Underground pipeline augmented reality dynamic mapping method and system

Abstract

The invention discloses an underground pipeline augmented reality dynamic mapping method and system. The method comprises the steps that dynamic mapping equipment pose data, environment data and underground pipeline data are obtained and preprocessed; and performing dynamic calibration on the preprocessed pose data to obtain dynamically calibrated pose data, performing data verification on the dynamically calibrated pose data, the preprocessed environment data and the preprocessed underground pipeline data, and if verification fails, performing abnormal labeling. If the verification is passed, obtaining a corresponding pipeline three-dimensional model according to the pose data passing the verification, and carrying out LOD optimization and multi-factor coupling adaptive deformation compensation on the obtained pipeline three-dimensional model; and performing AR rendering on the pipeline three-dimensional model after LOD optimization and multi-factor coupling adaptive deformation compensation, generating an underground pipeline augmented reality dynamic mapping image, and outputting the image to a user terminal for display. According to the method, submeter-level-precision underground pipeline high-precision dynamic AR visual mapping in a complex environment can be realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of augmented reality (AR), and particularly relates to an underground pipeline augmented reality dynamic mapping method and system. BACKGROUND

[0002] In the scene of urban underground space digital management and pipeline construction and maintenance, underground pipeline augmented reality dynamic mapping is usually achieved through multi-source sensor positioning and AR visualization. However, the positioning error of the traditional GNSS positioning can reach more than 5 meters in a complex urban environment, and the positioning accuracy is insufficient; the existing AR pipeline superposition method has a display offset of more than 20 cm when the device moves, and the virtual-real registration is distorted; the BIM model loading delay is generally more than 500 ms, and the data update lags, which cannot meet the real-time requirement. SUMMARY

[0003] To solve the problems in the prior art, the application provides an underground pipeline augmented reality dynamic mapping method and system, which reduces the multi-source positioning data space-time fusion error, pipeline model dynamic loading delay and virtual-real registration angle error through space-time fusion of multi-source positioning data, dynamic calibration of positioning based on motion state recognition, and real-time multi-factor coupled adaptive deformation compensation of the pipeline three-dimensional model, and realizes high-precision dynamic AR visualization mapping of underground pipelines with sub-meter accuracy in a complex environment.

[0004] The application adopts the following technical solutions.

[0005] The first aspect of the application provides an underground pipeline augmented reality dynamic mapping method, including the following steps: S1: obtaining device pose data, environment data and underground pipeline data respectively; S2: preprocessing the pose data, environment data and underground pipeline data respectively to obtain preprocessed device pose data, environment data and underground pipeline data; S3: dynamically calibrating the preprocessed pose data to obtain dynamically calibrated pose data, and performing data verification with the preprocessed environment data and underground pipeline data, if the verification fails, performing abnormal marking, if the verification passes, entering S4; S4: obtaining the corresponding pipeline three-dimensional model according to the verified pose data and performing LOD optimization and multi-factor coupled adaptive deformation compensation on the obtained pipeline three-dimensional model; S5: performing AR rendering on the pipeline three-dimensional model after the LOD optimization and multi-factor coupled adaptive deformation compensation, generating an underground pipeline augmented reality dynamic mapping image and outputting the image to a user terminal for display.

[0006] Preferably, in S1, the pose data includes device positioning information, pose information, and three-dimensional space, environmental features, and auxiliary positioning data; The environmental data includes environmental image data and magnetic field data; The underground pipeline data includes 3D models of the pipelines, scanned data of underground facilities, and pipeline attribute information.

[0007] Preferably, in S2, spatiotemporal alignment of the pose data includes: performing coordinate system transformation on the data; aligning the clocks of the multi-source sensors using PPS pulse signals to achieve hardware-level synchronization; performing software-level compensation and prediction of the pose information through linear interpolation; and synchronizing the timestamps of the multi-source sensors using a timestamp synchronization model. Denoise the environmental data; Perform topology analysis on underground pipeline data to check and correct abnormal data.

