected (e.g., point A at the tunnel entrance, whose BIM absolute coordinates and geodetic coordinates have been accurately calibrated through construction surveying). The system sets point A as the starting reference for chained calculations, initializes the current topology point to point A, the next topology point to point B, and the subsequent topology point to point C. The odometer cumulative offset vector and IMU cumulative angle are both initialized to zero.

[0056] In some embodiments, if the trolley is equipped with an initial positioning aid (such as a laser rangefinder aligned with known tunnel cross-sectional features), initialization can be completed by automatically matching the nearest topological point without manual intervention.

[0057] S3. Real-time acquisition of displacement increments from the odometer and angle increments from the inertial measurement unit. Based on the displacement increments and angle increments, starting from the currently confirmed topology point, incremental calculations are performed along the current topology edge to calculate the cumulative offset distance and cumulative angle change of the construction machinery on the current topology edge in real time.

[0058] The incremental extrapolation process is explained in detail below.

[0059] During the movement of the rock drilling rig, the odometer (an optical encoder linked to the drive wheels of the walking mechanism, with an output frequency of 100Hz and a displacement resolution better than 1cm) outputs the displacement increment δd_k (where k is the sampling number) for each sampling cycle in real time. The IMU (a nine-axis MEMS inertial measurement unit, including a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer, with gyroscope zero-bias stability of approximately 10° / h, angular resolution better than 0.1°, and an output frequency of 200Hz) outputs the angle increment δθ_k (change in yaw angle around the Z-axis) and the pitch angle increment δφ_k in real time.

[0060] In the incremental extrapolation iteration of the k-th sampling period:

[0061] (i) Update the cumulative offset distance: δd_k;

[0062] (ii) Update the cumulative heading angle change: θ_acc(k) = θ_acc(k-1) + δθ_k;

[0063] (iii) Update the cumulative pitch angle change: φ_acc(k) = φ_acc(k-1) + δφ_k;

[0064] (iv) Calculate the odometer cumulative offset vector:

[0065] ΔX(k) = δ_acc(k) × cos(θ_AB + θ_acc(k)) × cos(φ_AB + φ_acc(k))

[0066] ΔY(k) = δ_acc(k) × sin(θ_AB + θ_acc(k)) × cos(φ_AB + φ_acc(k))

[0067] ΔZ(k) = δ_acc(k) × sin(φ_AB + φ_acc(k))

[0068] Where θ_AB and φ_AB are the preset heading angle and pitch angle of the current topological edge A→B.

[0069] In this embodiment, at a sampling frequency (100Hz) and a travel speed of 1m / s, the displacement increment per sampling cycle is approximately 1cm, the angle increment is close to zero on the straight section, and the cumulative angle change is approximately 0.75° per meter in the 15° turning section (B→C).

[0070] S4. By monitoring distance conditions, angle conditions, and visual confirmation conditions, confirm whether the construction machinery has reached the next topology point.

[0071] Combination Figure 3 After each incremental calculation iteration, three conditions are checked simultaneously:

[0072] (i) Distance condition: The cumulative offset distance δ_acc(k) is greater than the preset length L_AB (20m) of the current topology edge A→B, or further, the preset length L_AB (20m) of the current topology edge A→B is set to -5% tolerance (20-20*5%=19m) as the judgment value to cover the cumulative deviation of the odometer slip and the system deviation caused by wheel diameter wear;

[0073] (ii) Angle condition: The absolute value of the cumulative heading angle change |θ_acc(k)| exceeds the preset corner angle of 15° between the current topology edge A→B and the next topology edge B→C, or further, set a tolerance of -2° and use 13° as the judgment value to cover IMU angle drift;

[0074] (iii) Visual confirmation conditions: The visual sensor (depth camera such as IntelRealSense D455, RGB-D mode, working distance 0.3~20m, frame rate 30fps, or LiDAR such as Velodyne PuckLite 16-line) mounted on the front end of the rock drilling rig successfully identifies and matches the preset feature marker at point B. The feature marker type can be: (a) a highly reflective target sprayed on the tunnel wall (retroreflection coefficient >1000 cd / lx / m²); (b) a preset geometric pattern, such as a black and white checkerboard pattern with a side length of 20cm; (c) corner features of the tunnel structure, such as the corner features of the bypass opening, which are matched after being converted from BIM geometric data into a 2D projection template.

[0075] All three conditions must be met simultaneously; if any one condition is not met, the system continues incremental calculation on the current topology edge (returning to step S3). This method utilizes complementary failure modes of three types of sensors to achieve safety redundancy.

