Unmanned aerial vehicle inspection method and equipment for reservoir bank slope defects and medium
By constructing a 3D real-scene model and using the TSP algorithm to generate the shortest flight path, the problems of path redundancy and high energy consumption in UAV inspection were solved, realizing efficient and accurate inspection of reservoir bank slopes and reducing risks and energy consumption.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THREE GORGES JINSHAJIANG CHUANYUN HYDROPOWER DEV CO LTD
- Filing Date
- 2026-02-04
- Publication Date
- 2026-05-12
AI Technical Summary
Existing drone inspection technology cannot efficiently and accurately detect defects on reservoir bank slopes, and its path planning is redundant, energy consumption is high, and it is easy to miss key defects.
By acquiring defect location data of the reservoir bank slope, a three-dimensional real-scene model is constructed. The shortest flight path is generated using the TSP algorithm, and the UAV is controlled to inspect along the optimal route. Combined with the three-dimensional coordinates and defect type classification, an independent shortest path route is generated.
It enables efficient, accurate, and low-energy-consumption inspection of reservoir bank slopes, reduces repeated take-off and landing and hovering time, lowers the risk of drones crashing into mountains or exceeding power limits, and improves the safety and efficiency of inspections.
Smart Images
Figure CN122015853A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of unmanned aerial vehicle (UAV) defect inspection technology, and in particular to a UAV inspection method, equipment and medium for reservoir bank slope defects. Background Technology
[0002] As a crucial component of reservoir dams, reservoir bank slopes are constantly exposed to wind, sun, rain, and water level fluctuations, making them highly susceptible to defects such as cracking and spalling of shotcrete, localized damage and blockage of drainage ditches, and surface seepage and calcium precipitation. If these defects are not detected and addressed promptly, they can lead to slope instability, landslides, and even jeopardize dam safety. Therefore, regular inspections of reservoir bank slopes are essential.
[0003] Traditional inspection methods rely on manual on-site inspections or drone aerial photography of the entire area. The former is inefficient and risky, while the latter involves a large amount of data collection and time-consuming post-processing, making it difficult to quickly assess the safety status of slopes.
[0004] Existing UAV path planning methods are mostly based on two-dimensional planar models, failing to consider the large elevation differences (tens to hundreds of meters) of reservoir bank slopes and the irregular three-dimensional distribution of defect points. This results in redundant flight paths, high energy consumption, and a tendency to miss critical defect points. Therefore, there is an urgent need for a highly efficient UAV inspection method that utilizes three-dimensional terrain to target existing defect points and provide optimal paths. Summary of the Invention
[0005] To address the aforementioned issues, this invention provides a method, equipment, and medium for unmanned aerial vehicle (UAV) inspection of reservoir bank slope defects, enabling intelligent UAV inspection of reservoir bank slope defects and improving the efficiency of reservoir bank slope monitoring.
[0006] This invention provides a method for unmanned aerial vehicle (UAV) inspection of defects on reservoir bank slopes, the specific technical solution of which is as follows: S1: Obtain data on the locations of defects that have appeared on the reservoir bank slope, divide the defect locations according to the defect type, and number each defect location. S2: In the pre-established 3D real-world model of the reservoir bank slope, query and record the 3D coordinates of each type of defect point; S3: For the set of three-dimensional coordinates of the same type of defect points, construct the TSP model and solve for the shortest flight path covering all defect points of that type. S4: Convert the shortest flight path into an executable flight path for the UAV and download it to the UAV flight control system. Control the UAV to perform inspections on the corresponding defect points and collect image data according to the flight path.
[0007] Furthermore, in step S1, the defect location data is obtained by fusing manual inspection record data with UAV aerial image data.
[0008] Furthermore, in step S1, the defect locations are divided into the following three categories: cracked or broken sprayed concrete, partially damaged or blocked drainage ditches, and apparent water seepage and calcium precipitation.
[0009] Furthermore, the construction of the 3D reality model and the acquisition of the 3D coordinates of each defect point are carried out in the following specific process: S201: Collect and acquire overall three-dimensional data of the target reservoir bank slope; S202: Construct a three-dimensional real-scene model of the reservoir bank slope based on the collected three-dimensional data; S203: Mark the identified defect locations in the three-dimensional reality model; S204: Query and record the coordinates of each defect location with each number in each category.
[0010] Furthermore, the TSP model is constructed using the defect location as the necessary node and the UAV take-off and landing point as the starting and ending nodes, where the distance between nodes is a three-dimensional Euclidean distance and the objective function is to minimize the total path length.