[0008] Preferably, in S3, dynamic calibration is triggered when a pre-stored positioning marker is detected. An improved extended Kalman filter algorithm is used to dynamically calibrate the pose data, generating calibrated pose data. The specific details of using the improved extended Kalman filter algorithm to dynamically calibrate the pose data are as follows:

[0009]

[0010]

[0011] in, , These are the posterior state estimate and the prior state estimate at time k, respectively; The pose data sensor measures the position at time k. Kalman gain; An observation model for a GNSS receiver unit; is the observation model of the IMU sensor; x is the state vector; The observation matrix; It is the magnetic field compensation factor; This represents the original geomagnetic field strength. This is the reference intensity of the geomagnetic field. It is the unit of geomagnetic field strength; It is a 3-order identity matrix; Represents the norm.

[0012] Preferably, the improved extended Kalman filter algorithm adaptively adjusts the window length for data processing based on the device's movement speed, as shown in the formula:

[0013] in, Real-time motion speed; The baseline speed threshold is set according to the type of work scenario. This is the initial window length; This is the floor function.

[0014] Preferably, in S4, the corresponding area's 3D pipeline model is extracted from the underground pipeline database based on the verified pose data; and the data query and loading efficiency is improved through spatial index optimization, progressive transmission, and BIM model lightweighting; the pipeline database triggers incremental updates when the deviation between the pipeline position detected by the lidar and the pipeline position recorded in the database exceeds a threshold.

[0015] Preferably, in S4, the LOD optimization of the pipeline 3D model includes: dynamically adjusting the model accuracy according to the line-of-sight distance, parsing the connection semantics of the model and optimizing and adjusting it.

[0016] Preferably, in S4, the multi-factor coupled adaptive deformation compensation model for the pipeline 3D model is as follows:

[0017] in: The pipeline coordinates are after adaptive deformation compensation due to multi-factor coupling; The basis functions are cubic B-spline functions. These are deformation parameters; Environmental weighting factors are the result of multi-factor coupling. represents the original 3D coordinates of the i-th control point in the pipeline 3D model; n represents the number of control points.

[0018] Preferably, environmental weighting factor The following multi-factor coupling model is used for calculation:

[0019] in, T ref , B ref , p ref An adaptive threshold for scene temperature, magnetic field strength, and humidity; Δ T Δ B Δ p These represent the deviations from the reference values ​​for temperature, magnetic field strength, and humidity, respectively; the sensitivity coefficient α+β+γ=1.

[0020] Preferably, in S5, an AR rendering engine is used to perform virtual-real fusion rendering based on ambient lighting to achieve sub-pixel-level alignment display between the pipeline 3D model and the environmental scene image.

[0021] A second aspect of the present invention provides an augmented reality dynamic mapping system for underground pipelines, comprising a dynamic mapping device, wherein the dynamic mapping device includes a pose data acquisition unit, an environmental data acquisition unit, an underground pipeline database, a processing unit, and an AR unit; The pose data acquisition unit, environmental data acquisition unit, and underground pipeline database are respectively used to acquire pose data, environmental data, and underground pipeline data of the dynamic mapping device; The processing unit is used to preprocess the pose data, environmental data, and underground pipeline data respectively to obtain preprocessed equipment pose data, environmental data, and underground pipeline data; to dynamically calibrate the preprocessed pose data to obtain dynamically calibrated pose data, and to perform data verification together with the preprocessed environmental data and underground pipeline data. If the verification fails, an anomaly is marked. If the verification passes, the corresponding pipeline 3D model is obtained based on the verified pose data, and the obtained pipeline 3D model is subjected to LOD optimization and multi-factor coupled adaptive deformation compensation. The AR unit is used to perform AR rendering on the pipeline 3D model after LOD optimization and multi-factor coupling adaptive deformation compensation, generate underground pipeline augmented reality dynamic mapping images, and output them to the user terminal for display.

[0022] A third aspect of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method.

