An unmanned aerial vehicle automatic navigation method and system for power transmission line inspection

Abstract

The present application relates to the technical field of unmanned aerial vehicle path planning, and particularly relates to an unmanned aerial vehicle automatic navigation method and system for power transmission line inspection. The method comprises the following steps: collecting terrain elevation data and wind speed fluctuation signals, generating a line corridor boundary elevation distribution map, analyzing the influence of terrain slope on conductor sag, and determining the curvature limitation adjustment requirement; re-planning the flight trajectory in combination with obstacle position information to generate a detection path sequence; calculating the height mutation value and attitude angle deviation value, judging whether there is an acceleration impact inhibition requirement, and generating a vegetation intrusion area inspection risk score; determining the cross-span distance risk quantization index to generate a flight control instruction sequence; verifying whether the path meets the boundary integrity standard to generate an automatic navigation path sequence. The present application solves the problems of unstable unmanned aerial vehicle inspection path, insufficient coverage and poor safety under complex terrain conditions, and realizes stable, continuous and efficient automatic generation of the power transmission line inspection path.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) path planning technology, and in particular to an automatic navigation method and system for UAVs used in power line inspection. Background Technology

[0002] As power systems continue to expand, transmission lines are increasingly extending into complex terrain areas such as mountains, hills, valleys, and forests, making the operating environment increasingly complex. In complex terrain, conductor sag changes significantly, and factors such as vegetation intrusion and wind speed fluctuations substantially increase the difficulty of transmission line operation and inspection. Traditional methods relying on manual inspection or ground equipment are inefficient, labor-intensive, and pose significant safety hazards in high-risk areas, making them insufficient to meet the demands of modern power systems for high-frequency, high-precision inspections. Existing UAV inspection technologies mostly employ preset flight paths or simple terrain obstacle avoidance strategies. Their path planning is primarily based on tower locations or static spatial coordinates of conductors, lacking comprehensive consideration of the overall terrain characteristics of the line corridor, conductor sag changes, and dynamic environmental factors. In complex terrain, flying at a single altitude or simply following a shape can easily lead to sudden changes in UAV path curvature, attitude instability, and excessive flight energy consumption, thus affecting inspection quality and flight safety. Furthermore, power transmission lines sway under wind conditions, causing a certain spatial displacement in the critical components such as vibration dampers and hardware. Existing technologies mostly plan inspection paths based on static design positions, making it difficult to effectively cover the actual distribution range of the lines and their components under dynamic operating conditions, easily leading to missed inspections or blind spots. Simultaneously, in complex terrain, drones need to frequently switch altitude levels. Existing technologies lack a systematic assessment and smoothing mechanism for the acceleration impact and attitude deviation caused by sudden altitude changes, resulting in insufficient path continuity and inspection stability.

[0003] Therefore, this invention is proposed, which can comprehensively utilize the elevation distribution of the line corridor, the changes in terrain slope, the sag curve of the conductor, and real-time wind field information to perform refined modeling and dynamic optimization of the inspection path. While ensuring the safety and stability of the path, it can improve the coverage integrity and inspection efficiency of the transmission line and its key components. Summary of the Invention

[0004] This invention provides an automatic navigation method and system for UAVs used in power transmission line inspection, which significantly improves the path stability, coverage integrity and operational safety of UAVs in power transmission line inspection in complex terrain.

[0005] In a first aspect, the present invention provides an automatic navigation method for unmanned aerial vehicles (UAVs) used for power transmission line inspection, the method comprising:

[0006] Step S1: Collect terrain elevation data and wind speed fluctuation signals of the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary; calculate the terrain slope change between adjacent towers based on the distribution map, and analyze the influence of terrain slope on the sag curve of the conductor by combining height layer switching and terrain following mode, and determine the path curvature limit adjustment requirements when the UAV inspects the conductor.

[0007] Step S2: Based on the obstacle location information obtained by the obstacle avoidance sensor, replan the UAV flight trajectory, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in combination with the wind speed fluctuation signal to generate a detection path sequence.

[0008] Step S3: Extract height stratification switching points from the detection path sequence, calculate the height mutation value between adjacent path segments and the UAV attitude angle deviation value, determine whether the height mutation triggers acceleration impact suppression requirements, and generate an inspection risk score for the vegetation invasion area based on the spatial distribution of the height stratification switching points.

[0009] Step S4: Based on the inspection risk score and wind speed fluctuation data, determine the cross-crossing distance risk quantification index; if the risk quantification index exceeds the preset critical value, perform smooth transition processing on the altitude layer switching point to generate a continuous flight control command sequence.

[0010] Step S5: By simulating the coverage and energy consumption data of the flight control command sequence, determine whether the boundary integrity standard is met, and generate the final automatic navigation path sequence based on the determination result.

[0011] As a preferred embodiment of the present invention, step S1, generating a refined elevation distribution map of the railway corridor, includes:

[0012] Elevation sampling data of complex terrain is collected by lidar, and wind speed fluctuation compensation signals are obtained by airborne sensors. The elevation data around the tower positioning markers is initially filtered using a preset threshold to separate the influence of terrain undulations and wind speed fluctuations. A refined elevation distribution map of the route corridor is generated based on the filtered elevation data. The wind speed fluctuation compensation signals are time-domain aligned to ensure the synchronization of elevation data and wind speed data. Noise points are further removed through multi-layer threshold filtering to obtain a more accurate elevation distribution map of the route corridor.

[0013] As a preferred embodiment of the present invention, step S1, determining the conductor path curvature constraint adjustment requirements, includes:

[0014] Based on the elevation distribution map of the line corridor, the elevation difference and horizontal distance between adjacent towers are calculated to obtain the terrain slope changes. Combining the height layer switching strategy and terrain following mode, the impact of different slope sections on the spatial shape of the conductor sag curve is analyzed, and the curvature of the UAV's following path for conductor inspection is calculated. The calculated path curvature is compared with the safe curvature range to determine the path curvature limit adjustment requirements for corresponding sections. For sections where the terrain slope change exceeds the preset threshold, the corresponding curvature adjustment parameters are recorded, and a layered curvature constraint model is introduced. Differentiated curvature constraints are set according to different height levels. Through iterative optimization, the adjusted path curvature is matched with the terrain features and conductor sag characteristics to generate optimized path curvature limit adjustment requirements.

[0015] As a preferred embodiment of the present invention, step S2, generating the adjusted detection path sequence, includes:

[0016] Obstacle location information is obtained through obstacle avoidance laser scanning data; the flight trajectory is replanned based on the optimized path curvature constraint adjustment requirements; a path smoothing algorithm is used to reduce abrupt changes in trajectory curvature; wind speed fluctuation compensation signals collected in real time by an airborne wind speed sensor are introduced, and the smoothed flight trajectory is adaptively corrected and its position and attitude changes in space are dynamically adjusted through the wind speed fluctuation compensation signals, and an adjusted detection path sequence is generated, so that the path avoids obstacles and adapts to wind field changes.

[0017] As a preferred embodiment of the present invention, step S3, determining the acceleration impact suppression requirement and generating a risk score for vegetation intrusion area inspection, includes:

[0018] Height stratification switching points are extracted from the detection path sequence. The corresponding height abrupt change value is calculated based on the vertical height difference between adjacent height stratification switching points. Combined with pitch and roll angle data collected by the airborne attitude sensor, the attitude angle deviation value of the UAV during the height switching process is calculated. The height abrupt change value is compared with a preset height abrupt change threshold to determine whether the height abrupt change triggers acceleration impact suppression requirements. A preliminary vegetation intrusion area inspection risk score is generated based on the judgment result and the distribution of height stratification switching points. The risk score quantifies the impact of vegetation intrusion on inspection safety.

[0019] As a preferred embodiment of the present invention, step S4, determining the crossover distance risk quantification index, includes:

[0020] Based on the risk score of the vegetation intrusion area inspection combined with the critical value of the cross-crossing distance stability; and the comprehensive analysis of wind speed fluctuation compensation data; the cross-crossing distance risk quantification index is determined by using the height layer switching point time interval calculation method; the risk quantification weight is adjusted for different wind speed fluctuation intensities; and the final cross-crossing distance risk quantification index is generated.

[0021] As a preferred embodiment of the present invention, step S4, generating a continuous sequence of flight control commands, includes:

[0022] When the crossover distance risk quantification index exceeds the preset critical value, attitude angle deviation correction technology is used to adjust the flight attitude of the UAV during the altitude layer switching process for the altitude layer switching point. By continuously correcting the pitch angle and roll angle of the UAV, the flight trajectory corresponding to the altitude layer switching point is smoothly transitioned. The flight trajectory after attitude angle deviation correction and smoothing is transformed into continuous flight control commands, generating a continuous flight control command sequence.

