Methods, systems, media and program products for engineering inspection based on drones

Abstract

The method, system, medium and program product for detecting engineering projects based on unmanned aerial vehicles (UAVs) are related to the fields of intelligent monitoring devices and intelligent data processing. The method includes: controlling the UAV to fly along the first detection route in the fast scanning mode to obtain the first detection data of all engineering areas; determining each suspected risk location where each suspected risk problem occurs and each suspected risk problem corresponding to each suspected risk problem according to the first detection data; determining the second flight speed set based on each suspected risk location and each suspected risk problem level; controlling the UAV to switch to the fine observation mode and fly along the second detection route to obtain the second detection data of each suspected risk location, and the fine observation mode corresponds to the second flight altitude and the second flight speed set; determining the risk location where the risk problem exists in each suspected risk location according to the second detection data. The implementation of this method not only ensures the accuracy of engineering detection, but also improves the detection efficiency.

Description

Technical Field

[0001] This application relates to the fields of intelligent monitoring devices and intelligent data processing, and particularly to a method, system, medium, and program product for engineering detection based on unmanned aerial vehicles (UAVs). Background Art

[0002] With the development of technology, the application of UAV technology has become increasingly widespread, especially in the fields of intelligent detection devices and intelligent data processing. For example, in tunnels, due to the high height or complex conditions of the tunnels, inspectors cannot enter the tunnels or lift inspection equipment to conduct a comprehensive structural inspection of the tunnel project; in tall concrete structures, it is also impossible to set up inspection platforms for crack detection or other related performance inspections of the outer facades. In these cases, the application of UAVs can greatly improve the safety and efficiency of inspection work, especially in areas that are not easily accessible or dangerous.

[0003] In related technologies, UAV detection technology usually involves remotely controlling the UAV to fly back and forth twice above the engineering area according to a preset flight route, continuously shooting or collecting data to obtain the engineering area for inspection.

[0004] Although the UAV detection technology in related technologies can achieve engineering area coverage and provide basic data collection, it has some limitations and defects, especially in terms of accuracy and efficiency. Since the UAV uses a preset flight route and fixed flight parameters throughout the detection process, when dealing with large-scale or complex engineering projects, it will consume a large amount of battery life and flight time, resulting in low efficiency. At the same time, due to the use of a preset flight route and fixed flight parameters, it may not be flexible or detailed enough when dealing with areas that require special attention, resulting in low detection accuracy of the engineering area. Summary of the Invention

[0005] This application provides a method, system, medium, and program product for engineering detection based on UAVs, which can improve the detection efficiency while ensuring the accuracy of engineering detection.

[0006] In a first aspect, the present application provides a method for detecting a project based on a drone, which is applied to a detection system. The method includes: controlling the drone to fly along a first detection route in a fast scanning mode to obtain first detection data of all project areas. The fast scanning mode corresponds to a first flight speed and a first flight height, and multiple shooting points in the first detection route are determined by the maximum scanning area corresponding to the first flight height and the area of all project areas; determining, according to the first detection data, each suspected risk location where each suspected risk problem occurs and each suspected risk problem level corresponding to each suspected risk problem; determining a second flight speed set based on each suspected risk location and each suspected risk problem level. The second flight speed set includes multiple second flight speeds respectively corresponding to each suspected risk location. The higher the suspected risk problem level corresponding to a suspected risk location, the slower the second flight speed corresponding to the suspected risk location; controlling the drone to switch to a fine observation mode and fly along a second detection route to obtain second detection data of each suspected risk location. The fine observation mode corresponds to a second flight height and the second flight speed set, and the second detection route is determined according to the distances between each suspected risk location and the second flight speeds corresponding to each suspected risk location; determining, according to the second detection data, the risk locations where risk problems exist among each suspected risk location.

[0007] In the above embodiments, the drone performs two rounds of inspections on the engineering area in two different flight modes to achieve efficient full coverage of the engineering area and targeted inspection of suspected risk locations. In the fast scanning mode, a relatively high first flight altitude and a relatively fast first flight speed are adopted. Since the higher the flight altitude, the larger the corresponding scanning area, the fewer the shooting points of the drone on the first inspection route. While ensuring full coverage of the engineering area, the flight path is also saved. Combined with the relatively fast first flight speed, the flight time is reduced and the inspection efficiency is improved. Through the first inspection data, the inspection system can determine suspected risk problems, and thus control the drone to perform a secondary inspection on the suspected risk locations corresponding to the suspected risk problems in the fine observation mode. In the fine observation mode, the second flight altitude of the drone is slightly lower than the first flight altitude, and the second flight speed is slower than the first flight speed to obtain more detailed and clear second inspection data. When the drone inspects the suspected risk locations, it will consider the suspected risk level corresponding to the suspected risk problems of the suspected risk locations, and adopt different second flight speeds to inspect the suspected risk locations according to the suspected risk level. A high suspected risk level requires a slower second flight speed, while a low suspected risk level can be inspected through a relatively faster second flight speed, so as to highlight the targeted inspection of the suspected risk locations and improve the accuracy of the inspection. The second inspection route is based on the distances between the suspected risk locations and the second flight speeds corresponding to the suspected risk locations, thereby optimizing the total flight duration and improving the efficiency of the flight inspection. Compared with the method in the related art of remotely controlling the drone to fly back and forth twice above the engineering area according to a preset flight route to continuously shoot or collect data, the related art adopts a fixed flight route and flight parameters, and uses the same inspection mode for all engineering areas, and cannot adjust the second inspection targeted according to the results of the first inspection, and it is difficult to ensure the inspection efficiency and accuracy. While this solution optimizes the first inspection route and the second inspection route, scans the key areas more finely, improves the accuracy of the inspection, and at the same time greatly reduces the flight mileage and flight time, and significantly improves the inspection efficiency.

[0008] In connection with some embodiments of the first aspect, in some embodiments, controlling the drone to switch to the fine observation mode and fly along the second detection route to obtain the second detection data of each suspected risk location specifically includes: determining the flight duration of the drone between each suspected risk location according to the distance between each suspected risk location and the second flight speed corresponding to each suspected risk location; taking any one of the suspected risk locations as the flight starting point of the drone, determining the next flight starting point with the shortest flight duration from the flight starting point until all suspected risk locations are traversed, obtaining an alternative second detection route and adding it to the set of alternative second detection routes, where the number of alternative second detection routes in the set of alternative second detection routes is the same as the number of suspected risk locations; determining the second detection route as the alternative second detection route with the shortest total flight duration in the set of alternative second detection routes; controlling the drone to switch to the fine observation mode and fly along the second detection route to obtain the second detection data of each suspected risk location.

