Tunnel crack detection method and tunnel crack detection system

Abstract

The invention discloses a tunnel crack detection method and a tunnel crack detection system, and belongs to the field of tunnel safety detection.The tunnel crack detection method comprises the steps that a tunnel inspection robot is utilized to conduct inspection in a tunnel to obtain complete three-dimensional data in the tunnel so as to construct a three-dimensional map in the tunnel, and the position of each crack area in the three-dimensional map is determined; determining an inspection tour area corresponding to each crack area and executing path planning, and generating an optimal inspection tour path of the unmanned aerial vehicle; the optimal inspection path enables the unmanned aerial vehicle to sequentially pass through each inspection area and then leave the tunnel after entering the tunnel under the condition that obstacle avoidance conditions are met, and the total time is shortest; enabling the unmanned aerial vehicle to move according to the optimal inspection path, and scanning the crack region by using the depth camera when the unmanned aerial vehicle moves to one inspection region so as to obtain fine three-dimensional data of the crack region; and performing analysis processing on the fine three-dimensional data of the crack area to complete tunnel crack detection. According to the invention, the efficiency and precision of long tunnel crack detection can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of tunnel safety detection, and more specifically, relates to a tunnel crack detection method and a tunnel crack detection system. Background Art

[0002] Tunnel crack detection is crucial to ensure the safety of tunnel structures. Traditional manual inspections are inefficient, prone to missed inspections, and have high inspection costs and risks. Currently, track-mounted inspection robots are gradually replacing traditional manual inspections. They are equipped with high-definition cameras, lidars, and various sensors to improve inspection efficiency and safety. However, track-mounted inspection robots move and inspect on fixed tracks, which limits their coverage and detection accuracy. At the same time, track-mounted inspection robots may find problems during inspections, and manual re-inspections are still required to ensure accuracy and completeness. However, for some complex or more detailed evaluation situations such as cracks that require multi-angle observation, defects in complex structures such as bends or intersections, etc.

[0003] The key to tunnel crack detection is to accurately obtain three-dimensional data of the crack area. UAVs, with their high flexibility and strong maneuverability, can carry various sensors such as high-definition cameras and infrared sensors into tunnels, and easily and quickly reach places that are difficult for humans and tunnel inspection robots to reach for inspection. The inspection range is wider, and compared with tunnel inspection robots, the three-dimensional data obtained by UAV inspections is more accurate, which is more conducive to the accurate detection of tunnel cracks.

[0004] Due to the above advantages, drones have been increasingly widely used in tunnel inspections. However, with the continuous development of transportation, most of the current tunnels are long tunnels. Long tunnels are long, and with continuous use, multiple cracks often appear in the same tunnel. This makes drones face the problem of short battery life, requiring frequent battery replacement or charging, and low inspection efficiency, which will affect the normal and safe use of tunnels. In addition, the space in long tunnels is closed and the lighting conditions are poor, which will affect the flight stability of drones and the quality of three-dimensional data acquisition, reducing the accuracy of detection. Summary of the invention

[0005] In view of the defects and improvement needs of the prior art, the present invention provides a tunnel crack detection method and a tunnel crack detection system, the purpose of which is to improve the efficiency and accuracy of crack detection in long and large tunnels.

[0006] To achieve the above object, according to one aspect of the present invention, a tunnel crack detection method is provided, comprising:

[0007] Use a tunnel inspection robot to inspect the tunnel from the tunnel entrance to obtain complete three-dimensional data of the tunnel interior;

[0008] Use the complete 3D data inside the tunnel to build a 3D map of the tunnel, detect crack areas in the tunnel, and determine the location of each crack area in the 3D map;

[0009] Determine the inspection area corresponding to each crack area; the inspection area satisfies: when the drone equipped with the depth camera is located in the inspection area, the crack area is within the field of view of the depth camera;

[0010] Path planning is performed based on the three-dimensional map and the positions of each inspection area in the three-dimensional map to generate the optimal inspection path for the drone. The optimal inspection path allows the drone to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the drone entering the tunnel to leaving the tunnel is the shortest.

[0011] The drone moves along the optimal inspection path, and each time it moves to an inspection area, it uses a depth camera to scan the crack area to obtain detailed three-dimensional data of the crack area;

[0012] Perform analysis and processing on the detailed 3D data of the crack area to complete the tunnel crack detection;

[0013] The analysis and processing includes at least one of the following: three-dimensional reconstruction of the crack area, identification of the crack type, extraction of the three-dimensional shape of the crack, assessment of the degree of crack danger, and prediction of crack development trend.

