Unmanned aerial vehicle control parameter determination method, device, equipment, medium and product

Abstract

The invention provides an unmanned aerial vehicle control parameter determination method and device, equipment, a medium and a product. The method comprises the steps that after a target power line bounding box is determined according to an initial power line image, the relative height between the unmanned aerial vehicle and a power line is determined according to a preset distance, the camera focal length corresponding to the initial power line image and the width of the target power line bounding box. And when the rotation angle of the target power line bounding box is greater than a preset angle, determining a displacement error and a yaw angle error at the current moment according to the relative height between the unmanned aerial vehicle and the power line, the rotation angle of the target power line bounding box, the displacement error at the previous moment and the yaw angle error at the previous moment. And finally, determining control parameters of the unmanned aerial vehicle according to the displacement error and the yaw angle error at the current moment. The problem of low navigation accuracy of the unmanned aerial vehicle is solved.

Description

Technical Field

[0001] The present application relates to the field of power inspection technology, and in particular to a method, device, equipment, medium and product for determining control parameters of a drone. Background Art

[0002] Overhead transmission lines are crucial components of power systems, and their safe and stable operation is crucial for ensuring the transmission of electricity. Therefore, regular inspections are necessary to check whether these lines are affected by factors such as climatic conditions, surrounding animal and human activities, and other factors.

[0003] In the existing technology, the operation mode based on drone inspection of overhead transmission lines is widely used. The drone takes inspection images in real time and transmits them back to the ground station, so that the ground station sends back control instructions. The control instructions can enable the drone to move along the power lines between towers.

[0004] However, in the prior art, when a drone encounters a tower and needs to turn, there is a problem of low drone navigation accuracy. Summary of the Invention

[0005] The embodiments of the present application provide a method, apparatus, device, medium, and product for determining control parameters of a drone, to solve the problem of low drone navigation accuracy.

[0006] In a first aspect, an embodiment of the present application provides a method for determining control parameters of a drone, comprising:

[0007] Determine a target power line bounding box according to the initial power line image, where the power lines contained in the target power line bounding box are straight lines;

[0008] Determine the relative height between the drone and the power line based on the preset distance, the camera focal length corresponding to the initial power line image, and the width of the target power line bounding box;

[0009] When the rotation angle of the target power line bounding box is greater than a preset angle, the displacement error and yaw angle error at the current moment are determined based on the relative height between the UAV and the power line, the rotation angle of the target power line bounding box, the displacement error at the previous moment, and the yaw angle error at the previous moment. The rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image.

[0010] Determine the UAV control parameters based on the displacement error and yaw angle error at the current moment.

[0011] In one possible implementation, the ratio between the preset distance and the width of the target power line bounding box is calculated; and the product of the ratio and the camera focal length is determined as the relative height between the drone and the power line.

[0012] In one possible implementation, the formula Calculate the displacement error and yaw angle error at the current moment;

[0013] Among them, Δx is the displacement error component in the x-axis direction of the drone coordinate system, Δy is the displacement error component in the y-axis direction of the drone coordinate system, Δz is the displacement error component in the z-axis direction of the drone coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, θ is the rotation angle and h r is the relative height between the drone and the power line.

[0014] In one possible implementation, the initial power line image is input into a power line recognition model, and multiple initial power line bounding boxes output by the power line recognition model are obtained. The power lines contained in the initial power line bounding boxes are straight lines. The power line recognition model is pre-trained based on multiple sample power line images and sample power line bounding boxes corresponding to each sample power line image; the initial power line bounding box with the largest width is determined as the target power line bounding box.

[0015] In one possible implementation, the formula Determine the drone control parameters;

[0016] Among them, K px , K py , K pz and K pθ Both are used to indicate the preset proportional coefficient, h set is the preset relative height between the UAV and the power line, v is the target speed, and ω is the target angular velocity. The target speed and yaw angle are UAV control parameters.

[0017] In one possible implementation, the deviation displacement of the UAV relative to the power line is determined based on a preset distance, the horizontal coordinates of the center point of the target power line bounding box, the coordinates of the center point of the initial power line image, and the width of the target power line bounding box; the deviation angle of the UAV relative to the power line is determined based on the endpoint coordinates of the power line contained in the target power line bounding box; the displacement error and yaw angle error at the current moment are determined based on the relative height between the UAV and the power line, the deviation displacement of the UAV relative to the power line, the deviation angle of the UAV relative to the power line, the displacement error at the previous moment, and the yaw angle error at the previous moment; and the UAV control parameters are determined based on the displacement error and yaw angle error at the current moment.

[0018] In one possible implementation, according to the formula as well as Determine the deviation displacement of the UAV relative to the power line;

[0019] Among them, dy is the component of the deviation displacement in the y-axis direction in the UAV coordinate system, dx is the component of the deviation displacement in the x-axis direction in the UAV coordinate system, and c i is the center point coordinate of the initial power line image, cx is the horizontal coordinate of the center point of the target power line bounding box, d r is the preset distance, w is the width of the target power line bounding box, and h is the actual length corresponding to each pixel in the initial power line image.

[0020] In one possible implementation, according to the formula Determine the displacement error and yaw angle error at the current moment;

[0021] Among them, Δx is the displacement error component of the x-axis direction in the UAV coordinate system, Δy is the displacement error component of the y-axis direction in the UAV coordinate system, Δz is the displacement error component of the z-axis direction in the UAV coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, φ is the deviation angle and h r is the relative height between the drone and the power line.

