Unmanned aerial vehicle mixed path inspection method and system for distribution network tower and line
By using a hybrid path inspection method involving drones, combined with dynamic path planning and obstacle avoidance decision-making, efficient and accurate inspection of poles and lines has been achieved. This solves the problems of repetitive paths and data fragmentation in existing technologies, and improves inspection efficiency and accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-22
- Publication Date
- 2026-03-27
AI Technical Summary
In existing drone inspection technologies, the independent inspection of poles and lines leads to numerous overlapping paths, fragmented data, and blind spots, making it difficult to efficiently and accurately monitor power distribution facilities.
A hybrid path inspection method using unmanned aerial vehicles (UAVs) is adopted, which integrates pole and tower circumferential inspection with line-along inspection through multi-class target detection. Combined with dynamic path planning and obstacle avoidance decision-making, hybrid path inspection of poles and lines is achieved, and precise positioning and data correlation are performed using image acquisition equipment and sensor data.
It improved inspection efficiency and accuracy, eliminated invalid paths, connected data, covered inspection blind spots, and achieved efficient and accurate defect location.
Smart Images
Figure CN121742491A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of unmanned aerial vehicle autonomous inspection, and particularly relates to a mixed path inspection method and system for unmanned aerial vehicles for distribution network towers and lines. BACKGROUND
[0002] With the increasing demand for construction and maintenance of distribution networks, traditional manual inspection methods face many challenges, such as low inspection efficiency, long time consumption, high personnel safety risk, complex working environment, and are easily affected by weather, terrain and other factors, resulting in the accuracy and timeliness of the inspection data being unable to be effectively guaranteed. In addition, due to the wide distribution of distribution network towers and lines, the coverage of manual inspection is limited, and it is difficult to ensure efficient and accurate monitoring of the operation status of each facility.
[0003] Current unmanned aerial vehicle inspection technology has gradually become an effective solution for power inspection, but the existing unmanned aerial vehicle inspection systems generally adopt a single target inspection mode, i.e., tower inspection and line inspection are independent of each other - tower inspection only revolves around a preset hovering point or a fixed flight path, and line inspection only flies parallel to the conductor. This single mode has the following core defects: first, there are many repeated and ineffective paths, resulting in increased inspection time and high overall cost; second, the data is fragmented, and the detection data of the tower and the line lacks correlation, making it difficult to quickly locate the transition zone defects such as the connection part of the conductor and the tower; third, there are inspection blind spots, such as key transition areas such as the connection fittings of the conductor and the tower, and the insulator mounting point, which are easily missed, seriously affecting the inspection accuracy. The mixed path inspection mode proposed by the present application organically integrates tower revolving inspection and line along-path inspection, and realizes dynamic switching of the path through multi-classification target detection, which not only eliminates ineffective paths, but also connects data, covers transition zone blind spots, and significantly improves inspection efficiency and accuracy. SUMMARY
[0004] The technical problem to be solved by the present application is that in view of the above problems existing in the prior art, the present application provides a mixed path inspection method and system for unmanned aerial vehicles for distribution network towers and lines, which has the advantages of simple implementation method, low cost, high inspection efficiency, and high automation level, can realize autonomous mixed path inspection of distribution network towers and lines using unmanned aerial vehicles, improve the efficiency, automation level and accuracy of distribution line inspection, and facilitate fast and accurate defect location.
[0005] To solve the above technical problems, the technical solution proposed by the present application is as follows: A mixed path inspection method for unmanned aerial vehicles for distribution network towers and lines, comprising the following steps: Step S01. Acquire an image of the area to be inspected and perform multi-classification target detection, and determine the initial flight direction of the unmanned aerial vehicle according to the position of the tower in the image in the detection result; Step S02. Real-time obstacle avoidance data is received during the flight of the unmanned aerial vehicle to the tower, and flight control is performed according to the obstacle avoidance data; Step S03. When the unmanned aerial vehicle approaches the tower, the three-axis speed of the unmanned aerial vehicle is adjusted to gradually center the position of the tower in the image, and the inspection image acquisition height is determined according to the size of the tower detection frame in the image and the height detection data; Step S04. A circumferential trajectory is generated by circumferential movement around the tower at a specified radius at the determined inspection image acquisition height, the unmanned aerial vehicle is controlled to fly around the tower according to the circumferential trajectory to perform multi-angle image acquisition of the tower, real-time obstacle detection is performed, and the acquisition angle that avoids the risk of obstacles is controlled according to the real-time obstacle detection result; Step S05. The line detection result is obtained according to the acquired image, and the line direction is analyzed, and the yaw angle of the unmanned aerial vehicle is adjusted to control the unmanned aerial vehicle to fly to the next tower.