[0014] Furthermore, before the tunnel inspection robot starts to inspect in the tunnel from the tunnel entrance, the method further includes: obtaining a relative posture between the tunnel inspection robot and the tunnel entrance at the tunnel entrance, recorded as an entrance reference relative posture;

[0015] Furthermore, before the UAV moves along the optimal inspection path, it also includes: obtaining the relative posture of the UAV and the tunnel entrance at the tunnel entrance, recording it as the entrance alignment relative posture, and aligning the entrance alignment relative posture to the entrance reference relative posture, so that the tunnel inspection robot and the UAV share a unified three-dimensional map coordinate system.

[0016] Furthermore, the tunnel crack detection method provided by the present invention further includes: setting a visual mark at the tunnel entrance in advance;

[0017] Furthermore, the entrance reference relative position and the entrance alignment relative position are both determined with the aid of visual markers.

[0018] Furthermore, before determining the inspection area corresponding to each crack area, the method further includes: determining the blind area of ​​the tunnel inspection robot in the tunnel and the position of the blind area in the three-dimensional map according to the three-dimensional map;

[0019] Path planning is performed based on the three-dimensional map and the positions of each blind spot in the three-dimensional map to generate a supplementary inspection path for the drone; the supplementary inspection path allows the drone to enter the tunnel from the tunnel entrance, pass through each blind spot in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the drone entering the tunnel to leaving the tunnel is the shortest;

[0020] The drone is moved along the supplementary inspection route, and each time the drone moves to a blind area, the drone uses a depth camera to scan the blind area to obtain fine three-dimensional data of the blind area as supplementary depth information of the blind area;

[0021] The three-dimensional map is updated using supplementary depth information from blind areas of vision.

[0022] According to another aspect of the present invention, there is provided a tunnel crack detection system, comprising: a crack preliminary positioning module, a crack precision scanning module, a communication module and a cloud computing server;

[0023] The crack preliminary positioning module includes: a tunnel inspection robot and a mobile edge computing server mounted on the tunnel inspection robot;

[0024] The tunnel inspection robot is used to inspect inside the tunnel to obtain complete three-dimensional data inside the tunnel and send it to the edge computing server;

[0025] The mobile edge computing server is used to construct a three-dimensional map of the interior of the tunnel using the complete three-dimensional data of the interior of the tunnel, detect crack areas in the tunnel, determine the position of each crack area in the three-dimensional map, and send the three-dimensional map and the position of each crack area in the three-dimensional map to the crack precision scanning module;

[0026] The crack precision scanning module includes a drone equipped with a depth camera and a posture calculation unit mounted on the drone;

[0027] The drone is used to scan the target within the field of view of its depth camera to obtain detailed three-dimensional data of the target;

[0028] The posture calculation unit is used to determine the inspection area corresponding to each crack area; the inspection area satisfies: when the UAV is located in the inspection area, the crack area is within the field of view of the depth camera;

[0029] The posture calculation unit is also used to perform path planning according to the three-dimensional map and the positions of each inspection area in the three-dimensional map, and generate an optimal inspection path for the UAV; the optimal inspection path enables the UAV to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the UAV entering the tunnel to leaving the tunnel is the shortest;

[0030] The posture calculation unit is also used to make the UAV move according to the optimal inspection path, and each time the UAV moves to an inspection area, it uses the depth camera to scan the crack area to obtain fine three-dimensional data of the crack area and send it to the cloud computing server;

[0031] The cloud computing server is located outside the tunnel and is used to perform analysis and processing on the fine three-dimensional data of the crack area to complete the tunnel crack detection; the analysis and processing includes at least one of the following: three-dimensional reconstruction of the crack area, identification of crack types, extraction of three-dimensional shapes of cracks, assessment of crack danger levels, and prediction of crack development trends;

[0032] The communication module is used to realize the communication between the crack preliminary positioning module and the crack precise scanning module, and between the crack precise scanning module and the cloud computing server.

[0033] Further, the communication module includes: a wireless access point arranged in the tunnel, a WIFI communication module arranged in the preliminary positioning module and the crack precision scanning module;

[0034] In addition, the communication between the crack preliminary positioning module and the crack precise scanning module is realized through a wireless access point and a WIFI communication module.

[0035] Furthermore, the communication module further includes: a 5G communication module provided in the crack precision scanning module and the cloud service calculator;

[0036] In addition, the communication between the crack precision scanning module and the cloud service computer is realized through the 5G communication module.

[0037] Furthermore, the tunnel inspection robot is also equipped with a vehicle-mounted positioning sensor, and the drone is also equipped with an airborne positioning sensor;

[0038] The vehicle-mounted positioning sensor is used to obtain the relative posture of the tunnel inspection robot and the tunnel entrance at the tunnel entrance, which is recorded as the entrance reference relative posture and sent to the crack precision scanning module;

[0039] In addition, the airborne positioning sensor is used to obtain the relative posture of the UAV and the tunnel entrance at the tunnel entrance before the UAV moves according to the optimal inspection path, recorded as the entrance alignment relative posture, and align the entrance alignment relative posture to the entrance reference relative posture.

