Beidou and MARG integrated underground pipeline detection positioning system and method

Abstract

The application provides a Beidou and MARG fused underground pipeline detection positioning system and method, and belongs to the field of underground pipeline detection technology.The system comprises a handheld pipeline detector, a Beidou multi-source fusion positioning module, a mobile device and a GNSS antenna.The nine-axis MARG sensor is used to identify the motion mode of the device in real time, and the Beidou / MARG fusion positioning algorithm considering the flexible coupling motion of the human and the instrument is combined to synchronously trigger the Beidou positioning when the pipeline detector senses the signal peak value, so that the real-time collection and fusion of the pipeline plane position, buried depth and strike data are realized.The LSTM network is used for motion discrimination, and the circular fitting positioning method is used to optimize the pipeline center solution in the quasi-static state.The application solves the problem that the traditional method needs to separate the measurement and post-processing data, significantly simplifies the operation process, and improves the dynamic measurement accuracy and efficiency in the complex urban environment.

Description

Technical Field

[0001] This invention relates to the field of underground pipeline detection technology, and in particular to an underground pipeline detection and positioning system and method integrating BeiDou and MARG. Background Technology

[0002] Underground urban pipelines, including those for water supply, rainwater, sewage, gas, electricity, fiber optic cables, and communications, are an important component of urban infrastructure. The information required for pipeline measurement includes pipeline type, burial depth, direction, coordinates in the CGCS2000 coordinate system, manhole cover dimensions, etc., and must be collected and stored in a database according to prescribed formats. Among these, burial depth, direction, and coordinates in the CGCS2000 coordinate system require measurement using specialized instruments.

[0003] The most commonly used instrument for pipeline detection is the handheld pipeline detector, which can effectively detect underground pipeline data. The handheld pipeline detector receives reflected signals from the pipeline and displays the pipeline's burial depth and direction on the screen. It provides sound and vibration alerts to the operator regarding changes in the signal field strength, allowing the surveyor to determine the most accurate pipeline location. After determining the location, the surveyor marks the location and uses GNSS-RTK equipment to measure the planar position, obtaining the pipeline's coordinates in the high-precision CGCS2000 coordinate system.

[0004] Accurately determining pipeline distribution using handheld pipeline detectors and GNSS-RTK equipment generally involves three steps.

[0005] The first step is to use a handheld pipeline detector to detect and measure the pipeline's burial depth and direction parameters on the ground. Combined with existing data or manual exploration, the pipeline type, function, manhole cover size, attribute information, and calibration data of the unit to which it belongs are uniformly entered into the pipeline detection and recording software, and the pipeline location is marked on the ground.

[0006] The second step is to use GNSS-RTK equipment to accurately measure the above-mentioned ground calibration points to obtain the plane coordinates X and Y in the CGCCS2000 coordinate system;

[0007] The third step involves data integration and entry. Planar coordinates in the CGCCS2000 coordinate system are exported from the GNSS-RTK equipment's handheld device, and information such as pipeline burial depth, routing parameters, and pipeline type are exported from the pipeline detection and recording software. All data and information are then manually entered into a unified manner using a computer, and relevant data and report documents are compiled.

[0008] Currently, the method for detecting underground pipelines still relies on a combination of handheld pipeline detectors and GNSS-RTK satellite positioning receivers to detect the three-dimensional coordinates of underground pipelines. This requires two measurement operations to integrate the three-dimensional coordinates of the pipeline (planar position X, Y in the CGCCS2000 coordinate system + burial depth H). This method is inefficient and has poor accuracy, resulting in cumbersome procedures and huge consumption of manpower and resources in actual measurement work. Summary of the Invention

[0009] The purpose of this invention is to provide an underground pipeline detection and positioning system and method that integrates BeiDou and MARG, which solves the problem that traditional pipeline detection data acquisition requires post-processing and integration of pipeline detector data and GNSS data, simplifies the underground pipeline detection operation procedure, and improves the accuracy of handheld underground pipeline dynamic measurement in complex urban environments.