[0006] Further, in step S01, the initial flight direction of the unmanned aerial vehicle is determined according to the position of the tower in the image in the detection result, including: gradually correcting the flight angle of the unmanned aerial vehicle through the position of the center point of the tower detection frame in the image to determine the initial flight direction.
[0007] Further, in step S02, flight control is performed using an obstacle avoidance mode or an adaptive model, wherein when flight control is performed using the obstacle avoidance mode, the corresponding obstacle avoidance level is determined according to the closest distance between the unmanned aerial vehicle and the obstacle in the forward channel, and the response strategy corresponding to the obstacle avoidance level is responded to; when flight control is performed using the adaptive model, the flight angle of the unmanned aerial vehicle and the pitch angle of the image acquisition device are adjusted adaptively according to the position of the tower, and the flight speed is adjusted according to the change in the pitch angle of the image acquisition device, when the unmanned aerial vehicle approaches the tower, the pitch angle of the unmanned aerial vehicle is increased, and the flight speed is controlled according to the change in the angle.
[0008] Further, in the flight control using the adaptive model, a PD algorithm is used to adjust the heading angle of the unmanned aerial vehicle and the pitch angle of the image acquisition device according to the deviation of the current tower detection frame in the image from the center, so as to keep the center position of the detection frame in the image, wherein the calculation expression of the adjustment amount of the unmanned aerial vehicle yaw angle and the pitch angle of the image acquisition device is:
[0009]
[0010] wherein, Kp is the proportional gain for calculating the unmanned aerial vehicle yaw angle, Kd is the differential gain for calculating the unmanned aerial vehicle yaw angle, is the adjustment amount of the unmanned aerial vehicle yaw angle, an adjustment amount of the pitch angle of the image acquisition device, calculating a proportional gain for the pitch angle of the image acquisition device, calculating a differential gain for the pitch angle of the image acquisition device.
[0011] Further, in step S03, adjusting the three-axis speed of the UAV so that the position of the tower in the image is gradually centered includes: when the pitch angle of the UAV reaches 90 degrees, calculating the deviation of the current position of the tower in the image from the center according to the PID control method, and adjusting the three-axis speed of the UAV to gradually center the tower according to the deviation, wherein the calculation expression of the three-axis speed of the UAV is:
[0012] wherein, is a proportional gain coefficient, is an integral gain coefficient, is a differential gain coefficient, is the X-axis speed, is the Y-axis speed.
[0013] Further, in step S03, the inspection image acquisition height is determined according to the following formula :
[0014]
[0015] wherein, is the height data detected by the height detection sensor at time t, is a weight, is the visual height estimate at time t, is the focal length of the image acquisition device, is the actual width of the tower, is the pixel width of the frame tower detection box at time t.
[0016] Further, in step S04, the linear speed of flying around the tower is also determined according to the following formula:
[0017] ,
[0018] wherein, is the radius, angular velocity, is the set speed of flying around the tower, is the current angle of the UAV, is the angle of the UAV before flying around the tower.
[0019] Further, in step S05, the UAV yaw angle is adjusted according to the line direction analysis result: the center point of each parallel line detection frame is calculated , and the tower head is taken as a reference point , the line direction angle is calculated according to the line connecting the reference point and the center point of the parallel line detection frame , the UAV yaw angle is adjusted according to the line direction angle until it is 0.
[0020] The UAV hybrid path inspection system comprises a UAV carrying an image acquisition device, and further comprises an inspection device connected with the UAV, wherein the inspection device comprises a processor and a memory, the memory is used to store a computer program, and the processor is used to execute the computer program to perform the above method.
[0021] A computer readable storage medium storing a computer program, the computer program is executed to realize the method as described above.
[0022] Compared with the prior art, the advantages of the present application are as follows: 1. The present application realizes the "rough positioning-accurate calibration" of the tower in stages, first determines the initial flight direction of the UAV according to the position of the tower, performs flight control in combination with dynamic path planning and obstacle avoidance decision during the UAV flying to the tower to roughly position the tower, and performs accurate calibration when the UAV approaches the tower by using the position of the tower in the image, so as to accurately position the position of the tower, realize the autonomous inspection of the distribution line UAV, and improve the inspection efficiency and quality of the distribution line.