[0022] In a second aspect, an embodiment of the present application provides a device for determining control parameters of a drone, comprising: a first determination module, configured to determine a target power line bounding box based on an initial power line image, wherein the power lines contained in the target power line bounding box are straight lines;

[0023] a second determination module, configured to determine a relative height between the UAV and the power line based on a preset distance, a camera focal length corresponding to the initial power line image, and a width of a target power line bounding box;

[0024] a third determining module, configured to determine, when the rotation angle of the target power line bounding box is greater than a preset angle, a displacement error and a yaw angle error at a current moment based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at a previous moment, and the yaw angle error at a previous moment, wherein the rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image;

[0025] The fourth determination module is used to determine the control parameters of the UAV according to the displacement error and yaw angle error at the current moment.

[0026] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor;

[0027] Memory stores computer-executable instructions;

[0028] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0029] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementation methods of the first aspect.

[0030] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0031] The embodiments of the present application provide a method, device, equipment, medium and product for determining the control parameters of a drone. First, a target power line bounding box that marks the power line is determined based on an initial power line image. Then, the relative height between the drone and the power line is determined based on a preset distance, the camera focal length corresponding to the initial power line image, and the width of the target power line bounding box, so that the distance between the drone and the power line can be evaluated based on the height. Then, when the rotation angle of the target power line bounding box is greater than the preset angle, it indicates that the drone is flying to the tower and needs to turn. Therefore, the displacement error and yaw angle error at the current moment are determined based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at the previous moment, and the yaw angle error at the previous moment. Finally, the drone control parameters can be calculated based on the displacement error and yaw angle error, and the drone can be controlled, solving the problem of low drone navigation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0033] Figure 1 Schematic diagram of the drone inspection scenario provided in the embodiment of this application Figure 1 ;

[0034] Figure 2 Schematic diagram of the drone inspection scenario provided in the embodiment of this application Figure 2 ;

[0035] Figure 3 A flow chart of a method for determining control parameters of a drone provided in an embodiment of the present application Figure 1 ;

[0036] Figure 4A flow chart of a method for determining control parameters of a drone provided in an embodiment of the present application Figure 2 ;

[0037] Figure 5 A drone control parameter determination system provided in an embodiment of the present application;

[0038] Figure 6 A schematic diagram of a transmission line directional detection network provided in an embodiment of the present application;

[0039] Figure 7 Schematic diagram of the target power line boundary box provided in the embodiment of the present application Figure 1 ;

[0040] Figure 8 Schematic diagram of the target power line boundary box provided in the embodiment of the present application Figure 2 ;

[0041] Figure 9 A schematic diagram of the process of extracting power lines from a target power line boundary frame provided in an embodiment of the present application;

[0042] Figure 10 Schematic diagram of the results of a method for determining drone control parameters provided in an embodiment of the present application Figure 1 ;

[0043] Figure 11 Schematic diagram of the results of a method for determining drone control parameters provided in an embodiment of the present application Figure 2 ;

[0044] Figure 12 A schematic diagram of the structure of a device for determining control parameters of a drone provided in an embodiment of the present application;

[0045] Figure 13 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0046] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0047] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0048] First, let’s explain the terms involved in this application:

[0049] Tension-resistant linear towers: These are used to support power lines and can provide the necessary mechanical tension in the straight section to ensure the stability and safety of the conductors. They are usually installed in the straight section of the power line.

[0050] Tension-resistant corner pole tower: refers to a special pole tower used to support power lines and handle line turning. It can not only withstand the tension of the conductors, but also adapt to changes in line direction to ensure the safety and stability of current transmission.

[0051] Figure 1 Schematic diagram of the drone inspection scenario provided in the embodiment of this application Figure 1 ,like Figure 1 As shown, the drone is patrolling above a field where power transmission lines are erected. After the drone inspects the power lines, it needs to change its direction when it encounters a tower so that it can continue inspecting the next section of the power lines.

[0052] Figure 2 Schematic diagram of the drone inspection scenario provided in the embodiment of this application Figure 2 ,like Figure 2 As shown in the figure, there are five towers: ①, ③, and ⑤ are straight tension towers, and ② and ④ are angle tension towers. The spacing between each tower is 100 meters, for a total length of approximately 400 meters. After the drone completes its inspection between towers ① and ②, it needs to change its direction and rotate toward tower ③, continuing to inspect the power lines between towers ② and ③. This will ultimately inspect the power line area between towers ① and ⑤.

[0053] Combining the above scenarios, it can be seen that in existing technologies, when using drones to inspect overhead transmission lines, the navigation required for the drone inspection not only needs to consider the inspection route between towers, but also the inspection route that crosses a tower and turns to the next tower. However, existing technologies only use drones to capture inspection images in real time and transmit them back to a ground station, which then calculates control parameters and returns control commands. These control commands are designed for drones flying along power lines and cannot handle situations where the drone needs to turn after reaching a tower. Therefore, existing technologies suffer from low drone navigation accuracy.

[0054] In this regard, the inventor believes that in the case where a turn is required when inspecting a pole tower, this can be determined by marking the bounding box of the power line in the power line image. If the angle between the bottom edge of the bounding box and the bottom edge of the power line image is too large, it means that the direction of the power line facing the drone has changed. Therefore, the corresponding drone may have encountered a pole tower and needs to turn to the next pole tower so that it can continue to inspect along the power line. The direction of the drone can be corrected by using the angle between the bottom edge of the bounding box and the bottom edge of the power line image, so that the direction of the drone is the same as the direction of the power line. At the same time, when the drone turns, the relative height between the drone and the power line must also be considered to prevent the drone from contacting the power line. In this way, the displacement error and yaw angle error that the drone needs to move are calculated, and then the drone control parameters are calculated, so that the drone turns to the next pole tower and continues to inspect along the power line, solving the problem of low drone navigation accuracy.

