Unmanned aerial vehicle control method for power transmission tower line dismantling
By using grid division and similarity analysis when the drone removes the cable, the misjudgment problem caused by tower blocking is solved, which reduces energy consumption and increases the battery life of the drone.
Patent Information
- Application Number
- CN202510787847.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-13
- Publication Date
- 2025-07-25
- Estimated Expiration
- 2045-06-13
AI Technical Summary
When existing drones remove cables, errors occur when determining whether to obtain the second aerial image because the tower body is blocked by vegetation, which increases the energy consumption of the drone.
By setting up grid areas and demarcating molecular grids, controlling the drone's flight altitude to acquire images, using similarity analysis and standard shape replacement mechanisms, we can judge whether the flight altitude needs to be adjusted to avoid misjudgments caused by occlusion.
It reduces the additional flight missions caused by misjudgment by drones, reduces energy consumption, and improves the robustness and battery life of judgment logic.
Smart Images

Figure CN120370981A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) control, and particularly to a UAV control method for removing transmission tower lines. Background Art
[0002] With the continuous breakthrough of UAV technology and the continuous decline in cost, its applications in various fields such as surveying, inspection, and logistics are accelerating. By using UAVs to recover cables, the automation and intelligence of high-altitude operations can be realized, which can greatly improve the operation efficiency and reduce the risk of manual operations.
[0003] Before formally removing the cable, the UAV first collects high-resolution images and depth images of the target area. After the collection is completed, the images are classified pixel by pixel through a pre-trained convolutional neural network (such as a semantic segmentation model based on U-Net or DeepLab), accurately marking the spatial position and orientation of the cable, and fusing the depth image with the segmentation result to reconstruct the three-dimensional curve of the cable in the real scene.
[0004] In a Chinese patent application with the patent publication number CN119295678B, a method for constructing a three-dimensional space model for removing transmission lines is disclosed. In its solution, it mainly includes correcting a three-dimensional model with a second aerial image to obtain a three-dimensional real scene model and obtaining a final three-dimensional model based on the three-dimensional real scene model and power line modeling. However, when determining whether to obtain the second aerial image, since some parts of the tower body may be blocked by vegetation, etc., errors occur when judging whether to obtain the second aerial image, which in turn leads to controlling the UAV to fly to the second height to take the second aerial image, increasing the energy consumption of the UAV. Summary of the Invention
[0005] The purpose of the present invention is to provide a UAV control method for removing transmission tower lines to solve the above technical problems.
[0006] The purpose of the present invention can be achieved by the following technical solutions: A UAV control method for removing transmission tower lines includes the following steps: Set a grid area and divide it into sub-grids; move the grid area so that the projection point of the center point of the power tower pole on the ground is at the center point of any sub-grid, control the flight height of the UAV to be the first height, obtain the image within the grid area, and obtain the target shape corresponding to the sub-grid based on the image; During the surveying process, control the UAV to move forward along the planned route, and the aerial image collected at the center point of the wide side of the grid area is recorded as the target image; If the area of the minimum circumscribed rectangle covering all the power tower pole areas in the target image , then take the minimum bounding rectangle of a single power tower pole in the target image, denoted as the comparison shape; and obtain the projection point of the center point of the power tower pole in the target image on the ground, denoted as the standard point, and denote the target shape corresponding to the sub-grid where the standard point is located as the standard shape; When the similarity between the comparison shape and the corresponding standard shape is greater than a preset value, adjust the flight altitude of the drone to the second altitude to obtain a second aerial image; When the similarity between the standard shape and the comparison shape is less than or equal to the preset value, replace the corresponding comparison shape with the standard shape, and obtain the area S1 of the new minimum bounding rectangle; If , then adjust the flight altitude of the drone to the second altitude to obtain a second aerial image; If S1 ∈ [f, g], then control the drone to survey the next grid area along the current survey route; After the survey is completed, construct a three-dimensional curve of the cable in the real scene according to the survey information for the demolition of the transmission line.