[0023] 2. The present application analyzes the line direction according to the images detected during the tower circling, adjusts the UAV yaw angle according to the line direction, so that the UAV can automatically connect to fly to the next tower, realizes the automatic transition from tower inspection to wire tracking inspection and the intelligent connection control of wire topology perception, thereby realizing the full-automatic hybrid path inspection of the tower and the line, improving the efficiency and automation level of the distribution line inspection, avoiding invalid and repeated flight paths, and effectively associating the inspection data of the tower and the line through the hybrid path inspection, thereby facilitating the defect positioning according to the inspection data, and further avoiding the inspection blind area and improving the inspection accuracy based on the hybrid path inspection mode. BRIEF DESCRIPTION OF DRAWINGS
[0024] Figure 1 is the implementation flowchart of the UAV hybrid path inspection method of the distribution network tower and line of the present embodiment. DETAILED DESCRIPTION
[0025] The present application is further described below in conjunction with the accompanying drawings and specific preferred embodiments, but it will be understood that the scope of the present application is not limited to the specific embodiments.
[0026] As shown in the embodiment of the method for mixed path inspection of a power grid tower and a line by a UAV, the steps include: Figure 1 Step S01. Take-off stage: acquire images of the area to be inspected and perform multi-classification target detection, and determine the initial flight direction of the UAV according to the position of the tower in the images in the detection results.
[0027] In the take-off stage, the video stream images of the area to be inspected collected by the UAV in real time are acquired, and the multi-classification target detection is performed by using the detection model YOLOX-1 to identify the types of towers, tower heads, insulators, etc., and the initial flight direction of the UAV is determined according to the position of the tower in the detection results. Through multi-classification target detection, pre-data can be provided for subsequent tower-circulating inspection, auxiliary tower positioning can be provided, and the environment can be perceived to optimize the flight path.
[0028] Specifically, when the initial flight direction of the UAV is determined according to the position of the tower in the detection results, the flight angle of the UAV can be gradually corrected through the position of the center point of the tower detection frame in the image to determine the initial flight direction. For example, the detection model YOLOX-1 can be used to infer the images of the area to be inspected, and output multi-classification target detection results such as the types of towers, tower heads, insulators, etc. The flight angle of the UAV can be gradually corrected through the position of the center point of the tower detection frame in the image to determine the initial flight direction.
[0029] Further, the take-off position can also be recorded for use as GPS data for the subsequent return process.
[0030] Step S02. Tower positioning stage: the UAV receives obstacle avoidance data in real time during the flight to the tower, and performs flight control according to the obstacle avoidance data.
[0031] In the tower positioning stage, the UAV flies towards the tower direction, and the UAV will receive obstacle avoidance data in real time during the flight to the tower, and make path planning and obstacle avoidance decisions according to the obstacle avoidance data to realize flight control. The obstacle avoidance data can be obtained by an obstacle avoidance module, which can be composed of a combination system of visual SLAM and monocular depth estimation. The obstacle avoidance data issued by the obstacle avoidance module is the nearest distance of the obstacle in the forward channel within a specified range (such as 1m x 1m) centered on the UAV body.
[0032] As an optional implementation, when the UAV flies to the tower in the obstacle avoidance mode, the corresponding obstacle avoidance level is determined according to the closest distance between the UAV and the obstacle in the front channel, and the response strategy corresponding to the obstacle avoidance level is responded to. For example, a quantization system with five levels of obstacle avoidance states can be configured according to the distance range from the obstacle, as shown in Table 1. When a level 3 or above obstacle is detected, three-dimensional path re-planning is started, and different response strategies are used for each level. When it is level 1, the UAV is controlled to stop immediately, when it is level 2, the point position is issued to the obstacle avoidance module, the flight path is planned, when it is level 3, the speed is maintained and the visual positioning is started, when it is level 4, the warning prompt is issued and the monitoring is maintained, and when it is level 5, the normal cruise is maintained. The configuration of the obstacle avoidance level and the response strategy can be configured according to actual needs.
[0033] Table 1: Obstacle avoidance mode
[0034] As an optional implementation, when the flight control is performed by using the adaptive model, the flight angle of the UAV and the pitch angle of the image acquisition device (such as a gimbal camera) can be adaptively adjusted according to the position of the tower, and the flight speed can be adjusted according to the change of the pitch angle of the image acquisition device. When the UAV approaches the tower, the pitch angle of the UAV is increased, and the flight speed is controlled according to the angle change.