[0023] As a preferred embodiment of the present invention, step S5, generating the final automatic navigation path sequence, includes:

[0024] The simulation results of vibration damper offset coverage and energy consumption are obtained from the flight control command sequence; it is determined whether the simulation results meet the preset line corridor boundary integrity standard; based on the determination results, an automatic navigation path sequence for small-scale fine inspection of transmission lines in complex terrain is generated, wherein the line corridor boundary integrity standard is used to characterize the effective coverage of the inspection path on key components and spatial range of the transmission line corridor.

[0025] Secondly, the present invention also provides an unmanned aerial vehicle (UAV) automatic navigation system for power transmission line inspection, for implementing the above-mentioned method, the system comprising:

[0026] The terrain and wind field sensing unit is used to collect terrain elevation data and wind speed fluctuation signals in the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary. Based on the distribution map, the terrain slope change between adjacent towers is calculated. Combining the height layer switching and terrain following mode, the influence of terrain slope on the conductor sag curve is analyzed to determine the path curvature limit adjustment requirements when the UAV inspects the conductor.

[0027] The obstacle avoidance path planning unit is used to replan the UAV flight trajectory based on the obstacle position information obtained by the obstacle avoidance sensor, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in combination with the wind speed fluctuation signal to generate a detection path sequence.

[0028] The hierarchical risk assessment unit is used to extract height hierarchical switching points from the detection path sequence, calculate the height mutation value between adjacent path segments and the UAV attitude angle deviation value, determine whether the height mutation triggers the acceleration impact suppression requirement, and generate an inspection risk score for the vegetation invasion area based on the spatial distribution of the height hierarchical switching points.

[0029] The control command generation unit is used to determine the cross-crossing distance risk quantification index based on the inspection risk score and wind speed fluctuation data; if the risk quantification index exceeds the preset critical value, the altitude layer switching point is smoothed to generate a continuous flight control command sequence.

[0030] The navigation path acquisition unit is used to determine whether the boundary integrity standard is met by simulating the coverage and energy consumption data of the flight control command sequence, and to generate the final automatic navigation path sequence based on the determination result.

[0031] Thirdly, the present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0032] The beneficial effects of this invention are as follows:

[0033] This invention collects terrain elevation data and wind speed fluctuation signals from the inspection area, generates a refined elevation distribution map of the line corridor boundary, and analyzes the sag characteristics of the conductor by combining terrain slope changes between adjacent towers, height layer switching, and terrain following modes. This determines reasonable path curvature constraint adjustment requirements from the source, providing constraints for trajectory planning that conform to the terrain and conductor physical characteristics. Under the guidance of these constraints, obstacle avoidance sensors are used to acquire obstacle location information, and the UAV flight trajectory is replanned. Curvature continuity processing eliminates abrupt changes in the trajectory, while wind speed fluctuation signals are used for dynamic correction, thereby generating a detection path sequence that balances obstacle avoidance safety, curvature smoothness, and environmental adaptability. By extracting height layer switching points from the detection path sequence and calculating height abrupt changes and attitude angle deviations, the invention can accurately identify flight dynamic changes caused by terrain undulations or obstacle avoidance, and determine... The system identifies whether there is a need to suppress acceleration impacts. It also generates a risk score for vegetation intrusion areas by combining the spatial distribution of altitude stratification switching points, extending path analysis from a geometric level to a safety risk level. Furthermore, it integrates the inspection risk score with wind speed fluctuation data to form a quantitative indicator of cross-crossing distance risk. When the risk exceeds a threshold, a smooth transition is implemented at the altitude stratification switching points, generating a continuous sequence of flight control commands to effectively suppress attitude changes and acceleration impacts, thus improving flight stability. By simulating the coverage and energy consumption of the flight control command sequence, the system verifies whether the path meets the line corridor boundary integrity standard, and outputs the final automatic navigation path sequence accordingly. Through the synergy of these technical solutions, a closed-loop optimization from environmental perception, path planning, risk assessment to control verification is achieved, significantly improving the path stability, coverage integrity, and operational safety of UAVs in power line inspections in complex terrain. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of an automatic navigation method for a drone used for power transmission line inspection, as shown in the embodiment.

[0036] Figure 2 This is a flowchart illustrating the method for generating risk scores for vegetation intrusion areas in this embodiment.

[0037] Figure 3 This is a schematic diagram illustrating the effect of smooth transition processing on the flight trajectory corresponding to the altitude layer switching point in the embodiment;

[0038] Figure 4 This is a structural diagram of an unmanned aerial vehicle (UAV) automatic navigation system for power transmission line inspection, as shown in the embodiment. Detailed Implementation

[0039] This invention provides an automatic navigation method and system for unmanned aerial vehicles (UAVs) used for power line inspection. The terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0040] For ease of understanding, the specific process of the embodiments of the present invention will be described below, such as... Figure 1 As shown in the figure, an automatic navigation method for UAVs used in power transmission line inspection according to an embodiment of the present invention includes:

[0041] Step S1: Collect terrain elevation data and wind speed fluctuation signals of the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary; calculate the terrain slope change between adjacent towers based on the distribution map, and analyze the influence of terrain slope on the sag curve of the conductor by combining height layer switching and terrain following mode, and determine the path curvature limit adjustment requirements when the UAV inspects the conductor.

[0042] In step S1, a refined elevation distribution map of the route corridor is generated, including:

[0043] Elevation sampling data of complex terrain is collected by lidar, and wind speed fluctuation compensation signals are obtained by airborne sensors. The elevation data around the tower positioning markers is initially filtered using a preset threshold to separate the influence of terrain undulations and wind speed fluctuations. A refined elevation distribution map of the route corridor is generated based on the filtered elevation data. The wind speed fluctuation compensation signals are time-domain aligned to ensure the synchronization of elevation data and wind speed data. Noise points are further removed through multi-layer threshold filtering to obtain a more accurate elevation distribution map of the route corridor.

[0044] Specifically, in order to achieve high-precision generation of UAV automatic navigation paths during power transmission line inspection under complex terrain conditions, it is first necessary to obtain a line corridor elevation distribution map that can truly reflect the terrain characteristics of the power transmission line corridor and is less affected by environmental interference. To this end, a data processing workflow combining multi-source data acquisition, phased processing and progressive refinement was constructed, so that the obtained elevation distribution map can provide a reliable data foundation for subsequent path curvature analysis, risk assessment and flight control command generation.

[0045] When drones conduct inspections along power transmission lines, a lidar system mounted on the drone platform continuously scans the complex terrain areas where the power transmission lines are located, acquiring elevation sampling data containing terrain undulation information. Simultaneously, onboard sensors on the drone collect wind speed fluctuation compensation signals in real time to reflect the dynamic impact of external wind fields on the drone and the lidar ranging process during the inspection. In actual power transmission line inspection scenarios, such as mountainous or hilly areas, the lidar can generate point cloud data with high spatial resolution, for example, a point cloud density of approximately 10 sampling points per square meter. The wind speed fluctuation signal collected by the onboard anemometer covers a range of, for example, 0–15 m / s, thus ensuring the integrity and continuity of both elevation and wind speed data under complex terrain and variable weather conditions. To avoid misinterpretation of elevation measurement results due to wind speed disturbances, the collected elevation sampling data is processed using the tower positioning markers as a reference. The system divides the space into regions and focuses on filtering elevation data points around pole positioning markers. Specifically, a preset height deviation threshold, such as 5 meters, is set. The elevation data points are compared with the average terrain curve of the corresponding area. When the deviation of a data point from the average terrain curve exceeds the height deviation threshold, it is determined that the deviation is mainly caused by wind speed disturbance or measurement noise, and the data point is discarded. Data points that do not exceed the threshold are retained to represent the true terrain undulation characteristics. Through this filtering process, abnormal elevation changes caused by wind speed fluctuations can be effectively separated while maintaining the integrity of terrain slope and undulation information, thereby significantly reducing the impact of noise on subsequent analysis. For example, under strong wind conditions, wind speed disturbance may cause an offset of about 2 meters in elevation measurement results. After the height deviation threshold filtering process, the accuracy of terrain undulation identification can be improved, laying a reliable foundation for subsequent path planning and vegetation encroachment risk assessment.