[0007] As a further solution of the present invention: obtaining the target shape includes: Set a grid area, divide the grid area into several sub-grids, the length and width of the sub-grids are both a preset first length; determine the projection point A of the center point of the power tower pole on the ground, move the grid area so that the projection point A is at the center point of the sub-grid i, keep the flight altitude of the drone at the first altitude, collect the image within the grid area at the center point of the wide side of the grid area, and mark the minimum bounding rectangle of the power tower pole in the image, denoted as the target shape.
[0008] As a further solution of the present invention: obtaining the target shape further includes: Start from the starting point of the preset second altitude range, select several altitudes at a preset altitude interval, denoted as target altitudes, and obtain the target shape when the flight altitude of the drone is any one of the target altitudes.
[0009] As a further solution of the present invention: obtaining the similarity includes: Obtain the area difference C1 between the comparison shape and the corresponding standard shape, and obtain the area C2 of the overlapping part between the comparison shape and the corresponding standard shape, and calculate the similarity P = C2 / C1.
[0010] As a further solution of the present invention: if the area difference C1 is less than the preset area difference threshold, then the similarity P is equal to the value of C2.
[0011] As a further solution of the present invention: after obtaining the second aerial image, keep the flight altitude of the drone at the second altitude, and after obtaining a new aerial image, determine whether to adjust the flight altitude of the drone according to the target shape of the target altitude that is the same as the second altitude.
[0012] As a further solution of the present invention: during the surveying and mapping process, after the target area is divided into grids, the aspect ratio of the grid area is 2, and the width is a preset second length.
[0013] Beneficial effects of the present invention: Compared with the prior art: 1) The present invention establishes a reference shape database of power tower poles at different heights through sub-grid division and standard shape pre-acquisition mechanism. When judging whether to trigger secondary surveying and mapping, the similarity analysis (area difference and coincidence degree) between the comparison shape and the standard shape is introduced. When the actual measured area is abnormal due to the tower body being blocked by vegetation, if the similarity is lower than the preset value, the system can identify it as "occlusion interference", and replace the actual measured shape with the unobstructed standard shape to recalculate the area, avoiding misjudging as "insufficient resolution" due to local occlusion and starting secondary surveying and mapping. This mechanism effectively filters out false anomalies caused by occlusion, reduces the additional flight tasks of the drone due to misjudgment, and improves the robustness of the judgment logic; 2) The present invention, through the calibration mechanism before surveying and mapping, makes a small translation adjustment to the grid area, so that the projection points of the same power tower pole fall on the centers of different sub-grids in sequence, and the minimum circumscribed rectangles are extracted at each position to construct a multi-group "target shape" sample library. During actual measurement, the system matches the comparison shape obtained in real time with the sample library, which can effectively compensate for the contour changes caused by factors such as the actual position offset of the tower pole and the distortion of the shooting angle. Even if the contour of the tower body is partially blocked by vegetation or affected by light and shadow and is incomplete, the system can accurately evaluate whether the current imaging quality meets the requirements through the feature matching of the sample library, so as to correctly judge whether it is necessary to adjust the flight altitude for supplementary shooting, significantly reducing the misjudgment probability caused by occlusion or perspective problems, avoiding the drone from flying repeatedly due to invalid commands, and reducing unnecessary energy consumption; 3) The present invention adjusts the flight altitude of the drone using the linear correction formula in the background technology, and sets the switched second altitude as the new reference altitude. During the subsequent surveying and mapping process, only when significant deviations occur in consecutive multiple grid areas, will the altitude be adjusted again. This mechanism abandons the traditional frequent adjustment mode of "independent lifting and lowering for each grid", avoids the push-brake cycle generated by the frequent altitude switching of the drone, and reduces the loss of the battery and the motor. By maintaining a relatively stable flight state, the drone can cover most of the surveying and mapping area with a more efficient path, effectively extending the endurance time while ensuring the modeling accuracy, and improving the overall operation fluency and economy. Description of the Drawings
[0014] The present invention will be further described below with reference to the accompanying drawings.