[0009] In the above embodiments, when planning the second detection route, the detection system comprehensively considers the distance between each suspected risk location and the second flight speed corresponding to each suspected risk location. Since the distances between each pair of suspected risk locations are different, the energy consumption of the drone caused thereby is also different. At the same time, since the flight speeds at each suspected risk location are different, when the drone flies from one suspected risk location to another, it needs to perform uniformly accelerated linear motion, uniformly decelerated linear motion, or uniform linear motion, and the energy consumption of the drone caused thereby is also different. Considering the energy consumption caused by both of these factors, the method of determining the second detection route by the shortest flight duration is adopted. Taking any one of the suspected risk locations as the flight starting point, determining the next flight starting point with the shortest distance from the flight starting point until all suspected risk locations are traversed, obtaining multiple alternative second detection routes. Determining the alternative second detection route with the shortest total flight duration among the multiple alternative second detection routes as the second detection route, thereby further optimizing the flight efficiency. This way of dynamically planning based on specific detection requirements can significantly reduce the flight time, lower the energy consumption, and improve the detection efficiency compared with the single flight route preset manually in the related art.

[0010] In connection with some embodiments of the first aspect, in some embodiments, before the step of controlling the drone to fly along the first detection route in the fast scanning mode to obtain the first detection data of all engineering areas, the method further includes: determining each construction link and the corresponding construction end time point of each construction link according to the obtained engineering construction progress plan, and the construction link is carried out in the corresponding engineering area; determining multiple risk prevention detection time points according to each construction end time point, where the risk prevention detection time point is the construction end time point minus the preset duration, and is used to instruct the drone to perform risk prevention detection on all engineering areas.

[0011] In the above embodiments, the detection system scientifically and reasonably determines in advance the time points for the drones to perform risk preventive detections on all engineering areas according to the engineering construction progress plan, achieving active prevention and control closely integrated with the engineering construction progress. The intelligent planning of the risk prevention detection time points can, compared with the passive detection method in the related art, discover and eliminate risk hazards in advance to the greatest extent, reduce the risk of engineering accidents, and enhance the pertinence, initiative, and effectiveness of the detection.

[0012] Combined with some embodiments of the first aspect, in some embodiments, after the step of determining the risk positions with risk problems among the suspected risk positions according to the second detection data, the method further includes: controlling multiple drones to switch to the fixed-point observation mode and flying along different third detection routes to obtain the third detection data of the risk positions. The fixed-point observation mode corresponds to a third flight altitude set and a third flight speed set. The third flight altitude in the third flight altitude set is determined according to the rectification altitude of the risk position, the third flight speed in the third flight speed set is determined according to the risk level of the risk problem, and the third detection route is determined according to the edge curve of the rectification area of the risk position; determining the rectification situation of the risk problem before the construction end time point according to the third detection data.

[0013] In the above embodiments, after the detection system detects the risk positions with risk problems, it is necessary to rectify these risk positions. To understand the rectification situation, multiple drones are set to perform risk rectification supervision on these risk positions in the fixed-point observation mode. The fixed-point observation mode corresponds to a third flight altitude and a third flight speed. Each drone has a different third detection route, third flight altitude, and third flight speed, which are determined by the risk level of the risk problem corresponding to the risk position detected by the drone, the rectification altitude of the risk position, and the edge curve of the rectification area of the risk position, so as to achieve detailed, accurate, and targeted observation and improve the accuracy of the detection. Different from the passive monitoring with a single preset parameter in the related art, the present application can obtain the rectification progress of different risk positions specifically by actively monitoring the rectification situation of the risk positions, so as to ensure that the rectification is completed before the construction end time point, without delaying the project progress, ensuring that the rectification is in place, and improving the risk governance effect.

[0014] Combined with some embodiments of the first aspect, in some embodiments, before the step of controlling multiple drones to switch to the fixed-point observation mode and flying along different third detection routes to obtain the third detection data of the risk positions, the method further includes: determining the rectification measures corresponding to the risk problem according to the risk problem corresponding to the risk position; determining the rectification area of the risk position based on the rectification scope corresponding to the rectification measures. The higher the risk level of the risk problem corresponding to the risk position during construction, the smaller the area corresponding to the rectification area; determining the number of drones according to the area corresponding to the risk position and the area corresponding to the rectification area.

[0015] In the above embodiments, when rectifying the risk positions, the detection system determines the rectification measures for the risk problems, and then divides the rectification areas accordingly. The edge curves of the rectification areas determine the third detection routes of the drones, with one drone corresponding to one third detection route. The rectification areas are related to the risk levels of the risk problems. The higher the risk level of the risk problem, the smaller the corresponding area of the rectification area and the smaller the detection range of the third detection route, which helps to focus on the rectification areas with high risk problem levels to avoid missed and false detections, improving the accuracy of detection. The detection system determines the number of drones based on the area corresponding to the risk position and the area corresponding to the rectification area, achieving an optimized allocation of resources. This method of allocating detection resources as needed, compared with the manual experience preset in the related art, can achieve precise delivery, effectively improve the detection efficiency, and reduce the monitoring cost.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of determining the number of drones according to the area corresponding to the risk position and the area corresponding to the rectification area, the method further includes: respectively determining the risk types of the risk problems corresponding to the rectification areas according to a preset risk classification table, where the risk classification table includes the corresponding relationship between the risk problems and the risk types; determining the drone models corresponding to each rectification area in a preset drone selection table, where the drone selection table includes the corresponding relationship between the risk types and the drone models; if the matching fails in the drone selection table according to the risk type, determining the performance parameters of the drone according to the risk type and setting the performance parameters of the drone.

[0017] In the above embodiments, the detection system realizes customized detection for different risk problems by matching the risk types with the drone models. This monitoring scheme for determining the drone types according to the risk types can ensure that the selected drone models can meet the monitoring requirements of the corresponding risk positions, and the monitoring effect is better. Compared with the related art that uses a single type of drone for monitoring, this scheme can greatly improve the pertinence of monitoring and make the monitoring more accurate.

[0018] In combination with some embodiments of the first aspect, in some embodiments, if the matching fails in the drone selection table according to the risk type, determining the performance parameters of the drone according to the risk type and setting the performance parameters of the drone specifically includes: analyzing the detection requirements corresponding to the risk type to determine the performance parameters of the drone, where the performance parameters include image resolution, flight altitude, and flight speed; calling the setting function in the drone control interface to configure the performance parameters to the drone.

[0019] In the above embodiments, an alternative solution is designed for the case where the risk type fails to match the UAV model. That is, by intelligently analyzing the detection requirements of the risk type, the performance parameters of the UAV are determined, and the interface is called to configure the parameters of the UAV, so as to achieve personalized monitoring of special risk types. This way of flexibly adjusting the UAV model for the monitoring solution realizes customized personalized monitoring for unconventional risks. Compared with the fixed monitoring mode of the related technology, it can improve the adaptability and pertinence of the monitoring, and also improve the accuracy of the monitoring.

[0020] In a second aspect, an embodiment of the present application provides a detection system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to cause the detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product containing instructions. When the computer program product runs on the detection system, it causes the detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions. When the instructions run on the detection system, they cause the detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the detection system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0025] 1. Since the present invention uses two different flight modes to perform two rounds of detection on the engineering area, the fast scanning mode adopts a higher flight altitude and a faster flight speed to achieve efficient full coverage, and the fine observation mode adopts a lower flight altitude and a slower flight speed to detect suspected risk positions specifically, and dynamically adjusts the second flight speed according to the level of suspected risk problems, so as to achieve targeted detection of suspected risk positions. Therefore, the present invention can greatly improve the detection efficiency while ensuring the detection accuracy, effectively solving the problem that the detection accuracy and efficiency are not high due to the use of fixed flight routes and fixed flight parameters in the related art, and thus realizing efficient and accurate engineering risk detection. In addition, the present invention also optimizes the first detection route and the second detection route, scans the key areas more finely, greatly reduces the flight mileage and flight time, and significantly improves the detection efficiency.