[0055] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0056] Figure 3 A flow chart of a method for determining control parameters of a drone provided in an embodiment of the present application Figure 1 ,like Figure 3 As shown, the method includes:

[0057] S301 : Determine a target power line bounding box according to an initial power line image.

[0058] The initial power line image refers to an image containing power lines and towers captured by an image acquisition device installed on the drone.

[0059] Exemplarily, the image acquisition device may be a camera, a still camera, etc.

[0060] The target power line bounding box is the rectangular area in the initial power line image where the power lines are calibrated. The power lines contained in the target power line bounding box are straight lines, and the direction of the target power line bounding box is consistent with the direction of the actual power lines.

[0061] It should be understood that when the direction of the actual power lines in the image changes, the direction of the target power line bounding box will also change accordingly, and the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image will also change accordingly.

[0062] Therefore, in practical applications, after the target power line bounding box is determined, the bounding box parameters such as the center point coordinates (cx, cy), height h, width w and rotation angle θ of the bounding box can be determined. The rotation angle is the angle between the bottom edge of the bounding box and the bottom edge of the initial power line image. The specific method can be Figure 7 and Figure 8 The embodiments shown are used for illustration only and will not be described in detail here.

[0063] In one possible implementation, the initial power line image is input into a power line recognition model, and multiple initial power line bounding boxes output by the power line recognition model are obtained, where the power lines contained in the initial power line bounding boxes are straight lines; the initial power line bounding box with the largest width is determined as the target power line bounding box.

[0064] Among them, the power line recognition model refers to a model that can delineate the area where the power lines are located based on the power line image. The power line recognition model is obtained by pre-training the model based on multiple sample power line images and the sample power line bounding boxes corresponding to each sample power line image.

[0065] In practical applications, the power line recognition model may identify multiple initial power line bounding boxes for an initial power line image, indicating that there are multiple power line areas in the image. In this case, the initial power line bounding box with the largest height is selected as the target power line bounding box. This is because a larger bounding box means that the power line it encircles occupies a larger viewing angle in the image, indicating that the drone is closer to this section of power line. At this time, it is necessary to give priority to the situation with a closer distance to ensure the safe flight of the drone. At the same time, the power lines under a larger viewing angle can also be used to reflect more inspection details.

[0066] S302: Determine the relative height between the UAV and the power line according to the preset distance, the camera focal length corresponding to the initial power line image, and the width of the target power line boundary box.

[0067] The preset distance refers to the distance between the leftmost power line and the rightmost power line.

[0068] The camera focal length corresponding to the initial power line image refers to the focal length of the camera used when capturing the initial power line image. Because the camera focal length affects the magnification and viewing angle of objects in the image, it can be used to calculate the distance and size of objects in the image.

[0069] The relative height between the drone and the power line refers to the height difference between the drone and the power line segment being inspected. This height difference changes dynamically as the inspection location moves and needs to be adjusted according to different inspection requirements.

[0070] In one possible implementation, the ratio between the preset distance and the width of the target power line bounding box is calculated; and the product of the ratio and the camera focal length is determined as the relative height between the drone and the power line.

[0071] In practical applications, the relative height between the drone and the power line can be calculated by the formula: Calculated, where h r is the relative height between the UAV and the power line, d r is the preset distance, d p is the pixel distance between the left and right ends of the power line, and f is the focal length of the camera. Since the drone is set at the center above the power line, d p Determined by the width w of the target power line bounding box, the relative height between the drone and the power line is determined by the formula: Calculated.

[0072] S303. When the rotation angle of the target power line bounding box is greater than a preset angle, determine the displacement error and yaw angle error at the current moment based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at the previous moment, and the yaw angle error at the previous moment.

[0073] The preset angle refers to a preset rotation angle threshold value, which is used to distinguish between the situation where the drone is turning or flying in a straight line.

[0074] In actual applications, when the rotation angle is greater than the preset angle, it means that the drone has flown to the tower. At this time, the direction of the power line changes, so the direction of the drone must also change accordingly. At this time, the rotation angle is used as the prior input of the drone's yaw angle.

[0075] The rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image.

[0076] Displacement error refers to the difference between the actual displacement of the drone and its target displacement. Because the drone needs to follow the line and turn, the displacement error needs to be determined in order to control the drone.

[0077] Yaw error refers to the angular difference between the current heading of the drone and the target direction. The current heading of the drone can be understood as the direction its nose is pointing. Similarly, when the drone turns, the direction of the drone needs to be changed according to the yaw error.

[0078] In practical applications, the formula Calculate the displacement error and yaw angle error at the current moment;

[0079] Among them, Δx is the displacement error component in the x-axis direction of the drone coordinate system, Δy is the displacement error component in the y-axis direction of the drone coordinate system, Δz is the displacement error component in the z-axis direction of the drone coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, θ is the rotation angle and h r is the relative height between the drone and the power line.

[0080] That is, Δx t , Δy t and Δz t are the displacement errors at the current moment, Δx t is the displacement error component of the current moment in the x-axis direction in the drone coordinate system, Δy t is the displacement error component of the current moment in the y-axis direction of the UAV coordinate system, Δz t is the displacement error component of the current moment in the z-axis direction in the drone coordinate system; Δx t-1 , Δy t-1 and Δz t-1 are the displacement errors of the previous moment, Δx t-1 is the displacement error component of the previous moment in the x-axis direction in the drone coordinate system, Δy t-1 is the displacement error component of the displacement error at the previous moment in the y-axis direction in the drone coordinate system, Δz t-1 is the displacement error component of the previous moment in the z-axis direction in the UAV coordinate system.

