Real-time tree barrier detection method, device and equipment on power transmission line, medium and product

Abstract

The application discloses a real-time tree barrier detection method, device and equipment on a power transmission line, and a medium and product. The method comprises the following steps: in the process that a multi-rotor unmanned aerial vehicle flies along the extension direction of a target power transmission line, subscribing to three-dimensional point cloud data and positioning and attitude system data collected by a laser radar in real time through a data acquisition node, and sending the data to a tree barrier analysis node in real time; constructing a three-dimensional model of the target power transmission line and segmenting a conductor point cloud area according to the data received in real time by the tree barrier analysis node; identifying at least one barrier that meets a preset distance condition with the target power transmission line in the power transmission line point cloud area through the tree barrier analysis node, and generating a tree barrier information analysis report matched with the target power transmission line according to three-dimensional position information of each barrier and distance information of each barrier to the target power transmission line. The technical scheme of the embodiment of the application can significantly improve the intelligent level and operation efficiency of power transmission line inspection.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method, apparatus, equipment, medium, and product for real-time tree obstacle detection on power transmission lines. Background Technology

[0002] With the large-scale construction of ultra-high voltage transmission lines, the demand for detecting tree obstructions in transmission corridors is becoming increasingly prominent. Traditional manual inspection methods are limited by factors such as complex terrain and weather conditions, resulting in low efficiency and significant safety hazards, making it difficult to meet the requirements of modern intelligent power grid operation and maintenance.

[0003] While existing detection solutions exist, they still suffer from problems such as insufficient data processing efficiency, low target recognition accuracy, and limited spatial positioning precision. These methods struggle to establish accurate spatial structure models, hindering the rapid and accurate identification and location of potential risk points, thus limiting the improvement of detection effectiveness. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, medium, and product for real-time tree obstruction detection on power transmission lines, so as to achieve accurate and efficient identification of potential tree obstructions around power transmission lines.

[0005] According to one aspect of the present invention, a real-time tree obstacle detection method on a power transmission line is provided, executed by a robot operating system, the robot operating system including a data acquisition node deployed on a multi-rotor drone and a tree obstacle analysis node deployed on a processing platform, the multi-rotor drone being equipped with a lidar, the method comprising:

[0006] During the flight of the multi-rotor UAV along the extension direction of the target power line, the data acquisition node subscribes to the real-time three-dimensional point cloud data and positioning and attitude system data collected by the lidar, and sends the subscribed data to the tree obstacle analysis node in real time.

[0007] Based on the real-time received 3D point cloud data and positioning and attitude system data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line that the multi-rotor UAV flies over, and segments the power transmission line point cloud region in the line point cloud model.

[0008] The tree obstacle analysis node identifies at least one obstacle in the power line point cloud area that meets the preset distance condition to the target power line, and generates a tree obstacle information analysis report that matches the target power line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power line.

[0009] According to another aspect of the present invention, a real-time tree obstacle detection device for power transmission lines is provided, configured in a robot operating system. The robot operating system includes a data acquisition node deployed on a multi-rotor drone and a tree obstacle analysis node deployed on a processing platform. The multi-rotor drone is equipped with a lidar system. The device includes:

[0010] The data acquisition module is used to subscribe to the real-time 3D point cloud data and positioning and attitude system data collected by the lidar through the data acquisition node during the flight of the multi-rotor UAV along the extension direction of the target power line, and send the subscribed data to the tree obstacle analysis node in real time.

[0011] The model building module is used to generate a 3D point cloud model of the target power transmission line that the multi-rotor UAV flies over, based on the real-time received 3D point cloud data and positioning and attitude system data through the tree obstacle analysis node, and to segment the power transmission line point cloud region in the line point cloud model.

[0012] The identification and analysis module is used to identify at least one obstacle in the power transmission line point cloud area that meets the preset distance condition with respect to the target power transmission line through the tree obstacle analysis node, and generate a tree obstacle information analysis report that matches the target power transmission line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power transmission line.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform a real-time tree obstacle detection method on a power transmission line according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement a real-time tree obstacle detection method on a power transmission line as described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the steps of the method as described in any embodiment of the present invention.

[0019] The technical solution of this invention involves the following steps: During the flight of a multi-rotor UAV along the extension direction of a target power transmission line, a data acquisition node first subscribes in real-time to 3D point cloud data collected by a lidar system and positioning and attitude system data, and then sends this data to a tree obstacle analysis node. Subsequently, based on the received 3D point cloud data and positioning and attitude system data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line covered by the UAV's flight trajectory, and accurately segments the power transmission line point cloud region within this model. Finally, the tree obstacle analysis node identifies one or more obstacles within the power transmission line point cloud region that meet preset distance conditions from the target power transmission line. Based on the 3D position information of each obstacle and its distance from the target power transmission line, it automatically generates a tree obstacle information analysis report matching the target power transmission line. This novel real-time tree obstacle detection method on power transmission lines enables accurate identification of potential hazards, significantly improving detection efficiency compared to traditional manual inspections. Furthermore, through automated report generation and visualization analysis functions, it significantly reduces the workload and safety risks for maintenance personnel, providing reliable technical support for smart grid construction.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a real-time tree obstacle detection method on a power transmission line according to Embodiment 1 of the present invention;

[0023] Figure 2 This is a flowchart of another real-time tree obstacle detection method on a power transmission line provided in Embodiment 2 of the present invention;

[0024] Figure 3 This is a schematic diagram of a real-time tree obstacle detection scheme on a power transmission line in a specific scenario applicable to an embodiment of the present invention;

[0025] Figure 4 This is a schematic diagram of the structure of a real-time tree obstacle detection device on a power transmission line according to Embodiment 3 of the present invention;

[0026] Figure 5This is a schematic diagram of the structure of an electronic device that implements a real-time tree obstacle detection method on a power transmission line according to an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention 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 of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus 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 apparatus.