[0026] 2. Since the present invention determines the risk prevention detection time point in advance according to the engineering construction progress plan and realizes active prevention and control closely combined with the engineering construction progress, the present invention can discover and eliminate risk hazards in advance to the greatest extent, reduce the risk of engineering accidents, effectively solve the problem that passive detection in the related art often lags behind the occurrence of risks, and thus realize the whole-process dynamic risk early warning. Compared with passive detection, the intelligent planning of the present invention makes the detection more targeted, proactive and effective.

[0027] 3. Since the present invention, after detecting risk problems, adopts a fixed-point observation mode to control multiple drones to supervise the rectification of the risk positions, and dynamically sets different third detection routes, third flight altitudes and third flight speeds for each risk position, the present invention can achieve detailed, accurate and targeted rectification observation, effectively solve the limitation of passive monitoring with a single preset parameter in the related art, and thus realize the goal of ensuring that the rectification is in place and the project progress is not delayed. The present invention can not only obtain the detailed rectification progress of different risk positions, but also ensure that the rectification is completed before the construction end time point, greatly improving the risk management effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flowchart of a method for detecting an engineering project based on an unmanned aerial vehicle in an embodiment of the present application;

[0029] Figure 2 is another flowchart of a method for detecting an engineering project based on an unmanned aerial vehicle in an embodiment of the present application;

[0030] Figure 3 is a schematic structural diagram of a physical device of a detection system in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to be limiting of the present application. As used in the specification of the present application, the singular expressions "a", "an", "above", "the" and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations of one or more of the listed items.

[0032] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as suggesting or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more.

[0033] As the scale of infrastructure and housing construction projects continues to expand, the demand for monitoring project quality and safety is growing. However, traditional manual inspection methods are inefficient and cannot meet the needs of all-weather inspection of large-scale projects. In recent years, the application of drone technology in the field of intelligent monitoring devices and intelligent data processing has brought new opportunities for this. As a mobile inspection platform, drones can be equipped with sensors such as cameras and thermal imagers to conduct fast and efficient risk detection of construction projects. However, as the scale of inspection increases, there are still challenges in how to intelligently plan the flight routes and inspection strategies of drones to improve the accuracy and efficiency of inspections while ensuring full coverage. If this problem can be solved, the value of drones in the field of intelligent monitoring devices and intelligent data processing will be greatly enhanced.

[0034] In response to the above-mentioned detection needs, the relevant technology adopts the method of preset flight routes and fixed flight parameters when inspecting the facades of buildings. Specifically, the technicians will manually plan the altitude, speed and route of the drone's flight in advance, and pre-program it into the drone for execution. For example, determine that the drone will fly horizontally at a speed of 10 meters per second and 10 meters away from the building, and scan the facade of the entire building twice according to the pre-set S-shaped flight route. This method can basically achieve full coverage detection of the building's facade, but due to the use of fixed routes and parameters, the second detection cannot be adjusted in a targeted manner according to the results of the first detection, and areas with suspected problems cannot be inspected in a focused manner, resulting in poor detection results. At the same time, the preset flight route cannot be dynamically optimized according to the actual situation, resulting in the drone's flight distance and time being too long, and the detection efficiency is not high.

[0035] To address the deficiencies in related technologies, this solution realizes the intelligent autonomous planning of drones, improving the detection accuracy and efficiency. For example, when this solution is applied to the facade inspection of a 30-story building with an exterior facade area of approximately 5,000 square meters. To achieve full coverage, this solution first determines the first flight altitude of the drone to be half of the building's floors, i.e., 15 floors, and the first flight speed to be 15 meters per second, and intelligently plans the first inspection route of the drone. After a quick scan, the detection system discovers that the surface temperature of a wall area at the top of the building is abnormally low based on the first inspection data collected. Through the color change of the thermal image, it can be judged that there is likely a problem of wall leakage here. Therefore, this solution determines that the drone will lower its flight altitude to the 5th floor near this area during the second inspection and assigns a lower and slower flight speed for this suspected problem area for focused observation. Finally, the detection system automatically optimizes and generates the second inspection route, enabling the drone to only conduct a secondary inspection on key areas, significantly shortening the flight mileage. Through two rounds of inspections, this solution achieves a comprehensive and efficient scan of the entire building's exterior facade, while specifically examining the suspected problem areas to ensure the detection quality. The entire detection time is reduced by 30% compared to related technologies. The intelligent planning of flight parameters and flight routes in this solution improves the efficiency and effectiveness of building exterior facade inspections.

[0036] Compared with the method in related technologies of remotely controlling a drone to make two round trips over the building facade according to a preset flight route to continuously capture or collect data, related technologies adopt a fixed flight route and flight parameters, using the same detection mode for all areas and being unable to adjust the second inspection targeted at the results of the first inspection. It is difficult to guarantee the detection efficiency and accuracy. However, this solution optimizes the first inspection route and the second inspection route, scans the key areas more precisely, improves the detection accuracy, and at the same time greatly reduces the flight mileage and flight time, significantly enhancing the detection efficiency.

[0037] The following introduces the relevant information about the unmanned aerial vehicle and the detection system in this solution: An unmanned aerial vehicle (UAV) is an aircraft that can fly without an on-board crew. The UAV provides a mobile platform that can carry various sensors and detection devices, such as cameras, infrared cameras, lidar, etc., enabling the UAV to reach areas that are difficult for humans to access or perform dangerous tasks, such as detecting the steel bar cover thickness at high altitudes and tunnel engineering inspections. The UAV can be manually operated by a remote control or automatically navigated through a preset program. In the embodiments of this application, the UAV establishes a communication connection with the detection system, and the connection method can be through a wireless wide area network, satellite communication, or a dedicated radio frequency, which is not limited herein. It receives the control instructions sent by the detection system to execute tasks and obtains real-time data to feedback to the detection system. The detection system can receive the data transmitted by the UAV and process it, or remotely control the UAV to control its flight. The detection system can be a computer device, such as a mobile phone or a smart tablet, which is not limited herein.

[0038] For ease of understanding, the method provided in this embodiment will be described in terms of its process in combination with the above scenario. Please refer to Figure 1 , which is a schematic flowchart of a method for detecting an engineering project based on an unmanned aerial vehicle in the embodiments of this application.