[0035] Specifically, in the flight control using the adaptive model, the PD algorithm can be used to adjust the heading angle of the UAV and the pitch angle of the image acquisition device according to the deviation of the current tower detection frame in the image from the center, so as to keep the center position of the detection frame in the image. The definition of the position deviation can be expressed as: (1) wherein, is the center coordinate of the current frame detection frame, is the center coordinate of the picture.
[0036] The calculation expression of the adjustment amount of the yaw angle of the UAV and the pitch angle of the image acquisition device can be expressed as: (2) (3) wherein, is the proportional gain for calculating the yaw angle of the UAV, is the differential gain for calculating the yaw angle of the UAV, is the adjustment amount of the yaw angle of the UAV, is the adjustment amount of the pitch angle of the image acquisition device, is the proportional gain for calculating the pitch angle of the image acquisition device, is the differential gain for calculating the pitch angle of the image acquisition device.
[0037] Further, the image acquisition device can also be configured with an angle adjustment range during the pitch angle adjustment process, for example, the angle limit of the gimbal can be configured as -90° to 45°.
[0038] Step S03. Precise positioning adjustment phase: when the unmanned aerial vehicle approaches the tower, the three-axis speed of the unmanned aerial vehicle is adjusted to gradually center the position of the tower in the image, and the inspection image acquisition height is determined according to the size of the tower detection frame in the image and the height detection data.
[0039] In this embodiment, the precise positioning adjustment phase is entered when the unmanned aerial vehicle approaches the tower. In this phase, the position of the tower target detection frame in the image and the IMU data of the gimbal on the unmanned aerial vehicle are used to accurately calibrate the tower position, and the size of the tower detection frame in the image and the height detection data are used to determine the inspection image acquisition height, i.e. the photographing height of the gimbal camera. When the pitch angle reaches -90°, the deviation of the current tower position from the center in the image is calculated, and the three-axis speed of the unmanned aerial vehicle is adjusted to gradually center the tower.
[0040] As an optional implementation, adjusting the three-axis speed of the unmanned aerial vehicle to gradually center the position of the tower in the image includes: When the pitch angle of the unmanned aerial vehicle reaches 90 degrees, the deviation of the current tower position from the center in the image is calculated according to the PID control method, and the three-axis speed of the unmanned aerial vehicle is adjusted to gradually center the tower according to the deviation, wherein the calculation expression of the three-axis speed of the unmanned aerial vehicle is: (4) wherein, Kp is a proportional gain coefficient, Ki is an integral gain coefficient, Kd is a differential gain coefficient, Vx is the X-axis speed, Vy is the Y-axis speed.
[0041] As an optional implementation, the inspection photographing height can be determined according to the size of the tower detection frame in the image and the height detection data according to the following formula : (5) (6) wherein, Ht is the height data detected by the height detection sensor at time t, w is a weight, Hvis is the visual height estimation at time t, f is the focal length of the image acquisition device, Wt is the actual width of the tower, Wvis is the pixel width of the tower detection frame at time t.
[0042] Step S04. Tower-circulating detection phase: a circumnavigating track is generated according to a specified radius around the tower at a determined image collection height, the UAV is controlled to fly around the tower according to the circumnavigating track to collect multi-angle images of the tower, and real-time obstacle detection is performed, and the collection angle at which there is a risk of obstacle avoidance is controlled according to the real-time obstacle detection result.
[0043] In the tower-circulating detection phase, the UAV is controlled to make circumnavigating movement around the tower at a fixed height with a certain radius and maintain a certain linear velocity to collect multi-angle images, so as to realize fine inspection of the tower.
[0044] As an optional implementation, the linear velocity of the tower-circulating flight can be determined as follows: (7) , (8) wherein, is the radius, is the angular velocity, is the set tower-circulating speed, is the current angle of the UAV, is the angle of the UAV before the tower-circulating flight.
[0045] Considering that the overhead distribution tower has simple structure and small volume, five photographing points per tower can meet the demand of multi-view fine inspection, and the photographing points can be specifically at 0°, 90°, 180°, 270° and directly above the tower, respectively, and the yaw angle and the gimbal pitch angle need to be ensured to be directed to the tower head of the tower to ensure that the tower head of the tower is within the image collection range.
[0046] In the tower-circulating process, the obstacle avoidance module detects the obstacle risk of the target point in real time, and if the avoidance return signal of a certain angle has an obstacle risk, the angle is skipped and the next angle is used for image collection. After each image collection is completed, the tower position information is stored, and the inspection result is marked, for example, including storing the tower number, GPS information, inspection photos, etc.