[0046] After initial filtering and obtaining relatively stable elevation data, a two-dimensional or three-dimensional elevation distribution model of the transmission line corridor is constructed using this filtered elevation data to mark the spatial boundary range of the corridor, thus forming a preliminary elevation distribution map of the transmission line corridor. Simultaneously, to ensure a one-to-one correspondence between the elevation data and the wind speed fluctuation compensation signal in the time dimension, this embodiment further performs time-domain alignment processing on the wind speed fluctuation compensation signal and the elevation sampling data. Specifically, by reading the timestamp information of each type of data, the wind speed signal is resampled to ensure its time series is consistent with the time series of the elevation sampling data. When time inconsistencies exist, linear interpolation is used to compensate for missing or misaligned time points, and the synchronization results are verified by setting a time difference threshold (e.g., less than 0.1 seconds) to ensure that the wind speed data accurately reflects the environmental conditions at the corresponding time. This time-domain alignment processing effectively reduces data fusion errors caused by sampling frequency differences or time drift, enabling the wind speed fluctuation compensation signal to play a truly effective corrective role in subsequent elevation refinement and path analysis.

[0047] After completing the time-domain alignment process, to further improve the accuracy and reliability of the elevation distribution map of the railway corridor, a multi-layer threshold filtering mechanism is introduced based on the above filtering to refine the elevation data layer by layer. Specifically, the first layer of threshold filtering first detects deviations in individual elevation data points. When the height difference between a data point and the regional mean exceeds the first threshold (e.g., 3 meters), it is identified as an isolated noise point and removed. Subsequently, the second layer of threshold filtering based on neighborhood density analyzes the neighborhood around each elevation data point. When the density of data points in its neighborhood is lower than the second threshold (e.g., per...), the data points are removed. When the data point is 2 square meters, it is considered to lack spatial continuity and is also removed. Finally, in the third layer of filtering, the wind speed fluctuation compensation signal that has been aligned in the time domain is introduced into the threshold setting. The height deviation threshold is dynamically adjusted according to the intensity of wind speed fluctuation. For example, the height deviation threshold is tightened to 2 meters to further remove the elevation fluctuation caused by the remaining wind speed disturbance. Through the above multi-layer progressive threshold filtering process, the elevation data can be finely screened at different scales and with different sources of interference, so that the proportion of noise points is significantly reduced, while preserving the true terrain structure features to the greatest extent.

[0048] Through the above technical solution, after completing multi-layer threshold filtering, the retained elevation data is integrated to generate a more accurate elevation distribution map of the transmission line corridor. The elevation distribution map can not only truly reflect the terrain undulations and boundary features of the transmission line corridor under complex terrain conditions, but also maintain high data reliability and spatial continuity under complex scenarios such as wind speed disturbance, dense vegetation cover or high altitude, laying the foundation for obtaining high-precision automatic cruise paths.

[0049] Further, in step S1, determining the conductor path curvature constraint adjustment requirements includes:

[0050] Based on the elevation distribution map of the line corridor, elevation sampling data corresponding to the positioning marks of adjacent towers are extracted, and the elevation difference and horizontal distance between adjacent towers are calculated to obtain the terrain slope change. Combining the height layer switching strategy and terrain following mode, the influence of different slope sections on the spatial shape of the conductor sag curve is analyzed. Specifically, the terrain slope is mapped to the conductor tension change, and the curvature of the UAV following the conductor inspection path is calculated. The calculated path curvature is compared with the safe curvature range to determine the path curvature limit adjustment requirements for the corresponding sections. For sections where the terrain slope change exceeds the preset threshold, the corresponding curvature adjustment parameters are recorded, and a layered curvature constraint model is introduced. Differentiated curvature constraints are set according to different height levels. Through iterative optimization, the adjusted path curvature is matched with the terrain features and conductor sag characteristics to generate optimized path curvature limit adjustment requirements.

[0051] Specifically, after generating the elevation distribution map of the transmission line corridor, in order to enable UAVs to conduct stable and continuous inspections along the transmission line under complex terrain conditions and to avoid path mismatches or flight risks caused by sudden terrain changes or conductor sag variations, it is necessary to further analyze terrain changes based on the elevation distribution map of the transmission line corridor to determine the curvature constraint adjustment requirements of the conductor path. Specifically, elevation sampling point data corresponding to the positioning marks of each tower are extracted from the elevation distribution map of the transmission line corridor, and the elevation data between adjacent towers are segmented according to the direction of the transmission line. By calculating the elevation difference between the positioning marks of adjacent towers and their projection distance in the horizontal direction, the terrain slope change value of the corresponding section is obtained, thereby forming a slope data sequence reflecting the degree of terrain undulation along the line. The slope data not only characterizes the overall undulation trend of the terrain, but also provides a quantitative basis for subsequent conductor sag analysis.

[0052] After obtaining terrain slope change data, the impact of slope changes on the sag curve of the guide wire was further analyzed by combining the altitude layer switching strategy and terrain following mode used in the UAV inspection process. Altitude layer switching refers to dividing the UAV inspection path into multiple altitude levels based on terrain altitude changes and switching between different levels to adapt to sudden terrain changes. Terrain following mode refers to the UAV following the terrain contour as closely as possible to the terrain during inspection, thereby ensuring the accuracy of the inspection of the guide wire and its associated components. Specifically, based on the aforementioned slope data, potential factors affecting the sag curve of the guide wire are first identified. Sections where sag is significantly affected, such as when the slope exceeds a preset threshold (e.g., 10 degrees), are identified as sag-sensitive sections. Subsequently, the tension variation of the conductor sag curve is calculated based on the slope value of this section. The conductor sag curve characterizes the spatial sag shape of the conductor under the combined action of gravity and tension. Its geometric shape can be described using a catenary model, where the conductor is considered a uniform flexible body. Ignoring lateral loads such as wind loads, its sag shape in the vertical plane can be approximated by the catenary equation.

[0053]

[0054] in, Represents the distance coordinates along the horizontal direction. This indicates the vertical displacement of the conductor at the corresponding position relative to the lowest point. For hyperbolic cosine functions, the parameters are... ρ is the catenary constant, used to characterize the proportional relationship between the tension in a conductor and the weight per unit length. Its physical meaning can be expressed as:

[0055]

[0056] in, This represents the horizontal tension component of the conductor at its lowest point. Let g be the gravitational force per unit length of the conductor. Therefore, the parameter... It is directly related to the tension of the conductor; when the tension increases, As the value increases, the sag curve tends to flatten; when the tension decreases, As the value decreases, the sag of the sag curve increases. By mapping the aforementioned changes in terrain slope to variations in spatial height difference and horizontal distance between adjacent towers, the tension variation trend of the conductor in different slope sections can be further derived, and the parameters in the catenary model can be dynamically adjusted accordingly. This allows for the acquisition of sag curve shapes that match actual terrain conditions. This not only establishes a clear mathematical correspondence between terrain slope changes and conductor sag characteristics, but also enables UAV inspection paths to more realistically match the spatial distribution of conductors under complex terrain conditions.

[0057] Building upon this, further analysis of the curvature deviation of the sag curve in the altitude switching region is conducted by considering the location of the altitude layer switching point. Specifically, in mountainous or hilly scenarios with significant slope changes, if the UAV directly follows the path at a lower altitude level, the slope change may amplify the curvature deviation of the sag curve. However, by switching to a higher altitude level at an appropriate location, the stretching effect of the terrain slope on the sag curve can be reduced, thereby mitigating the risk of abrupt changes in path curvature. Simultaneously, in terrain-following mode, the impact of slope changes on the stability of the sag curve needs to be evaluated to ensure that the sag curve remains within the preset safe curvature range after path adjustments. For example, during power transmission line inspection, if the analysis results show that slope changes may cause the sag curve to deviate from its design state by more than 5%, it can be determined that the accuracy parameters of the terrain-following mode need to be improved to avoid detection blind spots during UAV inspections. The system addresses attitude instability issues. It also records the mapping relationship between terrain slope values ​​and corresponding sag deviations for each segment, providing structured data support for subsequent path optimization. After analyzing the slope-sag relationship, the curvature of the UAV's inspection path along the guide wire is calculated to determine the adjustment requirements for curvature constraints. Specifically, the UAV inspection path is treated as a continuous curve, and its curvature is calculated. By calculating the first and second derivatives of the path function, the curvature values ​​corresponding to each path point are obtained, and these curvature values ​​are compared with the safe curvature range obtained from sag analysis. When the curvature value of a certain path segment approaches or exceeds the safe threshold, it is determined that the segment has a curvature constraint adjustment requirement. For segments with drastic slope changes, the corresponding curvature adjustment parameters, such as the maximum allowable curvature limit or curvature change rate limit, need to be recorded further for use as constraints in subsequent path replanning or smoothing.