[0039] S101. Control the UAV to fly along the first detection route in the fast scanning mode to obtain the first detection data of all engineering areas. The fast scanning mode corresponds to the first flight speed and the first flight height. The multiple shooting points in the first detection route are determined by the maximum scanning area corresponding to the first flight height and the area of all engineering areas;

[0040] Based on the required image resolution and the scope of all engineering areas, the detection system needs to preset the first flight height and the first flight speed of the UAV in the fast scanning mode. A suitable first flight height can maximize the scanning area of each flight while ensuring image quality. The determination of the first flight height is usually related to the resolution of the sensor carried by the UAV. Resolution refers to the size of the smallest object that the sensor can distinguish and is usually inversely proportional to the first flight height. The detection system calculates the corresponding maximum scanning area according to the first flight height. The first flight speed should balance image quality and flight efficiency to ensure high-quality image data can still be obtained at high speed. It should be noted that due to the use of a relatively high flight height and a relatively fast flight speed, a high-resolution camera needs to be used for shooting and set to the automatic exposure mode to ensure clear images can be obtained under different lighting conditions.

[0041] Based on the focal length of the camera lens of the UAV and the size of the sensor, calculate the field of view (FOV) of the camera. The formula is as follows: d is the diagonal size of the camera sensor (or available width / height, depending on the requirements), and f is the focal length of the lens. The field of view of a camera is usually expressed as an angle, which determines the width and height of the scene that the camera can capture at a specific distance, and describes the visual range that the camera lens can cover in a single shot, that is, the maximum scan area corresponding to the first flight altitude. The detection system calculates the maximum scan area by using a simple geometric relationship between the first flight altitude and the field of view angle of the camera. The formula is as follows: (Assuming that the visual range that the camera lens can cover in a single shot is a square).

[0042] The detection system calculates the area of ​​all engineering areas and compares it with the visual range that can be covered by the drone camera lens in a single shot (i.e., the maximum scanning area corresponding to the first flight altitude) to determine the total number of shooting points of the drone. To ensure the continuity and overlap of the shooting coverage, the detection system can set the overlap rate between shooting points by adjusting the shooting point spacing. The shooting point spacing can be calculated by subtracting the corresponding overlapping part from the width and height of the maximum scanning area. For example, if a 30% overlap rate is selected, each shooting point should cover 30% of the area of ​​the adjacent points. The detection system enters the boundaries of the engineering area in the GIS software or flight planning tool, sets the shooting point spacing, and automatically generates a shooting point grid covering the entire area. The detection system converts these shooting points into the first detection route of the drone, ensuring that the drone flies in the order of these shooting points, thereby automatically optimizing the flight path to reduce flight time and power consumption.

[0043] For example, the engineering area monitored by the detection system is a single building in a construction site, which covers an area of ​​1,000 square meters. The detection system first sets a higher flight altitude, such as 10 meters, based on the area of ​​the engineering area. The higher the flight altitude, the larger the area that can be scanned and covered per unit time. At the same time, with a higher flight altitude, a faster flight speed is set, such as 10 meters per second. The faster the flight speed, the shorter the time it takes to scan the entire area. Assuming that the horizontal field of view angle is 60° and the vertical field of view angle is 45°, the ground length covered by the camera in the horizontal direction obtained by substituting into the above formula is 11.54 meters, and the ground length covered in the vertical direction is 8.28 meters, and the maximum scanning area is 95.57 square meters. Therefore, under the conditions of the first flight altitude of 10 meters, the horizontal field of view of 60°, and the vertical field of view of 45°, the drone's camera can cover an area of ​​approximately 95.57 square meters. Therefore, by dividing the area of ​​all engineering areas by the maximum scanning area corresponding to the first flight altitude, the integer part is obtained to obtain 11 shooting points. The first detection route passes through these 11 shooting points at 10 meters per second to obtain the first detection data of all engineering areas.

[0044] The first detection data mainly consists of various forms of raw data such as pictures, videos, and thermal imaging data collected by the on-board sensing devices of the unmanned aerial vehicle, providing a basis for subsequent detection of suspected risk locations by the unmanned aerial vehicle.

[0045] S102. According to the first detection data, determine each suspected risk location where each suspected risk problem occurs and each suspected risk problem level corresponding to each suspected risk problem;

[0046] The detection system has pre-constructed a risk problem feature library, which contains visual feature information corresponding to various risk problems. As shown in Table 1 below, it is an example of a risk problem feature library:

[0047]

[0048]

[0049] Table 1

[0050] After obtaining the first detection data, the detection system can use image processing algorithms to first detect and extract the visual features of all engineering areas, such as edge features, texture features, shape features, and pattern recognition features.

[0051] Edge features describe areas in an image where the brightness changes significantly, usually indicating the outline of an object or the boundary of a shape. Edge detection is usually achieved by finding the places where the gradient of the image brightness is the largest. Common edge detection algorithms include: Sobel operator: used to detect horizontal and vertical edges in an image; Canny edge detector: through a more complex algorithm, it can provide clearer and continuous edges.

[0052] Texture features describe the recurring patterns or local structures in an image, reflecting the properties of the object surface or the distribution of objects in the scene. Texture analysis can help distinguish different materials or states, such as unpaved and paved walls. Methods for extracting texture features include: Gray Level Co-occurrence Matrix (GLCM): analyzing the statistical distribution of pixel pairs to reflect the roughness, contrast, etc. of the texture; Local Binary Pattern (LBP): a simple and efficient texture feature that compares the size of each pixel with its surrounding pixels.

[0053] Shape features are used to describe and identify geometric shapes in an image, such as circles, squares, polygons, etc. For example, identifying mechanical equipment, safety signs, etc. Shape feature extraction methods include: Contour detection: extracting closed contours after edge detection; Hough transform: used to detect parametric shapes such as straight lines and circles in an image.

[0054] Pattern recognition features involve matching objects in an image with known patterns or templates to identify specific types of mechanical equipment, safety signs, etc. The extraction methods of pattern recognition features include: Template matching: comparing an image region with a preset template; Machine learning classifiers: such as support vector machines (SVMs) or deep learning models, which can identify complex patterns and objects through training.

[0055] Then, the detection system will match and compare the visual features of the extracted engineering areas with the visual feature information corresponding to each risk problem in the risk problem feature library one by one. If the visual features of a certain engineering area and the visual feature information corresponding to a certain risk problem exceed the preset similarity threshold in the similarity algorithm comparison, then the detection system can determine that there is a suspected risk problem in this engineering area. When the drone collects the first detection data, including images and the corresponding geographical marks, when the detection system determines that there is a suspected risk problem in this engineering area through the image, it will also determine the suspected risk location where the suspected risk problem appears, and can determine the suspected risk problem level corresponding to the suspected risk problem according to the risk problem feature library.

[0056] Suppose it is determined through the first detection data obtained by the drone that: the visual feature is a long straight line or an irregular line detected by the Canny algorithm. According to the above method, the suspected risk problem is determined to be "crack", the suspected risk location is the second building on the north side of the construction site, and the suspected risk problem level is medium; the visual feature is an irregular dark area shown by GLCM analysis with abnormal reflection characteristics. According to the above method, the suspected risk problem is determined to be "ponding", the suspected risk location is near the construction site entrance, and the suspected risk problem level is medium.

