Yaw angle measurement and correction method and device for underground pipeline detector
By combining single IMU gyroscope integration and laser spot displacement in the underground pipeline inspection instrument, a screen coordinate system is established and real-time compensation is performed, which solves the problems of yaw angle drift and observation instability, and realizes the stability of yaw angle measurement and the accuracy of three-dimensional trajectory reconstruction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-04
- Publication Date
- 2026-03-27
AI Technical Summary
In the existing technology, the yaw angle measurement of underground pipeline inspection instruments is easily affected by MEMS gyroscope drift. Furthermore, under conditions such as miniaturization, assembly errors, and camera shake, the observation of the image sensor is unstable, resulting in poor yaw angle correction effect.
The yaw angle integral of a single IMU gyroscope is used as the main measured value. The yaw angle observation obtained by laser spot displacement-baseline geometry is used as an external constraint. By establishing a screen coordinate system and performing real-time compensation, and using complementary filtering or Kalman filtering for correction, the stability of the yaw angle measurement is improved.
It effectively suppressed yaw angle drift, improved the stability of 3D trajectory reconstruction, reduced the impact of coordinate system instability and camera shake on measurement, and enhanced robustness under varying lighting and exposure conditions.
Smart Images
Figure CN121739969A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of pipeline spatial positioning and attitude measurement technology, and in particular relates to a yaw angle measurement and correction method and device for underground pipeline inspection instruments. Background Technology
[0002] In underground pipeline inspection and 3D trajectory reconstruction, attitude angles (especially yaw angles) are key quantities for correlating mileage information with spatial direction. In engineering, the yaw angle is often obtained by integrating the angular velocity of an IMU gyroscope. The inertial measurement unit (IMU) described in this invention can be an integrated MEMS inertial measurement chip or module, typically containing at least a three-axis gyroscope and a three-axis accelerometer, and optionally integrating a three-axis magnetometer and a temperature sensor, for outputting measurement data such as angular velocity, acceleration, and magnetic field.
[0003] However, MEMS gyroscopes have errors such as zero bias and random walk, and their angular velocity will drift as it integrates over time, with the yaw angle drift usually being the most significant. This is the fundamental reason why it is difficult to output a stable yaw angle over a long period of time by relying solely on gyroscope integration.
[0004] For pitch and roll angles, when the instrument is stationary or quasi-stationary and the linear acceleration is small, the gravitational acceleration vector measured by the triaxial accelerometer can be used as an absolute reference to correct for pitch and roll angles. However, yaw angle is a rotation about the direction of gravity, and the accelerometer's measurement of the gravity vector remains unchanged under this rotation. Therefore, the gravitational accelerometer cannot provide an absolute observation of the yaw angle; under single IMU conditions, the yaw angle still mainly relies on gyro integration, which is prone to long-term drift.
[0005] Existing technologies also include methods that utilize image sensors to acquire laser spot displacement and fuse it with IMU information. Furthermore, some schemes use image sensors to acquire laser spot displacement and calculate the pipe deflection / yaw angle observations accordingly. Some schemes further fuse this visual angle observation with the IMU-calculated angle using weighted or Kalman gain to obtain a more stable angle estimate. However, under conditions of miniaturization of pipe inspection instruments, assembly errors, slight camera shake, and changes in lighting or reflection, these schemes are prone to problems such as unstable correspondence between image coordinates and screen physical coordinates, and anomalies in spot extraction due to deformation and exposure. These issues lead to decreased reliability of the yaw angle observation, thus affecting the correction effect of the single IMU yaw angle. Summary of the Invention
[0006] The purpose of this invention is to solve the problems mentioned in the background art and to propose a single IMU yaw angle measurement and correction method and device for underground pipeline inspection instruments: using the yaw angle integral of the single IMU gyroscope as the main measured value, and introducing the yaw angle observation obtained by laser spot displacement-baseline geometry as an external constraint, to achieve periodic or continuous suppression of yaw angle drift and improve the stability of three-dimensional trajectory reconstruction, so as to solve engineering problems such as the difficulty in establishing a stable coordinate system under miniaturized assembly conditions, and the unstable observation caused by camera jitter and spot deformation.