[0058] Building upon this, to avoid overly conservative or aggressive path adjustments due to relying solely on a single curvature threshold, a hierarchical curvature constraint model is further introduced to optimize curvature constraint adjustment requirements. This hierarchical curvature constraint model, based on a highly hierarchical structure, divides the UAV inspection path into multiple levels, including low, middle, and high levels, and sets different curvature constraints for different levels. For example, stricter curvature constraints are applied to low-level paths close to terrain or guide lines, while relatively smaller curvature variations are allowed in high-level paths. During the construction of the hierarchical curvature constraint model, slope data, initial path curvature, and sag deviation data are used as inputs. This is achieved by constructing a path curvature... The loss function is set as the target of matching the terrain features, and the curvature adjustment parameters are solved using an iterative optimization method to minimize the deviation between the path curvature and the terrain features. In practical applications, the above-mentioned hierarchical curvature constraint model can also be dynamically extended by combining real-time data such as wind speed fluctuations. This allows for continuous optimization of path curvature constraint adjustment requirements in small-scale, refined inspection scenarios of power transmission lines in complex terrain. The application of this hierarchical curvature constraint model can significantly improve the matching degree between the UAV inspection path and the actual terrain features, enhance path stability and inspection safety, and provide a reliable curvature constraint basis for subsequent obstacle avoidance, risk assessment, and flight control command generation.

[0059] Step S2: Based on the obstacle location information obtained by the obstacle avoidance sensor, replan the UAV flight trajectory, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in conjunction with the wind speed fluctuation signal to generate a detection path sequence; specifically including:

[0060] Obstacle location information is obtained through obstacle avoidance laser scanning data; the flight trajectory is replanned based on the optimized path curvature constraint adjustment requirements; a path smoothing algorithm is used to reduce abrupt changes in trajectory curvature; wind speed fluctuation compensation signals collected in real time by an airborne wind speed sensor are introduced, and the smoothed flight trajectory is adaptively corrected and its position and attitude changes in space are dynamically adjusted through the wind speed fluctuation compensation signals, and an adjusted detection path sequence is generated, so that the path avoids obstacles and adapts to wind field changes.

[0061] Specifically, after determining the path curvature constraint adjustment requirements, in order to enable the UAV to safely avoid obstacles and stably adapt to dynamic wind field environments while meeting curvature constraints during power line inspections in complex terrain, it is necessary to replan the UAV's flight trajectory based on obstacle location information obtained from obstacle avoidance sensors, and generate an adjusted inspection path sequence for insulator string detection based on this. Specifically, during the UAV's inspection along the power line, the obstacle location information within the inspection area is first obtained using an obstacle avoidance laser scanning device mounted on the UAV. This obstacle avoidance laser scanning device preferably uses lidar, which calculates the distance to surrounding objects by periodically emitting laser pulses and measuring their return time, thereby obtaining... Point cloud data containing spatial coordinate information is used to improve the collected laser scanning point cloud data. Clustering algorithms are applied to group the point cloud data to distinguish between terrain background points and potential obstacle points. In this embodiment, the K-means clustering algorithm is preferred. By iteratively calculating the centroid of each cluster and minimizing the intra-cluster variance, the point cloud data is divided into multiple obstacle clusters. Subsequently, the corresponding bounding box information is extracted from each obstacle cluster to obtain the spatial location, size, and relative height parameters of the obstacles. For example, in the scenario of power transmission line inspection in complex terrain, the above processing can effectively separate obstacles such as trees and rocks and determine their positional relationship with the towers or conductors, thereby providing accurate obstacle avoidance constraints for subsequent path replanning.

[0062] After obtaining obstacle location information, the flight trajectory of the UAV is replanned based on the aforementioned path curvature constraint adjustment requirements. Specifically, the path curvature constraint adjustment requirements are quantified into the maximum allowable curvature value, and a feasible flight space model is constructed in conjunction with the current flight status of the UAV. In the feasible flight space model, the inspection area is discretized into a grid map, and the obstacle-occupied area and the passable area are marked on the map. Then, the flight trajectory is replanned using a path search algorithm. In this embodiment, the A* algorithm is preferred. By introducing a cost evaluation mechanism with Euclidean distance as the heuristic function, the flight path with the minimum cost is searched under the conditions of satisfying obstacle avoidance constraints and curvature constraints. The A* algorithm comprehensively considers the cost of the already traveled path and the estimated remaining cost during the search process, thereby avoiding getting trapped in a local optimum path and ensuring that the generated flight trajectory has better safety and efficiency overall. For example, when wind speed fluctuations or terrain undulations cause the curvature of the original path to exceed the preset threshold, the path replanned by the A* algorithm, although its length increases, its curvature level is significantly reduced, thereby improving the flight stability of the UAV in complex environments and reducing energy consumption risks.

[0063] After completing the path replanning based on obstacle avoidance and curvature constraints, to avoid the adverse effects of sharp corners or curvature abrupt changes in the path on the UAV's attitude control, further curvature continuity processing is applied to the replanned flight trajectory. Specifically, key path points are selected as control points from the replanned trajectory, and a path smoothing algorithm is used to interpolate the path to obtain a continuous and smooth flight trajectory. The Bezier curve interpolation method is preferred, as it effectively eliminates curvature abrupt changes in the trajectory by constructing parameterized curves and ensuring their continuity at the first and second derivative levels. This transforms paths with obvious corners into trajectories with continuous curvature changes, enabling the UAV to maintain stable flight during inspection missions and reducing attitude jitter and acceleration. The speed impact creates stable flight conditions for subsequent high-precision insulator string inspection. To further address the impact of dynamic environmental factors such as sudden wind speed changes and local turbulence on the flight stability of the UAV during transmission line inspection in complex terrain, a real-time wind field adaptation correction mechanism is introduced into the flight trajectory to achieve dynamic adjustment of the flight trajectory. Specifically, during the UAV's inspection mission, wind speed data in the inspection area is collected in real time by a wind speed sensor installed on the UAV. The wind speed data includes the wind speed magnitude and its fluctuation information over time. Since the real-time collected wind speed data is easily affected by factors such as measurement noise, instantaneous airflow disturbances, and changes in UAV attitude, in this embodiment, the Kalman filter algorithm is preferably used to fuse the collected wind speed data. By establishing a wind speed state prediction model and combining it with real-time measurements from wind speed sensors, the wind speed estimation results are continuously corrected during the iterative process of prediction and updating, thereby minimizing the estimation error and obtaining the optimal wind speed estimate that can represent the wind field state of the current inspection area. This enables real-time correction of the wind speed. The input data of the wind speed state prediction model includes historical wind speed measurement data, current wind speed observation data, and corresponding timestamp information. The wind speed state is predicted and updated through a state transition equation, thereby outputting the optimal wind speed state estimate data corresponding to the location of the UAV at a future time point. The optimal wind speed state estimate data includes the wind speed magnitude and its trend information after filtering and correction, which is used for subsequent wind field model construction and dynamic correction of the flight trajectory.

[0064] After obtaining the wind speed data processed by Kalman filtering, a wind field model is established based on the wind speed data. Specifically, the spatial area where the UAV inspection path is located is taken as the modeling range. This area is spatially discretized to form a wind speed field model composed of multiple grid cells, and the optimal wind speed estimate is mapped to the corresponding spatial grid cell. Subsequently, the discretized wind speed field is processed using the finite difference method. By calculating the wind speed difference between adjacent grid cells, the wind speed gradient in space is approximately solved, thereby obtaining wind speed gradient information that reflects the direction and intensity of wind speed change. The wind speed gradient is used to identify wind speed abrupt change areas or potential turbulence areas within the inspection area, providing a basis for subsequent dynamic trajectory correction. After the wind field model and wind speed gradient calculation are completed, the UAV flight trajectory is dynamically adjusted according to the wind speed gradient. When the wind speed gradient or wind speed value exceeds the preset threshold, it is determined that there is a strong risk of wind field interference in the current path segment, and the flight trajectory is adjusted accordingly based on the direction and magnitude of the wind speed gradient. For example, when the wind speed exceeds 5 meters per second, the drone's flight trajectory is offset by about 2 meters in the horizontal or vertical direction to avoid turbulent areas with drastic wind speed changes. In actual mountainous power transmission line inspection scenarios, when the wind speed suddenly changes from 3 m / s to 7 m / s, the Kalman filter is used to predict the wind speed change trend, and the wind speed gradient information calculated by the finite difference method is combined to correct the flight trajectory. This can effectively reduce the vibration amplitude of the UAV caused by wind and improve the detection accuracy of targets such as insulator strings. In another embodiment, for strong wind environments, when the detected wind speed reaches or exceeds 10 m / s, the execution frequency of wind field adaptation correction is increased, for example, the trajectory correction frequency is increased to 5 times per second, and the trajectory offset is increased accordingly to about 3 meters. This allows the UAV to respond to wind field changes more quickly, thereby maintaining the stability of the flight path under strong wind conditions and avoiding a large deviation of the flight attitude from the inspection target. Through the above methods, the flight trajectory is continuously updated during the inspection process, so that the UAV's flight path can adapt to wind field changes in real time.