[0057] S103. Based on each suspected risk location and each suspected risk problem level, determine a second flight speed set. The second flight speed set includes multiple second flight speeds corresponding to each suspected risk location respectively. The higher the suspected risk problem level corresponding to a suspected risk location, the slower the second flight speed corresponding to this suspected risk location;

[0058] After determining the suspected risk location, the detection system will determine the second flight speed when the drone reaches this suspected risk location according to the suspected risk problem level of the suspected risk problem at the suspected risk location. The flight speed is related to the suspected risk problem level. The detection system pre-defines the mapping relationship between the suspected risk problem level and the second flight speed. The following gives an example: High risk level: slow flight speed (e.g., 2 m / s)

[0059] Medium risk level: medium flight speed (e.g., 4 m / s)

[0060] Low risk level: fast flight speed (e.g., 6 m / s)

[0061] Continuing with the example in step S102, the suspected risk location: the second building on the north side of the construction site, the suspected risk problem level: medium, the second flight speed when the drone flies to the second building on the north side of the construction site is 4 m / s; the suspected risk location: near the construction site entrance, the suspected risk problem level: medium, the second flight speed when the drone flies near the construction site entrance is 4 m / s.

[0062] S104. Control the drone to switch to the fine observation mode and fly according to the second detection route to obtain the second detection data of each suspected risk location. The fine observation mode corresponds to a set of second flight heights and second flight speeds. The second detection route is determined according to the distances between the suspected risk locations and the second flight speeds corresponding to the suspected risk locations.

[0063] To enable the drone to fly according to the second detection route and collect the second detection data in the fine observation mode, the detection system needs to determine the second flight height and second flight speed of the drone in the fine observation mode. In step S103, the set of second flight speeds has been determined and will not be elaborated here. The second flight height is lower than the first flight height, requiring the drone to fly lower to obtain clearer images or data.

[0064] There are also many ways to determine the second detection route. It can be determined by the shortest path method or other methods, which are not limited here.

[0065] Optionally, generally, controlling the drone to switch to the fine observation mode and fly according to the second detection route to obtain the second detection data of each suspected risk location can be achieved in the following way: Determine the flight duration of the drone between each suspected risk location according to the distances between the suspected risk locations and the second flight speeds corresponding to the suspected risk locations; Take any one of the suspected risk locations as the flight starting point of the drone, and determine the next flight starting point with the shortest flight duration from the flight starting point until all suspected risk locations are traversed, obtaining an alternative second detection route and adding it to the set of alternative second detection routes. The number of alternative second detection routes in the set of alternative second detection routes is the same as the number of suspected risk locations; Determine the second detection route as the alternative second detection route with the shortest total flight duration in the set of alternative second detection routes; Control the drone to switch to the fine observation mode and fly according to the second detection route to obtain the second detection data of each suspected risk location.

[0066] In the embodiment of the present application, the determination of the second detection route follows the principle of the shortest flight duration. Since the levels of the suspected risk problems at each suspected risk position are different, the second flight speeds of the UAV flying to each suspected risk position are different. When the UAV flies from one suspected risk position to another suspected risk position, it needs to perform uniformly accelerated linear motion, uniformly decelerated linear motion, or uniform linear motion, and the resulting energy loss of the UAV is also different. The distances between each pair of suspected risk positions are different, and the resulting energy loss of the UAV is also different. Considering the energy losses caused by both of these factors, the second detection route is determined in the way of the shortest flight duration. Taking any suspected risk position as the flight starting point, the next flight starting point with the shortest distance from the flight starting point is determined until all suspected risk positions are traversed, and multiple alternative second detection routes are obtained. The alternative second detection route with the shortest total flight duration is determined from the multiple alternative second detection routes as the second detection route.

[0067] Suppose there are now five suspected risk positions A, B, C, D, and E, and the corresponding second flight speeds are Va = 5m / s, Vb = 3m / s, Vc = 1m / s, Vd = 5m / s, and Ve = 2m / s respectively. The distance between A and B is 5m, the distance between A and C is 10m, the distance between A and D is 20m, the distance between A and E is 50m, the distance between B and C is 10m, the distance between B and D is 12m, the distance between B and E is 3m, the distance between C and D is 7m, the distance between C and E is 50m, and the distance between D and E is 30m. When the UAV moves from the suspected risk position A to the suspected risk position B, the second flight speed changes from 5m / s to 3m / s, and it performs uniformly decelerated linear motion; when the UAV moves from the suspected risk position B to the suspected risk position A, the second flight speed changes from 3m / s to 5m / s, and it performs uniformly accelerated linear motion; when the UAV moves from the suspected risk position A to the suspected risk position D, the second flight speed remains unchanged, and it performs uniform linear motion.

[0068] According to the formula a represents the acceleration of the uniformly variable linear motion from the starting suspected risk position to the ending suspected risk position, V 初 represents the second flight speed corresponding to the starting suspected risk position, V 末 represents the second flight speed corresponding to the ending suspected risk position, s represents the distance between the starting suspected risk position and the ending suspected risk position, and t represents the time required from the starting suspected risk position to the ending suspected risk position.

[0069]

[0070] When the starting suspected risk position and the ending suspected risk position are swapped, the time used is the same, which will not be listed here. Next, taking any one of the suspected risk positions as the flight starting point, determine the next flight starting point that is the shortest distance from the flight starting point. After traversing all the suspected risk positions, multiple alternative second detection routes are obtained:

[0071] The first one: When taking the suspected risk position A as the flight starting point, determine the next flight starting point with the shortest flight duration from the flight starting point. Successively determine the next flight starting points as B, E, D, and C, and the total flight duration is 13.36 s; The second one: When taking the suspected risk position B as the flight starting point, determine the next flight starting point with the shortest flight duration from the flight starting point. Successively determine the next flight starting points as E, D, C, and A, and the total flight duration is 15.41 s; The third one: When taking the suspected risk position C as the flight starting point, determine the next flight starting point with the shortest flight duration from the flight starting point. Successively determine the next flight starting points as D, B, E, and A, and the total flight duration is 20.82 s; The fourth one: When taking the suspected risk position D as the flight starting point, determine the next flight starting point with the shortest flight duration from the flight starting point. Successively determine the next flight starting points as C, A, B, and E, and the total flight duration is 8.09 s; The fifth one: When taking the suspected risk position E as the flight starting point, determine the next flight starting point with the shortest flight duration from the flight starting point. Successively determine the next flight starting points as B, A, C, and D, and the total flight duration is 8.09 s;

[0072] Thus, five alternative second detection routes are determined. Among them, the fourth alternative second detection route and the fifth alternative second detection route have the shortest total flight time. Any one of these two alternative second detection routes is used as the second detection route.

[0073] S105. Determine the risk positions with risk problems among the suspected risk positions according to the second detection data.

[0074] Referring to step S102, it is possible to detect that there are indeed risk problems at the suspected risk positions according to the second detection data, and determine the risk positions with risk problems.

[0075] Since the present invention adopts two different flight modes to conduct two rounds of detection on the engineering area, the fast scanning mode uses a higher flight altitude and a faster flight speed to achieve efficient full coverage, and the fine observation mode uses a lower flight altitude and a slower flight speed to detect suspected risk positions specifically, and dynamically adjusts the second flight speed according to the level of suspected risk problems, so as to achieve targeted detection of suspected risk positions. Therefore, the present invention can greatly improve the detection efficiency while ensuring the detection accuracy, effectively solving the problem that the detection accuracy and efficiency are not high due to the use of fixed flight routes and fixed flight parameters in the related art, and thus realizing efficient and accurate engineering risk detection. In addition, the present invention also optimizes the first detection route and the second detection route, scans the key areas more finely, and greatly reduces the flight mileage and flight time, and significantly improves the detection efficiency.