[0007] To achieve the objective of this invention, a yaw angle measurement and correction method for underground pipeline inspection instruments is disclosed, comprising the following steps:
[0008] Step 1: Acquire gyroscope data containing at least the yaw axis angular velocity using a single inertial measurement unit (IMU), and perform time integration on the angular velocity to obtain the main measured value of the yaw angle;
[0009] Step 2: Acquire image frames from the camera onto the imaging screen, which includes a central laser spot and five positioning points distributed around the perimeter of the imaging screen.
[0010] Step 3: Detect five positioning points in each image frame, and use the stable geometric relationship of the five positioning points to establish or update the screen coordinate system, and calculate the transformation relationship between camera shake, setup deviation or image coordinate system and screen coordinate system.
[0011] Step 4: Extract the center pixel coordinates of the laser spot in each image frame, convert the spot pixel coordinates into screen coordinates based on the transformation relationship obtained in step (3), and obtain the spot displacement with the spot position of the reference frame or reference time period as the zero point.
[0012] Step 5: Based on the calibration parameters from pixel scale to physical scale, convert the spot screen displacement into physical displacement, and establish a geometric relationship with the fixed baseline distance from the laser to the imaging screen to calculate the yaw angle observation.
[0013] Step 6: Use the yaw angle observation to periodically or continuously correct the yaw angle main measurement value obtained in Step 1, and output the corrected yaw angle.
[0014] Furthermore, in step 1, the yaw angle master measurement value is obtained through quaternion attitude update or equivalent integration, and the quaternion is normalized to suppress the accumulation of numerical error.
[0015] Furthermore, in step 3, the origin of the screen coordinate system is taken as the geometric center of the imaging screen or the equivalent center determined by the five positioning points, and the direction of the screen coordinate axis is determined by the zero orientation defined by at least one of the five positioning points; the two-dimensional similarity transformation or homography transformation is obtained using the five positioning points to align the current frame coordinates to the reference frame coordinates, thereby compensating for camera shake and assembly deviation in real time.
[0016] Furthermore, in step 4, the center pixel coordinates of the light spot are calculated using the brightness-weighted centroid method, and after threshold segmentation, morphological processing and connected component screening are combined to remove false targets; and when light spot saturation diffusion or deformation is detected, the centroid of the core region or the ellipse fitting center is used as the center of the light spot.
[0017] Furthermore, in steps 2 to 5, when the geometric constraints of the five positioning points do not meet the preset threshold or the spot extraction fails, or when the spot morphology index is deformed and exceeds the threshold, the corresponding image frame is determined to be an invalid frame and the yaw angle correction is not updated.
[0018] Furthermore, in step 5, the calibration from pixel scale to physical scale is obtained through a calibration ring on the imaging screen or the spacing between positioning points at a known distance; the yaw angle observation is calculated by the horizontal component of the physical displacement of the light spot and the distance to the fixed baseline according to the arctangent relationship, or by using a linear approximation relationship under small angle conditions.
[0019] Furthermore, in step 6, complementary filtering or Kalman filtering is used for correction, and the gyroscope zero bias is used as the state quantity estimate to further suppress yaw angle drift.
[0020] Furthermore, the weight of the yaw angle observation is adaptively adjusted based on the amplitude of the light spot displacement or the consistency index of the positioning point; the larger the displacement amplitude or the higher the consistency, the greater the weight.
[0021] To achieve the objectives of this invention, a single IMU yaw angle measurement and correction device for underground pipeline inspection instruments is also disclosed, comprising: a single inertial measurement unit (IMU) for outputting three-axis angular velocity data and three-axis acceleration data, and optionally outputting three-axis magnetic field data; a laser for projecting a laser spot onto an imaging screen; an imaging screen with five positioning points around its perimeter; a camera for acquiring image frames containing the five positioning points and the laser spot; a processor and a memory storing a program that, when executed by the processor, implements a yaw angle measurement and correction method for underground pipeline inspection instruments; wherein the geometric baseline length between the laser and the imaging screen is a fixed value or a calibrable value.
[0022] Furthermore, the imaging screen includes a ring structure or a scale reference structure for pixel scale calibration; the five positioning points are any one or any combination of positioning holes, positioning notches, reflective marks or highlight marks; the camera is an endoscopic camera or a miniature camera for pipeline inspection; the processor performs invalid frame rejection and outputs a continuous yaw angle sequence under valid frames for pipeline three-dimensional trajectory reconstruction.