[0010] To achieve the above objectives, this invention provides a method for underground pipeline detection and positioning that integrates BeiDou and MARG, comprising the following steps:

[0011] Step S1: Connect the BeiDou multi-source fusion positioning module to the handheld pipeline detector. When the handheld pipeline detector detects the peak value of the pipeline sensing signal, it triggers the BeiDou multi-source fusion positioning module to synchronously collect the raw data of the MARG sensor and receive the BeiDou ground-based differential augmentation data through the GNSS positioning antenna.

[0012] Step S2: The BeiDou multi-source fusion positioning module executes the BeiDou / MARG fusion pipeline planar position positioning algorithm, which takes into account the flexible coupling motion between the person and the pipeline detector, to achieve the positioning of the underground pipeline planar position;

[0013] Step S3: Simultaneously collect, input, and store pipeline detection data and BeiDou positioning data using pipeline detection data acquisition software in the mobile device.

[0014] Preferably, in step S2, the pipeline planar location algorithm based on the BeiDou / MARG fusion, which takes into account the flexible coupling motion between the person and the pipeline detector, includes:

[0015] Step S21: Transform the raw data from the MARG sensor into a global coordinate system centered on the inertial navigation system using a rotation matrix, and preprocess the IMU data using a Vondark low-pass filter to separate high-frequency noise;

[0016] Step S22: Construct a MARG attitude fusion self-calibration algorithm based on quaternions, gyroscope zero bias, and accelerometer zero bias as state parameters, construct measurement equations with acceleration and magnetic heading angle, and calibrate the MARG sensor zero bias error.

[0017] Step S23: Perform time-domain feature reduction on the data from the BeiDou satellite positioning sensor and the nine-axis MARG sensor, and classify them into three motion modes: static, quasi-static, and swaying.

[0018] Step S24: Based on the LSTM network, construct the cross-entropy cost function for motion mode discrimination.

[0019] Preferably, in step S21, the raw data from the MARG sensor is transformed to a global coordinate system centered on the inertial navigation system using a rotation matrix, as follows:

[0020] ;

[0021] in, This represents the rotation matrix from the sensor coordinate system to the navigation coordinate system of the BeiDou multi-source fusion positioning module. , and Both represent Euler rotation angles. , , All represent sensor observations in the navigation coordinate system. , , All of these represent sensor observations in the carrier coordinate system.

[0022] Preferably, the IMU data is preprocessed using a Vondark low-pass filter to separate high-frequency noise. The specific process is as follows:

[0023] Raw IMU data was acquired using the sliding window method. :

[0024] ;

[0025] in, , , , , , These represent the output values ​​of the IMU's three-axis accelerometer and three-axis gyroscope, respectively.

[0026] Constructing a second-order difference matrix Used to represent changes in curvature;

[0027] The NumPy function from the linear algebra library is used to solve the system of linear equations to obtain smoothed IMU data. :

[0028] ;

[0029] in, Represents the identity matrix. Indicates the smoothing parameter. This indicates transpose.

[0030] Preferably, motion mode discrimination includes:

[0031] When the pipeline detector is determined to be static, record the current position;

[0032] When the pipeline detector is determined to be quasi-static, the circle fitting positioning method is used to obtain the pipeline center position;

[0033] When the pipeline detector is determined to be shaking, positioning data from the BeiDou multi-source fusion positioning module is not acquired.

[0034] This invention also provides an underground pipeline detection and positioning system integrating BeiDou and MARG, comprising:

[0035] Handheld pipeline detectors are used to detect the burial depth and direction of underground pipelines;

[0036] The BeiDou multi-source fusion positioning module is fixedly connected to the handheld pipeline detector and includes a BeiDou satellite positioning sensor and a nine-axis MARG sensor.

[0037] The mobile device connects to the BeiDou multi-source fusion positioning module via a serial cable and has built-in pipeline detection data acquisition software.

[0038] The GNSS positioning antenna is mounted on the top of the handheld pipeline detector and is connected to the BeiDou multi-source fusion positioning module via an RF cable.