[0011] Furthermore, the number of nodes in the TSP model is less than or equal to the first threshold. In the construction of the TSP model, if the number of defect points is greater than the first threshold, the defect points are grouped according to the defect category, so that the number of nodes in each group is less than or equal to the first threshold, thus obtaining an independent TSP model for each group. Then, the model is solved separately to obtain multiple sub-routes.
[0012] Furthermore, the methods also include: Multiple inspection data points were acquired, and point-to-point changes in close-up images of the same defect location at different inspection periods were detected to evaluate slope stability.
[0013] Furthermore, the point-to-point change detection includes: automatically pairing images from different periods according to defect numbers, measuring the change in crack width or the expansion of seepage area through pixel-level alignment, and triggering a safety warning when the change exceeds a preset threshold.
[0014] The present invention also provides a drone inspection device for reservoir bank slope defects. The drone inspection device for reservoir bank slope defects includes: a memory, a processor, and a drone inspection program for reservoir bank slope defects stored in the memory and executable on the processor. When the drone inspection program for reservoir bank slope defects is executed by the processor, it implements the steps of the drone inspection method for reservoir bank slope defects described above.
[0015] The present invention also provides a storage medium storing a drone inspection program for reservoir bank slope defects, wherein when the drone inspection program for reservoir bank slope defects is executed by a processor, the steps of the drone inspection method for reservoir bank slope defects described above are implemented.
[0016] The beneficial effects of this invention are as follows: 1. This invention first identifies the locations of existing defects, then uses three-dimensional coordinates to calculate the shortest path for similar defects using the TSP algorithm. It then controls a drone to inspect along the optimal route, narrowing the inspection target from the entire area to the identified defect locations. Furthermore, it upgrades path calculation from two-dimensional to three-dimensional, solving the problems of large data volume in full-area collection, high energy consumption due to redundancy in two-dimensional paths, and easy omissions in existing technologies. This invention achieves efficient, accurate, and low-energy intelligent inspection of reservoir bank slopes.
[0017] 2. After classifying the defects into three categories—cracking / damage to sprayed concrete, damage / clogging of drainage ditches, and calcium efflorescence—the optimal path is solved independently for each category to avoid conflicts between the flight altitude and shooting angle required for different defects, thereby reducing repeated take-off and landing and hovering time.
[0018] 3. Based on the coordinates of the defect points, the coordinates can be directly queried in the 3D real scene model for positioning. This allows the generated flight path to automatically avoid excessive climbs or dives when the optimal path is solved by the TSP algorithm, reducing the risk of the drone crashing into mountains or exceeding its power limits. Attached Figure Description
[0019] Figure 1 This is a schematic diagram of the overall process of the method of the present invention.
[0020] Figure 2 This is a flowchart illustrating the process of obtaining defect locations, classifying types, and numbering defects according to the present invention.
[0021] Figure 3 This is a schematic diagram of the defect point coordinate acquisition process of the present invention. Detailed Implementation
[0022] The technical solutions in the embodiments of the present invention are clearly and completely described in the following description. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0023] In the description of the embodiments of the present invention, it should be noted that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is conventionally placed during use, or the orientation or positional relationship in which those skilled in the art conventionally understand it during use. This is only for the convenience of describing the present invention and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of the present invention. Furthermore, the terms "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0024] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.
[0025] Example 1 Embodiment 1 of the present invention discloses a method for unmanned aerial vehicle (UAV) inspection of defects on reservoir bank slopes, such as... Figure 1 As shown, the details are as follows: S1: Obtain data on the locations of defects that have appeared on the reservoir bank slope, divide the defect locations according to the defect type, and number each defect location. like Figure 2 As shown, the specific process is as follows: S101: Collect the defect locations identified by preliminary manual inspections of the target reservoir bank slope; generally, manual inspections can only reach areas of the slope with horse trails or steps, and cannot record defects on steep cliff slopes or areas inaccessible to personnel; in practice, manual inspections can generally cover 60% to 70% of the reservoir bank slope; Therefore, for steep cliffs or slopes that are inaccessible to personnel, drones are used for close-range aerial photography to inspect for defects in these areas. For example, multi-rotor drones combined with visible light cameras can be used to inspect various defects on the slope from an aerial perspective. Drone inspections are efficient and cover more areas, generally achieving a coverage rate of over 98% within the slope area. By combining the results of manual inspections and drone inspections, and cross-verifying them, the locations of defects on the reservoir bank slopes were determined, and these defects were then classified. Specifically, in this embodiment, the defect points are divided into the following three categories: cracked or broken sprayed concrete, partially damaged or blocked drainage ditches, and surface seepage and calcium precipitation.
