Method and system for inspecting tunnels

Abstract

The present invention relates to a method and system for inspecting tunnels, comprising: calculating at least one datum related to at least one distance inspected by at least one transport unit in the tunnel via at least one distance sensor; capturing at least one image via at least one image sensor; analysing the at least one image via at least one processing module in order to obtain at least one parameter of the inspected distance; obtaining at least one 3D parametric model from the at least one distance travelled and at least one image via at least one analysis module; and linking the at least one 3D parametric module to the at least one parameter, via the at least one analysis module, in order to obtain at least one digital twin of the at least one inspected distance of the tunnel.

Description

Method and system for tunnel inspection DESCRIPTIVE MEMORANDUM

[0001] The present invention relates to a method and system for tunnel inspection, which allows for the monitoring of the condition of tunnels, for example, mining tunnels or similar, where the walls are reinforced by fastening elements such as bolts. In this way, the method and system of the invention allows for the generation of a three-dimensional digital twin through the use of artificial intelligence in combination with on-site inspections.

[0002] In this sense, the method of the invention essentially comprises the steps of: indicating at least one reference point in a tunnel where the inspection begins using at least one conveyor, through at least one distance sensor; indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement, through at least one image sensor; indicating information regarding at least one theoretical cross-section of the tunnel, through at least one image sensor; calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel, through at least one distance sensor; capturing at least one image, through at least one image sensor; analyzing the at least one image obtained through at least one processing module, to obtain at least one parameter of the inspected distance;to obtain at least one parametric three-dimensional model, from at least one distance traveled and at least one image, through at least one analysis module; and to link the at least one parametric three-dimensional model with the at least one parameter, through the at least one analysis module, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

[0003] Furthermore, the system of the invention essentially comprises at least one processing module, transported by at least one conveyor, comprising: at least one distance sensor, for: i) indicating at least one reference point in a tunnel where the inspection begins by means of the at least one conveyor; and ii) calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel; at least one image sensor, for: i) indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement; ii) indicating information regarding at least one theoretical cross-section of the tunnel; and iii) capturing at least one image; wherein the at least one processing module analyzes the at least one image to obtain at least one parameter of the inspected distance;and at least one analysis module, to: i) obtain at least one parametric three-dimensional model, from at least one distance traveled and at least one image; and ii) link the at least one parametric three-dimensional model with at least one parameter, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

[0004] Based on the method and system of the invention, it is possible to reduce inspection times in tunnels, improving productivity in sectors such as mining, where work stoppages due to safety problems in tunnel reinforcements are to be avoided at all costs. BACKGROUND

[0005] In the mining industry, tunnel reinforcement is a key process that allows for safe mineral extraction operations for workers. Generally, these tunnels are inspected by quality control teams who walk through them, verifying the condition of every meter traveled and every bolt installed. Any issues regarding the quality of one or more bolts must be rectified as quickly as possible to avoid payment delays from clients to the construction companies.

[0006] In this sense, current quality teams only have their eyes on the ground, mining lamps, sheets of paper and calculation tables to carry out inspections and keep track of this process over time, which makes the inspections take a considerable amount of time (a mining tunnel can be more than 4,000 km long), and increases the risk of error by placing all the responsibility on what each quality inspector observes from the visual inspection.

[0007] Therefore, there is a need not only for a system and method that improves the accuracy and delivery of information regarding tunnel fortification inspections through automation, but also for a solution that allows these inspection tasks to be carried out more quickly, so that all the necessary information is available to enable better decision-making by those in charge of maintaining these tunnels.

[0008] In the field of patents, there are solutions that address methods and systems for tunnel inspection. For example, the Korean patent KR102396675B1 describes a 3D tunnel position estimation and mapping system for autonomous underground mining robots using a LiDAR sensor, which generates accurate 3D tunnel maps, as well as a method for doing so.According to the document, the method comprises: a step of using a horizontal LiDAR sensor installed on an autonomous underground mining robot to detect a horizontal tunnel wall and perform a pattern comparison on the detected horizontal tunnel wall to calculate the robot's matching accuracy and heading angle; a step of using values ​​measured by an inertial measurement unit (IMU) sensor and an encoder to estimate the robot's position; and a step of merging the estimated position information of the robot and a vertical tunnel section mapped by data detected through a vertical LiDAR sensor to generate a 3D map.