[0039] The positioning system uses a BeiDou / MARG fusion pipeline planar position positioning algorithm that takes into account the flexible coupling motion of the person and the pipeline detector to obtain the pipeline planar position in real time, and collects and stores the data synchronously with the handheld pipeline detector.

[0040] Preferably, the feature is the integration of pipeline detection data acquisition software:

[0041] The motion modality classification unit identifies three states—static, quasi-static, and swaying—based on data from a nine-axis MARG sensor.

[0042] The positioning trigger unit triggers BeiDou positioning when the pipeline detector senses a peak signal.

[0043] The data fusion unit integrates pipeline burial depth, direction, and planar coordinate data in real time.

[0044] Therefore, the underground pipeline detection and positioning system and method integrating BeiDou and MARG as described above have the following beneficial technical effects:

[0045] (1) Efficient data acquisition and fusion: By connecting the Beidou multi-source fusion positioning module with the handheld pipeline detector, the raw data of the MARG sensor and the Beidou ground-based differential augmentation data are collected simultaneously when the peak value of the pipeline sensing signal is detected. This realizes the real-time acquisition and fusion of pipeline plane position, burial depth and direction data, avoiding the cumbersome steps of post-processing and integration of pipeline detector data and GNSS data in the traditional method, and improving the data acquisition efficiency.

[0046] (2) High-precision positioning: The pipeline plane position positioning algorithm that integrates Beidou / MARG takes into account the flexible coupling motion of people and pipeline detectors can effectively solve the impact of human movement on positioning accuracy. Combined with LSTM network for motion discrimination and the circle fitting positioning method for quasi-static state to optimize pipeline center calculation, the dynamic measurement accuracy in complex urban environments is significantly improved, and the positioning results are more accurate and reliable.

[0047] (3) Intelligent motion mode recognition: The motion mode of the equipment is identified in real time using a nine-axis MARG sensor. By constructing a MARG attitude fusion self-calibration algorithm based on quaternions, gyroscope zero bias, and accelerometer zero bias as state parameters and classifying motion modes, the static, quasi-static, and shaking states of the pipeline detector are accurately identified, providing a basis for precise positioning. Attached Figure Description

[0048] Figure 1 This is a structural diagram of the underground pipeline detection and positioning system integrating BeiDou and MARG of the present invention;

[0049] Figure 2 This is a flowchart of the underground pipeline detection and positioning method integrating BeiDou and MARG according to the present invention. Detailed Implementation

[0050] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0052] Example 1

[0053] like Figure 1 As shown, the underground pipeline detection and positioning system integrating BeiDou and MARG includes:

[0054] Handheld pipeline detectors are used to detect the burial depth and direction of underground pipelines;

[0055] The BeiDou multi-source fusion positioning module is fixedly connected to the handheld pipeline detector and includes a BeiDou satellite positioning sensor and a nine-axis MARG sensor.

[0056] The mobile device (Android) connects to the Beidou multi-source fusion positioning module via a serial cable and has built-in pipeline detection data acquisition software.

[0057] The GNSS positioning antenna (miniaturized GNSS positioning antenna) is installed on the top of the handheld pipeline detector and connected to the Beidou multi-source fusion positioning module via an RF cable.

[0058] Among them, the underground pipeline detection and positioning system uses a BeiDou / MARG fusion pipeline plane position positioning algorithm that takes into account the flexible coupling motion of people and pipeline detectors to obtain the pipeline plane position in real time, and collects and stores the data synchronously with the handheld pipeline detector.

[0059] Pipeline detection data acquisition software integration:

[0060] The motion modality classification unit identifies three states—static, quasi-static, and swaying—based on data from a nine-axis MARG sensor.

[0061] The positioning trigger unit triggers BeiDou positioning when the pipeline detector senses a peak signal.

[0062] The data fusion unit integrates pipeline burial depth, direction, and planar coordinate data in real time.