[0076] If the odometer slips (the drive wheels spin freely on a muddy road), the odometer may falsely report displacement, causing the distance condition to be met prematurely. However, the IMU can detect that the heading angle has not changed accordingly (the vehicle's heading remains unchanged when slipping), and the angle condition is not met, thus preventing incorrect updates.

[0077] If the IMU accumulates false angles due to gyroscope zero bias in a long straight tunnel, but the visual sensor can detect whether the feature markers are actually in the field of view, it will prevent erroneous refresh when the visual conditions are not met.

[0078] If the visual sensor's field of view is obstructed by dust / water mist, the joint dead reckoning calculation by the odometer and IMU can still maintain the positioning. The system will not refresh the image and will continue to calculate along the current topology edge until the visual conditions are met after the field of view is restored, at which point a refresh will be performed.

[0079] Triple condition monitoring ensures the reliability of this method even when sensor performance fluctuates: when a sensor degrades in the short term (such as slight slippage of the odometer or IMU angle drift), deviations within the tolerance range will not cause refresh failure; when the sensor degrades severely and exceeds the tolerance range, the system refuses to refresh under uncertain conditions, avoiding amplification of positioning deviation caused by incorrect topology point switching.

[0080] S5. When it is confirmed that the construction machinery has reached the next topology point, execute the topology point refresh operation: move the previous point, current point and subsequent point in the chain model forward in sequence, reset the odometer cumulative offset distance and the cumulative value of the inertial measurement unit based on the coordinates of the new current topology point, and continue to perform incremental calculations along the next topology edge.

[0081] In this embodiment, when all three conditions are met simultaneously—namely, distance ≥ 19m, angle ≥ 13°, and the feature marker at point B is visually identified—the following operations are performed:

[0082] (a) Topological point role rotation: Rotate A→B→C in the current chain model to B→C→D—point B becomes the previous point from the current point, point C becomes the current point from the next point, and point D (read from the preloaded chain model) becomes the new next point;

[0083] (b) Reset cumulative values: The cumulative offset distance δ_acc of the odometer and the cumulative angle changes θ_acc and φ_acc of the IMU are both reset to zero;

[0084] (c) Calculation benchmark update: The current calculation benchmark coordinates are changed from the coordinates of point A to the coordinates of point B (BIM absolute coordinates);

[0085] (d) Loading the next topological edge parameters: Read the topological edge parameters (L_BC=15m, θ_BC=15°, φ_BC=+2°) of segment B→C from the chain model for the next round of incremental calculation.

[0086] After the refresh operation is completed, the system uses point B as the new calculation benchmark and continues to execute the cycle from step S3 to step S5 along the B→C segment. In the B→C segment, the odometer starts accumulating from zero, and the IMU angle change also starts accumulating from zero. Any residual errors accumulated in the A→B segment (odometer slip error of 0.6~1.0m, IMU angle drift of approximately 0.02°) are completely discarded and do not propagate to the B→C segment.

[0087] In the aforementioned bifurcated tunnel scenario, when the construction machinery is estimated to be near point B and the triple conditions are met, the system queries the daily work tasks issued by the construction task management system (such as "Work area: left side of the main tunnel, C1 direction"), automatically updates the current chain window from A→B→C (C is the topology point of the nearest unbranched segment) to B→C1→D1, and loads the topology edge parameters of the C1 branch.

[0088] This "event-driven" discrete reset strategy uses topological points as anchors, performing a sensor zero-reference reset only once when passing through a topological point, rather than continuously accessing BIM frame by frame for global correction. The fundamental difference between this and the "time-driven" continuous correction method in existing BIM fusion positioning technologies lies in three aspects:

[0089] (a) Triggering mechanism: The present invention uses "passing through the topology point" as the event triggering condition (once every 20m), while the existing method uses "arrival of each laser frame" as the time triggering condition (once every 0.1 seconds);

[0090] (b) BIM access mode: The present invention only needs to read the BIM coordinates of the next topology point at the moment of refresh (point query), while the existing method needs to perform ICP or NDT registration (global matching) between the current laser point cloud and the global BIM point cloud every frame.

[0091] (c) Computational resource consumption: Under the conditions of 10Hz frame rate, 20m segment length and 1m / s travel speed, the existing method requires 200 global BIM registrations (10Hz×20s) in the A→B segment. The present invention adopts a periodic calculation once every 5 seconds, and only performs about 4 positioning calculations in the same segment, reducing the BIM matching frequency by about 50 times.