[0081] It should be understood that when the rotation angle of the target power line boundary box is greater than the preset angle, it means that the yaw angle deviation of the drone is large at this time, so it is necessary to correct the inspection direction and temporarily stop the position adjustment in the x and y directions.

[0082] S304: Determine the control parameters of the UAV according to the displacement error and yaw angle error at the current moment.

[0083] Among them, drone control parameters refer to the input quantities for controlling the flight of the drone, including position control parameters, such as the horizontal position and the vertical position, and attitude control parameters, such as the yaw angle.

[0084] In one possible implementation, the formula Determine the drone control parameters;

[0085] Among them, K px , K py , K pz and K pθ Both are used to indicate the preset proportional coefficient, h setis the preset relative height between the UAV and the power line, v is the target speed, and ω is the target angular velocity. The target speed and yaw angle are UAV control parameters.

[0086] Among them, v x is the speed in the x-axis direction in the UAV coordinate system; v y is the velocity in the y-axis direction in the UAV coordinate system; v z is the speed in the z-axis direction in the UAV coordinate system; ω z It is the angular velocity around the z-axis in the drone coordinate system, thereby controlling the yaw angle.

[0087] It should be understood that when controlling the drone according to the drone control parameters, it is determined in real time whether the rotation angle determined during the flight of the drone is within the threshold. If the angle is adjusted to within the threshold, the rotation angle is adjusted according to the threshold. Figure 4 The method of the embodiment is controlled.

[0088] An embodiment of the present application provides a method for determining drone control parameters. First, a target bounding box containing straight power lines is determined through an initial power line image, thereby delineating the power line area in the captured image. Then, the relative height between the drone and the power line is calculated using a preset distance, camera focal length, and target bounding box length. When the rotation angle of the target power line bounding box is greater than a preset angle, the error value at the current moment can be updated by combining the relative height, rotation angle, and displacement error and yaw angle error at the previous moment. This allows the drone control parameters to be adjusted based on these errors to optimize the flight path and direction, thereby coping with situations where the drone needs to turn when facing a pole tower, thereby improving the accuracy of drone navigation.

[0089] In practical applications, the PID strategy is used to dynamically adjust the speed and angular velocity of the drone. Here, the MAVLink protocol is implemented in ROS (MAVLink on ROS, MAVROS) function package to establish communication between the Robot Operating System (ROS) and the Micro Air Vehicle Link (MAVLink) protocol. In the message field of the / mavros / setpoint_raw / local topic, the coordinate system frame is set to the body coordinate system and type_mask is set to 1991. x 、v y 、v z 、ω z Four input quantities are used to control the movement of the drone, and the specific error is Δx t , Δy t , Δz t , Δψ tThe exponential moving weighted average method is used here. Its core idea is to perform weighted averaging on historical data with exponentially decreasing weights, so that recent data has a greater impact on the average value, while the impact of long-term data gradually weakens.

[0090] Figure 4 A flow chart of a method for determining control parameters of a drone provided in an embodiment of the present application Figure 2 ,like Figure 4 As shown, this embodiment Figure 3 Based on the embodiment, a method for determining the control parameters of a drone is described in detail, specifically describing the case where the rotation angle of the target power line bounding box is less than or equal to a preset angle. The method includes:

[0091] S401 : Determine the deviation displacement of the UAV relative to the power line according to a preset distance, the horizontal coordinate of the center point of the target power line bounding box, the coordinate of the center point of the initial power line image, and the width of the target power line bounding box.

[0092] Among them, the horizontal coordinate of the center point of the target power line bounding box is Figure 3 One of the bounding box parameters in the embodiment can be specifically Figure 8 The embodiments shown are used for illustration only and will not be described in detail here.

[0093] The center point coordinates of the initial power line image are also obtained by Figure 8 The illustrated embodiment is described.

[0094] In practical applications, the deviation displacement of the UAV relative to the power line is calculated according to the formula as well as Calculated, where dy is the component of the displacement in the y-axis direction in the drone coordinate system, dx is the component of the displacement in the x-axis direction in the drone coordinate system, and c i is the center point coordinate of the initial power line image, cx is the horizontal coordinate of the center point of the target power line bounding box, d r is the preset distance, d p is the pixel distance between the left and right ends of the power line, H is the actual length of each pixel in the initial power line image, and w is the width of the target power line boundary box. Since the drone is set at the center above the power line, d p is determined by the width w of the target power line bounding box. Therefore, in one possible implementation, according to the formula as well as Determine the deviation displacement of the drone relative to the power lines.

[0095] S402: Determine the deviation angle of the UAV relative to the power line based on the endpoint coordinates of the power line contained in the target power line bounding box.

[0096] In one possible implementation, the arc tangent of the power line's endpoints (x1, y1) and (x2, y2) is calculated to obtain the corresponding deflection angle. After removing line segments with significantly different deflection angles, the average deflection angle of the remaining valid line segments is calculated to determine the deviation angle φ of the drone relative to the power line. Tukey's Honestly Significant Difference Test (HSD) can be used to remove line segments with significantly different deflection angles.

[0097] S403. Determine the displacement error and yaw angle error at the current moment based on the relative height between the UAV and the power line, the deviation displacement of the UAV relative to the power line, the deviation angle of the UAV relative to the power line, the displacement error at the previous moment, and the yaw angle error at the previous moment.