[0009] In this regard, document KR102396675B1, while describing a system that allows mapping the walls of a tunnel in an automated way, does not disclose or suggest anything about monitoring the condition of the fortification elements placed on the tunnel walls.

[0010] Another example is European patent application EP3933756A1, which describes a method for controlling a building element. Using an image capture device and a computer device, the method comprises the steps of assigning a visual feature to each of a first and a second region of the building element; taking a first plurality of images of the visual features using the image capture device at a first time; taking a second plurality of images of the visual features using the image capture device at a second time; and having the computer device process the images from the first and second plurality of images to derive at least one of the distances between the two visual features and the relative orientations of the two visual features within the first plurality of images and within the second plurality of images; and have the computer device deduce a state of the building element from at least one of the difference between the distances and the difference between the relative orientations.

[0011] Upon examining the main features of this patent application, it can be observed that, while it describes some particularities of the invention, such as taking photographs to be processed by a computer device to determine the condition of a structural element, such as a reinforcement bolt, or the use of LiDAR-type technologies, instead of GPS technology, to accurately determine the location within the mine, there are other fundamental features of the invention that are not described in this patent application. Among them, the following can be mentioned: 1) Only one photograph of each area to be traversed in the tunnel is needed to detect the reinforcement bolts and their condition.2) It is possible to monitor the tunnel through multiple mobile devices, which inspect different sections of the tunnel, which can be processed by the technology software to build the digital twin through the different inspected sections.

[0012] In view of the above, the need for a technological solution that addresses the shortcomings of state-of-the-art solutions is evident, since none of them adequately handle the automated monitoring of tunnels, specifically the reinforcement elements that protect their walls from collapses, in order to have a quick and accurate analysis of the condition of each of them in the different sections of interest within the tunnel, which can be visualized by anyone located outside the tunnel through the generation of a digital twin.

[0013] This and other advantages associated with other aspects of the technology are described in more detail below. DESCRIPTION OF THE INVENTION

[0014] The invention relates to a method and system for tunnel inspection, which increases the speed at which inspections are carried out, while maintaining the level of information pressure on the condition of the fortification elements arranged on the tunnel walls.

[0015] According to a first preferred embodiment of the invention, the method for tunnel inspection comprises: a) indicating at least one reference point in a tunnel where the inspection begins by means of at least one conveyor, through at least one distance sensor comprising at least one processing module; b) indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement, through at least one image sensor comprising the at least one processing module; c) indicating information regarding at least one theoretical cross-section of the tunnel, through the at least one image sensor; d) calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel, through the at least one distance sensor; e) capture at least one image, through at least one image sensor; f) analyze the at least one image obtained through at least one processing module, to obtain at least one parameter of the inspected distance; g) obtain at least one parametric three-dimensional model, from the at least one distance traveled and from at least one image, through at least one analysis module; and h) link the at least one parametric three-dimensional model with the at least one parameter, through the at least one analysis module, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

[0016] The present invention proposes to carry out tunnel inspections using a processing module arranged as a mobile device, comprising an image sensor, such as a camera, capable of recognizing fasteners installed in the tunnel walls, such as plates and reinforcing bolts, through the use of algorithms trained with artificial intelligence. Once this data has been collected, a digital twin is constructed, allowing for the optimized, rapid, and clear observation of potential defects in the tunnel walls.

[0017] This allows operators to inspect the tunnel as they walk through it, as well as divide the operators into teams that walk through different sections of the tunnel, which will increase the speed and reliability of data collection.

[0018] According to another embodiment of the invention, step a) further comprises: turning on at least one processing module; locating and turning on at least one distance sensor at the at least one reference point of the tunnel; and turning on at least one image sensor.