[0065] Finally, by integrating the outputs of obstacle avoidance, path replanning, curvature continuity processing, and wind field dynamic correction, an adjusted insulator string detection path sequence is formed. This path sequence consists of multiple consecutive path points, each of which has comprehensively considered the distribution of obstacles, path curvature limitations, and the impact of wind speed fluctuations. This ensures that the UAV can safely avoid obstacles and stably approach the conductors and their associated components to complete the detailed detection of key targets such as insulator strings during the inspection of power transmission lines in complex terrain. It has higher practicality and reliability under complex terrain and dynamic environmental conditions.

[0066] Step S3: Extract height stratification switching points from the detection path sequence, calculate the height abrupt change values ​​between adjacent path segments and the UAV attitude angle deviation values, determine whether the height abrupt change triggers acceleration impact suppression requirements, and generate an inspection risk score for the vegetation intrusion area based on the spatial distribution of the height stratification switching points; Figure 2 As shown, it specifically includes:

[0067] Height stratification switching points are extracted from the detection path sequence. The corresponding height abrupt change value is calculated based on the vertical height difference between adjacent height stratification switching points. Combined with pitch and roll angle data collected by the airborne attitude sensor, the attitude angle deviation value of the UAV during the height switching process is calculated. The height abrupt change value is compared with a preset height abrupt change threshold to determine whether the height abrupt change triggers acceleration impact suppression requirements. A preliminary vegetation intrusion area inspection risk score is generated based on the judgment result and the distribution of height stratification switching points. The risk score quantifies the impact of vegetation intrusion on inspection safety.

[0068] Specifically, after obtaining the insulator string detection path sequence for transmission line inspection, in order to further identify the potential impact of vegetation intrusion on the safety of UAV inspection under complex terrain conditions, the above insulator string detection path sequence is analyzed in depth. Height layer switching points are extracted from it, and inspection risk scores for vegetation intrusion areas are generated accordingly. Specifically, the above insulator string detection path sequence is first traversed. It consists of multiple path nodes arranged in time or space order. Each path node contains corresponding spatial coordinate information. By comparing the height information of adjacent path nodes, the location nodes in the path that switch from one height level to another are identified, thereby determining the location coordinates of the height layer switching points. The above height layer switching points are used to characterize the key locations where the UAV undergoes significant height changes due to terrain undulations or obstacle avoidance requirements during the inspection process.

[0069] After obtaining the altitude layer switching points, the altitude change value and attitude angle deviation value are calculated based on these points. Specifically, based on the spatial coordinate information of the altitude layer switching points, the vertical altitude difference between adjacent altitude layer switching points is calculated, and this height difference is used as the altitude change value to quantify the altitude change amplitude experienced by the UAV during altitude layer switching. For example, under complex terrain conditions, if the altitudes of two adjacent switching points are 150 meters and 180 meters respectively, the corresponding altitude change value is 30 meters. Simultaneously, attitude data collected by the UAV's onboard attitude sensors is used to analyze the attitude changes of the UAV during altitude switching. By performing Euler angle transformation or equivalent attitude calculation on the attitude data, the corresponding attitude angle deviation value is obtained, which characterizes the deviation between the actual flight attitude of the UAV and the attitude of the ideal inspection path. The attitude angle deviation value is then fused with the wind speed fluctuation compensation signal, and the attitude correction value is adjusted according to real-time wind speed changes. Dynamic adjustments are made to reflect the impact of environmental factors on flight attitude stability. This allows for a comprehensive assessment of the UAV's attitude stability during altitude transitions while quantitatively characterizing altitude changes, thereby improving the accuracy of path analysis results and reducing the risk of flight deviations caused by abrupt altitude changes. After calculating the altitude change value and attitude angle deviation correction value, it is further determined whether the altitude change triggers an acceleration shock suppression requirement. Specifically, the altitude change value is compared with a preset altitude change threshold. When the altitude change value exceeds the threshold, it is determined that the UAV may experience a large acceleration change at the corresponding altitude layer transition point, which may lead to aircraft vibration or attitude instability. In this case, it is determined that acceleration shock suppression measures need to be implemented for the flight process corresponding to the altitude layer transition point. This transforms continuous path geometric characteristics into a qualitative or quantitative assessment of flight dynamic performance, providing a basis for subsequent risk scoring.

[0070] Based on this, a preliminary risk score for vegetation intrusion areas is generated according to the judgment results of whether abrupt changes in altitude trigger acceleration impact suppression requirements and the spatial distribution of altitude stratification switching points along the path. Specifically, the judgment results of acceleration impact suppression requirements are incorporated into the risk score calculation as a weighting factor reflecting the degree of path abrupt change risk. At the same time, the distribution density of altitude stratification switching points within a unit path length is statistically analyzed to characterize the degree to which the UAV frequently adjusts its altitude within a specific path segment. By combining the above weighting factors with the distribution density of altitude stratification switching points, a preliminary risk score for vegetation intrusion areas is obtained. This score can reflect the frequency of path altitude changes caused by vegetation intrusion, terrain undulations, and other factors under complex terrain conditions, as well as the potential inspection risks arising therefrom. For example, in densely vegetated areas, because the UAV needs to frequently adjust its altitude to avoid vegetation, the identified altitude stratification switching points are densely distributed, thus significantly increasing the preliminary risk score and indicating a high inspection safety risk in the area.

[0071] Step S4: Based on the risk score of the vegetation invasion area inspection and the wind speed fluctuation data, determine the cross-crossing distance risk quantification index; if the risk quantification index exceeds the preset critical value, then perform smooth transition processing on the height layer switching point to generate a continuous flight control command sequence.

[0072] In step S4, the quantitative indicators for cross-crossing distance risk are determined, including:

[0073] Based on the risk score of the vegetation intrusion area inspection combined with the critical value of the cross-crossing distance stability; and the comprehensive analysis of wind speed fluctuation compensation data; the cross-crossing distance risk quantification index is determined by using the height layer switching point time interval calculation method; the risk quantification weight is adjusted for different wind speed fluctuation intensities; and the final cross-crossing distance risk quantification index is generated.

[0074] Specifically, after obtaining the risk score for vegetation intrusion area inspection, to further assess the safety of UAVs crossing conductors, towers, or terrain obstacles when performing power line inspection tasks under complex terrain conditions, it is necessary to determine a quantitative index of crossing distance risk based on the aforementioned vegetation intrusion area inspection risk score. Specifically, a preliminary vegetation intrusion area inspection risk score based on height-layer switching point analysis is first obtained. This risk score characterizes the potential risk level of the path due to factors such as vegetation intrusion and terrain undulations, with a numerical range of 0 to 10, where a higher score indicates a higher risk level. The aforementioned vegetation intrusion area risk score is then combined with a preset critical value for crossing distance stability. This critical value is preferably obtained through statistical analysis of historical power line inspection data and is used to characterize the critical distance at which vegetation or other obstacles may affect conductor stability within a specific distance range. For example, when the crossing distance stability is critical... When the threshold is set to 50 meters, it indicates that if there is a high risk of vegetation intrusion within this distance range, it may adversely affect the sag of the conductor or the flight stability of the UAV. By comparing and analyzing the standard risk score and standard critical value obtained after normalizing the inspection risk score of the vegetation intrusion area with the above-mentioned cross-crossing distance stability critical value, when the above-mentioned standard risk score exceeds the above-mentioned standard critical value, the corresponding path segment is marked as a potentially unstable area. For example, in the scenario of inspecting power transmission lines in complex terrain, when the standard score corresponding to the inspection risk score of the vegetation intrusion area of ​​a certain path segment is 0.5, while the standard critical value corresponding to the cross-crossing distance stability critical value is 0.4, it indicates that the vegetation in this area may have approached the conductor or the flight path of the UAV. It is necessary to comprehensively consider factors such as vegetation intrusion and conductor sag changes in the subsequent path generation process to avoid stability loss or safety risks during the crossing flight, so as to identify potential unstable crossing segments in the early stage.