[0026] S102: Each type of defect location is numbered with Arabic numerals. The numbering can be done sequentially according to elevation or distance from the dam abutment. The numbering order will not affect the quality of the subsequent inspection route.
[0027] S2: A 3D reality model of the reservoir bank slope is pre-built. Based on the 3D reality model, the 3D coordinates of each type of defect point are queried and recorded, such as... Figure 3 As shown, it specifically includes: S201: Collect and acquire overall three-dimensional data of the target reservoir bank slope; specifically, the aerial photograph taken by the camera covers a large area, while the range of the defect is generally small. That is to say, the coordinates of the defect point do not need to be very precise. Therefore, in this embodiment, data can be collected by a UAV. The flight altitude of the UAV can be set to 60m-100m, the flight path overlap rate can be set to 85% in the heading direction and 75% in the lateral direction, and the single-lens vertical photography mode can be used to obtain the three-dimensional aerial photography data of the entire slope.
[0028] S202: Based on the collected 3D data, use third-party software (such as DJI Terra or ContextCapture) to generate a 3D reality model and a digital orthophoto map (DOM) that meet the requirements. S203: Mark the identified defect locations in the 3D reality model; specifically, number the three major categories of defects and represent them with different colors, such as: cracking and detachment of sprayed concrete is represented by red; local damage and blockage of drainage ditches are represented by yellow; and apparent water seepage and calcium precipitation defects are represented by blue. Each of them was assigned a number in advance and then applied to the 3D reality model.
[0029] S204: Query the coordinates of each defect point of each category and number, including the three-dimensional coordinates of X (or longitude E), Y (or latitude N), and Z (or elevation U), and record the coordinates of each defect point of each category and number in their respective tables.
[0030] S3: For the set of three-dimensional coordinates of the same type of defect points, with the defect points as the necessary nodes and the UAV take-off and landing points as the starting and ending nodes, construct the Traveling Salesman Problem (TSP) model, where the distance between nodes is the three-dimensional Euclidean distance, the objective function is to minimize the total path length, and solve for the shortest flight path covering all defect points of this type. Specifically as follows: The objective function is defined as the total travel distance of the inspection path for defects on the reservoir bank slope. , express coordinates to The straight-line distance of the coordinates is generally unevenly distributed on the reservoir bank slope, and each inspection is assumed to be completed by only one drone. Therefore, there is only one moving object in this problem. Based on this, the inspection route is determined, and the following model assumptions are made: There are n necessary locations (excluding hangars or drone take-off and landing points); There are no multiple drones being inspected separately; Each necessary location only needs to be inspected once in a single inspection, and the inspection point is not returned to. The distance between two points is calculated using three-dimensional spatial distance, taking into account height differences.
[0031] The set of all points in the model , The elements in the array represent the defect numbers to be inspected, where 0 indicates the location of the hangar or drone take-off and landing point, thus creating a permutation. This minimizes the total inspection path length. Construct the following model: Decision function:
[0032] The target is its minimum value; in, Indicates defects Point to Defect Euclidean distance; The constraints are:
[0033] Ensure that the drone passes through each necessary location only once;
[0034] Excluding the hangar, the number of coordinates that the drone needs to inspect is: .
[0035] Solving the Traveling Salesman Problem involves a huge amount of computation, assuming the solution contains... The problem of travel agents at each point, then there is There are multiple possible path arrangements, meaning that the computation time can be very long when the number of inspection points is too large. In this embodiment, the enumeration method uses a recursive function to generate all possible path arrangements and calculates the length of each path. If the length of the current path is less than the known shortest path length, the shortest path is updated.
[0036] By employing an enumeration method to exhaustively enumerate all permutations and combinations of inspection points, and calculating the total length and energy consumption cost of each path one by one, the optimal inspection path scheme that meets the conditions is accurately identified after completing the comparison of all data, providing a reliable standard answer for subsequent algorithm comparison studies.