[0019] According to another embodiment of the invention, the method further comprises, before step d), starting the journey through the tunnel by means of at least one conveyor.

[0020] According to another embodiment of the invention, the method further comprises, prior to step e), sending at least one piece of data related to the at least one inspected distance from the at least one distance sensor to the at least one image sensor, through at least one antenna.

[0021] According to another embodiment of the invention, step f) further comprises: activating at least one image cropping and segmentation algorithm; activating at least one fixing element counting algorithm; activating at least one quality recognition and classification algorithm.

[0022] According to another embodiment of the invention, the method further comprises compiling the data obtained through at least one algorithm to obtain at least one parameter, which corresponds to one or more of: at least one image of the at least one distance traveled; at least one piece of data related to the distance inspected in the tunnel, delivered in a unit of measurement; at least one number for each at least one fastening element; and at least one quality classification for each at least one fastening element.

[0023] According to another embodiment of the invention, the method further comprises storing at least one parameter in at least one quality database.

[0024] According to another embodiment of the invention, the method further comprises deciding whether to inspect at least one additional distance, wherein, in the event of inspecting at least one additional distance, the method further comprises returning to step d).

[0025] According to another embodiment of the invention, the method further comprises, prior to step f), importing information from at least one quality database to at least one analysis database of at least one analysis module.

[0026] According to another embodiment of the invention, the method further comprises, prior to step g): extracting at least one distance traveled and at least one image, through at least one analysis module; and extracting at least one parameter related to at least one quality classification for each at least one fixing element, through at least one anaphysics module.

[0027] According to another embodiment of the invention, the method further comprises visualizing at least one digital twin of the at least one inspected distance through at least one visualization means.

[0028] According to another embodiment of the invention, the method further comprises: visualizing at least one parameter related to at least one quality classification for each at least one fastening element; and visualizing at least one image of the at least one image for each at least one inspected distance.

[0029] Furthermore, according to a second preferred embodiment of the invention, a tunnel inspection system is also described, comprising: - at least one processing module, transported by at least one conveyor, comprising: or at least one distance sensor, for: i. indicating at least one reference point in a tunnel where the inspection by means of the at least one conveyor begins; and ii. calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel; or at least one image sensor, for: i. indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement; ii. indicating information regarding at least one theoretical cross-section of the tunnel; and iii. capturing at least one image; wherein the at least one processing module analyzes the at least one image to obtain at least one parameter of the inspected distance; and - at least one analysis module, to: i. obtain at least one parametric three-dimensional model, from at least one distance traveled and at least one image; and ii. link at least one parametric three-dimensional model with at least one parameter, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

[0030] The system of the invention comprises three essential parts: the hardware part, the analysis part, and the three-dimensional modeling.

[0031] The hardware consists of the processing module, configured and arranged as a mobile device, which operators can quickly and easily transport through the tunnels to be inspected, for example, in a mine, having the freedom to take all the data they deem necessary.

[0032] Subsequently, the data obtained for each of the inspected sections of the tunnel are sent to an analysis module, which can also be located on the mobile device or remotely, in a data cloud, where they are processed, in order to generate the digital twin of each of the inspected sections, which the user can visualize through a visualization interface.

[0033] The three-dimensional model or digital twin is capable of updating in real time with the information provided by the mobile device, once it is able to connect with the cloud servers, thus allowing the user to anticipate failure events that may occur in any of the inspected sections, proceeding with the corresponding mitigation measures, which contributes to a lower number of payment withholdings and a lower risk of work stoppages due to failures or collapses, with the consequent risk of accidents for the operators working inside.

[0034] According to another embodiment of the invention, the at least one conveyor corresponds to at least one user. This user may be a tunnel operator or external personnel hired exclusively for inspection work.

[0035] According to another embodiment of the invention, the at least one transporter corresponds to at least one motorized land vehicle.

[0036] According to another embodiment of the invention, the at least one transporter corresponds to at least one unmanned aerial vehicle.

[0037] According to another embodiment of the invention, the at least one processing module further comprises at least one antenna, for sending at least one data related to the at least one inspected distance from the at least one distance sensor to the at least one image sensor.

