Laser positioning and deviation rectifying method applied to tunnel excavation

Abstract

The invention discloses a laser positioning and deviation rectifying method applied to tunnel excavation, which comprises the following steps that: an image acquisition device acquires an image of a laser point formed by irradiating a laser target by a laser emitting device in real time, analyzes the image through a pre-trained deep learning model, and calculates position data of the laser point; then comparing the real-time laser point position data with preset central position data to calculate the offset of the machine head of the tunnel excavating device; according to the offset and the depth data of the machine head of the tunnel excavating device, the adjustment amount of the machine head of the tunnel excavating device on the X axis, the Y axis, the Z axis and the angle is obtained through calculation, corresponding control signals are formed, and the position and the angle of the machine head are adjusted through a machine head control system. The system can adapt to different working environments, maintain high-precision and high-efficiency operation, and is suitable for equipment deviation detection and correction tasks requiring extremely high precision, so that a tunnel excavation device can complete the operation according to a correct excavation path, and the tunnel excavation quality is ensured.

Description

Technical Field

[0001] This invention relates to the field of controlling excavation equipment in tunnel or pipeline excavation, and particularly to a laser positioning and correction method applied to tunnel excavation. Background Technology

[0002] With rapid urbanization, the construction of urban underground passages is receiving increasing attention. Tunnel boring machines (TBMs), as a primary tunneling equipment, are widely used in urban underground engineering projects. During TBM excavation, factors such as geological conditions, underground pipelines, and existing projects like subways can cause the TBM to deviate from its planned route. Therefore, correcting deviations during TBM excavation is a crucial measure to ensure tunnel excavation quality and reduce accidents.

[0003] Existing TBM laser guidance systems generate laser beams using lasers, utilize optical principles and detectors to receive reflected laser signals, and achieve automatic guidance and positioning of the tunnel boring machine through positioning algorithms and control systems. Laser guidance systems offer advantages such as high precision, high efficiency, and automation, significantly improving the quality and speed of tunnel construction. However, the harsh environment during tunnel excavation, including significant dust and lighting conditions, can affect the clarity of the laser target image acquired by the TBM laser guidance system. This can lead to decreased positioning accuracy or even incorrect corrections, potentially causing accidents. Summary of the Invention

[0004] This invention provides a laser positioning and correction method for tunnel excavation. By acquiring and analyzing images of laser target points through an artificial intelligence model, the method can accurately identify laser points according to different site environments, and monitor and adjust the correction amount of the tunnel excavation device in real time.

[0005] The present invention adopts the following technical solution: A laser positioning and correction method applied to tunnel excavation includes a laser emitting device, a laser target mounted on the head of the tunnel excavator to receive the laser beam, and an image acquisition device for detecting the position of the laser point on the laser target. The correction method is as follows: The image acquisition device acquires images of the laser points formed by the laser emitting device illuminating the laser target in real time. The images are analyzed and the position data of the laser points are calculated using a pre-trained deep learning model. Then, the real-time laser point position data is compared with the preset center position data to calculate the offset of the tunnel excavator head. Based on the offset, the adjustment amount of the tunnel excavator head in the X, Y, Z axes and angle is calculated and corresponding control signals are generated. The head control system adjusts the preset tunneling trajectory of the head to correct the deviation.

[0006] Preferably, the image acquisition device is provided in multiple parts, which simultaneously acquire images of the laser point from different angles, and calculate the position of the laser point by triangulation.

[0007] Preferably, a light sensor for detecting ambient light is also provided at the location of the laser target. The deep learning model is trained based on the data collected by the light sensor so that the deep learning model can identify the location of the laser point under different ambient light conditions.

[0008] Preferably, the deep learning model analyzes the illumination of the laser target position in real time based on the image acquired by the image acquisition device, and adjusts the lighting devices in the environment accordingly.

[0009] Preferably, the system architecture of the deep learning model includes a data acquisition module, an image segmentation module, a deep learning analysis module, and a control and feedback module. The data acquisition module includes interfaces for connecting to an image acquisition device and a light sensor. The image segmentation module includes image preprocessing, image segmentation, and feature extraction. Image segmentation segments the target region and extracts the region of interest. Feature extraction includes extracting key features such as the shape, position, and brightness of the laser point. The deep learning analysis module uses a trained CNN model to identify and classify the key features, and compares the position of the laser point in the image with preset center position data to obtain the offset. The control and feedback module generates control commands based on the offset from the deep learning analysis module to adjust the position of the machine head to correct the offset, and adjusts the control commands in real time according to the changing trend of the laser point image to optimize the control strategy.