[0047] Step S05. Linkage flight phase: the line detection result is obtained according to the collected multi-angle images, and the line direction is analyzed, and the yaw angle of the UAV is adjusted according to the analysis result of the line direction to control the UAV to fly to the next tower.
[0048] In this embodiment, by analyzing the line direction based on the images detected during the tower circling process, the yaw angle of the UAV is adjusted according to the line direction, enabling it to automatically connect and fly to the next tower. This achieves automatic transition from tower inspection to conductor tracking inspection and automatic connection to the next tower inspection, thereby realizing fully automatic hybrid path inspection of towers and lines. This improves the efficiency and automation level of power distribution line inspection, avoids invalid and repetitive flight paths, and effectively associates tower and line inspection data through hybrid path inspection, making it easier to locate defects based on the inspection data. In addition, the hybrid path inspection mode can also avoid inspection blind spots and improve inspection accuracy.
[0049] As an optional implementation, the center point of the detection frame for each parallel line segment is calculated. And using the tower head as a reference point According to the reference point Center point of parallel line detection frame Calculate the direction and angle of the connecting lines. According to the angle of the route Adjust the drone's yaw angle until It is 0.
[0050] Specifically, the YOLOX-2 detection model can be used to infer the images acquired during the tower circumference process, output the line (conductor) detection results, perform line alignment analysis, and calculate the center point of each parallel conductor detection frame. Using the tower head as a reference point Using the reference point as an example, the traverse angle is calculated by connecting the reference point to the center point of the traverse. Adjust the drone's yaw angle according to the direction and angle of the guide wire until... The value is 0. When multiple parallel conductor segments are detected, the next tower to be flown to can be determined by selecting the branch direction.
[0051] As an optional implementation, the system can automatically return to base after completing a predetermined number of pole inspections. The return point can be determined based on the recorded takeoff position and GPS data.
[0052] In specific application embodiments, considering the characteristics of overhead power distribution lines, the inspection task is divided into multiple operational phases, including: Takeoff phase: using the YOLOX-1 detection model to infer and output multi-class target detection results to determine the initial flight direction; Tower positioning phase: during flight to the tower, obstacle avoidance data is integrated to determine the flight status, and obstacle avoidance mode or adaptive model is used for flight control; Precise positioning and adjustment phase: when approaching the tower, the tower position is precisely calibrated using target detection boxes and gimbal IMU data, and the inspection and photography height is determined by combining TOF sensor data; Around-the-tower inspection phase: generating a circumferential trajectory, taking multi-angle photos, performing real-time obstacle avoidance detection, storing tower position information, marking inspection results, and uploading a status report; Connecting flight phase: the YOLOX-2 detection model infers and outputs conductor detection results, performs conductor direction analysis, and controls the flight direction of the next tower; Return phase: automatically returning after completing the predetermined number of tower inspections.
[0053] In summary, this invention enables dynamic path planning and obstacle avoidance decision-making, freeing autonomous inspection from the limitations of non-RTK areas and expanding its application scope. By achieving "coarse positioning-fine calibration" in stages, combined with target detection box center calibration and multi-sensor data fusion, it can accurately determine the location of the target to be inspected. At the same time, it can realize intelligent connection control based on line topology perception, and achieve autonomous heading control between towers based on line route analysis. It can be applied to the scenario of multiple branch lines in the distribution network to quickly realize the autonomous hybrid path inspection of overhead power distribution lines by UAVs. It eliminates the need for a series of complicated line scanning, 3D modeling, and flight path planning, which can improve the efficiency and quality of power distribution line inspection and significantly reduce the dependence on high-precision UAV hardware and professional flight control personnel.
[0054] This embodiment also provides a UAV hybrid path inspection system, including a UAV equipped with an image acquisition device, and an inspection device connected to the UAV. The inspection device includes a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program to perform the method described above.
[0055] This embodiment further provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described above.
[0056] Those skilled in the art will understand that the above embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0057] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention should fall within the protection scope of the present invention.