[0098] In one possible implementation, according to the formula Determine the displacement error and yaw angle error at the current moment;

[0099] Among them, Δx is the displacement error component of the x-axis direction in the UAV coordinate system, Δy is the displacement error component of the y-axis direction in the UAV coordinate system, Δz is the displacement error component of the z-axis direction in the UAV coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, φ is the deviation angle and h r is the relative height between the drone and the power line.

[0100] S404: Determine the control parameters of the UAV according to the displacement error and yaw angle error at the current moment.

[0101] It should be understood that the rotation angle of the target power line boundary box at this time is less than or equal to the preset angle, which means that the inspection direction of the drone does not need to be adjusted. After determining the displacement error and yaw angle error at the current moment, proportional control can be directly adopted, that is, Figure 3 The target speed and yaw angle calculation method in the embodiment feeds back the drone control parameters to the drone's underlying flight control, thereby fine-tuning the drone's posture to achieve robust linear flight control.

[0102] The embodiment of the present application provides a method for determining the control parameters of a drone. When the rotation angle of the target power line boundary box is less than or equal to a preset angle, that is, when the orientation of the drone is consistent with the direction of the power line, the deviation displacement of the drone relative to the power line is calculated by the preset distance, the coordinates of the center point of the target power line boundary box and the image, and the width of the target power line boundary box, and the deviation angle is determined by the coordinates of the power line endpoint. Then, the displacement error and yaw angle error at the current moment are updated by combining the relative height, deviation displacement, deviation angle, displacement error at the previous moment, and yaw angle error at the previous moment. Finally, the control parameters of the drone are determined based on these errors to cope with the situation where the drone is flying along the power line, thereby improving the accuracy of drone navigation.

[0103] Figure 5 The present invention provides a drone control parameter determination system, such as Figure 5 As shown, the drone control parameter determination system includes a server, which includes an acquisition module 501, an identification module 502, an image processing module 503 and a tracking control module 504.

[0104] The acquisition module 501 is used to acquire and store the power line image captured by the camera in real time, and send the power line image to the recognition module 502 .

[0105] The recognition module 502 includes a power line orientation detection network (i.e., a power line recognition model), which is used to output the parameters of the initial power line bounding box according to the power line image, including the center point coordinates, width, length, and rotation angle. The specific power line orientation detection network can be Figure 6 The illustrated embodiment is described.

[0106] The image processing module 503 is used to select the initial power line bounding box, determine the target power line bounding box, and obtain the straight line segments in the power line image after processing the power line image with a rotation angle of the target power line bounding box less than or equal to a preset angle; the module is also used to calculate the posture error of the UAV based on the parameters of the target power line bounding box and the coordinates of the straight line segments, and then send it to the tracking control module 504. The posture error includes the relative height between the UAV and the power line, the deviation displacement of the UAV relative to the power line, and the deviation angle of the UAV relative to the power line.

[0107] The tracking control module 504 is used to calculate the drone control parameters based on the drone's posture error and send them to the drone.

[0108] Figure 6 A schematic diagram of a transmission line directional detection network provided in an embodiment of the present application is shown in FIG. Figure 6As shown in the figure, the power line orientation detection network identifies power lines in power line images. To enable real-time detection and recognition on the drone's edge computing device, the YOLOv11 algorithm with a single-stage neural network architecture was selected. This network extracts the specific location of power lines from the drone's FPV camera's real-time image and outputs (cx, cy, w, h, θ), representing the center coordinates, width, height, and rotation angle of the power line bounding box, respectively.

[0109] YOLOv11's model architecture consists of three components: the backbone network (Backbone), the neck architecture (Neck), and the head network (Head), which together achieve efficient and accurate object detection. The core of YOLOv11's backbone network is the C3k2 module, an evolution of the CrossStagePartial (CSP) bottleneck introduced in earlier versions. The C3k2 module optimizes information flow within the network because it contains smaller kernel convolutions. While retaining essential image feature processing capabilities, smaller kernel convolutions are faster and computationally cheaper than larger kernel convolutions. The neck architecture connects the backbone network to the rest of the system, collecting and combining information from different parts of the image. Compared to previous series of network architectures, YOLOv11 adds new components such as the C3k2 module, the Spatial Pyramid Pooling-Fast (SPPF) module, and the C2PSA mechanism. Among them, the SPPF module is designed to pool features from different areas of the image at different scales, improving the network's ability to capture objects of different sizes (especially small objects). The C2PSA mechanism is a multi-head attention mechanism embedded within the C2 mechanism, which can improve the model's sensitivity and accuracy to features, enabling it to more accurately capture the detailed features of the target. YOLOv11 uses a multi-scale prediction head to detect objects of different sizes, and inserts two depthwise separable convolutions (DWConv) into the classification detection head, which can significantly reduce the number of parameters and computational complexity. The detection head outputs the prediction results of three feature maps based on the different granularity levels in the image to ensure that smaller objects are detected with finer details.

[0110] In practical applications, considering that the transmission line area from a bird's-eye view has certain geometric characteristics, such as the inter-tower area is mostly rectangular, and the corner tower area is mostly rectangular with a certain rotation angle, it is considered to change the detection head based on the YOLOv11 network to an oriented bounding box (OBB) detection head, that is, to rotate the length and width of the bounding box to adapt to the transmission line area at different angles.