[0029] Example 1

[0030] Figure 1 This is a flowchart of a real-time tree obstacle detection method on a power transmission line provided in Embodiment 1 of the present invention. This embodiment is applicable to the real-time detection of potential tree obstacles around a high-voltage power transmission line when a multi-rotor drone equipped with a lidar performs autonomous inspection. The method can be executed by a real-time tree obstacle detection device on the power transmission line. This device can be implemented in hardware and / or software and is generally configured in a robot operating system. The robot operating system includes a data acquisition node deployed on the multi-rotor drone and a tree obstacle analysis node deployed on a processing platform. The multi-rotor drone is equipped with a lidar.

[0031] Correspondingly, such as Figure 1 As shown, the method includes:

[0032] S110. During the flight of the multi-rotor UAV along the extension direction of the target power line, the data acquisition node subscribes to the real-time three-dimensional point cloud data and positioning and attitude system data collected by the lidar, and sends the subscribed data to the tree obstacle analysis node in real time.

[0033] The data acquisition node can be understood as the core component of the ROS (Robot Operating System) deployed on the edge computing device of the drone, responsible for the unified management of sensor data input. Subscribing to the LiDAR can be understood as the LiDAR acquiring 3D environmental information by emitting laser beams and receiving reflected signals. The data acquisition node obtains the raw point cloud by subscribing to topics, and after parsing, it contains the 3D coordinates and reflection intensity information of each point, providing basic data for subsequent modeling.

[0034] 3D point cloud data can be understood as a set of discrete spatial points generated by LiDAR scanning, with each point containing 3D coordinates and attributes (such as reflection intensity). Positioning and attitude system data can be understood as RTK-GNSS (Real-Time Kinematic Global Navigation Satellite System) and IMU (Inertial Measurement Unit) data from the UAV, including parameters such as latitude and longitude, altitude, roll, pitch, and yaw. The obstacle analysis node can be understood as a core processing module deployed within an edge computing box, running on the ROS framework. For ease of explanation, positioning and attitude system data will be consistently referred to as POS (Position and Orientation System) data in the following text.

[0035] In this embodiment, the data acquisition node acquires 3D point cloud data generated by LiDAR scanning in real time through the ROS topic subscription mechanism, while simultaneously subscribing to the precise location and flight attitude information of the UAV provided by POS data. This node performs time synchronization processing on both types of data to ensure spatiotemporal alignment between the point cloud and the positioning information, and then transmits the synchronized data to the tree obstacle analysis node via the ROS communication interface for subsequent processing.

[0036] S120. Based on the real-time received 3D point cloud data and POS data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line that the multi-rotor drone flies over, and segments the power transmission line point cloud region in the line point cloud model.

[0037] In this embodiment, the tree obstacle analysis node receives synchronized point cloud and positioning data, and dynamically constructs a 3D point cloud model of the power transmission line based on the UAV flight trajectory. A density clustering algorithm is used to extract point cloud clusters with linear distribution characteristics from the model, and the consistency of the conductor direction is verified by combining the heading information provided by the positioning system, ultimately segmenting the precise point cloud region of the power transmission line. Here, a point cloud cluster can be understood as a set of local point clouds with specific spatial distribution characteristics separated from the lidar scanning data.

[0038] S130. Identify at least one obstacle in the power transmission line point cloud area that meets the preset distance condition with respect to the target power transmission line through the tree obstacle analysis node, and generate a tree obstacle information analysis report matching the target power transmission line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power transmission line.

[0039] In this embodiment, the tree obstacle analysis node, based on the established traverse point cloud spatial structure, executes a radius search algorithm to detect potential obstacle points around the traverse. For each detected obstacle point, its minimum distance to the traverse point cloud is calculated. When the distance is less than a preset safety threshold, the three-dimensional coordinates, minimum distance value, and type of obstacle of that point are recorded. Finally, the spatial information and distance data of all potential obstacle points are integrated to generate a standardized tree obstacle analysis report. The radius search algorithm can be understood as a nearest neighbor query method based on spatial indexing, which quickly retrieves all neighboring points within a specified range around the target point by setting a fixed search radius.

[0040] The technical solution of this invention involves the following steps: During the flight of a multi-rotor drone along the extension direction of a target power transmission line, a data acquisition node first subscribes in real-time to 3D point cloud data and POS data collected by a lidar sensor, and sends this data to a tree obstacle analysis node. Subsequently, based on the received 3D point cloud data and POS data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line covered by the drone's flight trajectory, and precisely segments the power transmission line point cloud region within this model. Finally, the tree obstacle analysis node identifies one or more obstacles within the power transmission line point cloud region that meet preset distance conditions from the target power transmission line. Based on the 3D position information of each obstacle and its distance from the target power transmission line, the node automatically generates a tree obstacle information analysis report matching the target power transmission line. This novel real-time tree obstacle detection method on power transmission lines enables accurate identification of potential hazards, significantly improving detection efficiency compared to traditional manual inspections. Furthermore, through automated report generation and visualization analysis, it significantly reduces the workload and safety risks for maintenance personnel, providing reliable technical support for smart grid construction.

[0041] Optionally, based on the above embodiments, segmenting the transmission line point cloud region in the line point cloud model using tree obstacle analysis nodes may include:

[0042] By analyzing tree barriers, the line point cloud model is divided into equally spaced blocks to obtain multiple line point cloud blocks.

[0043] By using tree barrier analysis nodes, clustering is performed on each point cloud point in each line point cloud block to obtain at least one cluster corresponding to each line point cloud block.

[0044] By analyzing tree barriers, the clusters with linear characteristics of point cloud points in each cluster of each line point cloud block are identified as candidate conductor clusters.

[0045] By analyzing tree barriers, target traverse clusters whose linear extension direction is consistent with the flight path direction of the multi-rotor UAV are identified in each candidate traverse cluster.

[0046] By clustering the identified target conductors into clusters through tree barrier analysis nodes, the point cloud region of the transmission line is obtained.