[0038] According to another embodiment of the invention, the at least one antenna corresponds to at least one module based on LoRA technology.

[0039] According to another embodiment of the invention, the system further comprises at least one quality database for storing at least one parameter.

[0040] According to another embodiment of the invention, the system further comprises at least one analysis database of at least one analysis module, into which information is imported from at least one quality database.

[0041] According to another embodiment of the invention, the import of information is done wirelessly.

[0042] According to another embodiment of the invention, the import of information is carried out through at least one cable.

[0043] According to another embodiment of the invention, the at least one analysis module comprises the at least one processing module.

[0044] According to another embodiment of the invention, at least one analysis module is hosted in a data cloud.

[0045] According to another embodiment of the invention, the system further comprises at least one visualization means, for visualizing one or more of: the at least one digital twin of the at least one inspected distance; the at least one parameter; and at least one image of the at least one image for each at least one inspected distance.

[0046] From the above, it is possible to observe that the present invention contributes to providing an improved solution to the problem related to the inspection of fixing elements in tunnels, where thanks to the use of algorithms trained with artificial intelligence, capable of detecting the fixing elements in the tunnel walls from images, the analysis of the inspected areas can be obtained more quickly, maintaining a high degree of precision, thus allowing the people in charge of tunnel maintenance to make decisions more quickly and effectively. BRIEF DESCRIPTION OF THE FIGURES

[0047] As part of the present invention, the following representative figures are presented, which show preferred configurations of the invention and, therefore, should not be considered as limiting the definition of the claimed subject matter.

[0048] Figure 1 shows a block diagram related to the start-up of the system and method of the invention, according to a preferred configuration.

[0049] Figure 2 shows a block diagram related to the stages of the method of the invention for taking data inside the tunnel, according to a preferred configuration.

[0050] Figure 3 shows a block diagram related to the stages of the method of the invention for generating and visualizing the digital twin of the tunnel, according to a preferred configuration.

[0051] Figure 4 shows a schematic of the system of the invention, according to a preferred configuration DETAILED DESCRIPTION OF THE FIGURES

[0052] With reference to the accompanying figures, Figures 1, 2 and 3 show the block diagrams corresponding to the tunnel inspection method of the invention, according to a preferred configuration.

[0053] Specifically, Figure 1 shows that the method begins in a stage (E1), where the person or people responsible for inspecting the tunnel travel to it, dividing into sections assigned to each of them. Subsequently, each person turns on the mobile device (E2), which in turn activates the processing module (E3). Next, each person must position themselves at the beginning of the tunnel section to be inspected (E4) to activate the distance sensor (E7), which then indicates a reference point in the tunnel where the inspection begins. (E8). In parallel with these actions, the system activates the antenna (E5) and the image sensor (E6), after which the image sensor indicates a theoretical quantity of the fixing elements on the tunnel walls per meter (E9), and provides information regarding at least one theoretical cross-section or profile of the tunnel (E10). Once all these actions have been carried out, the inspection begins and each of the personnel in charge (El 1) starts walking.

[0054] Regarding Figure 2, it shows the stages following the start of the inspection by each of the personnel responsible for these tasks (E1). Once the inspection begins, the distance sensor is activated (E2) to calculate data related to the distance inspected by the person in charge through the tunnel (E13). Subsequently, this information is sent via the antenna, equipped with LoRA technology, to the image sensor (E14). Then, an image of the inspected section is captured by the image sensor (E15). From this image, three algorithms are activated: one related to image cropping and segmentation (E16), another to counting the fasteners, such as plates and / or reinforcing bolts (E17), and a third to recognizing and classifying the quality of each of the fasteners (E8).Once the algorithms are running, the analyzed data (E9) is compiled to obtain the different parameters that will allow the creation of the tunnel's digital twin. These parameters are related to an image of the distance traveled by the inspector (E20), data related to the distance inspected in the tunnel (E21), provided in a unit of measurement, for example, meters, a number for each identified fastener (E22), and a quality rating for each fastener (E23). Upon completion of parameter generation, using the three algorithms trained with artificial intelligence, the parameters are stored in a quality database (E24), which can be hosted locally on a physical device or in a data cloud.Finally, the system detects (E25) whether the person in charge will continue inspecting the next unit of measurement defined for image generation, in which case it returns to stage (E12), or whether the person in charge has finished the inspection (E26).