[0092] This step also includes a safety redundancy mechanism: the feature markers identified by the visual sensor at point B contain directional information (such as arrow patterns and branch numbers), which are compared with the preset branch directions for the construction task. If the visual recognition result (such as detecting a feature marker pointing to the C2 cross passage) is inconsistent with the preset branch (C1 main tunnel), the system triggers a branch inconsistency alarm and prominently prompts the operator through the human-machine interface: "Visual recognition of C2 branch features does not match the preset C1 branch. Please confirm the direction of travel," to prevent construction machinery from mistakenly entering non-operating branches and causing safety accidents.

[0093] S6. Overlay the coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the cumulative angle compensation of the inertial measurement unit to obtain the real-time position of the construction machinery in the BIM coordinate system.

[0094] The absolute coordinates of construction machinery in the BIM coordinate system at any given time are calculated using the following formula:

[0095] P(t) = P_anchor + V_odom(t) + V_imu_comp(t)

[0096] in:

[0097] P(t) represents the calculated absolute coordinates (three-dimensional vector) of the construction machinery at the current moment, and the parameter t represents the number of times that have elapsed since the construction machinery reached the currently confirmed topology point.

[0098] P_anchor represents the absolute coordinate vector of the currently confirmed topology point (which has passed triple condition verification) in the BIM coordinate system;

[0099] V_odom(t) represents the odometer cumulative offset vector, V_odom(t) = [ΔX(t),ΔY(t),ΔZ(t)];

[0100] V_imu_comp(t) represents the three-dimensional position compensation vector caused by the IMU's heading angle drift, used to correct the influence of the heading angle drift caused by the gyroscope's zero bias on the direction of the odometer offset vector. The compensation term is calculated as follows: the zero-speed output during the IMU's stationary period (detected by the trolley's parking state) is used as the zero bias estimate, and the accumulated heading angle drift error within time t is obtained as δψ_drift(t), V_imu_comp(t)=δ_acc(t)×δψ_drift(t)×[sin(θ_AB+θ_acc(t)),-cos(θ_AB+θ_acc(t)), 0]^T.

[0101] The above calculation decomposes the positioning problem of construction machinery into three independent sub-problems: determining the nearest anchor point, incremental offset from the anchor point, and sensor drift compensation. After decoupling, each of the three sub-problems is solved using the most suitable sensor and algorithm, thus avoiding the high-dimensional calculation and coupling error propagation caused by joint state estimation in traditional multi-sensor fusion.

[0102] In this embodiment, a topology consistency verification step is also performed, specifically, in conjunction with... Figure 3 The system performs topology consistency checks at preset intervals, such as a time interval (every 5 seconds) or a distance interval (every 5 meters traveled), independent of the triple condition check and topology point refresh logic. The checks include:

[0103] (a) Geometric range verification: Check whether the vertical distance d_perp from the current calculated position P(t) to the current topological edge exceeds the preset geometric tolerance ε_geo (e.g., 1 meter). If d_perp > ε_geo, it indicates that the construction machinery may have deviated from the preset tunnel centerline (e.g., accidentally entered a cross passage or experienced lateral slippage).

[0104] (b) Angle compatibility check: Check whether the difference between the absolute value of the cumulative heading angle change |θ_acc(t)| and the preset heading angle of the current topology edge exceeds the preset angle tolerance ε_angle (e.g., 5°). If the difference > ε_angle, it indicates that the heading of the construction machinery may be inconsistent with the tunnel orientation;

[0105] (c) Timing Reasonableness Check (Optional): Check whether the dwell time on the current topology edge exceeds the upper limit of the normal passage time estimated based on the segment length and reasonable travel speed. If the timeout is severe, it may indicate that the system is stuck somewhere or has performed an unexpected construction operation.

[0106] Based on the degree of deviation, a three-level response is triggered:

[0107] (a) Minor deviation (exceeding tolerance but less than 2 times tolerance): record the warning log and continue to calculate, while shortening the next verification cycle to 1 second / 1 meter for intensive monitoring. If the subsequent verification returns to normal, restore the default cycle.

[0108] (b) Moderate deviation (2 to 5 times tolerance): Pause the output of positioning data to the construction control system, send the current IMU data into the zero-speed calibration process (if the trolley is detected to be stationary), and prompt the operator to confirm the current position through the human-machine interface;

[0109] (c) Severe deviation (>5 times tolerance): Triggers the complete relocation process - prompts the operator to manually move the construction machinery to the nearest known topology point (such as the nearest mileage marker or passing hole), re-execute the S2 initialization step, and restart the chain calculation based on the BIM coordinates of the topology point.