[0047] Generally, to prevent excessive computation, the point cloud model is divided into equally spaced blocks. Specifically, the point cloud data is divided at fixed intervals along the UAV's flight direction (i.e., the direction of the power line extension), with adjacent blocks retaining a fixed overlap to prevent missed detections. This block division method reduces the amount of data processed per cycle and can adapt to the uneven point cloud density caused by changes in conductor sag.

[0048] Generally, for the point cloud within each block, an improved DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is used for segmentation. This algorithm clusters densely distributed points into multiple clusters by calculating the Euclidean distance between them. For example, conductor point clouds will form elongated clusters due to their continuous distribution, while trees or noise will appear as clumps or scattered points. After clustering, each block may output multiple clusters, including conductor clusters, vegetation clusters, noise clusters, etc., which require further filtering. The Euclidean distance can be understood as a classic method for calculating the straight-line distance between two points, defined in three-dimensional space as the square root of the sum of the squares of the differences in each dimension of the coordinates of the two points.

[0049] Generally, tree barrier analysis nodes use principal component analysis to screen candidate conductor clusters with linear characteristics from each cluster. Specifically, the covariance matrix of each cluster is calculated and eigenvalues ​​are decomposed. When the largest eigenvalue is significantly larger than the other two eigenvalues ​​(the ratio is usually greater than 10:1), the cluster is determined to have linear characteristics, and its principal direction vector is recorded. This step effectively eliminates interference from non-conductor point clouds such as shrubs and birds, retaining point cloud clusters that are truly likely to represent transmission lines, laying the foundation for subsequent accurate identification.

[0050] Generally, the main direction vector of the candidate conductor cluster is checked for consistency with the real-time heading data of the UAV. By calculating the angle between the main direction of the cluster and the heading angle provided by the POS data, if the angle is less than a preset threshold (such as 15 degrees), the cluster is determined to be the target conductor cluster. This verification process makes full use of the trajectory characteristics of the UAV flying along the power line, which can effectively distinguish between parallel conductors and crossings (such as communication lines, tree branches, etc.), greatly improving the accuracy of conductor identification.

[0051] Generally, the identified target conductor clusters in all blocks are spatially stitched and optimized. First, the continuity of the spacing between adjacent conductor clusters in the blocks is checked. Then, a B-spline curve fitting algorithm is used to eliminate minor deviations at the joints, ultimately generating a complete and smooth transmission line point cloud region. The B-spline curve fitting algorithm is a mathematical method that approximates discrete data points using piecewise polynomial functions. Its core is to generate smooth and continuous spatial curves through control points.

[0052] Optionally, based on the above embodiments, the tree obstacle analysis node generates a tree obstacle information analysis report matching the target power line according to the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power line. This report may include:

[0053] By using a preset image generation algorithm through the tree obstacle analysis node, the three-dimensional position information of each obstacle is mapped in two dimensions to obtain a multi-angle image view corresponding to each obstacle.

[0054] The tree obstacle analysis node performs image recognition based on multi-angle image views of each obstacle, generating obstacle attribute information corresponding to each obstacle.

[0055] The tree obstacle analysis node generates a tree obstacle information analysis report that matches the target power line based on the three-dimensional location information of each obstacle, the distance information to the target power line, multi-angle image views, and obstacle attribute information.

[0056] Generally, the tree obstacle analysis node uses a pre-defined image generation algorithm to convert the 3D position information of each obstacle into a 2D image view. This process uses orthogonal projection and perspective projection techniques to map the 3D point cloud data onto the XY plane (top view), XZ plane (side view), and a cross-sectional view along the traverse plane, generating standardized images. These multi-angle views fully present the spatial relationship between obstacles and power lines, with hazard points highlighted in red and safety distance boundaries marked with green dashed lines. A scale bar and coordinate axis references are also added, providing intuitive visualization for subsequent analysis.

[0057] Generally, deep learning models are used to intelligently identify obstacles from multiple angles, extracting key features. Convolutional neural networks analyze morphological, textural, and spatial distribution features in the images to automatically classify obstacle types (such as trees, buildings, or temporary structures) and estimate their height, volume, and other attribute parameters. The identification process integrates multi-view information for cross-validation to ensure the accuracy of attribute judgments, ultimately outputting structured obstacle attribute information, including type labels, size estimates, and risk level assessments.

[0058] Generally, the tree obstacle analysis node integrates the three-dimensional coordinates, minimum distance values, multi-angle views, and attribute information of obstacles to generate a standardized XML (Extensible Markup Language) format analysis report. The report includes the latitude and longitude coordinates of the hazard point, distance measurements accurate to centimeters, the corresponding view file path, and a description of the classification attributes. The report is pushed to the power grid management platform in real time via HTTPS (Hypertext Transfer Protocol Secure), while simultaneously storing a timestamped file locally for subsequent querying, statistics, and work order issuance, achieving end-to-end digital management from detection to response.

[0059] Example 2

[0060] Figure 2 This is a flowchart of another real-time tree obstacle detection method on a power transmission line provided in Embodiment 2 of the present invention. This embodiment is based on and optimized from the above embodiments. Specifically, the operation of "generating a three-dimensional point cloud model of the target power transmission line through which the multi-rotor drone flies by, based on the real-time received three-dimensional point cloud data and POS data by the tree obstacle analysis node" has been refined.

[0061] Correspondingly, such as Figure 2 As shown, the method includes:

[0062] S210. During the flight of the multi-rotor UAV along the extension direction of the target power line, the data acquisition node subscribes to the real-time three-dimensional point cloud data and POS data collected by the lidar, and sends the subscribed data to the tree obstacle analysis node in real time.

[0063] S220: Through the tree obstacle analysis node, independent 3D point cloud data and independent attitude system data are received in a loosely coupled manner.