[0015] Figure 1 It is a schematic flowchart of a method for controlling an unmanned aerial vehicle for removing transmission tower lines according to the present invention. Specific embodiments
[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0017] Please refer to Figure 1 As shown, the present invention is a method for controlling an unmanned aerial vehicle for removing transmission tower lines, including the following steps: Set a grid area and divide it into sub - grids; move the grid area so that the projection point of the center point of the power tower pole on the ground is at the center point of any one sub - grid, control the flight altitude of the unmanned aerial vehicle to be the first altitude, obtain the image within the grid area, and obtain the target shape corresponding to the sub - grid based on the image. In a preferred embodiment of the present invention, obtaining the target shape includes: Set a grid area, divide the grid area into several sub - grids, the length and width of the sub - grids are both a preset first length; determine the projection point A of the center point of the power tower pole on the ground, move the grid area so that the projection point A is at the center point of sub - grid i, keep the flight altitude of the unmanned aerial vehicle at the first altitude, collect the image within the grid area at the center point of the wide side of the grid area, and mark the minimum circumscribed rectangle of the power tower pole in the image, denoted as the target shape. It should be noted that first, set a grid area and divide it into several sub - grids. For example, the length and width of the sub - grids are both set to 5 meters (the preset first length) to form a regular grid matrix; then determine the projection point A of the center point of the power tower pole on the ground, and translate the grid area through manual operation or program control so that the projection point A is accurately aligned with the center point of sub - grid i. When sub - grid i is the first sub - grid in the upper left corner, adjust the grid position so that point A coincides with the cross - center point of the sub - grid; then control the unmanned aerial vehicle to fly to the center point of the wide side of the grid area at the first altitude (such as 50 meters, the flight altitude under the lowest target accuracy), fly stably along the planned route and collect the aerial image within the grid area; in the obtained image, use an image recognition algorithm to automatically detect the outline of the power tower pole, and mark the minimum circumscribed rectangle that wraps the main body of the tower pole. For example, for a tower pole with a rectangular cross - section, the algorithm will identify the four edge vertices and generate a rectangular frame that closely adheres to the outline, and this rectangle is the target shape corresponding to sub - grid i. By aligning the projection point of the tower pole with the center point of the sub-grid, it can ensure that the tower pole is in a fixed relative position in each captured image, facilitating the subsequent establishment of a unified shape reference; the regular division of the sub-grid (such as a side length of 5 meters) can subdivide the grid area into quantifiable units, establishing a one-to-one correspondence between the position of the tower pole and the sub-grid, and avoiding shape analysis errors caused by position deviations; capturing images at a fixed height (such as 50 meters) can ensure the consistency of resolution during initial surveying and mapping, facilitating the comparison of image features at different heights subsequently; the process of annotating the minimum bounding rectangle (such as identifying the edge vertices of the tower pole) is to convert the visual features of the tower pole into computable geometric parameters, providing basic data for subsequent similarity judgment and area calculation. This series of operations jointly construct a standardized "position-shape" mapping system, enabling the system to establish a reliable target shape sample library based on precise alignment and feature extraction at the sub-grid level, so as to effectively identify interference factors such as occlusion and perspective distortion by comparing the shape with the sample library during actual measurement, providing a basis for accurately judging whether secondary surveying and mapping are required, and ultimately improving the energy efficiency of UAV operations; In a preferred case of this embodiment, obtaining the target shape further includes: Starting from the starting point of the preset second height range, select a number of heights at a preset height interval, denoted as target heights, and obtain the target shape when the flight height of the UAV is any one of the target heights; It should be noted that after obtaining the second aerial image, keep the flight height of the UAV at the second height, and after obtaining a new aerial image, judge whether to adjust the flight height of the UAV according to the target shape at the target height that is the same as the second height; In the specific implementation process, starting from the starting point of the preset second height range (such as 30 meters), several heights are sequentially selected as target heights (such as 30 meters, 