[0075] After identifying potentially unstable areas, to further incorporate real-time environmental factors into the cross-crossing risk assessment process, a comprehensive analysis is performed on the wind speed fluctuation compensation data superimposed on these potentially unstable areas. This allows the risk assessment results to dynamically reflect the actual changes in the inspection environment. Specifically, airborne sensors mounted on the drone acquire wind speed fluctuation compensation signals in real time, and compensation data for risk analysis is extracted from these signals. This compensation data includes at least the average wind speed and the wind speed fluctuation amplitude during the inspection process, used to characterize the overall intensity and variation characteristics of the wind field in the current inspection area. After extracting the wind speed compensation data, it is superimposed on the marked potentially unstable areas using a weighted fusion method. Wind speed compensation data and vegetation intrusion area inspection risk scores are comprehensively calculated. The vegetation intrusion area inspection risk score reflects the static risk characteristics caused by topographic relief and vegetation intrusion, while wind speed compensation data reflects the dynamic risk characteristics caused by real-time wind field changes. By normalizing the above different types of data and setting corresponding weight coefficients, such as setting the weight of wind speed compensation data to be less than that of vegetation intrusion area inspection risk score, the fusion result can maintain sensitivity to topographic and vegetation risks while appropriately reflecting the impact of real-time environmental factors. This results in a fusion result that reflects the comprehensive risk level of potentially unstable areas, effectively improving the risk assessment's adaptability to real-time environmental changes and avoiding judgment bias caused by relying solely on static risk information.

[0076] After integrating risk scoring and wind speed compensation data, a time interval analysis method for altitude layer switching points is further introduced to quantify the risk of crossing distances. Specifically, altitude layer switching points are identified from the adjusted insulator string detection path sequence, and the time interval between adjacent altitude layer switching points is calculated based on the UAV's flight speed and the path distance between adjacent switching points. This time interval reflects the frequency with which the UAV adjusts its altitude within a short period. When the time interval is small, it indicates that the UAV needs to frequently switch altitude levels, which may increase the risk of flight attitude changes and acceleration impacts. Therefore, when the time interval between adjacent altitude layer switching points is less than a preset threshold, the segment is determined to be a high-frequency switching segment, and the risk quantification index value is increased accordingly. Based on this, the product of the above fusion result and the altitude layer switching frequency factor is calculated and used as the risk quantification index. The switching frequency factor can be determined by multiplying the reciprocal of the time interval by a preset constant, so that the above risk quantification index can reflect the combined impact of path temporal characteristics and environmental risks.

[0077] In another implementation, to further enhance the adaptability of risk assessment to complex environments, a path curvature constraint factor can be introduced into the quantification process. When the path curvature exceeds a preset curvature threshold, the altitude layer switching frequency factor can be amplified to reflect the impact of curvature changes on crossing risks. This improves the sensitivity of risk assessment in scenarios with steep terrain or drastic wind speed fluctuations, allowing the risk quantification index to not only reflect the frequency of altitude changes but also comprehensively consider the impact of path geometry on flight stability. Furthermore, to enable the risk quantification index to dynamically respond to wind speed fluctuations of varying intensities... Furthermore, the risk quantification weights are adjusted based on the intensity of wind speed fluctuations. Specifically, wind speed fluctuations are divided into low, medium, and high levels, and the weight ratios of risk scores and wind speed compensation data are set for different levels. For example, in scenarios with small wind speed fluctuations, the weight of the risk score for vegetation intrusion areas can be appropriately increased; while in scenarios with large wind speed fluctuations, the weight of wind speed compensation data is increased, making the risk assessment results more consistent with the real-time environmental conditions. By dynamically adjusting the weights and recalculating the risk quantification indicators, it can be ensured that the obtained risk assessment results can truly reflect the comprehensive risk level under the current inspection environment.

[0078] Finally, by comparing the risk quantification results after weight adjustment and multi-factor fusion with the preset risk threshold, the final crossover distance risk quantification index is generated. When the risk quantification index exceeds the preset threshold, it is determined that the corresponding path segment has a high crossing risk, and smooth transition processing or other risk suppression measures need to be triggered in the subsequent path generation process; otherwise, the path segment is considered to meet the safety inspection requirements. Through the above technical solution, this embodiment realizes a multi-dimensional quantitative assessment of crossover distance risk, so that vegetation intrusion risk, wind speed fluctuation characteristics, and flight path temporal characteristics are comprehensively reflected in a unified framework, thereby providing a reliable and calculable risk decision basis for the automatic navigation path generation method for power transmission line inspection in complex terrain.

[0079] Further, in step S4, a continuous sequence of flight control commands is generated, including:

[0080] When the crossover distance risk quantification index exceeds the preset critical value, attitude angle deviation correction technology is used to adjust the flight attitude of the UAV during the altitude layer switching process for the altitude layer switching point. By continuously correcting the pitch angle and roll angle of the UAV, the flight trajectory corresponding to the altitude layer switching point is smoothly transitioned. The flight trajectory after attitude angle deviation correction and smoothing is transformed into continuous flight control commands, generating a continuous flight control command sequence.

[0081] Specifically, the aforementioned crossover distance risk quantification index is compared with the aforementioned preset critical value, wherein the aforementioned preset critical value is obtained through historical inspection data statistics and can be set under the condition of wind speed fluctuation compensation superimposed data, so as to reflect the system's higher requirements for track smoothness when wind disturbance intensifies; when the aforementioned risk quantification index exceeds the aforementioned preset critical value, it is determined that the altitude layer switching point in the UAV inspection path needs to be transitionally optimized to avoid instantaneous acceleration change, attitude overshoot and yaw oscillation caused by directly switching according to the original planned altitude layer.

[0082] After triggering transition optimization, the altitude layer switching points are identified and parameterized. Attitude angle deviation correction technology is used to achieve smooth trajectory transition and controllable tracking. These altitude layer switching points are key nodes where the inspection track switches between adjacent altitude layers. They are determined by the path generation module based on terrain elevation, spatial alignment of the route, obstacle safety clearances, and shooting angle constraints. Since altitude layer switching in complex terrain is often accompanied by curvature changes and velocity vector retargeting, the time interval between switching points is further analyzed to establish the correspondence between the switching point time scale, altitude change amplitude, and attitude response requirements. Based on this, the attitude angle deviation value is calculated, which is used to quantify the current flight attitude. The deviation from the ideal path attitude is determined by the smoothed reference trajectory tangent direction and the continuity of the desired trajectory curvature. The actual attitude is obtained from the pitch and roll angles output in real time by the flight controller, and the difference between the two constitutes the error signal for closed-loop control. This error signal is then input into a proportional-integral-derivative (PID) control algorithm for continuous correction: the proportional term is used to quickly suppress deviation growth and improve response speed near the switching point; the integral term is used to accumulate and eliminate steady-state deviations caused by continuous wind disturbances or changes in altitude; and the derivative term is used to feedforward and suppress the rate of change of deviation to reduce overshoot and oscillation risks. Through this closed-loop adjustment, the rate of change of acceleration of the UAV at the switching point is constrained within a set range, such as... Figure 3 As shown, the altitude change, which might otherwise appear as a broken line or a step, is transformed into a continuous change process that satisfies dynamic constraints, ensuring that the aircraft maintains controllable attitude margin and imaging stability throughout the switching process. Furthermore, to enhance environmental adaptability in complex terrain, real-time wind field adaptation correction is integrated into the altitude change detection. This involves using the wind speed fluctuation compensation signal as a dynamic gain factor for attitude correction and establishing a mapping relationship between wind speed and attitude angle deviation correction. This allows attitude correction to depend on geometric switching requirements and also to adaptively adjust with the intensity of wind field disturbances. For example, in complex terrain channel winds or leeward vortex areas, when the wind speed fluctuation compensation signal shows a wind speed of 5 m / s, the attitude angle deviation correction value is adjusted to 3° to improve the wind disturbance resistance of the switching segment and reduce the structural impact and track deviation caused by sharp turns. This achieves a smooth transition at altitude layer switching points and improves the safety and stability of the inspection flight.