[0023] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0024] Compared with the prior art, the beneficial effects of the present invention include at least the following: 1. This invention performs spatiotemporal alignment preprocessing on pose data and dynamic calibration on the preprocessed pose data. The spatiotemporal alignment preprocessing includes coordinate system transformation, hardware-level synchronization based on PPS pulse signals, software-level compensation prediction of pose data based on linear interpolation compensation, and timestamp synchronization using a timestamp synchronization model, resulting in a spatiotemporal fusion error of <0.3m for multi-source positioning data. The dynamic calibration uses an improved extended Kalman filter algorithm. When the lidar identifies a pre-stored marker, pose calibration is triggered. By introducing a magnetic field compensation factor, magnetic field interference is suppressed, and the data processing window length is adaptively adjusted according to the device's movement speed, dynamically linking the device's movement state with the data processing window. A tiered allocation of computing resources is achieved through a rounding function, which can adapt to the movement speed characteristics of different operating scenarios and improve stability.

[0025] 2. This invention performs topology analysis preprocessing on underground pipeline data, checks and corrects abnormal data such as connection errors, spatial conflicts, missing or incorrect attributes, and performs real-time incremental updates through data lightweighting and index optimization, so that the dynamic loading delay of the pipeline 3D model is less than 200ms.

[0026] 3. This invention performs LOD optimization on the pipeline 3D model and adaptive deformation compensation based on multi-factor coupling using environmental weight factors. It also performs virtual-real fusion rendering based on ambient lighting, achieving sub-pixel-level alignment between the pipeline 3D model and the environmental scene, resulting in a virtual-real registration angle error of <1°. By calculating environmental weight factors through multi-factor coupling modeling, and simultaneously integrating three key interference sources—temperature, magnetic field, and humidity—it better reflects the actual interference characteristics of complex underground environments. The multi-factor interference intensity is mapped to the [0,1] interval, achieving a smooth transition of the weight factors. Furthermore, the sensitivity coefficient can be dynamically adjusted according to the specific scene, improving compensation accuracy, optimizing computational efficiency, and demonstrating strong engineering feasibility. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, other embodiments obtained by those skilled in the art without creative effort are all within the protection scope of this invention.

[0029] Embodiment 1 of the present invention provides an augmented reality dynamic mapping method for underground pipelines, implemented based on a dynamic mapping device. The dynamic mapping device includes a pose data acquisition unit, an environmental data acquisition unit, an underground pipeline database, a processing unit, and an AR unit, such as... Figure 1 and 2 As shown, the method includes the following steps: S1: Acquire dynamic mapping device pose data, environmental data, and underground pipeline data based on the pose data acquisition unit, environmental data acquisition unit, and underground pipeline database, respectively; (1) The pose data acquisition unit integrates the following multi-source sensors to acquire the spatial pose data of the device: a. GPS module / GNSS receiver unit: Deploy a dual-frequency RTK-GNSS antenna array to support RTK differential positioning (frequency ≥ 10Hz, horizontal error < 2cm). b. IMU sensor / inertial measurement unit: including a three-axis MEMS gyroscope (range ±2000° / s) and an accelerometer (noise density 100μg / √Hz). c. LiDAR unit: Equipped with a 16-line rotating scanning head (angular resolution 0.1°). d.UWB positioning module, as an auxiliary positioning source, adopts TDoA ranging protocol (6-channel antenna array). (2) The environmental data acquisition unit uses an RGBD depth camera to scan environmental data and a magnetic field sensor to acquire magnetic field data.

[0030] (3) The underground pipeline database includes: a. BIM Model Library: Stores 3D models of pipelines in IFC format; b. Point cloud database: contains underground facility scan data; c. Attribute database: Records attributes such as pipe diameter and material.