35 meters, 40 meters, etc.) at a preset height interval (such as 5 meters). At each target height, the process of "moving the grid area to align the tower pole projection point with the center of the sub-grid → collecting images → annotating the minimum bounding rectangle" is repeated to obtain the target shape corresponding to the height. This is done to establish multiple sets of "height-shape" feature pairs at different flight heights, forming a target shape sample library covering multiple resolutions; when it is necessary to adjust to the second height during actual measurement (such as height correction triggered by abnormal area), the system can directly call the target shape pre-collected at this height as the new benchmark without having to re-collect temporarily, avoiding the secondary judgment delay caused by the lack of benchmark data after height switching; at the same time, after setting the new height as the benchmark, the subsequent surveying and mapping of continuous grids can be directly based on the target shape at this height for comparison, and only when significant deviations continuously occur will it be adjusted again, reducing the ineffective operation of "gradually probing up and down grid by grid"; by pre-constructing a multi-height feature library, an immediately available comparison benchmark is provided for dynamic height adjustment, enabling the UAV to quickly enter a stable surveying and mapping state after switching to the second height, avoiding frequent start-stop and height oscillation, and improving the coherence and energy efficiency of UAV operations; During the surveying and mapping process, the UAV is controlled to move forward along the planned route, and the aerial image collected at the center point of the wide side of the grid area is recorded as the target image; It should be noted that during the surveying and mapping process, after the target area is divided into grids, the aspect ratio of the grid area is 2, and the width is the preset second length; During the surveying and mapping process, the UAV moves forward at a constant speed according to the preset planned route (such as taking the line connecting the center points of the two wide sides of the grid area as the flight path). When it flies to the center point of the wide side of the grid area (such as the midpoint coordinates of the left wide side are (X1, Y1)), an image collection instruction is triggered, and the aerial image at this position is taken and recorded as the target image. For example, in hilly terrain, the UAV takes a panoramic image of the grid area containing multiple power tower poles in a stable posture; If the area of the minimum bounding rectangle covering all the power tower pole areas in the target image , then take the minimum bounding rectangle of a single power tower pole in the target image and record it as the comparison shape; and obtain the projection point of the center point of the power tower pole in the target image on the ground and record it as the standard point, and record the target shape corresponding to the sub-grid where the standard point is located as the standard shape; If the area S of the minimum circumscribed rectangle covering all the power tower pole areas in the target image calculated by the image analysis algorithm exceeds the preset interval [f, g] (e.g., S is less than f or greater than g) (the interval [f, g] is the preset area interval), then the image is further subdivided: identify the contours of each independent power tower pole, extract the minimum circumscribed rectangle of a single pole (e.g., for the first pole from the left in the image, generate its independent rectangular frame), denoted as the comparison shape; at the same time, through the conversion of geolocation data and the image coordinate system, determine the actual projection point coordinates of the center point of the pole on the ground (e.g., the longitude and latitude coordinates (Lon, Lat)), define it as the standard point, and match the corresponding sub-grid according to the coordinates of the standard point (e.g., the coordinates fall within the range of sub-grid i), and retrieve the target shape (i.e., the minimum circumscribed rectangle of the pole corresponding to the sub-grid at the first height) marked in the pre-acquisition stage of the sub-grid as the standard shape; When the similarity between the comparison shape and the corresponding standard shape is greater than the preset value, adjust the flight height of the drone to the second height to obtain the second aerial image; Another preferred embodiment of the present invention, obtaining the similarity includes: Obtain the area difference C1 between the comparison shape and the corresponding standard shape, and obtain the area C2 of the overlapping part between the comparison shape and the corresponding standard shape, and calculate the similarity P = C2 / C1; It should be noted that if the area difference C1 is less than the preset area difference threshold, the similarity P is equal to the value of C2; By calculating the area difference C1 between the comparison shape and the standard shape (i.e., |comparison shape area - standard shape area|) and the overlapping area C2 (i.e., the area of the overlapping region between the two), and using P = C2 / C1 as the quantitative index of