[0064] In this embodiment, the tree obstacle analysis node receives raw point cloud data from the LiDAR and positioning and POS data (i.e., loosely coupled) through independent ROS topic interfaces. The two types of data are logically linked through message timestamps rather than being transmitted synchronously. This separate data reception mechanism allows point cloud and positioning information to be transmitted asynchronously at their respective native frequencies. When a data stream experiences a brief delay or loss, it can continue to run based on the most recent valid data, avoiding the blockage of the entire processing flow due to a single sensor malfunction.

[0065] S230. Based on the flight path direction and independent attitude system data of the multi-rotor UAV during flight, the obstacle analysis node filters out local three-dimensional point cloud data located in the fan-shaped area directly in front of the flight direction of the multi-rotor UAV from the independent three-dimensional point cloud data.

[0066] In this embodiment, the tree obstacle analysis node first acquires the real-time heading angle, pitch angle, and roll angle data of the UAV from the POS data, and establishes a fan-shaped spatial detection area centered on the UAV based on these attitude parameters. For example, this fan-shaped area can be symmetrical about the heading angle, expanding 15 degrees to the left and right, and extending 50 meters radially to form a three-dimensional spatial cone. Through spatial coordinate transformation, the original point cloud data in the lidar coordinate system is converted to a local coordinate system centered on the UAV. A fast filtering algorithm is used to retain only the point cloud data located within this fan-shaped cone. At the same time, the vertical opening angle of the fan-shaped area is dynamically adjusted according to the current flight altitude of the UAV to ensure complete coverage of the spatial distribution area of ​​the power transmission line within a pitch range of -10 degrees to +10 degrees.

[0067] S240. By using the tree obstacle analysis node to perform data preprocessing on the local three-dimensional point cloud data according to the preset filtering and / or denoising algorithm, and generating a three-dimensional point cloud model of the target transmission line based on the preprocessed local three-dimensional point cloud data.

[0068] In this embodiment, a voxel grid filtering algorithm is first used to uniformly downsample the point cloud, compressing the data volume to a set value while preserving the geometric features of the traverse. Then, a statistical outlier removal algorithm is used to remove discrete noise points such as birds and dust. Finally, based on the incremental KD-Tree (K-Dimensional Tree), also known as the incremental K-dimensional spatial tree algorithm, a spatial index is constructed to realize fast spatial retrieval and topological relationship calculation of point cloud data, providing optimized three-dimensional point cloud basic data for subsequent traverse modeling.

[0069] S250. Segment the transmission line point cloud region in the line point cloud model.

[0070] S260. Identify at least one obstacle in the power line point cloud area that meets the preset distance condition with respect to the target power line through the tree obstacle analysis node, and generate a tree obstacle information analysis report that matches the target power line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power line.

[0071] During the flight of the multi-rotor UAV along the extension direction of the target power transmission line, the data acquisition node first subscribes in real time to the 3D point cloud data and POS data collected by the lidar, and sends these synchronized data to the tree obstacle analysis node. The tree obstacle analysis node receives independent point cloud data and positioning attitude data in a loosely coupled manner. Based on the UAV's flight path direction and attitude information, it filters out local 3D point cloud data in a fan-shaped area directly in front from the original point cloud. Subsequently, the local point cloud data is preprocessed by filtering and denoising. Data quality is improved by voxel mesh downsampling and statistical outlier removal, and a 3D point cloud model of the target power transmission line is constructed. Density clustering is used in the line model. The algorithm segments the precise point cloud region of the power transmission line and ultimately identifies obstacles within this region that are less than a preset threshold away from the target power transmission line. Based on the obstacle's 3D position, minimum distance value, and multi-angle views, a standardized tree obstacle information analysis report is generated. This novel real-time tree obstacle detection method on power transmission lines significantly improves inspection efficiency and accuracy. Compared to traditional manual inspection methods, it is much more efficient, while fully meeting the power industry's standard requirements for detection accuracy. In particular, through dynamic region filtering and incremental processing technology, it effectively reduces the amount of invalid data computation, enabling the system to achieve real-time processing on an edge computing platform and providing timely and reliable data support for UAV autonomous flight decisions.

[0072] Optionally, based on the above embodiments, the tree barrier analysis node can generate a real-time three-dimensional point cloud model of the target transmission line using preprocessed local three-dimensional point cloud data. This can include:

[0073] By employing an incremental K-dimensional spatial tree algorithm through tree obstacle analysis nodes, a three-dimensional point cloud model of the target transmission line is generated in real time based on preprocessed local three-dimensional point cloud data.

[0074] Generally, the obstacle analysis node employs the KD-Tree algorithm to perform real-time modeling of preprocessed local 3D point cloud data. This algorithm constructs a hierarchical index structure through recursive spatial partitioning, dynamically updating the point cloud model during UAV flight, and can efficiently process continuously input point cloud data streams. This incremental modeling approach avoids the latency caused by traditional batch processing, ensuring that the 3D point cloud model of the route remains synchronized with the current scanning area, providing an accurate spatial reference for subsequent obstacle detection.

[0075] Optionally, based on the above embodiments, identifying at least one obstacle in the power transmission line point cloud region that meets a preset distance condition from the target power transmission line through the tree obstacle analysis node may include:

[0076] Based on the K-dimensional tree index space pre-constructed during the generation of the three-dimensional point cloud model of the line, the tree obstacle analysis node retrieves all obstacle points in the environment surrounding the target transmission line in the point cloud region of the transmission line.

[0077] By analyzing tree obstacles, all obstacle points are grouped into at least one connected component, and each connected component is determined as a candidate entity.

[0078] By using the tree obstacle analysis node, at least one obstacle that meets the preset distance condition from the target power line is identified among the candidate entities.

[0079] Generally, the tree obstacle analysis node utilizes the pre-built KD-Tree spatial index to perform a fast neighborhood search within the power line point cloud region. Using a radius query algorithm, it expands outwards at a preset distance from the power line point cloud to retrieve all obstacle points within that spatial range. This spatial index-based retrieval method significantly improves efficiency compared to brute-force search, enabling the calculation of spatial relationships in large-scale point cloud data within a short time, ensuring real-time performance is met.