[0111] The core principle of the OBB detection head is to achieve accurate localization of tilted objects by adding a rotation angle prediction branch to the traditional object detection head. Specifically, the detection head takes the feature map extracted by the backbone network as input and generates multi-scale features through multiple convolutional layers. It predicts the object's class probability, center point coordinates, width and height, and rotation angle θ. The angle prediction branch can use either regression or classification: regression methods directly output continuous angle values, while classification methods discretize the angle into multiple intervals (e.g., 10° per category) and then optimize it using a cross-entropy loss. The detection head jointly optimizes the classification loss, localization loss, and angle loss, ultimately converting the decoded parameters (cx, cy, w, h, θ) into the coordinates of the four vertices of a rotated rectangle. To adapt to the post-processing of rotated boxes, the inference stage uses Rotated Non-Maximum Suppression (NMS) or overlap calculation based on the Rotated Intersection over Union (RIoU) to address the non-maximum suppression issue between tilted boxes and ensure a balance between detection accuracy and efficiency.

[0112] Among them, the regression method directly outputs continuous angle values, and the Sigmoid activation function (Sigmoid Activation Function, Sigmoid) or the hyperbolic tangent activation function (Hyperbolic Tangent Activation Function, Tanh) can be used to constrain the range.

[0113] Among them, for optimizing classification loss, you can use focal loss (FocalLoss); for positioning loss, you can use complete intersection over union loss (CIoU Loss) or generalized intersection over union loss (GIoU Loss); for angle loss, you can use mean squared error (MSE) or smooth L1 loss (Smooth L1 Loss, Smooth L1).

[0114] In practical applications, after the initial power line image is inspected by the power line directional detection network, a non-maximum suppression (NMS) operation is performed. This may ultimately result in 0 to 2 high-confidence power line bounding boxes. In the case of two bounding boxes, the strategy adopted is to prioritize the bounding box with the larger h based on their height. This allows the drone to adjust its posture based on the closer bounding box.

[0115] Figure 7 Schematic diagram of the target power line boundary box provided in the embodiment of the present application Figure 1 ,like Figure 7 As shown in the figure, the blue frame is the target power line bounding box, and the black line segment can represent the bottom edge of the image. The angle between the bottom edge and the bounding box is the rotation angle of the target power line bounding box, which is represented by θ. At this time, it can be seen that the drone has inspected the tower and needs to turn to continue inspecting the next section of the power line. This corresponds to the case where the rotation angle of the target power line bounding box is greater than the preset angle. In addition, the preset distance d r like Figure 7 As shown, this is the distance between the leftmost power line and the rightmost power line.

[0116] Figure 8 Schematic diagram of the target power line boundary box provided in the embodiment of the present application Figure 2 ,like Figure 8 As shown in Figure 2, the black box represents the initial power line image, and the blue box represents the target power line bounding box. The target power line bounding box in the initial power line image delineates the power lines in the image, where c i is the coordinate of the center point of the initial power line image, cx is the horizontal coordinate of the center point of the target power line bounding box, φ is the deviation angle of the drone relative to the power line, h is the height of the target power line bounding box, and w is the width of the target power line bounding box.

[0117] Figure 9 This is a schematic diagram of the process of extracting power lines in the target power line boundary frame provided by the embodiment of the present application, as shown in FIG. Figure 9 As shown, the corresponding situation at this time is that the rotation angle of the target power line bounding box is less than the preset angle, including:

[0118] S901, determining a target power line bounding box based on the initial power line image, thereby removing cluttered background outside the target power line bounding box;

[0119] S902, after converting the image of the area where the target power line boundary box is located into a grayscale image, perform Gaussian blur processing on the grayscale image using a 5*5 Gaussian kernel to remove noise in the image and make the image smoother;

[0120] S903, selecting a Canny Edge Detection Operator (Canny), an edge detection operator having greater advantages in edge location and response uniformity, to perform edge detection on the Gaussian blurred image to obtain an edge image;

[0121] S904: dilate the edge image to make the edge thicker, so as to fill the broken or empty parts in the image;

[0122] S905, perform an erosion operation, which can remove some fine edges and noise to a certain extent, making the main edges clearer;

[0123] S906 uses the Probabilistic Hough Transform (PHT) to extract line segments within the region. PHT introduces a random sampling mechanism, probabilistically selecting edge points to generate candidate line segments. Combined with local search and dynamic point removal, this significantly reduces computational complexity. PHT directly outputs finite-length line segments and filters out noise by setting a minimum segment length and a maximum allowable gap. This improves efficiency while maintaining detection accuracy, making it particularly suitable for real-time scenarios.

[0124] Figure 10 Schematic diagram of the results of a method for determining drone control parameters provided in an embodiment of the present application Figure 1 ,like Figure 10 As shown, the figure is based on Figure 2 The inspection simulation environment constructed (the length, width and height of the tower are 11.5m×6.5m×32.5m respectively, the distance between each two towers is 100m, the total length is about 400m, and the ideal height of the drone and the power line is h set The figure is a top view of the drone's patrol trajectory. From the top view, we can see that the drone basically moves along the path just above the connection area of ​​the transmission tower, moving from tower ⑤ to tower ①, with no obvious deviation in the X and Y directions.

[0125] In practical applications, a simulation environment can be built based on an open-source simulation platform. This platform needs to be suitable for simulating drones and sensor models such as cameras and lidar. Its physical model should include properties such as friction coefficient, collision volume, surface texture, and mass, and have good visualization effects.

[0126] For the tower model, a 330kV tension-type double-circuit tower was selected for simulation. Using 3D modeling software, a model of the 330kV tension-type double-circuit tower was constructed. The tower was constructed of angle steel, with 17 insulators per string. This model was then converted to a format supported by the simulation platform and imported into the platform.