similarity (if C1 = 0, then P directly takes C2, and at this time the two areas are the same, and the coincidence degree is the similarity), the shape difference can be converted into a computable numerical relationship. The advantage of this is that the area difference C1 reflects the absolute difference amplitude between the two, while the overlapping area C2 reflects the contour matching degree, and the ratio P of the two can comprehensively evaluate the similarity between the comparison shape and the standard shape. For example, if a comparison shape has a reduced area due to occlusion but a high coincidence degree of the contour main body with the standard shape (C2 is large, C1 is small, and the P value is high), it indicates that there may be problems with the actual size or resolution, and the height needs to be adjusted for supplementary shooting; if the area difference is large and the coincidence degree is low (C2 is small, C1 is large, and the P value is low), then it is probably occlusion interference, and it can be compensated by the standard shape. This quantitative judgment mechanism provides an objective decision-making basis for the system, avoids the ambiguity of subjective threshold judgment, enables the drone to accurately identify "true anomalies" and "occlusion misjudgments" based on data, and thus only starts secondary mapping when necessary, reducing unnecessary height adjustments and flight energy consumption; When the similarity between the standard shape and the comparison shape is less than or equal to a preset value, replace the corresponding comparison shape with the standard shape, and obtain the new minimum circumscribed rectangle area S1; If , adjust the flight altitude of the drone to the second altitude to obtain a second aerial image; If S1 ∈ [f, g], control the drone to survey the next grid area along the current survey route; It can be understood that when the similarity P (calculated by P = C2 / C1, for example, P = 0.8) between the comparison shape (such as the minimum circumscribed rectangle measured for a certain power tower due to partial occlusion) and the corresponding standard shape (the rectangle in the pre-acquired unoccluded state) is greater than the preset value (such as 0.7), it indicates that the difference in their contours is small, and the area anomaly is more likely caused by insufficient resolution at the current flight altitude. At this time, the system triggers an instruction to adjust the flight altitude of the drone to the second altitude (such as from 50 meters to 30 meters, and the specific value uses the calculation method of the file cited in the background art), and fly along the original route at the new altitude to collect a second aerial image. For example, in a sparse vegetation area, a clearer overall image of the tower can be obtained after the altitude is reduced. When the similarity P is less than or equal to the preset value (such as P = 0.4), it means that the comparison shape is significantly different from the standard shape due to factors such as occlusion. The system automatically replaces the measured comparison shape with the standard shape (such as the pre-acquired complete rectangle), and recalculates the new minimum circumscribed rectangle area S1 covering all power tower areas. For example, after correcting the area of the occluded tower to the area of the standard shape, recombine it with the measured shapes of other towers to calculate the global area. If the corrected S1 still exceeds the preset interval [f, g] (such as S1 > g), it is determined that there is a real area anomaly, and the altitude is adjusted to the second altitude for supplementary shooting; if S1 falls within the interval (such as f ≤ S1 ≤ g), it is considered that the image at the current altitude already meets the accuracy requirements, and the drone is controlled to continue the next grid survey; Distinguish the essential reasons for area anomalies through similarity judgment: directly adjust the altitude for high similarity cases (P > preset value) to meet the resolution requirements; for low similarity cases (P ≤ preset value), first use the standard shape to compensate for occlusion interference, and then evaluate the global area to avoid misjudging as a resolution problem due to local occlusion. For example, when a certain tower is occluded by trees resulting in a smaller measured area and a low coincidence with the standard shape, the calculation deviation caused by occlusion can be eliminated through the standard shape substitution, avoiding unnecessary descent and supplementary shooting of the drone. This hierarchical processing mechanism not only ensures an effective response to real resolution deficiencies but also filters out false anomalies caused by occlusion, enabling the drone to only adjust its altitude when necessary, reducing the number of ineffective flights and energy consumption; After the survey is completed, construct a three-dimensional curve of the cable in the real scene based on the survey information for the demolition of the transmission line.
[0018] The above has described in detail an embodiment of the present invention, but the above content is only a preferred embodiment of the present invention and cannot be considered as defining the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.