[0110] The added topology consistency check and triple condition check work in synergy. The triple condition check provides positive confirmation: confirming that the construction machinery has reached the next topology point; while the topology consistency check checks whether the calculated results are inconsistent with the BIM chain model. The triple condition check is only activated near the new topology point, while the topology consistency check runs periodically at any position on the entire topology edge. The two complement each other to form a continuous safety monitoring method.

[0111] This invention constructs a chain model using BIM topology key points as anchor points, uses odometers and inertial measurement units (IMUs) as the main calculation methods and visual confirmation as an auxiliary confirmation condition, and realizes event-driven discrete error reset at the topology point level. This significantly reduces the real-time access requirements and computing resource consumption of the global BIM model while ensuring accuracy. The BIM matching frequency is reduced from frame-by-frame (once per frame, 10Hz, 10 times per second) to a low-frequency period (e.g., once every 5 seconds), enabling rapid, fully autonomous, real-time, and continuous positioning of underground construction machinery in environments without GNSS.

[0112] Example 2 is a positioning system for underground engineering construction machinery, comprising:

[0113] The BIM processing module is used to extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model.

[0114] The initialization module is used to determine the initial topology point position when construction machinery enters the construction area, and to use the initial topology point as the starting point for chain calculation.

[0115] The calculation module is used to collect the displacement increment output by the odometer and the angle increment output by the inertial measurement unit in real time. Based on the displacement increment and angle increment, it calculates the increment along the current topological edge starting from the currently confirmed topological point, and calculates the cumulative offset distance and cumulative angle change of the construction machinery on the current topological edge in real time.

[0116] The condition detection module is used to confirm whether the construction machinery has reached the next topological point by detecting distance conditions, angle conditions, and visual confirmation conditions.

[0117] The topology refresh module is used to perform a topology refresh operation when it is confirmed that the construction machinery has reached the next topology point;

[0118] The coordinate calculation module calculates the absolute coordinates of the construction machinery in the following way: by superimposing the coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit, the real-time position of the construction machinery in the BIM coordinate system is obtained.

[0119] The verification module is used for topology consistency verification: it periodically checks the compatibility between the absolute coordinates of construction machinery and the chain topology model.

[0120] Example 3

[0121] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, the computer program, and a BIM topology chain model database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the electronic device is used for exchanging information between the processor and external devices such as odometers, IMUs, vision sensors, and human-machine interfaces. The communication interface of the electronic device is used for communicating with an external construction task management system or BIM database via a network connection. When the computer program is executed by the processor, it implements all the steps in the above-described method embodiments.

[0122] Example 4

[0123] A computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps in the above-described method embodiments. The computer-readable storage medium includes at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), flash memory, optical memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric memory (FRAM), phase-change memory (PCM), etc. Volatile memory may include random access memory (RAM), such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0124] Example 5

[0125] A computer program product includes a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0126] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

Claims

1. A method of positioning an underground construction machine, characterized by, Includes the following steps: S1. Extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model; the chain topological model consists of sequentially connected topological points and topological edges, each topological point records its absolute coordinates in the BIM coordinate system, and each topological edge records the distance between adjacent topological points; the chain model is arranged according to the construction progress order, and the topological points are sequentially the previous point, the current point, and the next point. S2. When construction machinery enters the construction area, determine the location of the initial topology point and use the initial topology point as the starting reference for chain calculation. S3. Real-time acquisition of displacement increments from the odometer and angle increments from the inertial measurement unit, including heading angle and pitch angle. Based on the displacement increments and angle increments, starting from the currently confirmed topology point, incremental calculations are performed along the current topology edge to calculate the cumulative offset distance and cumulative angle change of the construction machinery on the current topology edge in real time. S4. Confirm whether the construction machinery has reached the next topological point by detecting distance conditions, angle conditions and visual confirmation conditions. The distance condition is that the current cumulative offset distance is greater than the length of the current topological edge. The angle condition is that the current cumulative angle change exceeds the corner angle of the current topological edge. The visual confirmation condition is that the visual sensor recognizes the preset feature mark of the next topological point. S5. When it is confirmed that the construction machinery has reached the next topology point, execute the topology point refresh operation: move the previous point, current point and subsequent point in the chain topology model forward in sequence, reset the cumulative values ​​of the odometer and inertial measurement unit based on the coordinates of the new current topology point, and continue to perform incremental calculations along the next topology edge. S6. The absolute coordinates of the construction machinery are calculated as follows: The coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit are superimposed to obtain the real-time position of the construction machinery in the BIM coordinate system.