[0040] Furthermore, the tunnel crack detection system provided by the present invention further includes: a visual marker disposed at the tunnel entrance;

[0041] Furthermore, the entrance reference relative position and the entrance alignment relative position are both determined with the aid of visual markers.

[0042] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0043] (1) The present invention uses a tunnel inspection robot to obtain three-dimensional data in the tunnel, so as to establish a three-dimensional map in the tunnel and preliminarily detect all crack areas in the tunnel and the positions of each crack area in the three-dimensional map, and accordingly plan and generate the optimal inspection path for the drone, so that the drone can finally move to the vicinity of the crack area in sequence according to the optimal inspection path for precise scanning. Since the drone only needs to scan the crack area during the inspection process, and the path planning ensures that its inspection time is the shortest, the drone can use the depth camera carried on it to quickly collect accurate three-dimensional data of all crack areas, with high data quality, and no need to frequently replace batteries, and the inspection efficiency is high; when planning the inspection path for the drone, the present invention takes into account the obstacle avoidance problem, thereby improving the flight stability of the drone in the tunnel. In general, the present invention uses the tunnel inspection robot and the drone to work together to effectively improve the efficiency and accuracy of crack detection in tunnels, especially in long and large tunnels.

[0044] (2) In the preferred embodiment of the present invention, when the drone moves along the optimal inspection path, it will first align its relative posture relative to the tunnel entrance (entrance alignment relative posture) to the relative posture of the tunnel inspection robot relative to the tunnel entrance (entrance reference relative posture), thereby ensuring that the drone can accurately locate and navigate in the three-dimensional map constructed by the tunnel inspection robot, improving the control accuracy of its posture during the drone inspection process, and ensuring that the fine three-dimensional data collected by the drone can be accurately aligned and fused with the three-dimensional map constructed by the tunnel inspection robot. In a further preferred embodiment, the relative posture of the drone / tunnel inspection robot relative to the tunnel entrance is determined with the help of visual markers, thereby enabling the determination of the relative posture to be achieved quickly and accurately.

[0045] (3) In the preferred embodiment of the present invention, after establishing a three-dimensional map of the tunnel based on the three-dimensional data obtained by the tunnel inspection robot, the blind spots of the tunnel inspection robot in the tunnel are further determined, and a supplementary inspection path is planned for the blind spots. The accurate three-dimensional data of the blind spots are collected by drones as supplementary depth information of the blind spots, and the three-dimensional map of the tunnel is updated and improved using the supplementary depth information. Then, the optimal inspection path of the drone is generated based on the updated and improved three-dimensional map. The optimal inspection path makes more comprehensive use of the three-dimensional information in the tunnel, and can further improve the inspection efficiency and flight stability of the drone.

[0046] (4) The tunnel crack inspection system provided by the present invention adopts a collaborative architecture of cloud computing servers and edge computing servers. The edge computing servers are used to complete relatively simple calculations such as three-dimensional map establishment and preliminary tunnel crack detection, while the complex calculations for analyzing and processing precise three-dimensional data are completed by the cloud computing servers. This reduces a large amount of data transmission, thereby optimizing bandwidth utilization, that is, ensuring the real-time responsiveness and flexibility of the system. It also provides powerful cloud-based analysis capabilities, allowing the system to quickly identify potential risks and conduct in-depth assessments, greatly improving the data processing efficiency and intelligence level of the entire detection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A flow chart of a tunnel crack detection method provided by an embodiment of the present invention;

[0048] Figure 2 A functional block diagram of a tunnel crack detection system provided by an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of the specific structure of a tunnel crack detection system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0051] In the present invention, the terms "first", "second", etc. (if any) in the present invention and the accompanying drawings are used to distinguish similar objects but not necessarily to describe a specific order or sequence.

[0052] In order to solve the technical problem that the existing drone inspection method cannot efficiently and accurately realize crack detection in long tunnels, the present invention provides a tunnel crack detection method and a tunnel crack detection system. The overall concept is that while using a drone equipped with a depth camera to collect three-dimensional data of the crack area to ensure data collection accuracy, by pre-planning the inspection path, the drone does not need to scan other areas outside the crack area during the inspection process to improve the inspection efficiency, solve the endurance problem, achieve obstacle avoidance, and improve flight stability.

[0053] Based on the above concept, the present invention decomposes the entire inspection process into two stages: in the first stage, the tunnel inspection robot collects three-dimensional data in the tunnel to establish a three-dimensional map of the tunnel and preliminarily locate the positions of all crack areas in the tunnel in the three-dimensional map, providing a basis for planning the inspection path of the unmanned aerial vehicle (UAV); in the second stage, based on the three-dimensional map of the tunnel and the positions of each crack area in the three-dimensional map, path planning is carried out to minimize the inspection time of the UAV and achieve obstacle avoidance, an optimal inspection path is generated for the UAV, and the UAV is enabled to collect accurate three-dimensional data of each crack area in turn along the optimal inspection path for subsequent analysis and processing.