[0055] Finally, Figure 3 shows the steps involved in generating the digital twin after the inspection is completed by one or more of the tunnel inspectors (E26). If there is no internet connection in the tunnel, the inspector must return to a location with a connection (E27), where the information is extracted from the quality database (E28) and entered into the analysis module (E29). This module stores the information in an analysis database (30), which can be located within the analysis module or in the cloud, depending on the user's needs. Once the information is entered into the analysis module, it extracts the distance traveled by the inspector and the image(s) obtained from the inspected sections (E31). It also extracts the parameters related to the quality classification for each fastening element (E32).Using this information, the analysis module is able to obtain a parametric three-dimensional model of the inspected tunnel section (E33). This parametric three-dimensional model is then linked to the quality classification parameters (E34) to obtain the digital twin of the inspected tunnel section (E35). Visualization is also possible based on the quality classification parameters for each fastener element (E32). The parameters related to the quality classification for each fastening element (E36), as well as the images of the inspected tunnel sections (E37). Both the digital twin (E35), and the parameters related to the quality classification for each fastening element (E36), and the images of the inspected tunnel sections (E37) are accessible to the operators or decision-makers regarding the maintenance of the inspected tunnel, through a display device (E38), which may correspond to a computer, tablet or smartphone, for example.

[0056] Furthermore, Figure 4 shows the main components of the tunnel inspection system (1). Specifically, it can be seen that the system (1) comprises a processing module (10), which includes a distance sensor (11) and an image sensor (12). As mentioned previously, the distance sensor (11) allows the operator to specify at least one reference point in the tunnel where the inspection begins, as well as calculate at least one data point related to at least one distance inspected. The image sensor (12), on the other hand, allows the operator to specify a theoretical quantity of at least one tunnel fastener per unit of measurement; to provide information regarding at least one theoretical cross-section of the tunnel; and to capture at least one image.Through this information, the processing module (10) can obtain at least one parameter of the inspected distance, which will subsequently be used by the analysis module to obtain the digital twin.

Claims

CLAIMS 1. A method for tunnel inspection, CHARACTERIZED in that it comprises the steps of: a) indicating at least one reference point in a tunnel where the inspection begins by means of at least one conveyor, through at least one distance sensor comprising at least one processing module; b) indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement, through at least one image sensor comprising the at least one processing module; c) indicating information regarding at least one theoretical cross-section of the tunnel, through the at least one image sensor; d) calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel, through the at least one distance sensor; e) capturing at least one image, through the at least one image sensor;f) analyze at least one image obtained through at least one processing module, to obtain at least one parameter of the inspected distance; g) obtain at least one parametric three-dimensional model, from the at least one distance traveled and from at least one image, through at least one analysis module; and h) link the at least one parametric three-dimensional model with the at least one parameter, through the at least one analysis module, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

2. The method according to claim 1, CHARACTERIZED in that step a) further comprises: turning on the at least one processing module; locating and turning on the at least one distance sensor at the at least one reference point of the tunnel; and turning on the at least one image sensor.

3. The method according to any of claims 1 to 2, CHARACTERIZED in that it further comprises, before step d), starting the journey through the tunnel by means of at least one conveyor.

4. The method according to any of claims 1 to 3, CHARACTERIZED in that it further comprises, prior to step e), sending at least one piece of data related to the at least one inspected distance from the at least one distance sensor to the at least one image sensor, via at least one antenna.

5. The method according to any of claims 1 to 4, CHARACTERIZED in that step f) further comprises: activating at least one image cropping and segmentation algorithm; activating at least one fixing element counting algorithm; activating at least one quality recognition and classification algorithm.