[0031] S2: The processing unit preprocesses the pose data, environmental data, and underground pipeline data respectively to obtain preprocessed equipment pose data, environmental data, and underground pipeline data; (1) The processing unit performs spatiotemporal alignment on the multi-source positioning data: (1.1) Timestamp synchronization; data synchronization mechanisms include: a. Use PPS pulse signals to align the clocks of multi-source sensors to achieve hardware-level synchronization; b. Timestamp alignment uses linear interpolation to compensate for communication delays. The calculation model is as follows:

[0032] In the formula The timestamp after synchronization (unit: seconds) is used to unify the time base of multi-sensor data; The original timestamp of the GPS module (unit: seconds (s)) is obtained from the NMEA message of the GNSS receiver; The timestamp is a local timestamp for the IMU sensor (unit: seconds (s)). Clock drift exists (typically ±20ppm), and periodic calibration is required. The IMU sampling frequency (unit: Hz) must be matched with the communication baud rate (e.g., a maximum of 400Hz is supported when the baud rate is 115200bps).

[0033] c. Apply motion prediction algorithms to IMU data for software-level compensation prediction, as shown in the following model;

[0034] In the formula The predicted angle at time t (unit: radians (rad)) is used to compensate for sensor delay and improve the real-time performance of attitude estimation. The angle (unit: radians (rad)) is measured at time t-1 as a prediction benchmark, and the sampling frequency is required to be ≥100Hz to prevent integral drift. Instantaneous angular velocity (unit: rad / s) reflects the speed of equipment rotation and requires temperature compensation and zero-bias correction; The sampling time interval (unit: seconds (s)) determines the prediction step size, which must be matched with the sensor sampling rate (e.g., 200Hz corresponds to Δt=0.005s). Angular acceleration (unit: rad / s) 2 This reflects the rate of rotational change, which cannot be ignored in highly dynamic scenarios.

[0035] The present invention also includes a feature matching unit, which performs SIFT feature point comparison to assist in realizing augmented reality dynamic mapping.

[0036] (1.2) The coordinate system transformation chain is as follows: WGS84 (Geocentric Coordinate System) → UTM Coordinate System (Projected Coordinate System) → Local Tangent Plane Coordinate System → Equipment Coordinate System; (2) The processing unit denoises the environmental data.

[0037] (3) The processing unit performs topology analysis on the underground pipeline data, checks and corrects abnormal data such as connection errors, spatial conflicts, missing attributes or errors.

[0038] S3: The processing unit performs dynamic calibration on the preprocessed pose data to obtain dynamically calibrated pose data, and performs data verification together with the preprocessed environmental data and underground pipeline data. If the verification fails, an anomaly is marked. If the verification passes, proceed to S4. S301: The processing unit performs dynamic calibration on the preprocessed pose data through a dynamic calibration engine, wherein the dynamic calibration engine includes: (1) Closed-loop correction module based on underground features: When the lidar identifies a pre-stored positioning marker, pose correction is automatically triggered; the marker recognition threshold is set to 80% similarity. (2) Error compensation unit: The improved extended Kalman filter algorithm (improved EKF algorithm) is used to dynamically compensate for the positioning error of the pose data and generate calibrated pose data; In the standard EKF state update equation, the observation matrix Nonlinear observation function The Jacobian matrix does not consider the impact of geomagnetic disturbances on attitude estimation.

[0039] After the improvement of this invention, in the observation matrix Introducing a magnetic field compensation factor The improved extended Kalman filter algorithm is as follows:

[0040] , These are the posterior state estimate and the prior state estimate at time k, respectively; This is the state vector of the sensor measurement value, i.e., the sensor's measurement result at time k; Kalman gain; An observation model for a GNSS receiver unit; is the observation model of the IMU sensor; x is the state vector; The observation matrix; It is the magnetic field compensation factor; This represents the original geomagnetic field strength. This is the reference intensity of the geomagnetic field. It is the unit of geomagnetic field strength; It is a 3-order identity matrix; Represents the norm.

[0041] In the observation matrix Introducing a magnetic field compensation factor :

[0042]

[0043] in , which is the reference intensity of the geomagnetic field.

[0044] (3) A sliding window optimization algorithm is adopted. The error compensation unit adaptively adjusts the window length for data processing according to the equipment movement speed. The formula is:

[0045] in Real-time motion speed (unit: m / s); The baseline speed threshold (unit: m / s) is set according to the type of work scenario. The initial window length (in seconds); This is the floor function.