Claims
1. A method for controlling an unmanned aerial vehicle for removing transmission tower lines, characterized in that, It includes the following steps: Set a grid area and divide it into sub - grids; move the grid area so that the projection point of the center point of the power tower pole on the ground is at the center point of any sub - grid, control the flight altitude of the UAV to be the first altitude, obtain the image within the grid area, and obtain the target shape corresponding to the sub - grid based on the image. During the surveying and mapping process, control the UAV to move forward along the planned route, and record the aerial image collected at the center point of the wide side of the grid area as the target image. If the area of the minimum circumscribed rectangle covering all the power pole areas in the target image , then take the minimum circumscribed rectangle of a single power pole in the target image, denoted as the comparison shape; and obtain the projection point of the center point of the power pole in the target image on the ground, denoted as the standard point, and record the target shape corresponding to the sub-grid where the standard point is located as the standard shape; When the similarity degree between the comparison shape and the corresponding standard shape is greater than the preset value, adjust the flight altitude of the UAV to the second altitude to obtain the second aerial image. When the similarity degree between the standard shape and the comparison shape is less than or equal to the preset value, replace the corresponding comparison shape with the standard shape, and obtain the new minimum circumscribed rectangle area S1. If , adjust the flight altitude of the drone to the second altitude to obtain a second aerial image; If S1 ∈ [f, g], then control the UAV to survey the next grid area along the current surveying and mapping route. After the surveying and mapping is completed, construct the three - dimensional curve of the cable in the real scene according to the surveying and mapping information for the demolition of the transmission line.
2. The drone control method for transmission tower line demolition according to claim 1, wherein, Obtaining the target shape includes: Set a grid area, divide the grid area into several sub - grids, and the length and width of the sub - grid are both the preset first length; determine the projection point A of the center point of the power tower pole on the ground, move the grid area so that the projection point A is at the center point of sub - grid i, keep the flight altitude of the UAV at the first altitude, collect the image within the grid area at the center point of the wide side of the grid area, and mark the minimum circumscribed rectangle of the power tower pole in the image, denoted as the target shape.
3. The drone control method for transmission tower line demolition according to claim 2, wherein, Obtaining the target shape also includes: Start from the starting point of the preset second altitude range, select several altitudes at the preset altitude interval, denoted as the target altitudes, and obtain the target shape when the flight altitude of the UAV is any target altitude.
4. A method for controlling an unmanned aerial vehicle for removing transmission tower lines according to claim 1, characterized in that, Obtaining the similarity degree includes: Obtain the area difference C1 between the comparison shape and the corresponding standard shape, and obtain the area C2 of the overlapping part between the comparison shape and the corresponding standard shape, and calculate the similarity degree P = C2 / C1.
5. A method for controlling an unmanned aerial vehicle for removing transmission tower lines according to claim 4, characterized in that, If the area difference C1 is less than the preset area difference threshold, then the similarity degree P is equal to the value of C2.
6. The drone control method for transmission tower line demolition according to claim 3, characterized in that, After obtaining the second aerial image, keep the flight altitude of the UAV at the second altitude, and after obtaining the new aerial image, judge whether to adjust the flight altitude of the UAV according to the target shape at the target altitude that is the same as the second altitude.
7. A method for controlling an unmanned aerial vehicle for removing transmission tower lines according to claim 1, characterized in that, During the surveying and mapping process, after the target area is divided into grids, the aspect ratio of the grid area is 2, and the width is the preset second length.
Citation Information
Patent Citations
Method for identifying insect veins based on mesh region and system thereof
CN101937512A
Power transmission line channel digital surveying method
CN104933223A
Three-dimensional surveying and mapping system and method based on unmanned aerial vehicle
CN115792955A
Power line inspection control method and system based on unmanned aerial vehicle
CN116719339A
Surveying and mapping method for natural resource research
CN118031912A
Cited By
Distribution network unmanned aerial vehicle ground simulation flight method and equipment based on AI vision and medium
CN121008592A