6. The method according to claim 5, CHARACTERIZED in that it further comprises compiling the data obtained through at least one algorithm to obtain at least one parameter, which corresponds to one or more of: at least one image of the at least one distance traveled; at least one piece of data related to the distance inspected in the tunnel, delivered in a unit of measurement; at least one number for each at least one fastening element; and at least one quality classification for each at least one fastening element.

7. The method according to any of claims 1 to 6, CHARACTERIZED in that it further comprises storing at least one parameter in at least one quality database.

8. The method according to any of claims 1 to 7, CHARACTERIZED in that it further comprises deciding whether to inspect at least one additional distance, wherein, in the event of inspecting at least one additional distance, the method further comprises returning to step d).

9. The method according to claim 7, CHARACTERIZED in that it further comprises, prior to step f), importing information from at least one quality database to at least one analysis database of at least one analysis module.

10. The method according to claim 9, CHARACTERIZED in that it further comprises, prior to step g): extracting at least one distance traveled and at least one image, through at least one analysis module; and extracting at least one parameter related to at least one quality classification for each at least one fixing element, through at least one analysis module.

11. The method according to any of claims 1 to 10, CHARACTERIZED in that it further comprises visualizing at least one digital twin of the at least one inspected distance through at least one visualization means.

12. The method according to claim 10, CHARACTERIZED in that it further comprises: visualizing at least one parameter related to at least one quality classification for each at least one fastening element; and visualizing at least one image of the at least one image for each at least one inspected distance.

13. A tunnel inspection system according to the method of claims 1 to 12, CHARACTERIZED in that it comprises: - at least one processing module, transported by at least one conveyor, comprising: or at least one distance sensor, for: i. indicating at least one reference point in a tunnel where the inspection by means of the at least one conveyor begins; and ii. calculating at least one piece of data related to at least one distance inspected by the at least one conveyor in the tunnel; or at least one image sensor, for: i. indicating a theoretical quantity of at least one tunnel fixing element per unit of measurement; ii. indicating information regarding at least one theoretical cross-section of the tunnel; and iii. capturing at least one image; wherein the at least one processing module analyzes the at least one image to obtain at least one parameter of the inspected distance; and - at least one analysis module, to: i. obtain at least one parametric three-dimensional model, from at least one distance traveled and at least one image; and ii. link the at least one parametric three-dimensional model with at least one parameter, to obtain at least one digital twin of the at least one inspected distance of the tunnel.

14. The system according to claim 13, CHARACTERIZED in that the at least one conveyor corresponds to at least one user.

15. The system according to claim 13, CHARACTERIZED in that the at least one transporter corresponds to at least one motorized land vehicle.

16. The system according to claim 13, CHARACTERIZED in that the at least one transporter corresponds to at least one unmanned aerial vehicle.

17. The system according to any of claims 13 to 16, CHARACTERIZED in that the at least one processing module further comprises at least one antenna, for sending the at least one data relating to the at least one inspected distance from the at least one distance sensor to the at least one image sensor.

18. The system according to claim 17, CHARACTERIZED in that the at least one antenna corresponds to at least one module based on LoRA technology.

19. The system according to any of claims 13 to 18, CHARACTERIZED in that it further comprises at least one quality database for storing at least one parameter.

20. The system according to any of claims 13 to 19, CHARACTERIZED in that it further comprises at least one analysis database of at least one analysis module, into which information is imported from at least one quality database.

21. The system according to claim 20, CHARACTERIZED in that the import of information is done wirelessly.

22. The system according to claim 20, CHARACTERIZED in that the import of information is carried out through at least one cable.

23. The system according to any of claims 13 to 22, CHARACTERIZED in that the at least one analysis module is comprised of the at least one processing module.

24. The system according to any of claims 13 to 22, CHARACTERIZED in that at least one analysis module is hosted in a data cloud.

25. The system according to any of claims 13 to 24, CHARACTERIZED in that it further comprises at least one display means, for displaying one or more of: the at least one digital twin of the at least one inspected distance; the at least one parameter; and at least one image of the at least one image for each at least one inspected distance.

Images