[0080] Generally, tree obstacle analysis nodes aggregate retrieved discrete obstacle points into connected component entities. A clustering algorithm based on Euclidean distance is used to merge spatially adjacent point clouds into candidate entities, each representing a potential obstacle (such as trees, buildings, etc.). The clustering process considers point density and spatial continuity to avoid misclassifying discrete noise points as valid obstacles, while preserving the complete geometric features of real obstacles.

[0081] Generally, the tree obstacle analysis node performs precise distance calculations and risk assessments for each candidate entity. By calculating the minimum distance from all points in the entity to the conductor point cloud and comparing it with a preset safety threshold (e.g., 3.5 meters for a 110 kV line), the obstacles constituting actual tree obstacle hazards are finally determined. This process integrates geometric analysis, distance field calculations, and risk assessment models to ensure that the identification results not only comply with power safety regulations but also reflect the actual level of spatial threat.

[0082] Optionally, based on the above embodiments, the method of identifying at least one obstacle among the candidate entities that meets a preset distance condition from the target transmission line through the tree obstacle analysis node may include:

[0083] The tree barrier analysis node determines the center point of each candidate entity, and the vertical distance from the center point of each entity to the target transmission line is calculated based on the linear characteristics in the transmission line point cloud region.

[0084] Candidate entities with a vertical distance less than or equal to a preset distance threshold are obtained through tree obstacle analysis nodes and used as obstacles.

[0085] Generally, the tree obstacle analysis node first calculates the center point location for each candidate entity. The center point position is determined by calculating the mean of the 3D coordinates of the entity's point cloud. Simultaneously, considering the linear characteristics of the traverse point cloud, the shortest perpendicular distance from the center point to the traverse is calculated using orthogonal projection. This process considers the actual direction and curvature changes of the traverse in space, solving for the perpendicular distance from a point to a straight line or curve in space through vector operations, ensuring that the distance measurement results accurately reflect the true spatial relationship between the obstacle and the traverse. Orthogonal projection is a geometric transformation method that maps points in 3D space to a 2D plane; its core characteristic is maintaining the parallelism and proportional relationship of objects before and after projection.

[0086] Generally, the tree obstacle analysis node compares the vertical distance of each entity with a preset safety threshold for screening. A dynamic distance threshold is set based on the voltage level of the transmission line; when the vertical distance from the center point of an entity to the conductor is less than or equal to this threshold, it is determined to be a valid obstacle. This step comprehensively considers the requirements of power safety regulations and the actual spatial relationship, avoiding misjudgments due to improper measurement methods and ensuring the accuracy and reliability of hazard identification.

[0087] To facilitate understanding, specific application scenarios applicable to each embodiment of the invention are described. In this specific embodiment, in order to detect tree obstacles on power transmission lines in real time, this embodiment of the invention designs a complete scheme for real-time tree obstacle detection on power transmission lines.

[0088] Specifically, in Figure 3 The diagram illustrates a real-time tree obstacle detection scheme on power transmission lines according to an embodiment of the present invention. Specifically, this scheme uses a multi-rotor UAV as a power transmission line inspection platform, integrating a lidar sensor and edge computing devices to achieve efficient and intelligent autonomous inspection tasks. The edge computing box embeds and runs the ROS operating system, serving as the core framework of the system and coordinating key functions such as data acquisition, processing, decision-making, and control. The data acquisition nodes in the ROS system are deployed on the edge computing box, primarily responsible for the real-time reception and preprocessing of lidar point cloud data. It also includes a flight control node for interacting with the flight control system, generating and issuing attitude and position control commands to achieve real-time adjustment and control of the UAV's flight status.

[0089] During flight, the system continuously scans the space ahead using lidar to construct a high-precision 3D environment model in real time. By extracting and identifying regions with obvious linear features in the point cloud data, the system can accurately detect and extract the spatial location information of the power transmission lines.

[0090] By combining real-time heading information provided by the POS data system, the spatial relationship between the power line and the UAV is further analyzed. Based on the spatial position information of the power line and with the help of the PID (Proportional-Integral-Derivative) control algorithm, the UAV can autonomously adjust its heading and attitude, thereby achieving dynamic autonomous flight along the power transmission conductor.

[0091] Meanwhile, the system uses the ROS topic mechanism to publish LiDAR point cloud data and POS location information to the tree obstacle analysis node in real time. The tree obstacle analysis node adopts a loosely coupled strategy, receiving point cloud and location information separately, effectively reducing the impact of data delay or loss, thereby improving the robustness and processing stability of the overall system.

[0092] The tree obstacle analysis node first filters and denoises the received raw point cloud to remove background clutter and anomalous noise, improving the overall quality and usability of the point cloud data. This process helps eliminate discrete points and unstructured clutter caused by sensor errors, environmental interference, or changes in flight attitude, laying the foundation for subsequent processing. Subsequently, the system dynamically extracts the spatial range of the transmission line corridor consistent with the flight path based on POS data trajectory information. By constructing a spatial window with the POS data direction as the main axis, effective point cloud data within the corridor area is extracted, and a three-dimensional point cloud model around the line is built based on an incremental KD-Tree structure. This enables rapid indexing and efficient spatial retrieval of the point cloud data, providing structural support for subsequent conductor extraction and hazard identification.

[0093] Next, the system divides the channel point cloud data along the POS data trajectory at certain intervals, and processes the point cloud data in each small segment separately. For each segment, a density clustering algorithm is used for analysis to extract point cloud clusters with obvious linear structure characteristics as suspected conductor data. These conductor clusters are further distinguished from actual transmission conductors and possible crossing conductors by analyzing the consistency between their main direction and the flight direction of the POS data, ensuring the accuracy and stability of conductor identification.