[0010] Preferably, the offset of the laser point is calculated as follows: The coordinates of the preset center position are If the coordinates of the laser point are P(XY,Z), then the offset ΔP is expressed as: The depth data from the tunnel boring machine head is projected onto the corresponding plane equation of the camera via laser light in the image acquisition device. Where A, B, and C are the normal vector coordinates of the plane, and D is the plane intercept, the depth data Z can be derived.

[0011] Preferably, the adjustment amount of the tunnel excavation device head is: Laser point offset Adjustments to the X, Y, and Z axis positions: The adjustment amount of the angle, assuming the rotation angles around the X, Y, and Z axes are respectively... , , The adjustment formula is then: in, , , It is the angle adjustment coefficient.

[0012] Beneficial effects: Deep learning models, powered by artificial intelligence, can more accurately pinpoint the location of laser points and adapt to different working environments, such as changes in lighting or camera position. These models can adjust parameters or retrain in real time to maintain detection accuracy. The system is adaptable to various working environments, maintaining high precision and efficiency through self-learning and adaptive functions. It is suitable for tasks requiring extremely high precision in equipment offset detection and correction, ensuring that tunnel excavation equipment follows the correct excavation path and guarantees tunnel excavation quality. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. The accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] Figure 1 This is a flowchart of the process of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0017] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0018] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0019] In the description of this application, it should be noted that the terms "inner" and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product is conventionally placed during use. These terms are used only for the convenience of describing this application and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] In the description of this application, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] A laser positioning and correction method applied to tunnel excavation includes a laser emitting device, a laser target mounted on the head of the tunnel excavator for receiving the laser beam, and an image acquisition device for detecting the position of the laser point on the laser target. (Refer to...) Figure 1 The correction method is as follows: The image acquisition device acquires images of the laser points formed by the laser emitting device illuminating the laser target in real time. The image is analyzed and the position data of the laser points is calculated by a pre-trained deep learning model. Then, the real-time laser point position data is compared with the preset center position data to calculate the offset of the tunnel excavator head. Based on the offset and the depth data of the tunnel excavator head, the adjustment amount of the tunnel excavator head in the X, Y, Z axes and angle is calculated and corresponding control signals are generated. The head control system adjusts the position and angle of the head.

[0023] Deep learning models, powered by artificial intelligence, can more accurately pinpoint the location of laser points and adapt to different working environments, such as changes in lighting or camera position. These models can adjust parameters or retrain in real time to maintain detection accuracy. The system is adaptable to various working environments, maintaining high precision and efficiency through self-learning and adaptive functions. It is suitable for tasks requiring extremely high precision in equipment offset detection and correction, ensuring that tunnel excavation equipment follows the correct excavation path and guarantees tunnel excavation quality.

[0024] Current TBM laser guidance systems typically use a single image acquisition device to obtain the laser point's position on the target. This device captures an image of the laser point projected onto the normal plane it points to. If the tunnel boring machine's excavation direction deviates slightly, the laser's position on the target will change slightly. However, for a single image acquisition device, this change is not significant in the image on the normal plane it points to, affecting the accuracy of detecting the tunnel boring machine's deflection. Therefore, multiple image acquisition devices can be set up in different directions to simultaneously acquire images of the laser point from different angles. Since the amount of change in the laser point images acquired from different directions will differ, the positional relationship of the image acquisition devices at different angles and the corresponding changes in the laser point images can more accurately determine the laser point's position data, thereby accurately calculating the tunnel boring machine's head offset. Furthermore, the laser point position data from image acquisition devices at different angles can be cross-calibrated to eliminate errors.

[0025] Because the lighting environment inside the tunnel affects the clarity of the laser points captured by the image acquisition device, such as insufficient contrast which may produce noise and affect the shooting accuracy of the image acquisition device, a light sensor for detecting ambient light is installed at the laser target location to enable the deep learning model to accurately calculate or determine the image boundaries of the laser points under different ambient lighting conditions. The deep learning model is trained based on the data collected by the light sensor, allowing it to identify the location of the laser points under different ambient lighting conditions, thereby further improving the accuracy of detection and calculation.

[0026] The system architecture of a deep learning model includes a data acquisition module, an image segmentation module, a deep learning analysis module, and a control and feedback module. The data acquisition module includes Camera system: Multiple high-resolution cameras are used to capture images of the target area from different angles, ensuring that the acquired data has rich perspective and detail.

[0027] Laser emitting device: emits a laser beam, forming a laser spot on the target, which is captured by a camera.

[0028] Sensor interface: Integrated with external sensors (such as light sensors) to capture environmental data and enhance image preprocessing.

[0029] Image processing module includes Preprocessing includes denoising, image enhancement, and color correction to improve image quality.

[0030] Image segmentation: Using deep learning models to segment target regions and extract regions of interest from the image.

[0031] Feature extraction: Extract key feature points, such as the shape, position, and brightness of the laser point, to provide data support for subsequent analysis.