[0037] In practice, cracking, breakage, and spalling defects are relatively common on reservoir bank slopes, especially if the slope area is large, exceeding 1 km². 2 When there are more than 30 defect points (the algorithm supports efficient calculation of up to 30 points), it is best to divide them into multiple flight paths. This can ensure the rapid determination of the optimal flight path and ensure that the drone's battery power can successfully complete the aerial photography task of each flight path. Defects such as local damage, siltation, and apparent water seepage and calcium precipitation in drainage ditches are relatively few in number, generally less than 30 points. However, if there are more than 30 points in the application, they can also be divided into multiple flight paths to carry out intelligent drone inspection. Therefore, as a preferred embodiment, in the construction of the TSP model, if the number of defect points is greater than 30, the defect points are grouped according to the defect category, so that the number of nodes in each group is less than or equal to 30, so that each group has an independent TSP model, and then the model is solved separately to obtain multiple sub-routes. S4: Convert the shortest flight path into an executable flight path for the UAV and download it to the UAV flight control system; Specifically, a program can be written based on compiler software (such as Dev-C++) to calculate the optimal inspection path, and then MATLAB can be used to plot the image after the result is calculated.
[0038] According to the executable route, control the UAV to perform inspections on the corresponding defect locations and collect image data along the route; Specifically, the inspection equipment can be a multi-rotor drone with a visible light camera, and the inspection flight altitude can be set to 16-20 meters.
[0039] As a preferred embodiment, the method further includes: acquiring multiple inspection data; evaluating the stability of the slope after the data from the 2nd, 3rd, ... nth inspections are collected; and performing point-to-point change detection on close-up images of the same defect location in different inspection periods, including: automatically pairing images from different periods according to defect numbers; measuring the change in crack width or the expansion of seepage area through pixel-level alignment; judging whether the crack has widened, whether there is a collapse, and whether there is seepage damage based on the change; and thus evaluating the slope stability; and triggering a safety warning when the change exceeds a preset threshold.
[0040] Example 2 Embodiment 2 of the present invention discloses a drone inspection device for reservoir bank slope defects. The device can be a user equipment (UE) such as a mobile phone, smartphone, laptop, digital broadcast receiver, personal digital assistant (PDA), tablet computer (PAD), handheld device, vehicle-mounted device, wearable device, computing device, or other processing device connected to a wireless modem, or mobile station (MS), used to perform the drone inspection method for reservoir bank slope defects. The device may be referred to as a user terminal, portable terminal, desktop terminal, etc.
[0041] Typically, the device includes: at least one processor, a memory, and a drone inspection program for reservoir bank slope defects stored in the memory and executable on the processor, the drone inspection program for reservoir bank slope defects being configured to implement the steps of the drone inspection method for reservoir bank slope defects as described in Example 1.
[0042] The processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor can be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). The processor may also include a main processor and coprocessors. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. The processor may also include an AI (Artificial Intelligence) processor, which handles the computational operations of the drone inspection program for reservoir bank slope defects, enabling the drone inspection method for reservoir bank slope defects to learn autonomously, improving efficiency and accuracy.
[0043] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory are used to store at least one instruction, which is executed by a processor to implement the UAV inspection method for reservoir bank slope defects described in Embodiment 1.
[0044] In some embodiments, the terminal may also optionally include a communication interface and at least one peripheral device. The processor, memory, and communication interface can be connected via a bus or signal lines. Each peripheral device can be connected to the communication interface via a bus, signal lines, or a circuit board. Specifically, the peripheral device includes at least one of a radio frequency circuit, a display screen, and a power supply.
[0045] The communication interface can be used to connect at least one I / O (Input / Output) related peripheral device to the processor and memory. The communication interface is used via the peripheral device to receive movement trajectories and other data uploaded by the user from multiple mobile terminals. In some embodiments, the processor, memory, and communication interface are integrated on the same chip or circuit board; in other embodiments, any one or two of the processor, memory, and communication interface can be implemented on separate chips or circuit boards, which is not limited in this embodiment.
[0046] Radio frequency (RF) circuits are used to receive and transmit RF signals, also known as electromagnetic signals. RF circuits communicate with communication networks and other communication devices via electromagnetic signals, thereby acquiring the movement trajectories and other data of multiple mobile terminals. RF circuits convert electrical signals into electromagnetic signals for transmission, or convert received electromagnetic signals back into electrical signals. Optionally, RF circuits include: antenna systems, RF transceivers, one or more amplifiers, tuners, oscillators, digital signal processors, codec chipsets, user identity module cards, etc. RF circuits can communicate with other terminals through at least one wireless communication protocol. These wireless communication protocols include, but are not limited to: metropolitan area networks (MANs), various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks (WLANs), and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit may also include circuitry related to NFC (Near Field Communication), which is not limited in this embodiment.