[0127] Figure 11 Schematic diagram of the results of a method for determining drone control parameters provided in an embodiment of the present application Figure 2 ,like Figure 11 As shown, the figure is based on Figure 1 The inspection simulation environment constructed (the length, width and height of the tower are 11.5m×6.5m×32.5m respectively, the spacing between each two towers is 100m, the total length is about 400m, and the ideal height of the drone and the power line is h set The figure is a side view of the drone's patrol trajectory. From the side view, we can see that the drone's height is maintained at 10m square to the tower, that is, around 43m (the tower height is 32.5m). In the base 2 to base 4 tower area, the drone can also stably adjust its own height to maintain an ideal distance from the transmission line, meeting the expected requirements.

[0128] Figure 12 A schematic diagram of a structure of a device for determining a control parameter of a drone provided in an embodiment of the present application is shown in FIG12. Figure 12 As shown, the drone control parameter determination device 120 provided in this embodiment includes:

[0129] A first determining module 1201 is configured to determine a target power line bounding box based on the initial power line image, where the power lines contained in the target power line bounding box are straight lines;

[0130] A second determining module 1202 is configured to determine a relative height between the UAV and the power line based on a preset distance, a camera focal length corresponding to the initial power line image, and a width of a target power line bounding box;

[0131] a third determining module 1203 configured to determine, when the rotation angle of the target power line bounding box is greater than a preset angle, a displacement error and a yaw angle error at a current moment based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at a previous moment, and the yaw angle error at a previous moment, wherein the rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image;

[0132] The fourth determining module 1204 is configured to determine the control parameters of the UAV according to the displacement error and the yaw angle error at the current moment.

[0133] In a possible implementation, the second determination module 1202 is further configured to calculate a ratio between a preset distance and a width of a target power line bounding box; and multiply the ratio by the camera focal length to determine the relative height between the drone and the power line.

[0134] In a possible implementation, the third determining module 1203 is further configured to: Calculate the displacement error and yaw angle error at the current moment;

[0135] Among them, Δx is the displacement error component in the x-axis direction of the drone coordinate system, Δy is the displacement error component in the y-axis direction of the drone coordinate system, Δz is the displacement error component in the z-axis direction of the drone coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, θ is the rotation angle and h r is the relative height between the drone and the power line.

[0136] In one possible implementation, the first determination module 1201 is also used to input the initial power line image into the power line recognition model, obtain multiple initial power line bounding boxes output by the power line recognition model, the power lines contained in the initial power line bounding boxes are straight lines, and the power line recognition model is pre-trained based on multiple sample power line images and the sample power line bounding boxes corresponding to each sample power line image; the initial power line bounding box with the largest width is determined as the target power line bounding box.

[0137] In a possible implementation, the fourth determining module 1204 is further configured to: Determine the drone control parameters;

[0138] Among them, K px , K py , K pz and K pθ Both are used to indicate the preset proportional coefficient, h set is the preset relative height between the UAV and the power line, v is the target speed, and ω is the target angular velocity. The target speed and yaw angle are UAV control parameters.

[0139] In one possible implementation, the fourth determination module 1204 is further used to determine the deviation displacement of the UAV relative to the power line based on a preset distance, the horizontal coordinates of the center point of the target power line bounding box, the coordinates of the center point of the initial power line image, and the width of the target power line bounding box; determine the deviation angle of the UAV relative to the power line based on the endpoint coordinates of the power line contained in the target power line bounding box; determine the displacement error and yaw angle error at the current moment based on the relative height between the UAV and the power line, the deviation displacement of the UAV relative to the power line, the deviation angle of the UAV relative to the power line, the displacement error at the previous moment, and the yaw angle error at the previous moment; and determine the UAV control parameters based on the displacement error and yaw angle error at the current moment.

[0140] In a possible implementation, the fourth determining module 1204 is further configured to: as well as Determine the deviation displacement of the UAV relative to the power line;

[0141] Among them, dy is the component of the deviation displacement in the y-axis direction in the UAV coordinate system, dx is the component of the deviation displacement in the x-axis direction in the UAV coordinate system, and c i is the center point coordinate of the initial power line image, cx is the horizontal coordinate of the center point of the target power line bounding box, d r is the preset distance, w is the width of the target power line bounding box, and H is the actual length corresponding to each pixel in the initial power line image.

[0142] In a possible implementation, the fourth determining module 1204 is further configured to: Determine the displacement error and yaw angle error at the current moment;

[0143] Among them, Δx is the displacement error component of the x-axis direction in the UAV coordinate system, Δy is the displacement error component of the y-axis direction in the UAV coordinate system, Δz is the displacement error component of the z-axis direction in the UAV coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, φ is the deviation angle and h r is the relative height between the drone and the power line.

[0144] The drone control parameter determination device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and are not described in detail in this embodiment.

[0145] Figure 13 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 13 As shown, the electronic device 130 provided in this embodiment includes: at least one processor 1301 and a memory 1302. Optionally, the electronic device 130 further includes a communication component 1303. The processor 1301, the memory 1302, and the communication component 1303 are connected via a bus 1304.

[0146] During the specific implementation process, at least one processor 1301 executes the computer-executable instructions stored in the memory 1302, so that the at least one processor 1301 performs the above method.

[0147] The specific implementation process of the processor 1301 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0148] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0149] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0150] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0151] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0152] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0153] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0154] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0155] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, either through an interface, electrical, mechanical, or other means.