[0054] The following are examples.

[0055] Embodiment 1:

[0056] A tunnel crack detection method, such as Figure 1 As shown, including:

[0057] Use a tunnel inspection robot to inspect the tunnel from the tunnel entrance to obtain complete three-dimensional data of the tunnel interior;

[0058] Use the complete 3D data inside the tunnel to build a 3D map of the tunnel, detect crack areas in the tunnel, and determine the location of each crack area in the 3D map;

[0059] Determine the inspection area corresponding to each crack area; the inspection area satisfies: when the drone equipped with the depth camera is located in the inspection area, the crack area is within the field of view of the depth camera;

[0060] Path planning is performed based on the three-dimensional map and the positions of each inspection area in the three-dimensional map to generate the optimal inspection path for the drone. The optimal inspection path allows the drone to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the drone entering the tunnel to leaving the tunnel is the shortest.

[0061] The drone moves along the optimal inspection path, and each time it moves to an inspection area, it uses a depth camera to scan the crack area to obtain detailed three-dimensional data of the crack area;

[0062] Analyze and process the detailed 3D data of the crack area to complete the tunnel crack detection.

[0063] In order to improve the control accuracy of the posture of the drone during the inspection process, as a preferred implementation, in this embodiment, before the tunnel inspection robot starts the inspection in the tunnel from the tunnel entrance, the method further includes: obtaining the relative posture of the tunnel inspection robot at the tunnel entrance and the tunnel entrance, recorded as the entrance reference relative posture;

[0064] Furthermore, before the UAV moves along the optimal inspection path, it also includes: obtaining the relative posture of the UAV and the tunnel entrance at the tunnel entrance, recording it as the entrance alignment relative posture, and aligning the entrance alignment relative posture to the entrance reference relative posture, so that the tunnel inspection robot and the UAV share a unified three-dimensional map coordinate system.

[0065] In order to facilitate the acquisition of the relative position and posture of the drone / tunnel inspection robot relative to the tunnel entrance, this embodiment pre-sets the ArUco mark at the tunnel entrance; the ArUco mark is easy to quickly identify, has high robustness and low sensitivity to illumination and occlusion, and is suitable for precise positioning in dynamic environments such as drones / tunnel inspection robots. Drones / tunnel inspection robots are often equipped with positioning sensors. The positioning sensor uses the Harris algorithm to quickly locate image feature points, uses the SIFT algorithm to extract the descriptors of feature points, ensures the scale and rotation invariance of feature recognition, uses the FLANN algorithm for efficient feature matching, uses the RANSAC algorithm to filter erroneous matches, ensures recognition accuracy, and uses the depth information and matching results to perform PnP algorithm triangulation to obtain the position and posture of the ArUco mark in the coordinate system of the positioning sensor. For drones, the positioning sensor can be an airborne positioning sensor carried thereon, and for tunnel inspection robots, the positioning sensor can be a vehicle-mounted positioning sensor carried thereon.

[0066] It should be noted that the use of ArUco markers to determine the relative pose is only an optional implementation and should not be understood as the only limitation of the present invention. In some other implementations of the present invention, other visual markers such as AprilTag may also be used. In some other implementations of the present invention, visual markers may not be used, and the relative pose may be determined directly based on coordinate transformation or other methods.

[0067] Optionally, in this embodiment, the three-dimensional map is constructed by the Gmapping algorithm; the preliminary detection of the crack area in the tunnel is performed by a pre-trained detection model. It should be noted that the construction of the three-dimensional map and the preliminary detection method of the crack area are only exemplary descriptions and should not be understood as the only limitation of the present invention.

[0068] In practical applications, according to specific detection requirements, the analysis and processing performed on the fine three-dimensional data of the crack area includes at least one of the following:

[0069] The data is processed using 3D reconstruction technology to identify the general structure and the approximate location of the cracks and achieve preliminary 3D reconstruction;

[0070] After the initial reconstruction, data cleaning is performed to remove noise, and then feature enhancement technology is used to highlight the crack features and improve their visibility in the model;

[0071] Use 3D segmentation technology to accurately distinguish cracks from other surface features, so as to accurately segment cracks and achieve crack identification;

[0072] After cracks are identified, a deep learning algorithm is used to extract the three-dimensional shape of the cracks and their parameters, including length, width, and depth;

[0073] Based on the extracted 3D shape and parameters of the cracks, the crack types are identified and detailed 3D reconstruction is performed to comprehensively analyze the morphology and distribution of the cracks;

[0074] Conduct in-depth analysis of crack data, assess crack hazard level, predict crack development trend, evaluate maintenance needs, and simulate the potential impact of cracks on structural performance to provide scientific support for structural health detection and maintenance decisions.