[0094] After accurately identifying the traverse point cloud, the system further uses the traverse as the center and a KD-Tree spatial index structure to quickly retrieve suspected tree obstruction points within a certain spatial range around it. For each retrieved point, the minimum spatial distance between it and the traverse is calculated to determine whether it constitutes an actual spatial safety hazard. Once this distance is less than a preset threshold, the system identifies it as a tree obstruction hazard point and records its 3D coordinates, the nearest distance to the traverse, and related attribute information for subsequent statistical analysis and risk assessment.

[0095] To enhance the intuitiveness and understandability of the analysis results, the system employs image generation technology to project and transform the 3D point cloud data, generating 2D view images, including top views, side views, and cross-sectional views. These images can intuitively show the spatial distribution of tree obstruction points and their relative relationship with guide lines, facilitating manual verification and auxiliary judgment by maintenance personnel. They also provide a reference for subsequent hazard handling and maintenance scheduling.

[0096] Ultimately, the system automatically generates structured hazard report files, using standard XML format to record detailed information for each hazard point, including its 3D spatial coordinates, minimum distance value, type classification, and corresponding 2D image resource path. This standardized data output method enables standardized management of hazard information, facilitating subsequent data archiving, remote querying, and seamless integration with other power inspection systems or management platforms, thereby improving the automation level and efficiency of overall hazard monitoring and handling.

[0097] Furthermore, through the clever combination of the above steps, real-time tree obstruction detection on power transmission lines can be achieved, resulting in the following effective effects:

[0098] 1) By using a dynamically updated hierarchical spatial index structure, efficient modeling and real-time updating of the 3D point cloud of transmission lines are achieved, solving the latency problem caused by traditional batch processing methods and ensuring that the model is always synchronized with the current scanning area.

[0099] 2) Based on the improved clustering algorithm and principal component analysis, the linear point cloud region of the conductor is accurately segmented from the complex point cloud, effectively eliminating interference objects such as vegetation and buildings, and significantly improving the accuracy of conductor identification.

[0100] 3) Based on the real-time flight path of the UAV, the detection area is dynamically delineated. Through spatial search and distance calculation, obstacles less than the safety threshold from the guide wire are accurately identified, thus achieving precise location of potential hazards.

[0101] 4) Utilize image generation technology to generate multi-angle views and automatically output standard format reports, including hazard coordinates, distances, and image paths, supporting seamless integration with the power grid management system and significantly improving inspection efficiency.

[0102] Example 3

[0103] Figure 4 This invention provides a real-time tree obstacle detection device on a power transmission line, as described in Embodiment 3. The device is configured within a robot operating system, which includes a data acquisition node deployed on a multi-rotor drone and a tree obstacle analysis node deployed on a processing platform. The multi-rotor drone is equipped with a lidar system. Figure 4 As shown, the device includes:

[0104] The data acquisition module 410 is used to subscribe to the three-dimensional point cloud data and positioning and attitude system data collected in real time by the lidar through the data acquisition node during the flight of the multi-rotor UAV along the extension direction of the target power line, and send the subscribed data to the tree obstacle analysis node in real time.

[0105] The model building module 420 is used to generate a three-dimensional point cloud model of the target power transmission line that the multi-rotor UAV flies over, based on the real-time received three-dimensional point cloud data and positioning and attitude system data through the tree obstacle analysis node, and to segment the power transmission line point cloud region in the line point cloud model.

[0106] The identification and analysis module 430 is used to identify at least one obstacle in the power transmission line point cloud area that meets the preset distance condition with respect to the target power transmission line through the tree obstacle analysis node, and generate a tree obstacle information analysis report that matches the target power transmission line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power transmission line.

[0107] The technical solution of this invention involves the following steps: During the flight of a multi-rotor drone along the extension direction of a target power transmission line, a data acquisition node first subscribes in real-time to 3D point cloud data and POS data collected by a lidar sensor, and sends this data to a tree obstacle analysis node. Subsequently, based on the received 3D point cloud data and POS data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line covered by the drone's flight trajectory, and precisely segments the power transmission line point cloud region within this model. Finally, the tree obstacle analysis node identifies one or more obstacles within the power transmission line point cloud region that meet preset distance conditions from the target power transmission line. Based on the 3D position information of each obstacle and its distance from the target power transmission line, the node automatically generates a tree obstacle information analysis report matching the target power transmission line. This novel real-time tree obstacle detection method on power transmission lines enables accurate identification of potential hazards, significantly improving detection efficiency compared to traditional manual inspections. Furthermore, through automated report generation and visualization analysis, it significantly reduces the workload and safety risks for maintenance personnel, providing reliable technical support for smart grid construction.

[0108] Furthermore, based on the above embodiments, the model module 420 may further include:

[0109] The data receiving submodule is used to receive independent 3D point cloud data and independent attitude system data through the tree obstacle analysis node in a loosely coupled manner.

[0110] The filtering submodule is used to filter out local 3D point cloud data located in a fan-shaped area directly in front of the multi-rotor UAV's flight direction from the independent 3D point cloud data based on the flight path direction and independent attitude system data of the multi-rotor UAV during flight through the tree obstacle analysis node.

[0111] The line point cloud generation submodule performs data preprocessing on local 3D point cloud data through tree obstacle analysis nodes according to preset filtering and / or denoising algorithms, and generates a 3D point cloud model of the target transmission line based on the preprocessed local 3D point cloud data.

[0112] Based on the above embodiments, a model module 420 is constructed, specifically for:

[0113] By analyzing tree barriers, the line point cloud model is divided into equally spaced blocks to obtain multiple line point cloud blocks.

[0114] By using tree barrier analysis nodes, clustering is performed on each point cloud point in each line point cloud block to obtain at least one cluster corresponding to each line point cloud block.

[0115] By analyzing tree barriers, the clusters with linear characteristics of point cloud points in each cluster of each line point cloud block are identified as candidate conductor clusters.