[0156] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0157] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0158] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0159] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0160] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. A method for determining control parameters of an unmanned aerial vehicle, characterized in that: include: determining a target power line bounding box according to the initial power line image, wherein the power lines contained in the target power line bounding box are straight lines; Determining the relative height between the drone and the power line based on a preset distance, a camera focal length corresponding to the initial power line image, and a width of a bounding box of the target power line; When the rotation angle of the target power line bounding box is greater than a preset angle, determining the displacement error and yaw angle error at a current moment based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at a previous moment, and the yaw angle error at a previous moment, wherein the rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image; Determine the UAV control parameters based on the displacement error and yaw angle error at the current moment.

2. The method according to claim 1, characterized in that The determining the relative height between the UAV and the power line according to the preset distance, the camera focal length corresponding to the initial power line image, and the width of the target power line bounding box includes: calculating a ratio between the preset distance and a width of the target power line bounding box; The product of the ratio and the focal length of the camera is determined as the relative height between the drone and the power line.

3. The method according to claim 1 or 2, characterized in that The determining of the displacement error and the yaw angle error at the current moment based on the relative height between the UAV and the power line, the rotation angle of the target power line bounding box, the displacement error at the previous moment, and the yaw angle error at the previous moment includes: By formula Calculate the displacement error and yaw angle error at the current moment; Wherein, Δx is the displacement error component in the x-axis direction of the drone coordinate system, Δy is the displacement error component in the y-axis direction of the drone coordinate system, Δz is the displacement error component in the z-axis direction of the drone coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, θ is the rotation angle and h r is the relative height between the UAV and the power line.

4. The method according to claim 1 or 2, characterized in that The step of determining a target power line bounding box according to the initial power line image includes: Inputting the initial power line image into a power line recognition model, obtaining a plurality of initial power line bounding boxes output by the power line recognition model, wherein the power lines contained in the initial power line bounding boxes are straight lines, and the power line recognition model is pre-trained based on a plurality of sample power line images and a sample power line bounding box corresponding to each sample power line image; An initial power line bounding box with the largest width is determined as the target power line bounding box.

5. The method according to claim 3, characterized in that Determining the control parameters of the drone based on the displacement error and yaw angle error at the current moment includes: By formula Determining control parameters of the drone; Among them, K px , K py , K pz and K pθ Both are used to indicate the preset proportional coefficient, h set is the preset relative height between the UAV and the power line, v is the target speed, ω is the target angular velocity, and the target speed and the yaw angle are UAV control parameters.

6. The method according to any one of claims 1, 2 or 5, characterized in that: When the rotation angle of the target power line bounding box is less than or equal to the preset angle, the method further includes: Determining the deviation displacement of the drone relative to the power line based on the preset distance, the horizontal coordinate of the center point of the target power line bounding box, the coordinate of the center point of the initial power line image, and the width of the target power line bounding box; Determining a deviation angle of the UAV relative to the power line based on the endpoint coordinates of the power line contained in the target power line bounding box; Determine the displacement error and yaw angle error at a current moment based on the relative height between the UAV and the power line, the deviation displacement of the UAV relative to the power line, the deviation angle of the UAV relative to the power line, the displacement error at a previous moment, and the yaw angle error at a previous moment; Determine the UAV control parameters based on the displacement error and yaw angle error at the current moment.

7. The method according to claim 6, characterized in that The determining the deviation displacement of the UAV relative to the power line according to the preset distance, the horizontal coordinate of the center point of the target power line bounding box, the coordinate of the center point of the initial power line image, and the width of the target power line bounding box includes: According to the formula as well as Determining the deviation displacement of the UAV relative to the power line; Wherein, dy is the component of the deviation displacement in the y-axis direction in the UAV coordinate system, dx is the component of the deviation displacement in the x-axis direction in the UAV coordinate system, and c i is the center coordinate of the initial power line image, cx is the horizontal coordinate of the center point of the target power line bounding box, d r is the preset distance, w is the width of the target power line boundary box, and H is the actual length corresponding to each pixel in the initial power line image.

8. The method according to claim 7, characterized in that Determining the displacement error and yaw angle error at the current moment based on the relative height between the drone and the power line, the deviation displacement of the drone relative to the power line, the deviation angle of the drone relative to the power line, the displacement error at the previous moment, and the yaw angle error at the previous moment includes: According to the formula Determine the displacement error and yaw angle error at the current moment; Wherein, Δx is the displacement error component in the x-axis direction of the drone coordinate system, Δy is the displacement error component in the y-axis direction of the drone coordinate system, Δz is the displacement error component in the z-axis direction of the drone coordinate system, t is the current moment, t-1 is the previous moment, Δψ is the yaw angle error, β is the preset smoothing coefficient, φ is the deviation angle and h r is the relative height between the UAV and the power line.

9. A device for determining control parameters of an unmanned aerial vehicle, characterized in that: include: A first determining module is configured to determine a target power line bounding box based on the initial power line image, wherein the power lines contained in the target power line bounding box are straight lines; a second determining module, configured to determine a relative height between the drone and the power line based on a preset distance, a camera focal length corresponding to the initial power line image, and a width of a bounding box of the target power line; a third determining module, configured to determine, when the rotation angle of the target power line bounding box is greater than a preset angle, a displacement error and a yaw angle error at a current moment based on the relative height between the drone and the power line, the rotation angle of the target power line bounding box, the displacement error at a previous moment, and the yaw angle error at a previous moment, wherein the rotation angle of the target power line bounding box refers to the angle between the bottom edge of the target power line bounding box and the bottom edge of the initial power line image; The fourth determination module is used to determine the control parameters of the UAV according to the displacement error and yaw angle error at the current moment.

10. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.

12. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 8 when executed by a processor.

Images