[0075] In general, the tunnel crack detection method provided in this embodiment overcomes the limitations of the traditional single detection method through the air-ground collaborative work of the tunnel inspection robot and the drone. The tunnel inspection robot is responsible for building a three-dimensional map of the tunnel and preliminarily locating the crack area, while the drone performs close-range and accurate scanning, achieving comprehensive and efficient tunnel detection. This air-ground collaborative working mode significantly improves the coverage and accuracy of detection, effectively solves the blind spot problem in traditional tunnel detection, and can more comprehensively detect and identify potential risks in tunnel structures.

[0076] This embodiment uses a drone equipped with a depth camera to collect three-dimensional data of the crack area. Combined with the maneuverability and flexibility of the drone, it can perform high-precision close-range scanning of the crack area, improving the quality of image acquisition. This method not only improves the accuracy of crack detection, but also obtains the three-dimensional geometric features of the crack, providing rich data support for subsequent crack analysis and evaluation, and greatly improving the accuracy of tunnel structure health assessment.

[0077] The present invention adopts the detection strategy of "preliminary positioning + precise scanning", which effectively solves the problem of short endurance of drones in long tunnels. During the inspection process, the drone only needs to perform a close-range precise scan of the potential crack area based on the optimal inspection path generated by the drone's position in the three-dimensional map and the position of the crack area in the three-dimensional map, without a comprehensive and detailed inspection, which greatly reduces the workload and flight time of the drone and improves flight stability. This strategy not only improves the detection efficiency, but also significantly extends the effective working time of the drone, enabling it to play a greater role in long tunnels.

[0078] Embodiment 2:

[0079] A method for detecting cracks in a tunnel. This embodiment is similar to the above-mentioned embodiment 1, except that this embodiment takes into account that during the inspection process of the tunnel inspection robot, due to the existence of blind spots in the field of vision, the collected three-dimensional data may be missing, resulting in the three-dimensional map constructed based on the data being unable to accurately reflect the three-dimensional environment in the tunnel. Therefore, after initially establishing a three-dimensional map of the tunnel and initially locating the crack area, before determining the inspection area corresponding to each crack area, this embodiment also includes:

[0080] Determine the blind area of ​​the tunnel inspection robot in the tunnel and the position of the blind area in the three-dimensional map according to the three-dimensional map;

[0081] Path planning is performed based on the three-dimensional map and the positions of each blind spot in the three-dimensional map to generate a supplementary inspection path for the drone; the supplementary inspection path allows the drone to enter the tunnel from the tunnel entrance, pass through each blind spot in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the drone entering the tunnel to leaving the tunnel is the shortest;

[0082] The drone is moved along the supplementary inspection route, and each time the drone moves to a blind area, the drone uses a depth camera to scan the blind area to obtain fine three-dimensional data of the blind area as supplementary depth information of the blind area;

[0083] The three-dimensional map is updated using supplementary depth information from blind areas of vision.

[0084] After updating the three-dimensional map using the supplementary depth information, this embodiment determines the inspection area corresponding to each crack area, and performs path planning based on the updated three-dimensional map and the position of the inspection area in the three-dimensional map.

[0085] The blind spots of the tunnel inspection robot are areas where the sensors carried by the tunnel inspection robot cannot fully cover or where it is difficult to obtain accurate data due to factors such as the sensor's field of view and obstacles. These blind spots can be identified and located by performing point cloud data analysis on the three-dimensional map initially constructed by the tunnel inspection robot. Specifically, in point cloud data analysis, the blind spots of the tunnel inspection robot's field of view are usually characterized by the following features: missing point cloud data, poor point cloud data quality, or significantly lower point cloud density than the surrounding area.

[0086] In this embodiment, the remaining steps are the same as those in the above-mentioned embodiment 1. For the specific implementation method, reference may be made to the description in the above-mentioned embodiment 1.

[0087] This embodiment uses a drone to collect accurate three-dimensional data of blind spots as supplementary depth information for the blind spots, and uses the supplementary depth information to update and improve the three-dimensional map of the tunnel, and then generates an optimal inspection path for the drone based on the updated and improved three-dimensional map. The optimal inspection path makes more comprehensive use of the three-dimensional information in the tunnel, and can further improve the inspection efficiency and flight stability of the drone.