[0076] Based on the above intelligent planned flight scheme, this scheme can be further optimized, including the dynamic planning of detection strategies, the refined monitoring of risk management, etc. For example, in a subway tunnel engineering project, this scheme intelligently determines the time points for the unmanned aerial vehicle to conduct risk prevention detection according to the engineering construction progress plan, closely combines the detection with the progress of the construction link, realizes proactive prevention before the event, and eliminates potential hazards to the greatest extent. After two rounds of detection and discovery of risk problems, this scheme can quickly determine the risk positions and risk types. For example, two crack risks and one water accumulation risk are found. In this regard, this scheme allocates three unmanned aerial vehicles for monitoring according to the area of the risk position, and considering the structural safety nature of the crack risk, selects the unmanned aerial vehicle carrying lidar to conduct customized monitoring on it. The three unmanned aerial vehicles can simultaneously conduct 24-hour uninterrupted fixed-point observation on their respective risk positions, monitor the rectification effect, so as to ensure that all problems are rectified before the final acceptance and avoid delaying the opening to traffic.

[0077] After combining the above scenarios, the following is a further and more specific process description of the method provided in this embodiment. Please refer to Figure 2 , which is another process schematic diagram of the method for detecting an engineering project based on an unmanned aerial vehicle in an embodiment of the present application.

[0078] Before step S101, the following steps may or may not be executed, and this is not limited herein.

[0079] S201. Determine each construction link and the corresponding construction end time point of each construction link according to the obtained engineering construction progress plan, and the construction link is carried out in the corresponding engineering area;

[0080] The detection system obtains the project construction progress plan by accessing the database or through file import. Other acquisition methods are not listed one by one here. The project construction progress plan details different construction links of the project (such as foundation works, main structure construction, decoration works, etc.) and the expected start time and construction end time point of each construction link. The detection system analyzes the project construction progress plan and extracts the names of all construction links and the corresponding construction end time points.

[0081] Example of project construction progress plan: Construction link: Foundation and basic construction; Construction end time point: 2024-03-20; Project area: Main building area; Construction link: Main structure construction; Construction end time point: 2024-06-15; Project area: Main building area; Construction link: Roof installation; Construction end time point: 2024-07-10; Project area: Top of main building; In other embodiments, there are other construction links and corresponding construction end time points and project areas, which are not limited here.

[0082] It should be noted that different construction links correspond to different project areas. For example, the main structure construction corresponds to the main building area. Therefore, the detection system determines the specific project area corresponding to each construction link while extracting the construction link and the construction end time point.

[0083] S202. Determine multiple risk prevention detection time points according to each construction end time point. The risk prevention detection time point is the construction end time point minus the preset duration, and is used to indicate the drone to perform risk prevention detection on all project areas;

[0084] After obtaining the construction end time points of each construction link, the detection system will determine the risk prevention detection time points for the drone to perform risk prevention detection on all project areas before the construction end time points. The risk prevention detection time point is the construction end time point minus a preset duration, which can be set according to the project scale and complexity. Generally, it can be set to 1-2 weeks before the end time point. Taking the above example, if the preset duration is set to 10 days, for the foundation and basement construction that ends on March 20, 2024, the detection system can determine the risk prevention detection time point as March 20, 2024; for the decoration project that ends on June 15, 2024, its risk prevention detection time point can be determined as June 15, 2024. The detection system will determine multiple such risk prevention detection time points to instruct the drone to conduct a comprehensive scan of all project areas according to these time points to discover potential risk hazards. This preventive detection can avoid the situation of rework caused by detecting problems only after the project is completed, and better control the project quality and progress. It should be noted that the specific operation of the drone to perform risk prevention detection on all project areas is completed by step S101 and will not be elaborated here.

[0085] After step S105, the following steps can be executed or not, which is not limited here.

[0086] S203. Determine the rectification measures corresponding to the risk problem according to the risk problem corresponding to the risk location;

[0087] The detection system needs to give rectification measures for the risk problems at each risk location. The corresponding relationship between risk problems and rectification measures will be preset in the detection system, and systematic rectification suggestions will be given for each specific risk problem. The formulation of rectification measures will also consider the specific location and severity of the risk problem to guide the subsequent rectification work to be more targeted and effective. The rectification measures for each risk problem can be determined by referring to the example of the risk problem feature library in step S102. For example, when detecting the risk problem of "crack", the detection system will determine the rectification measures for this risk problem according to specification requirements, expert suggestions, etc. For example, structural assessment is required, and wall reinforcement and repair are necessary when necessary. When detecting the risk problem of "water accumulation", the rectification measure is to clean the drainage system and check the waterproof layer. It should be noted that the rectification measure is only the first step of the rectification work, and the specific rectification monitoring and feedback are completed in the subsequent steps.

[0088] S204. Determine the rectification area of the risk location based on the rectification scope corresponding to the rectification measure. The higher the risk problem level corresponding to the risk problem occurring at the risk location during construction, the smaller the area corresponding to the rectification area;

[0089] After formulating the rectification measures for the risk issues, the detection system needs to determine the specific rectification areas. The rectification area is the specific scope for implementing the rectification work, and the operating space required by the rectification measures needs to be considered. The boundary of the rectification area will determine the third detection route of the UAV. For example, for the "crack" risk issue that requires partial wall reinforcement, the rectification area is the wall area within a 10-meter radius around the crack; for the "ponding" risk issue that requires cleaning the blockage, the rectification area is the complete drainage system area around the ponding location. At the same time, the rectification area will also be adjusted according to the severity of the risk issue, that is, the risk issue level. For risk issues with a higher risk issue level, in order to monitor more carefully, the rectification area will be correspondingly reduced. For example, one UAV is only responsible for monitoring within a 5-meter radius around the crack. For issues with a lower severity, the scope of the rectification area can be appropriately expanded to reasonably plan the density of the rectification work and the meticulousness of the monitoring according to the risk issue level.

[0090] S205. Determine the number of UAVs according to the area corresponding to the risk location and the area corresponding to the rectification area;

[0091] After obtaining the rectification area, the detection system calculates the area of the rectification area and compares it with the area of the entire risk location. Generally speaking, the number of UAVs is related to the number of rectification areas. One UAV is responsible for monitoring one rectification area, and the edge curve of one rectification area corresponds to the third detection route of one UAV. Multiple UAVs monitor all rectification areas to cover the area corresponding to the risk location.

[0092] S206. Determine the risk type of the risk issue corresponding to the rectification area according to the preset risk classification table. The risk classification table includes the corresponding relationship between the risk issue and the risk type;

[0093] A risk classification table will be preset in the detection system. The risk classification table clarifies the risk types corresponding to different risk issues. For example, the "crack" issue belongs to the structural safety risk type, and the "ponding" issue belongs to the drainage facility risk type, etc. The detection system can directly find out which risk type the risk issue corresponding to each rectification area belongs to according to the risk classification table. The risk types can include structural safety, mechanical and electrical equipment, drainage facilities, safety signs, etc. Clarifying the risk type helps to select the appropriate UAV for monitoring according to the risk type later and lay a foundation for subsequent customized monitoring.