[0063] Example 2

[0064] like Figure 2 As shown, the underground pipeline detection and positioning method integrating BeiDou and MARG performs the following steps:

[0065] Step S1: Connect the BeiDou multi-source fusion positioning module to the handheld pipeline detector. When the handheld pipeline detector detects the peak value of the pipeline sensing signal, it triggers the BeiDou multi-source fusion positioning module to synchronously collect the raw data of the MARG sensor and receive the BeiDou ground-based differential augmentation data through the miniaturized GNSS positioning antenna.

[0066] Step S2: The BeiDou multi-source fusion positioning module executes the BeiDou / MARG fusion pipeline planar position positioning algorithm, which takes into account the flexible coupling motion between the person and the pipeline detector, to achieve the positioning of the underground pipeline planar position.

[0067] A BeiDou / MARG fusion pipeline planar positioning algorithm that takes into account the flexible coupling motion between humans and pipeline detectors includes:

[0068] Step S21: Transform the raw data from the MARG sensor to a global coordinate system centered on the inertial navigation system using a rotation matrix, as follows:

[0069] ;

[0070] in, This represents the rotation matrix from the sensor coordinate system to the navigation coordinate system of the BeiDou multi-source fusion positioning module. , and Both represent Euler rotation angles. , , All represent sensor observations in the navigation coordinate system. , , All of these represent sensor observations in the carrier coordinate system.

[0071] Rotation matrix By the definition of quaternions:

[0072] ;

[0073] in, , , , Both represent quaternions.

[0074] The IMU data is preprocessed using a Vondark low-pass filter to separate high-frequency noise. The specific process is as follows:

[0075] Raw IMU data was acquired using the sliding window method. :

[0076] ;

[0077] in, , , , , , These represent the output values ​​of the IMU's three-axis accelerometer and three-axis gyroscope, respectively.

[0078] Constructing a second-order difference matrix Used to represent changes in curvature;

[0079] The NumPy function from the linear algebra library is used to solve the system of linear equations to obtain smoothed IMU data. :

[0080] ;

[0081] in, Represents the identity matrix. Indicates the smoothing parameter. This indicates transpose.

[0082] Step S22: Construct a MARG attitude fusion self-calibration algorithm based on quaternions, gyroscope zero bias, and accelerometer zero bias as state parameters, construct measurement equations with acceleration and magnetic heading angle, and calibrate the MARG sensor zero bias error.

[0083] The system state equation is:

[0084] ;

[0085] in, Represents the state vector. This indicates that the gyroscope measures angular velocity. This indicates process noise.

[0086] ;

[0087] in, This represents noise in the quaternion process. and Both represent the zero-bias process noise covariance matrix. Represents the zero vector. Representing quaternions, This indicates that the gyroscope has zero bias. This indicates that the accelerometer has zero bias. This represents a skew-symmetric matrix.

[0088] The system measurement equation is:

[0089] ;

[0090] in, Represents the measurement vector. This indicates measurement noise.

[0091] ;

[0092] in, Represents a known geomagnetic field vector. Represents the gravitational acceleration matrix. Indicates accelerometer noise. This indicates environmental magnetic field interference noise.

[0093] Step S23: Perform time-domain feature reduction on the data from the BeiDou satellite positioning sensor and the nine-axis MARG sensor, and classify them into three motion modes: static, quasi-static, and swaying.

[0094] Let the time window length be The number of sampling points is ,but:

[0095] Beidou positioning sensor data matrix for:

[0096] ;

[0097] MARG Data Matrix Represented as:

[0098] ;

[0099] in, , , Represents acceleration in three directions. , , Represents the angular velocity in three directions. , , This indicates the magnetic field strength in three directions.

[0100] Calculate for each dimension Time-domain features (with acceleration) (For example)

[0101] ;

[0102] in, Indicates the first The acceleration of each sampling point This represents the average value of acceleration.

[0103] Constructing 14 channels 3D feature matrix for:

[0104] ;

[0105] in, , These represent the time-domain characteristics of BeiDou positioning sensor data and the time-domain characteristics of MARG data, respectively.

[0106] Step S24: Based on the LSTM network, construct the cross-entropy cost function for motion mode discrimination.