[0023] Compared with existing technologies, the significant advancements of this invention are: 1) By constructing yaw angle observations through laser spot displacement without relying on a magnetometer, the yaw angle drift of a single IMU is effectively constrained, improving long-term stability; 2) By establishing a screen coordinate system through five positioning points and compensating for camera shake and assembly deviations in real time, the spot displacement measurement is transformed from absolute position to relative displacement, reducing the influence of fixed geometric offsets; 3) Through circular scale calibration and rotation compensation, the mapping from pixel displacement to physical displacement and then to yaw angle observations has clear physical meaning, facilitating engineering calibration and reproduction; 4) In image processing, the use of quantile thresholding, morphology, and brightness-weighted centroids enhances robustness under reflection and exposure changes and achieves sub-pixel-level positioning.
[0024] To more clearly illustrate the functional characteristics and structural parameters of the present invention, further explanation is provided below in conjunction with the accompanying drawings and specific embodiments. Attached Figure Description
[0025] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:
[0026] Figure 1 This is a schematic diagram of the overall structure of the device of the present invention;
[0027] Figure 2 This is a schematic diagram of the imaging screen structure;
[0028] Figure 3 This is a flowchart of yaw angle measurement and correction. Detailed Implementation
[0029] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0030] A yaw angle measurement and correction method for underground pipeline inspection instruments includes the following steps:
[0031] Step 1: Acquire gyroscope data containing at least the yaw axis angular velocity using a single inertial measurement unit (IMU), and perform time integration on the angular velocity to obtain the main measured value of the yaw angle;
[0032] Step 2: Acquire image frames from the camera onto the imaging screen, which includes a central laser spot and five positioning points distributed around the perimeter of the imaging screen.
[0033] Step 3: Detect five positioning points in each image frame, and use the stable geometric relationship of the five positioning points to establish or update the screen coordinate system, and calculate the transformation relationship between camera shake, setup deviation or image coordinate system and screen coordinate system.
[0034] Step 4: Extract the center pixel coordinates of the laser spot in each image frame, convert the spot pixel coordinates into screen coordinates based on the transformation relationship obtained in step (3), and obtain the spot displacement with the spot position of the reference frame or reference time period as the zero point.
[0035] Step 5: Based on the calibration parameters from pixel scale to physical scale, convert the spot screen displacement into physical displacement, and establish a geometric relationship with the fixed baseline distance from the laser to the imaging screen to calculate the yaw angle observation.
[0036] Step 6: Use the yaw angle observation to periodically or continuously correct the yaw angle main measurement value obtained in Step 1, and output the corrected yaw angle.
[0037] Specifically, in step 1, the yaw angle master measurement value is obtained through quaternion attitude update or equivalent integration, and the quaternion is normalized to suppress the accumulation of numerical error.
[0038] Specifically, in step 3, the origin of the screen coordinate system is taken as the geometric center of the imaging screen or the equivalent center determined by the five positioning points, and the direction of the screen coordinate axis is determined by the zero orientation defined by at least one of the five positioning points; the two-dimensional similarity transformation or homography transformation is obtained using the five positioning points, and the current frame coordinates are aligned to the reference frame coordinates, thereby compensating for camera shake and assembly deviation in real time.
[0039] Specifically, in step 4, the center pixel coordinates of the light spot are calculated using the brightness-weighted centroid method, and after threshold segmentation, morphological processing and connected component screening are combined to remove false targets; and when light spot saturation diffusion or deformation is detected, the centroid of the core region or the ellipse fitting center is used as the center of the light spot.
[0040] Specifically, in steps 2 to 5, when the geometric constraints of the five positioning points do not meet the preset threshold or the spot extraction fails, or when the spot morphology index is deformed and exceeds the threshold, the corresponding image frame is determined to be an invalid frame and the yaw angle correction is not updated.
[0041] Specifically, in step 5, the calibration from pixel scale to physical scale is obtained through the calibration ring on the imaging screen or the spacing between positioning points at a known distance; the yaw angle observation is calculated by the horizontal component of the physical displacement of the light spot and the distance to the fixed baseline according to the arctangent relationship, or by using a linear approximation relationship under small angle conditions.
[0042] Specifically, in step 6, complementary filtering or Kalman filtering is used for correction, and the gyroscope zero bias is used as a state quantity estimate to further suppress yaw angle drift.
[0043] Specifically, the weight of the yaw angle observation is adaptively adjusted based on the amplitude of the light spot displacement or the consistency index of the positioning point; the larger the displacement amplitude or the higher the consistency, the greater the weight.