[0046] The principle of dynamic adjustment implemented by the above formula is as follows: When the device is moving at high speed (v > v0), increase the window length to improve positioning stability; when the device is moving at low speed (v < v0), decrease the window length to improve response speed. In this formula, when the speed exceeds v0, the window length increases proportionally, but forms a stepped change through the rounding function to avoid frequent adjustments. The recommended parameter values are as follows: Urban road scenario: v0 = 1.2 - 1.8 m / s; Underground pipe gallery scenario: v0 = 0.5 - 0.8 m / s; Recommended range: 2 - 5 seconds.

[0047] S302: The processing unit performs data verification on the dynamically calibrated pose data, preprocessed environmental data, and underground pipeline data. If the dynamically calibrated pose data, preprocessed environmental data, and underground pipeline data match, the verification passes; otherwise, the verification fails.

[0048] For example, if the environmental data corresponds to building image data, while the pose data and underground pipeline data are clearly not data belonging to the building scenario, the verification fails.

[0049] S4: The processing unit obtains the pipeline 3D model of the corresponding area based on the verified pose data and performs multi-level LOD optimization and multi-factor coupled adaptive deformation compensation on the pipeline 3D model, including: S401: Extract the pipeline 3D model of the corresponding area from the underground pipeline database according to the spatial coordinates in the verified pose data; and improve the data query and loading efficiency through spatial index optimization, progressive transmission, and BIM model lightweighting; When the deviation between the pipeline position detected by lidar and the pipeline position recorded in BIM exceeds the threshold Δ, an incremental update of the underground pipeline database is triggered.

[0050] S402: Perform multi-level LOD (Level of Detail) optimization on the pipeline 3D model, including: Dynamically adjusting the model accuracy according to the viewing distance, such as adaptively adjusting the model size or the number of patches according to the viewing distance, and loading the complete BIM model or the simplified wireframe model; It also includes pipeline data processing: Analyzing and adjusting the connection semantics of the BIM model, such as dynamically adjusting the connection relationship between multiple models according to the connection semantics when adjusting the model size to achieve topological relationship reconstruction; When the detected pipeline position deviation exceeds the threshold Δ, an incremental update of the database is triggered.

[0051] S403: Perform multi-factor coupled adaptive deformation compensation on the pipeline 3D model based on the environmental depth information to correct environmental interference. Specifically, improve the accuracy of model deformation through the deformation model based on B-spline and environmental factor compensation (geomagnetic anomaly, temperature drift, multipath effect, etc.). The multi-factor coupled adaptive deformation compensation model is:

[0052] in: The basis functions are cubic B-spline functions. An adaptive environmental weighting factor that is coupled with multiple factors (related to temperature, magnetic field strength, etc.). represents the original 3D coordinates of the i-th control point in the pipeline 3D model; n represents the number of control points. Calculations were performed using a multi-factor coupling model:

[0053] in T ref , B ref , p ref An adaptive threshold for scene temperature, magnetic field strength, and humidity; Δ T Δ B Δ p These represent the deviations of temperature, magnetic field strength, and humidity from the reference values, respectively; the coefficient α+β+γ=1.

[0054] By employing multi-factor coupled modeling (simultaneously integrating three key interference sources—temperature, magnetic field, and humidity—to better reflect the actual interference characteristics of complex underground environments), dynamic weight adaptation (mapping the intensity of multi-factor interference to the [0,1] interval to achieve a smooth transition of weight factors), and scenario adaptability design (the sensitivity coefficients α, β, and γ can be dynamically adjusted according to specific scenarios), the system achieves improved compensation accuracy, optimized computational efficiency, and strong engineering feasibility.

[0055] S5: The AR unit performs AR rendering on the pipeline 3D model after multi-level LOD optimization and multi-factor coupling adaptive deformation compensation, generates AR images, and outputs them to the user terminal for display. The AR unit employs an AR rendering engine to perform virtual-real fusion rendering based on ambient lighting, achieving sub-pixel-level alignment between the pipeline's 3D model and the surrounding environment. Specifically, it utilizes the PnP algorithm to achieve sub-pixel-level registration and geometric alignment.