[0116] By analyzing tree barriers, target traverse clusters whose linear extension direction is consistent with the flight path direction of the multi-rotor UAV are identified in each candidate traverse cluster.

[0117] By clustering the identified target conductors into clusters through tree barrier analysis nodes, the point cloud region of the transmission line is obtained.

[0118] Based on the above embodiments, a line point cloud submodule is generated, specifically used for:

[0119] By using the incremental K-dimensional spatial tree algorithm through the tree obstacle analysis node, a three-dimensional point cloud model of the target transmission line is generated in real time based on the preprocessed local three-dimensional point cloud data.

[0120] Furthermore, based on the above embodiments, the identification and analysis module 430 may further include:

[0121] The retrieval submodule is used to retrieve all obstacle points in the target transmission line point cloud area based on the K-dimensional spatial tree index space pre-constructed during the generation of the line's 3D point cloud model by the tree obstacle analysis node;

[0122] The candidate entity submodule is used to form at least one connected component from all obstacle points through the tree obstacle analysis node, and to determine each connected component as a candidate entity.

[0123] The identification results submodule is used to identify at least one obstacle among the candidate entities that meets the preset distance condition from the target power line through the tree obstacle analysis node.

[0124] Based on the above embodiments, the identification result submodule is specifically used for:

[0125] The tree barrier analysis node determines the center point of each candidate entity, and the vertical distance from the center point of each entity to the target transmission line is calculated based on the linear characteristics in the transmission line point cloud region.

[0126] Candidate entities with a vertical distance less than or equal to a preset distance threshold are obtained through tree obstacle analysis nodes and used as obstacles.

[0127] Based on the above embodiments, the identification and analysis module 430 is specifically used for:

[0128] By using a preset image generation algorithm through the tree obstacle analysis node, the three-dimensional position information of each obstacle is mapped in two dimensions to obtain a multi-angle image view corresponding to each obstacle.

[0129] The tree obstacle analysis node performs image recognition based on multi-angle image views of each obstacle, generating obstacle attribute information corresponding to each obstacle.

[0130] The tree obstacle analysis node generates a tree obstacle information analysis report that matches the target power line based on the three-dimensional location information of each obstacle, the distance information to the target power line, multi-angle image views, and obstacle attribute information.

[0131] The real-time tree obstacle detection device on power transmission lines provided in this embodiment of the invention can execute the real-time tree obstacle detection method on power transmission lines provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0132] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0133] Example 5

[0134] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0135] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0136] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0137] Processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as performing a real-time tree obstacle detection method on a power transmission line as described in any embodiment of the present invention, namely:

[0138] During the flight of the multi-rotor UAV along the extension direction of the target power line, the data acquisition node subscribes to the real-time three-dimensional point cloud data and positioning and attitude system data collected by the lidar, and sends the subscribed data to the tree obstacle analysis node in real time.

[0139] Based on the real-time received 3D point cloud data and positioning and attitude system data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line that the multi-rotor UAV flies over, and segments the power transmission line point cloud region in the line point cloud model.

[0140] The tree obstacle analysis node identifies at least one obstacle in the power line point cloud area that meets the preset distance condition to the target power line, and generates a tree obstacle information analysis report that matches the target power line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power line.

[0141] In some embodiments, the real-time tree obstacle detection method on a power transmission line as described in any of the embodiments of the present invention can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the real-time tree obstacle detection method on a power transmission line as described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the real-time tree obstacle detection method on a power transmission line as described in any of the embodiments of the present invention by any other suitable means (e.g., by means of firmware).

[0142] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0143] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0144] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0145] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0146] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0147] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0148] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0149] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for real-time tree obstruction detection on power transmission lines, characterized in that, Executed by a robot operating system, which includes a data acquisition node deployed on a multi-rotor drone and a tree obstacle analysis node deployed on a processing platform, the method includes: During the flight of the multi-rotor UAV along the extension direction of the target power line, the data acquisition node subscribes to the real-time three-dimensional point cloud data and positioning and attitude system data collected by the lidar, and sends the subscribed data to the tree obstacle analysis node in real time. Based on the real-time received 3D point cloud data and positioning and attitude system data, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line that the multi-rotor UAV flies through, and segments the power transmission line point cloud region in the line point cloud model. The tree obstacle analysis node identifies at least one obstacle in the power line point cloud area that meets the preset distance condition to the target power line, and generates a tree obstacle information analysis report matching the target power line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power line. Specifically, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line traversed by the multi-rotor UAV based on real-time received 3D point cloud data and positioning and attitude system data, including: The tree obstacle analysis node receives independent 3D point cloud data and independent attitude system data in a loosely coupled manner. Based on the flight path direction of the multi-rotor UAV and the independent attitude system data, the tree obstacle analysis node filters out local 3D point cloud data located in a fan-shaped area directly in front of the flight direction of the multi-rotor UAV from the independent 3D point cloud data. The tree obstacle analysis node preprocesses the local 3D point cloud data according to a preset filtering and / or denoising algorithm, and generates a 3D point cloud model of the target power transmission line based on the preprocessed local 3D point cloud data. Specifically, based on the flight path direction and independent attitude system data during the multi-rotor UAV's flight, the tree obstacle analysis node filters out local 3D point cloud data located within a fan-shaped area directly in front of the multi-rotor UAV's flight direction from the independent 3D point cloud data. This includes: A sector-shaped spatial detection region centered on the UAV is established based on attitude parameters, including the UAV's real-time heading angle, pitch angle, and roll angle data, and the vertical opening angle of the sector region is dynamically adjusted according to the UAV's current flight altitude. Specifically, the point cloud region of the transmission line is segmented in the line point cloud model through tree obstacle analysis nodes, including: The line point cloud model is divided into equally spaced blocks by the tree obstacle analysis node to obtain multiple line point cloud blocks. The tree obstacle analysis node then clusters the point cloud points within each line point cloud block to obtain at least one cluster corresponding to each block. Within each cluster of line point cloud blocks, the tree obstacle analysis node identifies clusters with linear characteristics as candidate conductor clusters. From these candidate conductor clusters, the tree obstacle analysis node identifies target conductor clusters whose linear extension direction aligns with the flight path of the multi-rotor UAV. Based on the identified target conductor clusters, the tree obstacle analysis node obtains the transmission line point cloud region.