[0044] A single IMU yaw angle measurement and correction device for an underground pipeline inspection instrument includes: a single inertial measurement unit (IMU) for outputting three-axis angular velocity data and three-axis acceleration data, and optionally outputting three-axis magnetic field data; a laser for projecting a laser spot onto an imaging screen; an imaging screen with five positioning points around its perimeter; a camera for acquiring image frames containing the five positioning points and the laser spot; and a processor and memory storing a program that, when executed by the processor, implements a yaw angle measurement and correction method for an underground pipeline inspection instrument; wherein the geometric baseline length between the laser and the imaging screen is a fixed value or a calibrable value.
[0045] Specifically, the imaging screen includes a ring structure or a scale reference structure for pixel scale calibration; the five positioning points are any one or any combination of positioning holes, positioning notches, reflective marks or highlight marks; the camera is an endoscope camera or a miniature camera for pipeline inspection; the processor performs invalid frame rejection and outputs a continuous yaw angle sequence under the valid frames for pipeline three-dimensional trajectory reconstruction.
[0046] Example
[0047] (1) Relationship between device structure and installation
[0048] like Figure 1 As shown, a single IMU yaw angle measurement and correction device for an underground pipeline detector includes: an IMU, a laser, an imaging screen, a camera, and a processing unit. The laser is fixed at one end of the detector, and the imaging screen is fixed at the other end, spaced apart from the laser, with a baseline distance L between them. The camera is aligned with the imaging screen to acquire image frames.
[0049] A circular dimensional reference structure is set on the imaging screen, and five positioning points are set around its perimeter. The five positioning points can be notches, holes, reflective stickers, or bright marks, and it is preferred to use an unequal interval layout to uniquely determine the zero orientation.
[0050] The IMU is a single inertial measurement unit, comprising at least a three-axis gyroscope and a three-axis accelerometer, and optionally a three-axis magnetometer. The gyroscope outputs angular velocity for attitude updates, the accelerometer provides gravity-constrained correction for pitch and roll when stationary or quasi-stationary, and the magnetometer can be used for heading reference in favorable magnetic environments, but may be disabled or have reduced weight under conditions of magnetic distortion in underground pipelines. Since yaw angle is primarily derived from gyro integrals and exhibits significant long-term drift, this invention uses visual observations for correction. While the accelerometer provides gravity-constrained correction for pitch and roll when stationary or quasi-stationary, it cannot provide a gravity reference for yaw angle. Although the magnetometer can provide heading reference, its stability is insufficient in the magnetic field distortion environment of underground pipelines. Therefore, this invention employs a spot displacement-based yaw observation for correction.
[0051] (2) Calculation of yaw angle master measurement value of single IMU
[0052] The processing unit reads the IMU angular velocity at a sampling period Δt and completes the integration of the yaw axis angular velocity to obtain the master yaw angle value. Quaternion updates are preferably used to avoid the singularity of direct Euler angle integration, and the quaternions are normalized after each update to reduce the accumulation of numerical errors (this can be achieved through Euler integration or higher-order numerical integration). This master value has a high-frequency dynamic response, but it will drift over time.
[0053] (3) Coordinate system establishment and real-time compensation driven by five positioning points
[0054] The core functions of the five positioning points are as follows: Coordinate system fixation and zero orientation definition: Taking the geometric center of the imaging screen as the origin, a certain positioning point (such as the positioning point numbered P1) is used to determine the zero orientation of the screen coordinate system, thereby determining the direction of the screen x-axis; then the orthogonal direction is used to determine the y-axis, realizing the direction compensation of "pixel coordinate system → screen coordinate system".
[0055] Camera shake / assembly deviation compensation: Five points are detected in each frame, and similarity transformation (rotation, translation, scaling) or homography transformation is solved with the five points in the reference frame. The current frame is aligned with the reference frame, thereby converting slight camera shake and assembly deviation into a compensable transformation.
[0056] Robustness and anomaly removal: Consistency checks are performed using geometric constraints of distance and angle between five points; when the geometric relationship does not meet the threshold or only a few points are detected, the frame is determined to be invalid and the correction update is skipped.
[0057] Stable numbering: It can be sorted by polar angle and the point with the largest u coordinate can be specified as P1, so that the point sequence remains consistent under changes in illumination or slight rotation.