[0032] The deep learning analytics module includes Deep Convolutional Neural Network (CNN): A trained CNN model performs feature recognition and classification, comparing the location of laser points in an image with a reference location.

[0033] Self-learning model: Using reinforcement learning algorithms, the model can continuously adjust its parameters in practical applications and optimize itself based on newly collected data to improve recognition accuracy.

[0034] Multi-task learning: The model can handle multiple related tasks, such as simultaneously performing laser point detection and adapting to environmental changes.

[0035] Control and Feedback Module: Real-time control: Based on deep learning analysis results, control commands are generated to adjust the position of the machine head to correct the offset.

[0036] Feedback mechanism: The system receives feedback data from sensors and models to optimize the control strategy in real time.

[0037] The following details the laser point offset and the methods for calculating the adjustment amounts of the tunnel excavator head in the X, Y, Z axes and angles based on the offset and the depth data of the tunnel excavator head.

[0038] 1. Calculation of laser point offset Let the coordinates of the laser point at the ideal position be... The actual detected laser point coordinates are P(XY,Z). The offset ΔP is expressed as: For a single-camera system, depth information (i.e., displacement along the Z-axis) is estimated by incorporating the geometric relationships of the laser point. Assuming the laser beam is projected onto the camera on a plane, the equation of the plane is: The changes in depth information Z can be derived from the known geometric relationships.

[0039] 2. Real-time location adjustment Based on the detected laser point offset The X, Y, and Z axis positions and angles of the machine head are adjusted through geometric calculations. The adjustment amount can be calculated using the following formula: Adjustments to the X, Y, and Z axis positions: in, , , It is an adjustment coefficient, determined by the physical characteristics of the system.

[0040] The adjustment amount of the angle, assuming the rotation angles around the X, Y, and Z axes are respectively... , , The adjustment formula is then: in, , , It is the angle adjustment coefficient.

[0041] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A laser positioning and deviation correction method applied to tunnel excavation, characterized in that, It includes a laser emitting device, a laser target mounted on the head of the tunnel boring machine for receiving the laser beam, and an image acquisition device for detecting the position of the laser point on the laser target; the correction method is as follows: The image acquisition device acquires images of the laser points formed by the laser emitting device illuminating the laser target in real time. It then analyzes the images using a pre-trained deep learning model to calculate the position data of the laser points. Next, it compares the real-time laser point position data with preset center position data to calculate the offset of the tunnel excavator head. Based on the offset and the depth data of the tunnel excavator head, it calculates the adjustment amounts of the tunnel excavator head in the X, Y, and Z axes and angles, generating corresponding control signals. The head control system then adjusts the preset tunneling trajectory of the head to correct the deviation.

2. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, The image acquisition device is equipped with multiple units, which simultaneously acquire images of the laser point from different angles, and calculate the position of the laser point using triangulation.

3. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, A light sensor for detecting ambient light is also installed at the location of the laser target. The deep learning model is trained based on the data collected by the light sensor, enabling the deep learning model to identify the location of the laser point under different ambient lighting conditions.

4. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, The deep learning model analyzes the illumination of the laser target position in real time based on the images acquired by the image acquisition device, and adjusts the lighting devices in the environment accordingly.

5. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, The system architecture of the deep learning model includes a data acquisition module, an image segmentation module, a deep learning analysis module, and a control and feedback module. The data acquisition module includes interfaces for connecting to the image acquisition device and the light sensor; The image segmentation module includes image preprocessing, image segmentation, and feature extraction. Image segmentation divides the target region and extracts the region of interest. Feature extraction includes extracting key features such as the shape, position, and brightness of the laser point. The deep learning analysis module identifies and classifies key features using a trained CNN model, and compares the position of the laser point in the image with the preset center position data to obtain the offset. The control and feedback module generates control commands based on the offset from the deep learning analysis module to adjust the position of the head to correct the offset, and adjusts the control commands in real time to optimize the control strategy based on the changing trend of the laser point image.

6. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, The offset of the laser point is calculated as follows: The coordinates of the preset center position are If the coordinates of the laser point are P(XY,Z), then the offset ΔP is expressed as: The depth data from the tunnel boring machine head is projected onto the corresponding plane equation of the camera via laser light in the image acquisition device. Where A, B, and C are the normal vector coordinates of the plane, and D is the plane intercept, the depth data Z can be derived.

7. The laser positioning and correction method for tunnel excavation according to claim 1, characterized in that, The adjustment range of the tunnel excavator head is: Laser point offset Adjustments to the X, Y, and Z axis positions: The adjustment amount of the angle, assuming the rotation angles around the X, Y, and Z axes are respectively... , , The adjustment formula is then: in, , , It is the angle adjustment coefficient.

Images