2. The method according to claim 1, characterized in that, Based on preprocessed local 3D point cloud data, the tree barrier analysis node generates a real-time 3D point cloud model of the target transmission line, including: By using the incremental K-dimensional spatial tree algorithm through the tree obstacle analysis node, a three-dimensional point cloud model of the target transmission line is generated in real time based on the preprocessed local three-dimensional point cloud data. Accordingly, at least one obstacle that meets a preset distance condition from the target transmission line is identified in the transmission line point cloud area through the tree obstacle analysis node, including: Based on the K-dimensional tree index space pre-constructed during the generation of the three-dimensional point cloud model of the line, the tree obstacle analysis node retrieves all obstacle points in the environment surrounding the target transmission line in the point cloud region of the transmission line. By analyzing tree obstacles, all obstacle points are grouped into at least one connected component, and each connected component is determined as a candidate entity. By using the tree obstacle analysis node, at least one obstacle that meets the preset distance condition from the target power line is identified among the candidate entities.

3. The method according to claim 2, characterized in that, By using tree obstacle analysis nodes, at least one obstacle that meets a preset distance condition from the target power line is identified among the candidate entities, including: The tree barrier analysis node determines the center point of each candidate entity, and the vertical distance from the center point of each entity to the target transmission line is calculated based on the linear characteristics in the transmission line point cloud region. Candidate entities with a vertical distance less than or equal to a preset distance threshold are obtained through tree obstacle analysis nodes and used as obstacles.

4. The method according to any one of claims 1-3, characterized in that, Based on the 3D location information of each obstacle and the distance information of each obstacle to the target power line, the tree obstacle analysis node generates a tree obstacle information analysis report that matches the target power line, including: By using a preset image generation algorithm through the tree obstacle analysis node, the three-dimensional position information of each obstacle is mapped in two dimensions to obtain a multi-angle image view corresponding to each obstacle. The tree obstacle analysis node performs image recognition based on multi-angle image views of each obstacle, generating obstacle attribute information corresponding to each obstacle. The tree obstacle analysis node generates a tree obstacle information analysis report that matches the target power line based on the three-dimensional location information of each obstacle, the distance information to the target power line, multi-angle image views, and obstacle attribute information.

5. A real-time tree obstruction detection device for power transmission lines, characterized in that, Configured within a robot operating system, the robot operating system includes a data acquisition node deployed on a multi-rotor drone and a tree obstacle analysis node deployed on a processing platform. The multi-rotor drone is equipped with a lidar system. The device includes: The data acquisition module is used to subscribe to the real-time 3D point cloud data and positioning and attitude system data collected by the lidar through the data acquisition node during the flight of the multi-rotor UAV along the extension direction of the target power line, and send the subscribed data to the tree obstacle analysis node in real time. Specifically, the tree obstacle analysis node generates a 3D point cloud model of the target power transmission line traversed by the multi-rotor UAV based on real-time received 3D point cloud data and positioning and attitude system data, including: The tree obstacle analysis node receives independent 3D point cloud data and independent attitude system data in a loosely coupled manner. Based on the flight path direction of the multi-rotor UAV and the independent attitude system data, the tree obstacle analysis node filters out local 3D point cloud data located in a fan-shaped area directly in front of the flight direction of the multi-rotor UAV from the independent 3D point cloud data. The tree obstacle analysis node preprocesses the local 3D point cloud data according to a preset filtering and / or denoising algorithm, and generates a 3D point cloud model of the target power transmission line based on the preprocessed local 3D point cloud data. Specifically, based on the flight path direction and independent attitude system data during the multi-rotor UAV's flight, the tree obstacle analysis node filters out local 3D point cloud data located within a fan-shaped area directly in front of the multi-rotor UAV's flight direction from the independent 3D point cloud data. This includes: A fan-shaped spatial detection area centered on the UAV is established based on attitude parameters, including the UAV's real-time heading angle, pitch angle, and roll angle data. The vertical opening angle of the fan-shaped area is dynamically adjusted according to the UAV's current flight altitude. A model module is constructed to generate a 3D point cloud model of the target power transmission line that the multi-rotor UAV flies through, based on the real-time received 3D point cloud data and positioning and attitude system data, through the tree obstacle analysis node. The power transmission line point cloud area is then segmented in the line point cloud model. The identification and analysis module is used to identify at least one obstacle in the power transmission line point cloud area that meets the preset distance condition with respect to the target power transmission line through the tree obstacle analysis node, and generate a tree obstacle information analysis report that matches the target power transmission line based on the three-dimensional position information of each obstacle and the distance information of each obstacle to the target power transmission line. The process of segmenting the transmission line point cloud region in the line point cloud model using a tree obstacle analysis node includes: dividing the line point cloud model into equally spaced blocks using the tree obstacle analysis node to obtain multiple line point cloud blocks; clustering the point cloud points in each line point cloud block using the tree obstacle analysis node to obtain at least one cluster corresponding to each line point cloud block; identifying clusters with linear characteristics of point cloud points as candidate conductor clusters in each cluster of line point cloud blocks using the tree obstacle analysis node; identifying target conductor clusters whose linear extension direction is consistent with the flight path direction of the multi-rotor UAV during flight using the tree obstacle analysis node; and obtaining the transmission line point cloud region based on the identified target conductor clusters using the tree obstacle analysis node.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the real-time tree obstacle detection method on the power transmission line according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the real-time tree obstacle detection method on the power transmission line as described in any one of claims 1-4.

8. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the real-time tree obstacle detection method on power transmission lines according to any one of claims 1-4.

Images