[0083] After completing the attitude angle deviation correction and wind field adaptation correction, the smoothed altitude layer switching point sequence is transformed into continuous flight control commands, thereby generating a continuous flight control command sequence for metal fitting corrosion trace detection and executing power transmission line inspection tasks. The smoothed trajectory output by the specific path generation module is no longer directly issued in the form of discrete inflection points, but is first resampled into a time-series reference point set that satisfies the control cycle. In each control cycle, the corresponding expected position coordinates, velocity vector, and optional acceleration constraint parameters are generated for the UAV. The position coordinates are used to constrain the UAV's spatial position tracking, and the velocity vector is used to constrain the heading continuity and shooting speed stability. The speed and velocity vectors, together with the aforementioned acceleration rate of change constraints, ensure the continuity and differentiability of the trajectory in the time domain. Meanwhile, the detection of metal fitting corrosion marks requires consistent image clarity and target identifiability. Therefore, during command generation, the velocity vector is kept to transition smoothly within the switching segment, and the attitude change rate is limited, making the change of the camera's line of sight relative to the target more continuous, thereby reducing the risk of missed detections caused by motion blur and sudden changes in viewpoint. The final generated flight control command sequence is used to control the UAV to perform complex terrain power line inspection tasks along a smooth and continuous inspection path, thereby improving the path continuity and imaging stability of metal fitting corrosion mark detection while ensuring the safety margin of the crossing segment.

[0084] Step S5: By simulating the coverage and energy consumption data of the flight control command sequence, determine whether the boundary integrity standard is met, and generate the final automatic navigation path sequence based on the determination result, specifically including:

[0085] The simulation results of vibration damper offset coverage and energy consumption are obtained from the flight control command sequence; it is determined whether the simulation results meet the preset line corridor boundary integrity standard; based on the determination results, an automatic navigation path sequence for small-scale fine inspection of transmission lines in complex terrain is generated, wherein the line corridor boundary integrity standard is used to characterize the effective coverage of the inspection path on key components and spatial range of the transmission line corridor.

[0086] Specifically, after generating a continuous sequence of flight control commands, to ensure that the inspection path can fully cover key components within the transmission line corridor under complex terrain conditions while also meeting the engineering requirements for controlled UAV energy consumption, the flight control command sequence is further simulated and analyzed to determine whether it meets the line corridor boundary integrity standard. Based on this, the final automatic navigation path sequence for actual inspection is generated. Specifically, the adjusted insulator string detection path sequence and its corresponding flight control command sequence are used as input data to simulate and analyze the possible offset of the vibration damper during the inspection process. The vibration damper is an auxiliary hardware installed on the transmission line, usually located in the conductor section near the tower, to suppress the vibration of the conductor under wind, thereby reducing the risk of fatigue damage to the conductor and its connecting hardware. Since the vibration damper swings with the conductor as a whole under wind, it may have a certain spatial offset during the inspection process. Therefore, during the automatic navigation path generation process, the possible offset of the vibration damper is simulated and analyzed. The offset region and the coverage of the inspection path help improve the inspection integrity and safety of key components of the conductor. The aforementioned anti-vibration hammer offset coverage rate is used to characterize the degree to which the generated inspection path covers the spatial offset region of the anti-vibration hammer that may be caused by factors such as wind speed disturbance. To improve the reliability of the evaluation results, the Monte Carlo simulation method generates multiple sets of random offset scenarios under given wind speed fluctuation conditions and terrain undulation constraints, and counts the coverage ratio of the flight path to the potential position of the anti-vibration hammer in each offset scenario, thereby obtaining the anti-vibration hammer offset coverage rate result. At the same time, the energy consumption of the flight control command sequence is calculated. By integrating the dynamic parameters such as speed and acceleration of each flight segment in the path sequence, the total energy consumption of the UAV when executing the inspection path is estimated to obtain the corresponding energy consumption calculation result. By jointly analyzing the coverage result and the energy consumption result, the performance of the inspection path in terms of both coverage integrity and flight efficiency can be evaluated simultaneously.

[0087] After obtaining the vibration damper offset coverage rate and energy consumption calculation results, the simulation results are compared with the preset line corridor boundary integrity standard. This standard characterizes the effective coverage of the inspection path over key components within the transmission line corridor, including but not limited to conductors, vibration dampers, and hardware, as well as their spatial range. Energy consumption constraints are also considered to avoid overly conservative or complex path planning. For example, if the simulated vibration damper offset coverage rate is not lower than a preset threshold and the corresponding energy consumption does not exceed a preset energy consumption limit, the flight control command sequence is deemed to meet the line corridor boundary integrity standard. Conversely, if the results do not meet the standard, the current path is considered insufficient in terms of coverage integrity or energy efficiency. If the result meets the line corridor boundary integrity standard... Under standard conditions, the final automatic navigation path sequence is generated directly based on the simulation analysis results. Specifically, the flight control command sequence verified through simulation is integrated with the refined elevation distribution map of the route corridor boundary and wind speed fluctuation compensation data. This ensures that the final generated automatic navigation path sequence can maintain complete coverage of the route corridor spatial boundary under complex terrain conditions and adapt to environmental changes such as different slopes and wind speed fluctuations. In high-slope terrain or scenarios with severe terrain undulations, the above path sequence can automatically switch to terrain following mode and, in conjunction with obstacle avoidance laser scanning data, cover the sag curves between adjacent towers. In areas with large wind speed fluctuations, flight stability is maintained through attitude angle deviation correction and dynamic compensation mechanisms, thereby improving inspection accuracy.

[0088] If the judgment result does not meet the above-mentioned line corridor boundary integrity standard, the current simulation result is used as feedback input, and steps S1-S4 are repeated to re-optimize the flight control command sequence and perform coverage and energy consumption simulation again until the above-mentioned integrity standard is met. Through the above-mentioned path generation mechanism based on simulation evaluation and iterative optimization, reliable generation and verification of automatic navigation paths are realized in the scenario of small-scale fine inspection of transmission lines in complex terrain, reducing the need for manual intervention and significantly improving inspection safety and overall operation efficiency.

[0089] This invention also provides an automatic navigation system for unmanned aerial vehicles (UAVs) used for power line inspection, for implementing the above-mentioned method, such as... Figure 4 As shown, the system includes:

[0090] The terrain and wind field sensing unit is used to collect terrain elevation data and wind speed fluctuation signals in the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary. Based on the distribution map, the terrain slope change between adjacent towers is calculated. Combining the height layer switching and terrain following mode, the influence of terrain slope on the conductor sag curve is analyzed to determine the path curvature limit adjustment requirements when the UAV inspects the conductor.

[0091] The obstacle avoidance path planning unit is used to replan the UAV flight trajectory based on the obstacle position information obtained by the obstacle avoidance sensor, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in combination with the wind speed fluctuation signal to generate a detection path sequence.

[0092] The hierarchical risk assessment unit is used to extract height hierarchical switching points from the detection path sequence, calculate the height mutation value between adjacent path segments and the UAV attitude angle deviation value, determine whether the height mutation triggers the acceleration impact suppression requirement, and generate an inspection risk score for the vegetation invasion area based on the spatial distribution of the height hierarchical switching points.

[0093] The control command generation unit is used to determine the cross-crossing distance risk quantification index based on the inspection risk score and wind speed fluctuation data; if the risk quantification index exceeds the preset critical value, the altitude layer switching point is smoothed to generate a continuous flight control command sequence.

[0094] The navigation path acquisition unit is used to determine whether the boundary integrity standard is met by simulating the coverage and energy consumption data of the flight control command sequence, and to generate the final automatic navigation path sequence based on the determination result.

[0095] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0096] In summary, this invention collects terrain elevation data and wind speed fluctuation signals from the inspection area, generates a refined elevation distribution map of the line corridor boundary, and analyzes the sag characteristics of the conductor by combining terrain slope changes between adjacent towers, height layer switching, and terrain following modes. This allows for the determination of reasonable path curvature constraint adjustment requirements from the source, providing constraints for trajectory planning that conform to the terrain and conductor physical characteristics. Under the guidance of these constraints, obstacle avoidance sensors are used to acquire obstacle location information, and the UAV flight trajectory is replanned. Curvature continuity processing eliminates abrupt changes in the trajectory, while wind speed fluctuation signals are used for dynamic correction, thereby generating a detection path sequence that balances obstacle avoidance safety, curvature smoothness, and environmental adaptability. By extracting height layer switching points from the detection path sequence and calculating height abrupt changes and attitude angle deviations, the invention can accurately identify flight dynamic changes caused by terrain undulations or obstacle avoidance. The system determines whether there is a need to suppress acceleration impacts. Simultaneously, it generates a risk score for vegetation intrusion areas by combining the spatial distribution of altitude stratification switching points, extending path analysis from a geometric level to a safety risk level. Furthermore, it integrates the inspection risk score with wind speed fluctuation data to form a quantitative indicator of cross-crossing distance risk. When the risk exceeds a threshold, a smooth transition is implemented at the altitude stratification switching points, generating a continuous sequence of flight control commands to effectively suppress attitude changes and acceleration impacts, thus improving flight stability. By simulating the coverage and energy consumption of the flight control command sequence, the system verifies whether the path meets the line corridor boundary integrity standard, and outputs the final automatic navigation path sequence accordingly. Through the synergy of these technical solutions, a closed-loop optimization from environmental perception, path planning, risk assessment to control verification is achieved, significantly improving the path stability, coverage integrity, and operational safety of UAVs in power line inspections in complex terrain.