[0088] Embodiment 3:

[0089] A tunnel crack detection system, such as Figure 2 As shown, it includes: a crack preliminary positioning module, a crack precise scanning module, a communication module and a cloud computing server;

[0090] The crack preliminary positioning module includes: a tunnel inspection robot and a mobile edge computing server mounted on the tunnel inspection robot;

[0091] The tunnel inspection robot is used to inspect inside the tunnel to obtain complete three-dimensional data inside the tunnel and send it to the edge computing server;

[0092] The mobile edge computing server is used to construct a three-dimensional map of the interior of the tunnel using the complete three-dimensional data of the interior of the tunnel, detect crack areas in the tunnel, determine the position of each crack area in the three-dimensional map, and send the three-dimensional map and the position of each crack area in the three-dimensional map to the crack precision scanning module;

[0093] The crack precision scanning module includes a drone equipped with a depth camera and a posture calculation unit mounted on the drone;

[0094] The drone is used to scan the target within the field of view of its depth camera to obtain detailed three-dimensional data of the target;

[0095] The posture calculation unit is used to determine the inspection area corresponding to each crack area; the inspection area satisfies: when the UAV is located in the inspection area, the crack area is within the field of view of the depth camera;

[0096] The posture calculation unit is also used to perform path planning according to the three-dimensional map and the positions of each inspection area in the three-dimensional map, and generate an optimal inspection path for the UAV; the optimal inspection path enables the UAV to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in turn while meeting the obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the UAV entering the tunnel to leaving the tunnel is the shortest;

[0097] The posture calculation unit is also used to make the UAV move according to the optimal inspection path, and each time the UAV moves to an inspection area, it uses the depth camera to scan the crack area to obtain fine three-dimensional data of the crack area and send it to the cloud computing server;

[0098] The cloud computing server is located outside the tunnel and is used to perform analysis and processing on the detailed three-dimensional data of the crack area to complete the tunnel crack detection;

[0099] The communication module is used to realize the communication between the crack preliminary positioning module and the crack precise scanning module, and between the crack precise scanning module and the cloud computing server.

[0100] like Figure 3 As shown, in this embodiment, the communication module includes a wireless access point, a WIFI communication module and a 5G communication module. The wireless access points are arranged at intervals inside the tunnel to achieve WiFi network coverage in the tunnel; the WIFI communication modules are arranged on the tunnel inspection robot and the drone respectively; the tunnel inspection robot communicates with the drone through the WIFI network, and the 5G communication module is mounted on the ground terminal device; the 5G network is composed of a 5G communication base station and a 5G router, the cloud computing server communicates with the ground terminal device through the 5G network, the ground terminal device communicates with the drone through the WIFI network, and the drone communicates with the cloud computing server through the WIFI network and the 5G network.

[0101] This embodiment also includes a visual marker set at the tunnel entrance to facilitate determining the relative posture of the tunnel inspection robot and the drone relative to the tunnel entrance, and to achieve alignment of the initial relative posture of the inspection. Optionally, in this embodiment, the visual marker set at the tunnel entrance is specifically an ArUco marker. It should also be noted that in some other embodiments of the present invention, other visual markers such as AprilTag may also be used.

[0102] Optionally, in this embodiment, the tunnel inspection robot is equipped with a three-dimensional laser radar, an inertial measurement unit, an odometer, a vehicle-mounted positioning sensor and a vehicle-mounted main controller; the three-dimensional laser radar is used to scan the basic three-dimensional data of the tunnel in real time; the inertial measurement unit is used to obtain the roll angle, pitch angle and heading angle of the inspection robot relative to the horizontal plane in real time; the odometer is used to measure the mileage of the tunnel inspection robot; the vehicle-mounted positioning sensor is used to detect the visual marks at the tunnel entrance to obtain the relative posture of the visual marks at the tunnel entrance and the tunnel inspection robot, which is recorded as the entrance reference relative posture; the vehicle-mounted main controller constructs a three-dimensional map of the interior of the tunnel by processing and fusing the data of the laser radar sensor, the inertial measurement unit and the odometer, and obtains the posture of the tunnel inspection robot in the map coordinate system of the three-dimensional map in real time.

[0103] Optionally, in this embodiment, in addition to the depth camera, the drone is also equipped with an airborne inertial measurement unit, an airborne positioning sensor and an airborne main controller, and includes a multi-rotor frame; wherein the multi-rotor frame is used to carry the depth camera, the airborne inertial measurement unit, the airborne positioning sensor and the airborne main controller, and provides vertical take-off and landing, hovering, and rapid maneuvering capabilities in narrow or complex environments; the depth camera is used to capture high-resolution depth images of the tunnel surface and structure to obtain accurate three-dimensional spatial information, supplement the three-dimensional map constructed by the tunnel inspection robot and obtain three-dimensional data of the crack area; the airborne inertial measurement unit is used to measure the angular velocity and acceleration of the corresponding coordinate axes in the three-dimensional space; the airborne positioning sensor is used to detect the visual mark at the tunnel entrance to obtain the relative posture of the visual mark at the tunnel entrance and the drone, recorded as the entrance alignment relative posture; the airborne main controller is used to receive positioning information and control the drone to complete the corresponding action; the posture calculation unit is used to combine the entrance reference relative posture, the entrance alignment relative posture and the posture of the tunnel inspection robot in the three-dimensional map to obtain the posture of the drone in the three-dimensional map.