[0047] The display screen is used to display the UI (User Interface). This UI can include graphics, text, icons, videos, and any combination thereof. When the display screen is a touch screen, it also has the ability to collect touch signals on or above the surface of the display. These touch signals can be input as control signals to a processor for processing. In this case, the display screen can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, the display screen can be a single screen, the front panel of the electronic device; in other embodiments, there can be at least two screens, respectively disposed on different surfaces of the electronic device or in a folded design; in still other embodiments, the display screen can be a flexible screen, disposed on a curved or folded surface of the electronic device. Furthermore, the display screen can be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The display screen can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).
[0048] A power supply is used to power the various components in an electronic device. The power supply can be alternating current (AC), direct current (DC), a disposable battery, or a rechargeable battery. When the power supply includes a rechargeable battery, the rechargeable battery can support wired or wireless charging. The rechargeable battery can also be used to support fast charging technology.
[0049] Example 3 Embodiment 3 of the present invention discloses a computer storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the UAV inspection method for reservoir bank slope defects described in Embodiment 1 above.
[0050] The storage medium is a readable storage medium, specifically it can be a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, or any other readable storage medium that can store program code.
[0051] This invention is not limited to the specific embodiments described above. The invention extends to any new feature or combination disclosed in this specification, as well as any new method or process step or combination disclosed herein.
Claims
1. A method for unmanned aerial vehicle (UAV) inspection of defects on reservoir bank slopes, characterized in that, include: S1: Obtain data on the locations of defects that have appeared on the reservoir bank slope, divide the defect locations according to the defect type, and number each defect location. S2: Establish a three-dimensional real-scene model of the reservoir bank slope, query and record the three-dimensional coordinates of each type of defect point; S3: For the set of three-dimensional coordinates of the same type of defect points, construct the TSP model and solve for the shortest flight path covering all defect points of that type. S4: Convert the shortest flight path into an executable flight path for the UAV and download it to the UAV flight control system. Control the UAV to perform inspections on the corresponding defect points and collect image data according to the flight path.
2. The UAV inspection method for reservoir bank slope defects according to claim 1, characterized in that, In step S1, the defect location data is obtained by fusing manual inspection record data with UAV aerial image data.
3. The UAV inspection method for reservoir bank slope defects according to claim 2, characterized in that, In step S1, the defect points are divided into the following three categories: cracked or broken sprayed concrete, partially damaged or blocked drainage ditches, and surface seepage and calcium precipitation.
4. The UAV inspection method for reservoir bank slope defects according to claim 1, characterized in that, The specific process for constructing the 3D reality model and obtaining the 3D coordinates of each defect point is as follows: S201: Collect and acquire overall three-dimensional data of the target reservoir bank slope; S202: Construct a three-dimensional real-scene model of the reservoir bank slope based on the collected three-dimensional data; S203: Mark the identified defect locations in the three-dimensional reality model; S204: Query and record the coordinates of each defect location with each number in each category.
5. The UAV inspection method for reservoir bank slope defects according to claim 1, characterized in that, The TSP model is constructed with the defect location as the necessary node and the UAV take-off and landing point as the starting and ending nodes. The distance between nodes is the three-dimensional Euclidean distance, and the objective function is to minimize the total path length.
6. The UAV inspection method for reservoir bank slope defects according to claim 5, characterized in that, The number of nodes in the TSP model is less than or equal to the first threshold. In the construction of the TSP model, if the number of defect points is greater than the first threshold, the defect points are grouped according to the defect category, so that the number of nodes in each group is less than or equal to the first threshold, so that each group of independent TSP models is obtained, and then solved separately to obtain multiple sub-routes.
7. The UAV inspection method for reservoir bank slope defects according to any one of claims 1-6, characterized in that, Also includes: Multiple inspection data points were acquired, and point-to-point changes in close-up images of the same defect location at different inspection periods were detected to evaluate slope stability.
8. The UAV inspection method for reservoir bank slope defects according to claim 7, characterized in that, The point-to-point change detection includes: automatically pairing images from different periods according to defect numbers, measuring the change in crack width or the expansion of seepage area through pixel-level alignment, and triggering a safety warning when the change exceeds a preset threshold.
9. A drone inspection device for reservoir bank slope defects, characterized in that, The drone inspection equipment for reservoir bank slope defects includes: a memory, a processor, and a drone inspection program for reservoir bank slope defects stored in the memory and executable on the processor. When the drone inspection program for reservoir bank slope defects is executed by the processor, it implements the steps of the drone inspection method for reservoir bank slope defects according to any one of claims 1-8.
10. A storage medium, characterized in that, The storage medium stores a drone inspection program for reservoir bank slope defects. When the drone inspection program for reservoir bank slope defects is executed by the processor, it implements the steps of the drone inspection method for reservoir bank slope defects as described in any one of claims 1-8.