2. The method for positioning underground construction machinery according to claim 1, characterized in that, The topological key points include bifurcation points, turning points, passing tunnels, and mileage markers. The mileage markers are auxiliary topological points that are uniformly interpolated at preset equal intervals along the tunnel advance direction.

3. The method for positioning underground construction machinery according to claim 1, characterized in that, The distance condition is that the current cumulative offset distance is greater than 95% of the length of the current topological edge, and the angle condition is that the current cumulative angle change exceeds the angle threshold, which is the corner angle of the current topological edge - 2°.

4. The method for positioning underground construction machinery according to claim 1, characterized in that, The preset feature markers of the topological points include at least one of geometric shape features, reflection identification features, or structural features.

5. A method for positioning underground construction machinery according to any one of claims 1 to 4, characterized in that, The process includes a topology consistency verification step, specifically periodically checking the compatibility between the absolute coordinates of the construction machinery and the chain topology model. The compatibility includes whether the vertical distance from the current absolute coordinate position to the current topology edge exceeds the preset geometric tolerance, whether the difference between the absolute value of the cumulative change in heading angle and the preset heading angle of the current topology edge exceeds the preset angle tolerance, and whether the dwell time on the current topology edge exceeds the upper limit of the normal passage time.

6. The method for positioning underground construction machinery according to claim 5, characterized in that, The topology consistency verification step sets up multi-level responses based on the degree of deviation of the compatibility test results. The multi-level responses include shortening the verification cycle and recording a warning log when there is a slight deviation, pausing the output of customized data when there is a moderate deviation, and prompting the manual movement of construction machinery to the nearest known topology point and reconfirmation of the initial topology point when there is a severe deviation.

7. A positioning system for underground engineering construction machinery, characterized in that, include: The BIM processing module is used to extract topological key points from the BIM model of the construction area and construct a three-point-one-line chain topological model. The chain topological model consists of sequentially connected topological points and topological edges. Each topological point records its absolute coordinates in the BIM coordinate system, and each topological edge records the distance between adjacent topological points. The chain model is arranged according to the construction progress order, and the topological points are sequentially the previous point, the current point, and the next point. The initialization module is used to determine the initial topology point position when construction machinery enters the construction area, and to use the initial topology point as the starting point for chain calculation. The calculation module is used to collect the displacement increment output by the odometer and the angle increment output by the inertial measurement unit in real time. The angles include the heading angle and the pitch angle. Based on the displacement increment and the angle increment, the module calculates the incremental displacement along the current topological edge starting from the currently confirmed topological point and calculates the cumulative offset distance and cumulative angle change of the construction machinery on the current topological edge in real time. The condition detection module is used to confirm whether the construction machinery has reached the next topological point by detecting distance conditions, angle conditions, and visual confirmation conditions. The distance condition is that the current cumulative offset distance is greater than the length of the current topological edge, the angle condition is that the current cumulative angle change exceeds the corner angle of the current topological edge, and the visual confirmation condition is that the visual sensor recognizes the preset feature mark of the next topological point. The topology refresh module is used to perform a topology refresh operation when it is confirmed that the construction machinery has reached the next topology point: the previous point, current point and subsequent point in the chain topology model are moved forward in sequence, the cumulative values ​​of the odometer and inertial measurement unit are reset based on the coordinates of the new current topology point, and incremental calculation is continued along the next topology edge; The coordinate calculation module calculates the absolute coordinates of the construction machinery in the following way: by superimposing the coordinates of the currently confirmed topology point, the cumulative offset vector of the odometer, and the three-dimensional position compensation vector caused by the heading angle drift of the inertial measurement unit, the real-time position of the construction machinery in the BIM coordinate system is obtained.

8. The positioning system for underground construction machinery according to claim 7, characterized in that, It includes a verification module for topology consistency verification: periodically checking the compatibility between the absolute coordinates of the construction machinery and the chain topology model. The compatibility includes whether the vertical distance from the current absolute coordinate position to the current topology edge exceeds the preset geometric tolerance, whether the difference between the absolute value of the cumulative change in heading angle and the preset heading angle of the current topology edge exceeds the preset angle tolerance, and whether the dwell time on the current topology edge exceeds the upper limit of the normal passage time.

9. An electronic device, characterized in that, The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the underground engineering construction machinery positioning method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the underground engineering construction machinery positioning method according to any one of claims 1 to 6.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the underground engineering construction machinery positioning method according to any one of claims 1 to 6.

Images