[0094] S207. Determine the UAV model corresponding to each rectification area according to the risk type in the preset UAV selection table. The UAV selection table includes the corresponding relationship between the risk type and the UAV model;

[0095] After identifying the risk types corresponding to the rectification areas, the detection system will search for matching UAV models in the preset UAV selection table. The UAV selection table gives recommended UAV models according to different risk types. For example, for structural safety risks, the UAV selection table may recommend UAV models equipped with lidar, which have the ability to finely scan the surface defects of building structures. For mechanical and electrical equipment risks, the UAV selection table recommends UAVs with thermal imaging functions to detect circuit fault points. The detection system determines the UAV model suitable for monitoring each rectification area by checking the UAV selection table.

[0096] Applying the UAV selection table to the scenario of tunnel vault detection, the detection system determines the UAV model suitable for monitoring each tunnel section by querying the UAV selection table. UAV models equipped with lidar can finely scan the surface defects of the tunnel vault structure; UAVs with thermal imaging functions can detect local water seepage points on the vault; UAVs with strong light illumination can provide additional lighting inside the tunnel, which helps to inspect the dark areas of the tunnel; small and highly maneuverable UAVs are small in size and can flexibly avoid obstacles such as pipelines in the tunnel. Due to the limited space in the tunnel, the UAV needs to maintain a higher flight altitude as much as possible without getting too close to the vault, so as to ensure a wider camera field of view and fully cover the vault surface. Therefore, the detection system will give priority to selecting UAV models with a higher flight altitude during selection. For example, if the tunnel height is 5 meters, a UAV with a maximum flight altitude of 2 meters cannot meet the detection requirements, while another UAV with a maximum flight altitude of 4 meters can maintain a sufficient flight altitude in the tunnel for vault scanning. In this way, the present solution can select the most suitable UAV model for vault surface detection in the tunnel.

[0097] This customized way of selecting UAVs can better meet the monitoring requirements of different risk types than a single model of UAV, and can improve the monitoring quality. When no matching item can be found in the UAV selection table, the UAV parameters can also be customized through step S208.

[0098] S208: If the matching fails in the UAV selection table according to the risk type, determine the performance parameters of the UAV according to the risk type and set the performance parameters of the UAV.

[0099] Optionally, generally, if the matching fails in the UAV selection table according to the risk type, determining the performance parameters of the UAV according to the risk type and setting the performance parameters of the UAV can be achieved through the following method: analyze the detection requirements corresponding to the risk type, determine the performance parameters of the UAV, and the performance parameters include image resolution, flight altitude and flight speed; call the setting function in the UAV control interface and configure the performance parameters to the UAV.

[0100] If a matching UAV model cannot be found in the UAV selection table according to the risk type, the detection system will analyze the monitoring requirements corresponding to the risk type and independently determine the performance parameters of the UAV. For example, for special structural cracks, a higher-resolution camera is required for image acquisition, and the detection system can set the parameters of a camera with a higher pixel count. For special drainage pipes, a smaller UAV is needed to enter the pipe interior, and the detection system can set relatively small body size parameters. The detection system can comprehensively consider the monitoring requirements of the risk type, determine performance parameters such as resolution, flight altitude, and body size, and configure these performance parameters to the actual UAV execution device through the UAV control interface. In this way, even when facing new risk types, customization can be achieved through intelligent analysis of their monitoring requirements to ensure the monitoring effect.

[0101] When detecting the tunnel vault, if the risk of water accumulation in the tunnel is detected, the UAV needs to have an underwater mode and use an aerial photography body and a waterproof camera to view the water accumulation below; if the risk of air quality pollution in the tunnel interior is detected, the UAV needs to be equipped with a gas sensor and set corresponding gas detection parameters to enable the UAV to monitor the concentrations of gases such as methane and carbon monoxide; if the risk of tunnel lighting facility failure is detected, the UAV needs to be equipped with a strong light illumination device and set a mode in which the head of the UAV can maintain stable strong light irradiation to view the dark areas of the tunnel; if the risk of broken and fallen pipelines in the tunnel interior is detected, the UAV needs to have an obstacle avoidance mode so that the UAV can autonomously avoid obstacles such as pipelines according to the environment. For other risk types in the tunnel, the detection system can determine the performance parameters of the UAV, set the corresponding function modes, sensor parameters, obstacle avoidance parameters, etc. of the UAV, and realize the function customization of the UAV so that it can meet the requirements of various detection tasks in the complex tunnel environment.

[0102] S209. Control multiple UAVs to switch to the fixed-point observation mode and fly according to different third detection routes to obtain the third detection data of the risk location. The fixed-point observation mode corresponds to a third flight altitude set and a third flight speed set. The third flight altitude in the third flight altitude set is determined according to the rectification altitude of the risk location, the third flight speed in the third flight speed set is determined according to the risk level of the risk problem, and the third detection route is determined according to the edge curve of the rectification area of the risk location;

[0103] After the preparation of the UAV model and the number of UAVs is completed, the detection system will control these UAVs to switch to the fixed-point observation mode and implement rectification monitoring on the risk locations. Each UAV will be configured with a unique third detection route, a third flight altitude, and a third flight speed. The third flight altitude will consider the rectification altitude. For example, a lower third flight altitude is required to monitor the basement. The third flight speed will be set according to the risk problem level. The higher the risk level, the slower the third flight speed, so as to monitor more carefully. The third detection route will bypass the edge curve of the rectification area, so that the implementation process of the rectification measures can be observed comprehensively. The detection system schedules multiple UAVs for on-demand monitoring in real time, which can flexibly and efficiently conduct fixed-point observation on the rectification process and is crucial for timely feedback on the rectification effect.

[0104] S210. Determine the rectification situation of the risk problem before the construction end time point according to the third detection data.

[0105] Finally, the detection system will collect the third detection data fed back by the UAVs, such as video images, thermal imaging data, etc. By analyzing the third detection data, the execution degree and effect of the rectification measures can be judged. For example, it can be analyzed whether the cracks in the image have been repaired and whether the drainage system is unobstructed, etc., to determine the rectification progress and quality of the risk problems before the construction end time point of the project. If it is found that the rectification effect is not good, supplementary measures can be taken in time to finally ensure that all risk hazards can be completely rectified before the construction end time, avoiding engineering quality problems, as well as additional rework costs and project delays.