[0104] Optionally, in this embodiment, the drone is specifically a six-rotor drone. In some other embodiments of the present invention, other types of drones may also be used. The depth camera carried by the drone can be a ToF (Time of Flight) camera. The ToF camera uses the time of flight principle to measure the distance between the target object and the camera. By emitting light waves and measuring the time required for them to reflect back, it generates a depth image and three-dimensional point cloud data. It has high detection accuracy and accurate depth map. It is not limited by ambient lighting conditions and can work normally even in tunnel environments with insufficient lighting. During the inspection process, each time the drone moves to an inspection area, it will use the ToF camera to perform a high-precision close-range scan of the tunnel surface where the crack area is located, and collect three-dimensional data of the crack area. The three-dimensional data describes the geometric characteristics of the cracks, such as the size, shape, and spatial distribution of the cracks.

[0105] Optionally, in this embodiment, after collecting accurate three-dimensional data of the crack area, the drone will package the three-dimensional data of the crack area and the position of the corresponding crack area in the map coordinate system of the tunnel three-dimensional map and mark them with a time stamp to obtain a detection data packet of the crack area, and then transmit the detection data packet to the cloud computing server in real time through the 5G network for further analysis and processing.

[0106] In this embodiment, after receiving the detection data packet of the crack area uploaded by the drone, the cloud service computer extracts the fine three-dimensional data of the crack area and performs analysis and processing to complete the tunnel crack detection; in practical applications, according to specific detection requirements, the cloud service computer performs at least one of the following analysis and processing on the fine three-dimensional data of the crack area: processing the data through three-dimensional reconstruction technology to identify the general structure and the approximate location of the crack, and realize preliminary three-dimensional reconstruction;

[0107] After the initial reconstruction, data cleaning is performed to remove noise, and then feature enhancement technology is used to highlight the crack features and improve their visibility in the model;

[0108] Use 3D segmentation technology to accurately distinguish cracks from other surface features, so as to accurately segment cracks and achieve crack identification;

[0109] After cracks are identified, a deep learning algorithm is used to extract the three-dimensional shape of the cracks and their parameters, including length, width, and depth;

[0110] Based on the extracted 3D shape and parameters of the cracks, the crack types are identified and detailed 3D reconstruction is performed to comprehensively analyze the morphology and distribution of the cracks;

[0111] Conduct in-depth analysis of crack data, assess crack hazard level, predict crack development trend, evaluate maintenance needs, and simulate the potential impact of cracks on structural performance to provide scientific support for structural health detection and maintenance decisions.

[0112] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A tunnel crack detection method, characterized in that: include: Using a tunnel inspection robot to inspect the tunnel from the tunnel entrance to obtain complete three-dimensional data of the interior of the tunnel; Using the complete three-dimensional data of the interior of the tunnel to construct a three-dimensional map of the interior of the tunnel, detecting crack areas in the tunnel, and determining the position of each crack area in the three-dimensional map; Determine the inspection area corresponding to each crack area; The inspection area satisfies that: when the drone equipped with the depth camera is located in the inspection area, the crack area is within the field of view of the depth camera; Performing path planning according to the three-dimensional map and the position of each inspection area in the three-dimensional map to generate an optimal inspection path for the UAV; The optimal inspection path enables the UAV to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in sequence while meeting obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the UAV entering the tunnel to leaving the tunnel is the shortest; The drone moves along the optimal inspection path, and each time the drone moves to an inspection area, the drone uses the depth camera to scan the crack area to obtain fine three-dimensional data of the crack area; Perform analysis and processing on the detailed 3D data of the crack area to complete the tunnel crack detection; The analysis and processing includes at least one of the following: three-dimensional reconstruction of the crack area, identification of the crack type, extraction of the three-dimensional shape of the crack, assessment of the degree of crack danger, and prediction of crack development trend.

2. The tunnel crack detection method according to claim 1, characterized in that: Before using the tunnel inspection robot to inspect in the tunnel from the tunnel entrance, the method further includes: obtaining a relative posture between the tunnel inspection robot and the tunnel entrance at the tunnel entrance, recorded as an entrance reference relative posture; Furthermore, before making the UAV move according to the optimal inspection path, it also includes: obtaining the relative posture of the UAV and the tunnel entrance at the tunnel entrance, recording it as the entrance alignment relative posture, and aligning the entrance alignment relative posture to the entrance reference relative posture, so that the tunnel inspection robot and the UAV share a unified three-dimensional map coordinate system.

3. The tunnel crack detection method according to claim 2, characterized in that: Also includes: Pre-setting visual markers at the tunnel entrance; Furthermore, the entrance reference relative posture and the entrance alignment relative posture are both determined with the aid of the visual marker.