[0097] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0098] If the integrated unit is implemented as 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, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic navigation method for unmanned aerial vehicles (UAVs) used for power transmission line inspection, characterized in that, The method includes: Step S1: Collect terrain elevation data and wind speed fluctuation signals of the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary; calculate the terrain slope change between adjacent towers based on the distribution map, and analyze the influence of terrain slope on the sag curve of the conductor by combining height layer switching and terrain following mode, and determine the path curvature limit adjustment requirements when the UAV inspects the conductor. Step S2: Based on the obstacle location information obtained by the obstacle avoidance sensor, replan the UAV flight trajectory, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in combination with the wind speed fluctuation signal to generate a detection path sequence. Step S3: Extract height stratification switching points from the detection path sequence, calculate the height mutation value between adjacent path segments and the UAV attitude angle deviation value, determine whether the height mutation triggers the acceleration impact suppression requirement, and generate an inspection risk score for the vegetation invasion area based on the spatial distribution of the height stratification switching points. Step S4: Based on the inspection risk score and wind speed fluctuation data, determine the cross-crossing distance risk quantification index; if the risk quantification index exceeds the preset critical value, perform smooth transition processing on the altitude layer switching point to generate a continuous flight control command sequence. Step S5: By simulating the coverage and energy consumption data of the flight control command sequence, determine whether the boundary integrity standard is met, and generate the final automatic navigation path sequence based on the determination result; The process of generating a refined elevation distribution map of the line corridor boundary includes: collecting elevation sampling data of complex terrain using lidar and acquiring wind speed fluctuation compensation signals using airborne sensors; using a preset threshold to perform preliminary filtering on the elevation data around the tower positioning markers to separate the effects of terrain undulations and wind speed fluctuations; and generating a refined elevation distribution map of the line corridor based on the filtered elevation data. The method for determining the risk quantification index of the crossing distance includes: combining the risk score of the vegetation intrusion area inspection with the critical value of the crossing distance stability; performing a comprehensive analysis by superimposing wind speed fluctuation compensation data; determining the risk quantification index of the crossing distance using the height layer switching point time interval calculation method; adjusting the risk quantification weight for different wind speed fluctuation intensities; and generating the final risk quantification index of the crossing distance.

2. The method as described in claim 1, characterized in that, Step S1, generating the refined elevation distribution map of the route corridor boundary, also includes: The wind speed fluctuation compensation signal is time-domain aligned to ensure the synchronization of elevation data and wind speed data; noise points are further removed through multi-layer threshold filtering to obtain the final elevation distribution map of the line corridor.

3. The method as described in claim 2, characterized in that, In step S1, the path curvature constraint adjustment requirements during UAV inspection along the guide wire include: Based on the elevation distribution map of the line corridor, the elevation difference and horizontal distance between adjacent towers are calculated to obtain the terrain slope changes. Combining the height layer switching strategy and terrain following mode, the impact of different slope sections on the spatial shape of the conductor sag curve is analyzed, and the curvature of the UAV's following path for conductor inspection is calculated. The calculated path curvature is compared with the safe curvature range to determine the path curvature limit adjustment requirements for corresponding sections. For sections where the terrain slope change exceeds the preset threshold, the corresponding curvature adjustment parameters are recorded, and a layered curvature constraint model is introduced. Differentiated curvature constraints are set according to different height levels. Through iterative optimization, the adjusted path curvature is matched with the terrain features and conductor sag characteristics to generate optimized path curvature limit adjustment requirements.

4. The method as described in claim 1, characterized in that, In step S2, a detection path sequence is generated, including: Obstacle location information is obtained through obstacle avoidance laser scanning data; the flight trajectory is replanned based on the path curvature limitation adjustment requirements; a path smoothing algorithm is used to reduce abrupt changes in trajectory curvature; wind speed fluctuation compensation signals collected in real time by an airborne wind speed sensor are introduced, and the smoothed flight trajectory is adapted and corrected through the wind speed fluctuation compensation signals, and the position and attitude changes in space are dynamically adjusted, and an adjusted detection path sequence is generated, so that the path avoids obstacles and adapts to wind field changes.

5. The method as described in claim 4, characterized in that, In step S3, it is determined whether a sudden change in altitude triggers an acceleration impact suppression requirement, and based on the spatial distribution of the altitude stratification switching points, an inspection risk score for the vegetation intrusion area is generated, including: Height stratification switching points are extracted from the detection path sequence. The corresponding height abrupt change value is calculated based on the vertical height difference between adjacent height stratification switching points. Combined with pitch and roll angle data collected by the airborne attitude sensor, the attitude angle deviation value of the UAV during the height switching process is calculated. The height abrupt change value is compared with a preset height abrupt change threshold to determine whether the height abrupt change triggers acceleration impact suppression requirements. A preliminary vegetation intrusion area inspection risk score is generated based on the judgment result and the distribution of height stratification switching points. The risk score quantifies the impact of vegetation intrusion on inspection safety.

6. The method as described in claim 1, characterized in that, In step S4, a continuous sequence of flight control commands is generated, including: When the crossover distance risk quantification index exceeds the preset critical value, attitude angle deviation correction technology is used to adjust the flight attitude of the UAV during the altitude layer switching process for the altitude layer switching point. By continuously correcting the pitch angle and roll angle of the UAV, the flight trajectory corresponding to the altitude layer switching point is smoothly transitioned. The flight trajectory after attitude angle deviation correction and smoothing is transformed into continuous flight control commands, generating a continuous flight control command sequence.

7. The method as described in claim 1, characterized in that, In step S5, the final autonavigation path sequence is generated, including: The simulation results of vibration damper offset coverage and energy consumption are obtained from the flight control command sequence; it is determined whether the simulation results meet the preset line corridor boundary integrity standard; based on the determination results, an automatic navigation path sequence for small-scale fine inspection of transmission lines in complex terrain is generated, wherein the line corridor boundary integrity standard is used to characterize the effective coverage of the inspection path on key components and spatial range of the transmission line corridor.

8. An automatic navigation system for unmanned aerial vehicles (UAVs) used for power transmission line inspection, for implementing the method as described in any one of claims 1-7, characterized in that, The system includes: The terrain and wind field sensing unit is used to collect terrain elevation data and wind speed fluctuation signals in the inspection area, and preprocess them to generate a refined elevation distribution map of the line corridor boundary. Based on the distribution map, the terrain slope change between adjacent towers is calculated. Combining the height layer switching and terrain following mode, the influence of terrain slope on the conductor sag curve is analyzed to determine the path curvature limit adjustment requirements when the UAV inspects the conductor. The obstacle avoidance path planning unit is used to replan the UAV flight trajectory based on the obstacle position information obtained by the obstacle avoidance sensor, perform curvature continuity processing on the replanned flight trajectory, and dynamically correct the flight trajectory in combination with the wind speed fluctuation signal to generate a detection path sequence. The hierarchical risk assessment unit is used to extract height hierarchical switching points from the detection path sequence, calculate the height mutation value between adjacent path segments and the UAV attitude angle deviation value, determine whether the height mutation triggers the acceleration impact suppression requirement, and generate an inspection risk score for the vegetation invasion area based on the spatial distribution of the height hierarchical switching points. The control command generation unit is used to determine the cross-crossing distance risk quantification index based on the inspection risk score and wind speed fluctuation data; if the risk quantification index exceeds the preset critical value, the altitude layer switching point is smoothed to generate a continuous flight control command sequence. The navigation path acquisition unit is used to determine whether the boundary integrity standard is met by simulating the coverage and energy consumption data of the flight control command sequence, and to generate the final automatic navigation path sequence based on the determination result.

9. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-7.

Images