[0056] Embodiment 2 of the present invention provides an augmented reality dynamic mapping system for underground pipelines, which adopts a progressive architecture of perception layer, data layer, processing layer, rendering layer and application layer, including a dynamic mapping device, which includes a pose data acquisition unit, an environmental data acquisition unit, an underground pipeline database, a processing unit and an AR unit; The pose data acquisition unit, environmental data acquisition unit, and underground pipeline database are respectively used to acquire pose data, environmental data, and underground pipeline data of the dynamic mapping device; More preferably, the multi-source localization module (pose data acquisition unit) of the perception layer includes: a. GPS receiver unit: Deploys a dual-frequency RTK-GNSS antenna array; b. Inertial Measurement Unit: Includes a three-axis gyroscope and an accelerometer; c. LiDAR unit: Equipped with a 16-line rotating scanning head; The environmental sensing module (environmental data acquisition unit) includes: a. RGBD camera: 1080p depth resolution @ 30fps; directly connected to the LiDAR unit via a time synchronization cable; b. Magnetic field sensor: range ±8 Gauss.

[0057] The pipeline database of the data layer includes: a. BIM Model Library: Stores IFC format pipeline data; b. Point cloud database: contains underground facility scan data; c. Attribute database: Records attributes such as pipe diameter and material.

[0058] The processing unit is used to preprocess the pose data, environmental data, and underground pipeline data respectively to obtain preprocessed equipment pose data, environmental data, and underground pipeline data; to dynamically calibrate the preprocessed pose data to obtain dynamically calibrated pose data, and to perform data verification together with the preprocessed environmental data and underground pipeline data. If the verification fails, an anomaly is marked. If the verification passes, the corresponding pipeline 3D model is obtained based on the verified pose data, and multi-level LOD optimization and multi-factor coupled adaptive deformation compensation are performed on the obtained pipeline 3D model. More preferably, the data fusion center (processing unit) of the processing layer includes: a. Spatiotemporal alignment unit: Enables coordinate system transformation; b. Error Compensation Unit: Running the improved EKF algorithm, the error compensation unit can establish a two-way data verification channel with the attribute database. The attribute database can ensure the rationality of the location calculation and the environmental adaptability by providing the geometric and physical properties of the pipeline. c. Feature matching unit: Performs SIFT feature point comparison; d. Multi-factor coupled adaptive deformation compensation module: Based on the B-spline deformation algorithm, the multi-factor coupled adaptive deformation compensation module can receive real-time interference data from the magnetic field sensor.

[0059] The AR unit is used to perform AR rendering on the pipeline 3D model after multi-level LOD optimization and multi-factor coupling adaptive deformation compensation, generate underground pipeline augmented reality dynamic mapping images and output them to the user terminal for display.

[0060] More preferably, the AR rendering engine of the rendering layer includes: a. Geometric Alignment Module: Employs the PnP solver; b. Lighting blending module: Supports HDR environment mapping.

[0061] In practice, the adaptive deformation compensation module with multi-factor coupling can also be deployed in the rendering layer.

[0062] Application layer user terminals include: a. AR head-mounted display: 52° FOV binocular display; b. Mobile control terminal: Equipped with a touch-screen interface; c. Collaboration Server: Supports data synchronization across multiple terminals.

[0063] The following are examples of applications in urban road scenarios: Deploy differential GNSS base stations (210) to establish a centimeter-level positioning reference network; The mobile terminal (100) includes: a Trimble R12 GNSS receiver, an Xsens MTi-680G inertial unit, and a Velodyne VLP-16 lidar; The dynamic calibration engine uses an update frequency of 15 frames per second, and the sliding window length is set to 5 seconds; The AR rendering engine supports multi-user collaborative annotation, and the annotation data is uploaded to the cloud database in real time.