4. The tunnel crack detection method according to any one of claims 1 to 3, characterized in that: Before determining the inspection area corresponding to each crack area, the method further includes: determining the blind area of ​​the tunnel inspection robot in the tunnel and the position of the blind area in the three-dimensional map according to the three-dimensional map; Path planning is performed according to the three-dimensional map and the positions of each blind area in the three-dimensional map to generate a supplementary inspection path for the UAV; the supplementary inspection path enables the UAV to enter the tunnel from the tunnel entrance, pass through each blind area in turn while meeting obstacle avoidance conditions, and then leave the tunnel through the tunnel exit, and the total time from the UAV entering the tunnel to leaving the tunnel is the shortest; The drone moves along the supplementary inspection path, and each time the drone moves to a blind area, the drone uses the depth camera to scan the blind area to obtain fine three-dimensional data of the blind area as supplementary depth information of the blind area; The three-dimensional map is updated using the supplementary depth information of the blind area of ​​vision.

5. A tunnel crack detection system, characterized in that: include: Crack preliminary positioning module, crack precise scanning module, communication module and cloud computing server; The crack preliminary positioning module includes: a tunnel inspection robot and a mobile edge computing server mounted on the tunnel inspection robot; The tunnel inspection robot is used to inspect in the tunnel to obtain complete three-dimensional data inside the tunnel and send it to the edge computing server; The mobile edge computing server is used to construct a three-dimensional map of the interior of the tunnel using the complete three-dimensional data of the interior of the tunnel, detect crack areas in the tunnel, determine the position of each crack area in the three-dimensional map, and send the three-dimensional map and the position of each crack area in the three-dimensional map to the crack precision scanning module; The crack precision scanning module includes a drone equipped with a depth camera and a posture calculation unit mounted on the drone; The drone is used to scan a target within the field of view of its depth camera to obtain fine three-dimensional data of the target; The posture calculation unit is used to determine the inspection area corresponding to each crack area; the inspection area satisfies: when the UAV is located in the inspection area, the crack area is within the field of view of the depth camera; The posture calculation unit is further used to perform path planning according to the three-dimensional map and the position of each inspection area in the three-dimensional map, and generate an optimal inspection path for the UAV; the optimal inspection path enables the UAV to enter the tunnel from the tunnel entrance, pass through each inspection area in the tunnel in sequence while meeting the obstacle avoidance condition, and then leave the tunnel through the tunnel exit, and the total time from the UAV entering the tunnel to leaving the tunnel is the shortest; The posture calculation unit is also used to enable the drone to move according to the optimal inspection path, and each time the drone moves to an inspection area, the depth camera is used to scan the crack area to obtain fine three-dimensional data of the crack area, and send it to the cloud computing server; The cloud computing server is located outside the tunnel and is used to perform analysis and processing on the fine three-dimensional data of the crack area to complete the tunnel crack detection; the analysis and processing includes at least one of the following: three-dimensional reconstruction of the crack area, identification of crack types, extraction of three-dimensional shapes of cracks, assessment of crack danger levels, and prediction of crack development trends; The communication module is used to realize the communication between the crack preliminary positioning module and the crack precise scanning module, and between the crack precise scanning module and the cloud computing server.

6. The tunnel crack detection system according to claim 5, characterized in that: The communication module includes: a wireless access point arranged in the tunnel, a WIFI communication module arranged in the preliminary positioning module and the crack precision scanning module; Furthermore, the communication between the crack preliminary positioning module and the crack precise scanning module is realized through the wireless access point and the WIFI communication module.

7. The tunnel crack detection system according to claim 6, characterized in that: The communication module further includes: a 5G communication module provided in the crack precision scanning module and the cloud service calculator; In addition, the communication between the crack precision scanning module and the cloud service calculator is achieved through the 5G communication module.

8. The tunnel crack detection system according to any one of claims 5 to 7, characterized in that: The tunnel inspection robot is also equipped with a vehicle-mounted positioning sensor, and the drone is also equipped with an airborne positioning sensor; The vehicle-mounted positioning sensor is used to obtain the relative posture of the tunnel inspection robot and the tunnel entrance at the tunnel entrance, record it as the entrance reference relative posture, and send it to the crack precision scanning module; In addition, the airborne positioning sensor is used to obtain the relative posture of the drone and the tunnel entrance at the tunnel entrance before the drone moves according to the optimal inspection path, recorded as the entrance alignment relative posture, and align the entrance alignment relative posture to the entrance reference relative posture.

9. The tunnel crack detection system according to claim 8, characterized in that: Also includes: visual markings provided at the entrance of the tunnel; Furthermore, the entrance reference relative posture and the entrance alignment relative posture are both determined with the aid of the visual marker.

Images