[0106] In the embodiments of the present application, since the risk prevention detection time point is determined in advance according to the project construction progress plan, realizing active prevention and control closely combined with the project construction progress, the present invention can discover and eliminate risk hazards in advance to the greatest extent, reduce the risk of engineering accidents, effectively solve the problem that passive detection in related technologies often lags behind the occurrence of risks, and thus realize the whole-process dynamic risk warning. Compared with passive detection, the intelligent planning of the present invention makes the detection more targeted, proactive and effective. After detecting the risk problem, the detection system uses the fixed-point observation mode to control multiple UAVs to conduct rectification supervision on the risk locations, and dynamically sets different third detection routes, third flight altitudes and third flight speeds for each risk location. Therefore, the present invention can achieve detailed, accurate and targeted rectification observation, effectively solve the limitations of passive monitoring with a single preset parameter in related technologies, and thus achieve the goal of ensuring that the rectification is in place and the project progress is not delayed. The present invention can not only obtain the detailed rectification progress of different risk locations, but also ensure that the rectification is completed before the construction end time point, greatly improving the risk management effect.

[0107] The detection system in the embodiments of the present invention application will be described from the perspective of hardware processing. Please refer toFigure 3 , which is a schematic structural diagram of an entity device of the detection system in the embodiment of the present application.

[0108] It should be noted that Figure 3 the structure of the detection system shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

[0109] As Figure 3 shown, the detection system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage section 308 into the random access memory (RAM) 303, such as executing the method described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0110] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. The drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that the computer program read from it can be installed into the storage section 308 as needed.

[0111] Specifically, according to the embodiments of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments of the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are executed.

[0112] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings.

[0114] Specifically, the detection system of this embodiment includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the drone detection method for engineering provided in the above embodiment is implemented.

[0115] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the detection system described in the above embodiment; or it may exist separately without being assembled into the detection system. The above storage medium carries one or more computer programs. When the above one or more computer programs are executed by a processor of the detection system, the detection system is enabled to implement the drone detection method for engineering provided in the above embodiment.

[0116] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0117] As used in the foregoing embodiments, depending on the context, the term "when" may be construed to mean "if" or "after" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "when determining" or "if detecting (the stated condition or event)" may be construed to mean "if determining" or "in response to determining" or "when detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)".

[0118] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented, and the processes can be completed by relevant hardware instructed by a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the foregoing method embodiments. The foregoing storage medium includes: various media that can store program codes, such as ROM or random access memory RAM, magnetic disks, or optical discs.

Claims

1. A method for detecting engineering projects based on drones, characterized in that: Applied to a detection system, the method comprises: Controlling the UAV to fly along a first detection route in a rapid scanning mode to obtain first detection data of all engineering areas, wherein the rapid scanning mode corresponds to a first flight speed and a first flight altitude, and a plurality of shooting points in the first detection route are determined by a maximum scanning area corresponding to the first flight altitude and the areas of all engineering areas; Determine, according to the first detection data, each suspected risk location where each suspected risk problem occurs and each suspected risk problem level corresponding to each suspected risk problem; Based on the suspected risk positions and the suspected risk problem levels, determining a second flight speed set, wherein the second flight speed set includes a plurality of second flight speeds corresponding to the suspected risk positions, respectively, and the higher the suspected risk problem level corresponding to a suspected risk position, the slower the second flight speed corresponding to the suspected risk position; Controlling the drone to switch to a fine observation mode and fly along a second detection route to obtain second detection data of each suspected risk position, wherein the fine observation mode corresponds to a second flight altitude and a second flight speed set, and the second detection route is determined according to the distance between each suspected risk position and the second flight speed corresponding to each suspected risk position; Determine, according to the second detection data, a risk position having a risk problem among the suspected risk positions; The controlling the UAV to switch to the fine observation mode and fly along the second detection route to obtain the second detection data of each suspected risk position specifically includes: determining the flight time of the UAV between the suspected risk positions according to the distance between the suspected risk positions and the second flight speed corresponding to the suspected risk positions; taking any one of the suspected risk positions as the flight starting point of the UAV, and determining the next flight starting point with the shortest flight time with the flight starting point, until all the suspected risk positions are traversed, and an alternative second detection route is obtained and added to the alternative second detection route set, and the number of alternative second detection routes in the alternative second detection route set is the same as the number of the suspected risk positions; determining the alternative second detection route with the shortest total flight time in the alternative second detection route set as the second detection route; and controlling the UAV to switch to the fine observation mode and fly along the second detection route to obtain the second detection data of each suspected risk position.

2. The method according to claim 1, characterized in that Before the step of controlling the UAV to fly along the first detection route in the fast scanning mode to acquire first detection data of all engineering areas, the method further includes: Determine each construction link and the construction end time point corresponding to each construction link according to the obtained construction progress plan, and the construction link is carried out in the corresponding project area; A plurality of risk prevention detection time points are determined according to each of the construction end time points, wherein the risk prevention detection time point is the construction end time point minus a preset duration, and is used to instruct the drone to perform risk prevention detection on all engineering areas.

3. The method according to claim 1, characterized in that After the step of determining the risk location having a risk problem among the suspected risk locations according to the second detection data, the method further includes: Controlling multiple drones to switch to a fixed-point observation mode and fly along different third detection routes to obtain third detection data of the risk location, wherein the fixed-point observation mode corresponds to a third flight altitude set and a third flight speed set, wherein a third flight altitude in the third flight altitude set is determined according to a rectification altitude of the risk location, a third flight speed in the third flight speed set is determined according to a risk level of the risk problem, and the third detection route is determined according to an edge curve of a rectification area of ​​the risk location; The rectification status of the risk problem before the completion time of the construction is determined based on the third detection data.

4. The method according to claim 3, characterized in that Before the step of controlling the plurality of drones to switch to the fixed-point observation mode and fly along different third detection routes to obtain the third detection data of the risk position, the method further includes: Determine the corrective measures corresponding to the risk issues according to the risk issues corresponding to the risk locations; Determine a rectification area for the risk location based on the rectification scope corresponding to the rectification measures, wherein the higher the risk level corresponding to the risk problem occurring at the risk location, the smaller the area corresponding to the corresponding rectification area; The number of drones is determined according to the area corresponding to the risk location and the area corresponding to the rectification area.

5. The method according to claim 4, characterized in that After the step of determining the number of drones according to the area corresponding to the risk location and the area corresponding to the rectification area, the method further includes: Determine the risk types of the risk issues corresponding to the rectification areas according to a preset risk classification table, wherein the risk classification table includes the corresponding relationship between risk issues and risk types; Determining the drone model corresponding to each rectification area in a preset drone selection table according to the risk type, wherein the drone selection table includes a correspondence between the risk type and the drone model; If the risk type fails to match in the drone selection table, the performance parameters of the drone are determined according to the risk type and the performance parameters of the drone are set.

6. The method according to claim 5, characterized in that If the risk type fails to match in the drone selection table, determining the performance parameters of the drone according to the risk type and setting the performance parameters of the drone specifically includes: Analyze the detection requirements corresponding to the risk type and determine the performance parameters of the drone, the performance parameters including image resolution, flight altitude and flight speed; Call the setting function in the drone control interface to configure the performance parameters on the drone.

7. A detection system, characterized in that: The detection system comprises: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code comprises computer instructions, and the one or more processors call the computer instructions so that the detection system executes the method as described in any one of claims 1 to 6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a detection system, the detection system is caused to execute the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is run on a detection system, the detection system is caused to perform the method according to any one of claims 1 to 6.

Images