Claims
1. A method for UAV hybrid path inspection of power distribution network towers and lines, characterized in that, The steps include: Step S01. Acquire images of the area to be inspected and perform multi-class target detection, and determine the initial flight direction of the UAV based on the position of the tower in the image in the detection results; Step S02. During the flight of the drone to the tower, it receives obstacle avoidance data in real time and performs flight control based on the obstacle avoidance data; Step S03. When the drone approaches the tower, adjust the drone's three-axis speed so that the tower's position in the image gradually becomes centered, and determine the inspection image acquisition height based on the size of the tower detection box in the image and the height detection data. Step S04. Generate a circular trajectory by performing a circular motion around the tower with a specified radius at the determined inspection image acquisition height. Control the UAV to fly around the tower according to the circular trajectory to acquire images of the tower from multiple angles and perform obstacle detection in real time. Control the acquisition angle to avoid the risk of obstacles based on the real-time obstacle detection results. Step S05. Obtain the line detection results based on the collected images and analyze the line direction. Adjust the yaw angle of the UAV according to the line direction analysis results to control the UAV to fly to the next base tower.
2. The method for UAV hybrid path inspection of power distribution network towers and lines according to claim 1, characterized in that, In step S01, determining the initial flight direction of the UAV based on the position of the tower in the image from the detection results includes: gradually correcting the flight angle of the UAV by using the position of the center point of the tower detection box in the image to determine the initial flight direction.
3. The method for UAV hybrid path inspection of power distribution network towers and lines according to claim 1, characterized in that, In step S02, flight control is performed using either obstacle avoidance mode or an adaptive model. When using obstacle avoidance mode, the corresponding obstacle avoidance level is determined based on the closest distance between the UAV and obstacles in the forward path, and the response is performed according to the response strategy corresponding to the obstacle avoidance level. When using an adaptive model, the flight angle of the UAV and the pitch angle of the image acquisition device are adaptively adjusted based on the position of the tower, and the flight speed is adjusted based on the change in the pitch angle of the image acquisition device. When the UAV approaches the tower, the pitch angle of the UAV is increased, and the flight speed is controlled based on the change in angle.
4. The method for UAV hybrid path inspection of power distribution network towers and lines according to claim 3, characterized in that, In the adaptive model-based flight control, the PD algorithm is used to adjust the UAV's heading angle and the image acquisition device's pitch angle based on the deviation of the current tower detection box's position from the center in the image, in order to maintain the detection box's centered position in the image. The calculation expressions for the UAV's yaw angle and the image acquisition device's pitch angle adjustment are as follows: in, Calculate the proportional gain for the unmanned yaw angle. Calculate the differential gain for the unmanned yaw angle. For the yaw angle adjustment of the drone, This refers to the adjustment amount of the pitch angle of the image acquisition device. Calculate the proportional gain for the pitch angle of the image acquisition device. Calculate the differential gain for the pitch angle of the image acquisition device.
5. The method for UAV hybrid path inspection of power distribution network towers and lines according to claim 1, characterized in that, In step S03, adjusting the three-axis speed of the drone to gradually center the position of the tower in the image includes: When the drone's pitch angle reaches 90 degrees, the deviation between the current position of the tower and the center in the image is calculated using the PID control method. Based on this deviation, the drone's three-axis speed is adjusted to gradually center the tower. The expression for calculating the drone's three-axis speed is as follows: in, This is the proportional gain coefficient. This is the integral gain coefficient. The differential gain coefficient, The velocity along the X-axis. This represents the Y-axis velocity.
6. The method for UAV hybrid path inspection of power distribution network towers and lines according to claim 1, characterized in that, In step S03, the inspection image acquisition height is determined according to the following formula. : in, The height data detected by the height detection sensor at time t. As weight, For real-time visual height estimation, The focal length of the image acquisition device. The actual width of the tower. The pixel width of the tower detection frame for each time frame.
7. The method for UAV hybrid path inspection of power distribution network towers and lines according to any one of claims 1 to 6, characterized in that, Step S04 also includes determining the linear velocity of the aircraft flying around the tower according to the following formula: , in, For radius, angular velocity, The set speed for circling the tower, The current angle of the drone. The angle of the drone before it circled the tower.
8. The method for UAV hybrid path inspection of power distribution network towers and lines according to any one of claims 1 to 6, characterized in that, In step S05, the yaw angle of the UAV is adjusted based on the route analysis results: the center point of the detection box for each parallel route segment is calculated. And using the tower head as a reference point According to the reference point Center point of parallel line detection frame Calculate the direction and angle of the connecting lines. According to the angle of the route Adjust the drone's yaw angle until It is 0.
9. A drone-based hybrid path inspection system, comprising a drone equipped with an image acquisition device, and an inspection device connected to the drone, wherein the inspection device includes a processor and a memory, the memory being used to store a computer program, characterized in that, The processor is used to execute the computer program to perform the method as described in any one of claims 1 to 8.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed, it implements the method as described in any one of claims 1 to 8.