[0064] The following are examples of applications in underground utility tunnel scenarios: Pre-configured UWB positioning base stations as auxiliary positioning sources; Activate the closed-loop correction submodule and set the marker recognition threshold to 80% similarity; The pipeline model loading adopts a progressive transmission strategy, prioritizing the loading of data in the current field of view.

[0065] Tests showed that the positioning accuracy of the present invention reached 0.2m in open areas, and the model loading delay was <200ms, which is significantly better than the existing technology.

[0066] Embodiment 3 of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0067] Embodiment 4 of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0068] Compared with the prior art, the beneficial effects of the present invention include at least the following: 1. This invention performs spatiotemporal alignment preprocessing on pose data and dynamic calibration on the preprocessed pose data. The spatiotemporal alignment preprocessing includes coordinate system transformation, hardware-level synchronization based on PPS pulse signals, software-level compensation prediction of pose data based on linear interpolation compensation, and timestamp synchronization using a timestamp synchronization model, so that the spatiotemporal fusion error of multi-source positioning data is <0.3m. The dynamic calibration adopts an improved extended Kalman filter algorithm. When the lidar identifies a pre-stored marker, pose calibration is triggered. By introducing a magnetic field compensation factor, magnetic field interference is suppressed, and the window length of data processing is adaptively adjusted according to the device's movement speed to improve stability.

[0069] 2. This invention performs topology analysis preprocessing on underground pipeline data, checks and corrects unreasonable data, and through data lightweighting and index optimization, performs real-time incremental updates, so that the dynamic loading delay of the pipeline 3D model is <200ms.

[0070] 3. This invention performs LOD optimization on the pipeline 3D model and adaptive deformation compensation based on multi-factor coupling using environmental weight factors. It also performs virtual-real fusion rendering based on ambient lighting, achieving sub-pixel-level alignment between the pipeline 3D model and the environmental scene, resulting in a virtual-real registration angle error of <1°. By calculating environmental weight factors through multi-factor coupling modeling, and simultaneously integrating three key interference sources—temperature, magnetic field, and humidity—it better reflects the actual interference characteristics of complex underground environments. The multi-factor interference intensity is mapped to the [0,1] interval, achieving a smooth transition of the weight factors. Furthermore, the sensitivity coefficient can be dynamically adjusted according to the specific scene, improving compensation accuracy, optimizing computational efficiency, and demonstrating strong engineering feasibility.

[0071] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0072] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0073] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0074] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for augmented reality dynamic mapping of underground pipelines, characterized in that, Includes the following steps: S1: Acquire the pose data of the dynamic mapping device, environmental data, and underground pipeline data respectively; S2: Preprocess the pose data, environmental data, and underground pipeline data respectively to obtain preprocessed equipment pose data, environmental data, and underground pipeline data; S3: Perform dynamic calibration on the preprocessed pose data to obtain dynamically calibrated pose data, and perform data verification together with the preprocessed environmental data and underground pipeline data. If the verification fails, an anomaly label is made. If the verification passes, proceed to S4. S4: Obtain the corresponding 3D model of the pipeline based on the verified pose data, and perform LOD optimization and multi-factor coupling adaptive deformation compensation on the obtained 3D model of the pipeline. S5: Perform AR rendering on the 3D pipeline model after LOD optimization and multi-factor coupling adaptive deformation compensation, generate augmented reality dynamic mapping images of underground pipelines, and output them to the user terminal for display.

2. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S1, the pose data includes device positioning information, pose information, and three-dimensional space, environmental features, and auxiliary positioning data; The environmental data includes environmental image data and magnetic field data; The underground pipeline data includes 3D models of the pipelines, scanned data of underground facilities, and pipeline attribute information.

3. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S2, the pose data is spatiotemporally aligned, including: performing coordinate system transformation on the data; aligning the clocks of the multi-source sensors using PPS pulse signals to achieve hardware-level synchronization; performing software-level compensation and prediction of pose information through linear interpolation; and synchronizing the timestamps of the multi-source sensors using a timestamp synchronization model. Denoise the environmental data; Perform topology analysis on underground pipeline data to check and correct abnormal data.

4. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S3, dynamic calibration is triggered when a pre-stored positioning marker is detected. An improved extended Kalman filter algorithm is used to dynamically calibrate the pose data, generating calibrated pose data. The specific details of using the improved extended Kalman filter algorithm to dynamically calibrate the pose data are as follows: in, , These are the posterior state estimate and the prior state estimate at time k, respectively; The pose data sensor measures the position at time k. Kalman gain; An observation model for a GNSS receiver unit; is the observation model of the IMU sensor; x is the state vector; The observation matrix; It is the magnetic field compensation factor; This represents the original geomagnetic field strength. This is the reference intensity of the geomagnetic field. It is the unit of geomagnetic field strength; It is a 3-order identity matrix; Represents the norm.

5. The augmented reality dynamic mapping method for underground pipelines according to claim 4, characterized in that: The improved extended Kalman filter algorithm adaptively adjusts the window length for data processing based on the device's movement speed, using the following formula: in, Real-time motion speed; The baseline speed threshold is set according to the type of work scenario. This is the initial window length; This is the floor function.

6. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S4, the corresponding area's 3D pipeline model is extracted from the underground pipeline database based on the verified pose data; and the data query and loading efficiency is improved through spatial index optimization, progressive transmission, and BIM model lightweighting; the pipeline database triggers incremental updates when the deviation between the pipeline position detected by the lidar and the pipeline position recorded in the database exceeds a threshold.

7. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S4, LOD optimization is performed on the pipeline 3D model, including: dynamically adjusting the model accuracy based on the line of sight, parsing the connection semantics of the model and optimizing and adjusting it.

8. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S4, the adaptive deformation compensation model for multi-factor coupling of the pipeline 3D model is as follows: in: These are the pipeline coordinates after deformation compensation; The basis functions are cubic B-spline functions. These are deformation parameters; For adaptive environmental weighting factors that are coupled with multiple factors; represents the original 3D coordinates of the i-th control point in the pipeline 3D model; n represents the number of control points.

9. The augmented reality dynamic mapping method for underground pipelines according to claim 8, characterized in that: Adaptive environmental weighting factors with multi-factor coupling The calculation model is as follows: in, T ref , B ref , ρ ref An adaptive threshold for scene temperature, magnetic field strength, and humidity; Δ T Δ B Δ ρ These represent the deviations from the reference values ​​for temperature, magnetic field strength, and humidity, respectively; the sensitivity coefficient α+β+γ=1.

10. The augmented reality dynamic mapping method for underground pipelines according to claim 1, characterized in that: In S5, an AR rendering engine is used to perform virtual-real fusion rendering based on ambient lighting, achieving sub-pixel-level alignment between the pipeline 3D model and the environmental scene image.

11. An augmented reality dynamic mapping system for underground pipelines, used to implement the method described in any one of claims 1-10, characterized in that, The system includes a dynamic mapping device, which includes a pose data acquisition unit, an environmental data acquisition unit, an underground pipeline database, a processing unit, and an AR unit. The pose data acquisition unit, environmental data acquisition unit, and underground pipeline database are respectively used to acquire pose data of the dynamic mapping device, environmental data, and underground pipeline data; The processing unit is used to preprocess the pose data, environmental data and underground pipeline data respectively to obtain preprocessed equipment pose data, environmental data and underground pipeline data; The preprocessed pose data is dynamically calibrated to obtain dynamically calibrated pose data, which is then verified together with the preprocessed environmental data and underground pipeline data. If the verification fails, anomaly labeling is performed. If the verification passes, the corresponding pipeline 3D model is obtained based on the verified pose data, and the obtained pipeline 3D model is optimized for LOD and adaptive deformation compensation with multi-factor coupling is performed. The AR unit is used to perform AR rendering on the pipeline 3D model after LOD optimization and multi-factor coupling adaptive deformation compensation, generate underground pipeline augmented reality dynamic mapping images, and output them to the user terminal for display.

12. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-10.

Images