[0107] Forgotten Gate: ;

[0108] Input Gate: ;

[0109] Candidate memories: ;

[0110] Memory update: ;

[0111] Output gate: ;

[0112] Hidden state: ;

[0113] in, Indicates the current time step The output of the forget gate, Indicates the current time step The input gate output, Indicates the current time step Candidate memories, Indicates the current time step Memory cells, Indicates the previous time step Memory cells, Indicates the current time step The output gate output, Indicates the current time step The hidden state, Indicates the previous time step The hidden state, Indicates the current time step. This represents the features after SLPP dimensionality reduction. This represents the hidden state dimension. , , , Both represent weight matrices. , , , Both represent bias terms. This represents the sigmoid function. This indicates element-wise multiplication.

[0114] The output layer is categorized, and the hidden state at the last time step is retrieved. Perform category prediction:

[0115] ;

[0116] in, This represents the unnormalized scores for the three sports categories. This represents the output layer weight matrix. This represents the output layer bias vector. , .

[0117] Hidden state of the last time step of the LSTM After applying the softmax activation function to the fully connected layer, the probability distributions of the three motion modes are obtained. :

[0118] ;

[0119] in, , , These correspond to the probabilities of being static, quasi-static, and swaying, respectively.

[0120] The cross-entropy cost function for the true label (one-hot encoding is) loss function for:

[0121] ;

[0122] in, Indicates category index, Indicates the first One-hot encoding of each label, Indicates the first The probability distribution of each label;

[0123] Batch loss:

[0124] ;

[0125] in, Indicates batch size, Indicates the sample index.

[0126] The cross-entropy loss function adjusts the LSTM parameters through backpropagation to improve the predicted probability. Approaching real labels .

[0127] A. When the pipeline detector is determined to be static, record the current position;

[0128] B. When the pipeline detector is determined to be quasi-static, the circle fitting positioning method is used to obtain the pipeline center position.

[0129] For each BeiDou multi-source fusion positioning module measurement point when the pipeline detector is in a quasi-static state Define the target error function as:

[0130] ;

[0131] in, Indicates the first Error of the measurement , Indicates the total number of measurements. and It is the radius of the circle (i.e., the center position of the pipeline). Let represent the radius. Minimizing the sum of squared errors across all measurement points using the objective function yields:

[0132] ;

[0133] Linearize the error circle equation to construct a matrix equation:

[0134] ;

[0135] Solving the above matrix equation using the least squares method yields... and The estimated value. Ultimately, This is the optimal estimated coordinate of the measured point.

[0136] C. When the pipeline detector is determined to be shaking, the positioning data of the Beidou multi-source fusion positioning module is not acquired.

[0137] Step S3: Simultaneously collect, input, and store pipeline detection data and BeiDou positioning data using pipeline detection data acquisition software in the mobile device.

[0138] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0139] Therefore, the present invention adopts the above-mentioned underground pipeline detection and positioning system and method that integrates Beidou and MARG, which solves the problem that traditional pipeline detection data acquisition requires post-processing and integration of pipeline detector data and GNSS data, simplifies the underground pipeline detection operation procedure, and improves the accuracy of handheld underground pipeline dynamic measurement in complex urban environments.