[0058] (4) Image processing: extraction of positioning points and light spot center
[0059] To stably extract the five positioning points and the central light spot under conditions of reflection, exposure variation, and noise interference, this invention provides the following implementation process (but is not limited thereto):
[0060] a) Preprocessing: grayscale conversion / contrast enhancement / noise reduction filtering;
[0061] b) Ring detection: Detect the inner and outer circles or the outer radius rout of the ring and the center (cx, cy), and construct a ring-shaped mask to limit the candidate region;
[0062] c) Localization point extraction: Statistical distribution of brightness within the annular zone, take the high quantile (e.g., 99th percentile) as the threshold for adaptive binarization, retaining only the brightest area; then perform morphological opening operation for noise reduction, connected component filtering, and use brightness-weighted centroids to achieve sub-pixel localization; finally, filter by geometric distribution and spacing constraints to obtain 5 localization points;
[0063] d) Spot center extraction: After threshold segmentation of the central region, the connected components of the spot are extracted. The center pixel coordinates of the spot are calculated by using the brightness-weighted centroid, which improves the robustness to spot edge deformation, saturation diffusion and local noise.
[0064] e) Spot deformation discrimination and robust center determination: Calculate morphological indicators (including but not limited to area A, equivalent radius, ellipse fitting major-minor axis ratio / eccentricity e, perimeter-area ratio, centroid difference d between the core region and the halo region, etc.) for the connected regions of the spot; when it is determined to be slightly deformed, the core region extracted with a higher threshold is preferably used as the brightness-weighted centroid as the spot center; when it is determined to be severely deformed, with ghosting or multi-peak structure, the frame is marked as invalid or the weight of the yaw angle observation of the frame is reduced.
[0065] (5) Use of spot movement distance: relative displacement, keyframes and invalid frames
[0066] To eliminate fixed installation deviations, this invention does not directly use the absolute position of the light spot, but instead uses a reference frame / static period to establish a zero point: for example, using the average value (u0, v0) of the center of the light spot in the initial static period as a reference point, the relative displacement (Δu, Δv) of each frame is calculated, and the displacement amplitude can be defined for key frame discrimination or weight adaptation.
[0067] If severe overexposure or reflection causes spot extraction failure, or if the five-point geometric constraints are not met, the frame will be marked as invalid and discarded, and the yaw angle correction will not be updated. Alternatively, if the spot morphology index exceeds the threshold (indicating obvious deformation / ghosting / multi-peak), the frame will be marked as invalid and discarded.
[0068] (6) Calculation of pixel displacement to yaw angle observation
[0069] The coordinates and parameters can be defined as follows: the image pixel coordinate system {I} is represented by (u,v), and the screen coordinate system {S} is represented by (xs,ys); L is the equivalent baseline length from the laser to the imaging screen, Rs is the physical radius of the ring, and rout is the outer radius of the image detection ring.
[0070] The pixel-to-physical-scale conversion can be achieved by taking the scale factor s = Rs / rout, and using the rotation matrix R(θ) estimated from the orientation of the positioning point to compensate for the angle θ between the pixel axis and the screen axis, thereby obtaining the displacement component of the light spot in the screen coordinate system.
[0071] The yaw angle observation is preferably constructed from the horizontal displacement component Δx and the baseline L:
[0072] General form: φ_vis = arctan(Δx / L);
[0073] Small angle approximation: φ_vis ≈ Δx / L.
[0074] The vertical displacement component can be used for consistency verification or error assessment.
[0075] The above-mentioned link can be used to form a visual yaw angle observation that can be directly fused with the IMU yaw angle state, which can be used to suppress gyro integral drift.
[0076] (7) Yaw angle correction output
[0077] The IMU yaw angle main measurement and the visual yaw angle observation are calibrated and fused, preferably using complementary filtering or Kalman filtering; when further drift suppression is required, the gyroscope zero bias can be used as a state variable and jointly estimated with the yaw angle to achieve long-term drift suppression caused by zero bias.
[0078] To enhance engineering robustness, the observation weights can be adaptively adjusted based on factors such as the amplitude of the light spot displacement, the geometric consistency of the positioning point, and the consistency of the light spot shape (deformation index): when the displacement is larger and the geometric consistency is better, the observation weights are higher; conversely, the weights are reduced or updates are skipped.
[0079] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0080] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A yaw angle measurement and correction method for underground pipeline inspection instruments, characterized in that, Includes the following steps: Step 1: Acquire gyroscope data containing at least the yaw axis angular velocity using a single inertial measurement unit (IMU), and perform time integration on the angular velocity to obtain the main measured value of the yaw angle; Step 2: Acquire image frames from the camera onto the imaging screen, which includes a central laser spot and five positioning points distributed around the perimeter of the imaging screen. Step 3: Detect the five positioning points in each image frame, and use the stable geometric relationship of the five positioning points to establish or update the screen coordinate system, and calculate the transformation relationship between camera shake, mounting deviation or image coordinate system and screen coordinate system. Step 4: Extract the center pixel coordinates of the laser spot in each image frame, convert the spot pixel coordinates into screen coordinates based on the transformation relationship obtained in step (3), and obtain the spot displacement with the spot position of the reference frame or reference time period as the zero point. Step 5: Based on the calibration parameters from pixel scale to physical scale, convert the spot screen displacement into physical displacement, and establish a geometric relationship with the fixed baseline distance from the laser to the imaging screen to calculate the yaw angle observation. Step 6: Use the yaw angle observation to periodically or continuously correct the yaw angle main measurement value obtained in Step 1, and output the corrected yaw angle.
2. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In step 1, the yaw angle master measurement value is obtained through quaternion attitude update or equivalent integration, and the quaternion is normalized to suppress the accumulation of numerical error.
3. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In step 3, the origin of the screen coordinate system is taken as the geometric center of the imaging screen or the equivalent center determined by the five positioning points, and the direction of the screen coordinate axis is determined by the zero orientation defined by at least one of the five positioning points; the two-dimensional similarity transformation or homography transformation is obtained using the five positioning points to align the current frame coordinates to the reference frame coordinates, thereby compensating for camera shake and assembly deviation in real time.
4. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In step 4, the center pixel coordinates of the light spot are calculated using the brightness-weighted centroid method, and after threshold segmentation, morphological processing and connected component screening are combined to remove false targets; when light spot saturation diffusion or deformation is detected, the centroid of the core region or the ellipse fitting center is used as the center of the light spot.
5. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In steps 2 to 5, when the geometric constraints of the five positioning points do not meet the preset threshold or the spot extraction fails, or when the spot morphology index is deformed and exceeds the threshold, the corresponding image frame is determined to be an invalid frame and the yaw angle correction is not updated.
6. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In step 5, the calibration from pixel scale to physical scale is obtained by the calibration ring on the imaging screen or the spacing between positioning points at a known distance; the yaw angle observation is calculated by the horizontal component of the physical displacement of the light spot and the distance to the fixed baseline according to the arctangent relationship, or by using a linear approximation relationship under small angle conditions.
7. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, In step 6, complementary filtering or Kalman filtering is used for correction, and the gyroscope zero bias is used as the state quantity estimate to further suppress yaw angle drift.
8. The yaw angle measurement and correction method for an underground pipeline inspection instrument according to claim 1, characterized in that, The weight of the yaw angle observation is adaptively adjusted based on the amplitude of the light spot displacement or the consistency index of the positioning point; the greater the displacement amplitude or the higher the consistency, the greater the weight.
9. A single IMU yaw angle measurement and correction device for an underground pipeline inspection instrument, said device being based on the yaw angle measurement and correction method for an underground pipeline inspection instrument according to any one of claims 1-8, characterized in that, include: A single inertial measurement unit (IMU) outputs triaxial angular velocity and triaxial acceleration data, and optionally triaxial magnetic field data; a laser projects a laser spot onto an imaging screen; the imaging screen has five positioning points around its perimeter; a camera acquires image frames containing the five positioning points and the laser spot; a processor and memory store a program that, when executed by the processor, implements a yaw angle measurement and correction method for an underground pipeline inspection instrument; wherein, the geometric baseline length between the laser and the imaging screen is a fixed or calibrable value.
10. A single IMU yaw angle measurement and correction device for an underground pipeline inspection instrument according to claim 9, characterized in that, The imaging screen includes a ring structure or a scale reference structure for pixel scale calibration; the five positioning points are any one or any combination of positioning holes, positioning notches, reflective marks or bright marks; the camera is an endoscope camera or a miniature camera for pipeline inspection; the processor performs invalid frame rejection and outputs a continuous yaw angle sequence under valid frames for pipeline three-dimensional trajectory reconstruction.