[0140] 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 them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting and positioning underground pipelines by fusing Beidou and MARG, characterized in that, Comprise the following steps: Step S1, the Beidou multi-source fusion positioning module is fixed with the handheld pipeline detector, when the handheld pipeline detector detects the pipeline induction signal peak value, the Beidou multi-source fusion positioning module synchronous acquisition MARG sensor original data is triggered, through GNSS positioning antenna receives Beidou ground-based differential enhancement data; Step S2, the Beidou multi-source fusion positioning module executes the Beidou / MARG fusion pipeline plane position positioning algorithm considering the flexible coupling motion of human and pipeline detector, realizes the positioning of underground pipeline plane position; Step S3, through the pipeline detection data acquisition software in the mobile device, pipeline detection data and Beidou positioning data are synchronously collected, input and stored; In step S2, the Beidou / MARG fusion pipeline plane position positioning algorithm considering the flexible coupling motion of human and pipeline detector, comprises: Step S21, the original data of MARG sensor is converted to the global coordinate system with inertial navigation as the center through the rotation matrix, and the IMU data is preprocessed by Vondark low-pass filter to separate high-frequency noise; Step S22, a MARG attitude fusion self-calibration algorithm based on quaternion, gyro zero offset and accelerometer zero offset as state parameters is constructed, a measurement equation is constructed by acceleration and magnetic heading angle, and MARG sensor zero offset error is calibrated; Step S23, the data of Beidou satellite positioning sensor and nine-axis MARG sensor are time domain feature reduced, and classified into three types of motion modes: static, quasi-static and shaking; Step S24, based on LSTM network, a cross-entropy cost function is constructed to distinguish motion modes; The IMU data is preprocessed by Vondark low-pass filter to separate high-frequency noise, and the specific process is as follows: Collecting original IMU data by using a sliding window method : ； wherein, , , , , , respectively represent output values of a three-axis accelerometer and a three-axis gyroscope of the IMU. Constructing a second difference matrix for representing a change in curvature; The linear algebra library NumPy functions are referenced to solve the linear equations and obtain the smoothed IMU data : ； wherein denotes the identity matrix, denotes a smoothing parameter, denotes the transpose; Motion mode discrimination includes: When the pipeline detector is judged to be static, the current position is recorded; When the pipeline detector is judged to be quasi-static, the pipeline center position is obtained by using circle fitting positioning method; For each Beidou multi-source fusion positioning module measurement point of the pipeline detector in quasi-static state The target error function is defined as: ； wherein, denotes the error of the first measurement, , denotes the total number of measurements, denotes the radius; the objective function to be minimized is the sum of the error squares of all measurements: ； The error circle equation is linearized to construct a matrix equation: ； The least square method is used to solve the above matrix equation, and the estimated values of and are obtained; finally, is the optimal estimated coordinate of the measured point. When the pipeline detector is judged to be shaking, the positioning data of the Beidou multi-source fusion positioning module is not obtained.

2. The Beidou and MARG integrated underground pipeline detection positioning method according to claim 1, characterized in that, In step S21, the original data of MARG sensor is converted to the global coordinate system with inertial navigation as the center through the rotation matrix, as follows: ； wherein, represents a rotation matrix from the sensor coordinate system of the multi-source fusion positioning module of Beidou to the navigation coordinate system, , and all represent Euler rotation angles, , , all represent sensor observation values in the navigation coordinate system, , , all represent sensor observation values in the carrier coordinate system.

3. A system for detecting and locating underground pipelines using a combination of Beidou and MARG, characterized in that, For executing the Beidou and MARG fusion underground pipeline detection positioning method as claimed in any one of claims 1-2, comprising: The handheld pipeline detector is used for detecting the buried depth and trend of underground pipeline; The Beidou multi-source fusion positioning module is fixed with the handheld pipeline detector, and contains Beidou satellite positioning sensor and nine-axis MARG sensor; The mobile device is connected with the Beidou multi-source fusion positioning module through serial port line, and is built-in pipeline detection data acquisition software; The GNSS positioning antenna is installed on the top of the handheld pipeline detector, and is connected with the Beidou multi-source fusion positioning module through radio frequency cable; Wherein, the positioning system obtains the pipeline plane position in real time through the Beidou / MARG fusion pipeline plane position positioning algorithm considering the flexible coupling motion of human and pipeline detector, and synchronously collects the handheld pipeline detector data into the database.

4. The Beidou and MARG integrated underground pipeline detection and positioning system according to claim 3, characterized in that, The pipeline detection data acquisition software integrates: The motion modal classification unit identifies three states of static, quasi-static and shaking according to the nine-axis MARG sensor data; The positioning trigger unit triggers the Beidou positioning when the pipeline detector senses the peak value of the signal; The data fusion unit integrates the pipeline burial depth, strike and plane coordinate data in real time.

Images