Unmanned aerial vehicle obstacle avoidance detection control method and system for high-voltage tower inspection

Abstract

The application relates to the technical field of obstacle avoidance detection control, in particular to a UAV obstacle avoidance detection control method and system for high-voltage pole tower inspection, which comprises the following steps: acquiring three-dimensional point cloud data and various types of monitoring data in real time during UAV driving; acquiring various types of ideal monitoring data of the UAV based on the distance and position of obstacles in the three-dimensional point cloud data and the target position coordinates of the UAV; constructing a driving deviation value based on the difference between the various types of monitoring data and the ideal monitoring data at the current moment, so as to determine whether the control parameters need to be adjusted; if so, constructing a comprehensive state value based on the change trend and discrete degree of the driving deviation value of the various types of monitoring data, and combining the driving deviation value to construct an adjustment coefficient at the current moment, so as to adjust various control parameters in the PID control algorithm. The application adjusts various control parameters adaptively, and improves the accuracy and stability of UAV obstacle avoidance detection control.

Description

Technical Field

[0001] This application relates to the field of obstacle avoidance detection and control technology, specifically to an obstacle avoidance detection and control method and system for UAVs used in high-voltage tower inspection. Background Technology

[0002] In the process of using drones for the inspection of high-voltage power transmission towers, obstacle avoidance control is a core element in ensuring the safety and stability of the inspection operation. High-voltage power transmission towers are mostly located in complex geographical environments such as mountainous areas and rivers, resulting in numerous static obstacles such as trees, guy wires, and hardware around the towers. Furthermore, the high-voltage conductors between the towers are thin-diameter and easily missed obstacles, posing a significant risk of collision between the drone and these obstacles during inspections. This can lead to equipment damage and even major power accidents such as short circuits and power outages. Therefore, precise obstacle avoidance control of drones is crucial for ensuring the efficiency, safety, and stability of the inspection process.

[0003] However, during high-voltage power pole inspection using drones, the complex environment causes instability in drone operation, affecting the efficiency and safety of the inspection. Furthermore, when facing strong winds and airflow disturbances, the drone experiences vibrations, impacting sensor accuracy and causing dynamic changes in its relative position to the poles and power lines during flight. When using traditional obstacle avoidance control algorithms, drones are easily affected by these environmental disturbances, resulting in significant deviations in obstacle avoidance detection and control response, ultimately leading to poor accuracy and stability in obstacle avoidance control. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method and system for obstacle avoidance detection and control of unmanned aerial vehicles (UAVs) for high-voltage tower inspection. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of this application provide an obstacle avoidance detection and control method for unmanned aerial vehicles (UAVs) used for high-voltage tower inspection, the method comprising the following steps:

[0006] Real-time acquisition of 3D point cloud data and various monitoring data during the drone's flight process;

[0007] Clustering is performed on the non-ground point cloud in the current 3D point cloud data. Based on the shortest distance between the UAV and each cluster at the current time, as well as the 3D spatial coordinates of the center of each cluster and the 3D grid map it maps, and combined with the target position coordinates of the UAV, the obstacle avoidance driving path at the current time is obtained, and then various ideal monitoring data of the UAV at the current time are obtained.

[0008] Based on the differences between various monitoring data at the current moment and their ideal monitoring data, the driving deviation values ​​of various monitoring data at the current moment are constructed to determine whether the control parameters of the UAV need to be adjusted at the current moment. When adjustment is required, based on the significance and dispersion of the increasing trend of the driving deviation values ​​of various monitoring data at all sampling moments between the current moment and the last adjustment moment, the comprehensive state value of various monitoring data at the current moment is constructed.

[0009] Based on the comprehensive state value and driving deviation value of all types of monitoring data at the current moment, the adjustment coefficient for the current moment is constructed, and the initial values ​​of various control parameters in the PID control algorithm at the current moment are adjusted accordingly.

[0010] Preferably, the method for obtaining the obstacle avoidance driving path at the current moment is as follows:

[0011] Based on the distance between the UAV's current spatial location and the three-dimensional spatial coordinates of each cluster, the shortest distance between the UAV and each cluster at the current moment is obtained.

[0012] The current 3D grid map, the shortest distance between the UAV and each cluster, the 3D spatial coordinates of the center of each cluster, and the target position coordinates of the UAV are used as inputs to the A* algorithm. Combined with a fusion path planning algorithm based on the dynamic window method, the algorithm outputs the obstacle avoidance path of the UAV at the current moment.

[0013] Preferably, the method for obtaining the shortest distance between the UAV and each cluster at the current moment is as follows:

[0014] Real-time acquisition of latitude and longitude coordinates and flight altitude of the drone during its flight;

[0015] Based on the latitude and longitude coordinates and flight altitude of the UAV at the current moment, the three-dimensional spatial coordinates of the UAV at the current moment are obtained, and the three-dimensional spatial coordinates of the UAV and all point clouds in each cluster at the current moment are transformed to the same geographic coordinate system.

[0016] Calculate the Euclidean distance between the UAV and the transformed 3D spatial coordinates of all point clouds in each obstacle, and record the minimum value among all Euclidean distances as the shortest distance between the UAV and each cluster at the current moment.

[0017] Preferably, the method for constructing the three-dimensional raster map is as follows:

[0018] The three-dimensional spatial coordinate data of all cluster centers at the current moment are mapped to a three-dimensional raster space of preset resolution using a raster mapping algorithm. Rasters containing cluster center points are marked as danger zones, and rasters not containing cluster center points are marked as safe zones, thus obtaining a three-dimensional raster map at the current moment.

[0019] Preferably, the driving deviation value of various monitoring data at the current moment refers to the absolute value of the relative error between various monitoring data at the current moment and the corresponding ideal data.

[0020] Preferably, the specific process of determining whether the control parameters of the UAV need to be adjusted at the current moment is as follows:

[0021] Calculate the mean value of the driving deviation of all types of monitoring data at the current moment;

[0022] If the average value is greater than or equal to the preset driving deviation threshold, it is determined that the control parameters of the drone need to be adjusted at the current moment; otherwise, it is determined that the control parameters of the drone do not need to be adjusted at the current moment.

[0023] Preferably, the method for constructing the comprehensive status value of the various monitoring data at the current moment is as follows:

[0024] The sampling time at which the control parameters of the UAV need to be adjusted is recorded as the target adjustment time;

[0025] The slope of the fitted curve corresponding to the driving deviation values ​​of various monitoring data at all sampling times between the current time and the previous target adjustment time is calculated.

[0026] The variance of the driving deviation values ​​of various monitoring data at all sampling times between the current time and the previous target adjustment time is statistically analyzed.

[0027] The overall state value of the various monitoring data at the current moment is positively correlated with the slope and the variance.

[0028] Preferably, the method for constructing the adjustment coefficient at the current moment is as follows:

[0029] Calculate the ratio between the comprehensive state value of each type of monitoring data at the current moment and the sum of the comprehensive state values ​​of all types of monitoring data;

[0030] Calculate the mean value of the driving deviation of various monitoring data at all sampling times between the current time and the previous target adjustment time, and calculate the ratio of the driving deviation value of various monitoring data at the current time to the mean value;

[0031] The adjustment coefficient at the current moment is positively correlated with both of the above ratios.

[0032] Preferably, the specific process of adjusting the initial values ​​of each control parameter in the PID control algorithm at the current moment is as follows:

[0033] Based on the preset control parameter mapping table, the adjustment coefficients at the current time are mapped to the scaling factors of each control parameter at the current time.

[0034] The specific formula for adjusting the initial values ​​of each control parameter is as follows: In the formula, This represents the adjusted value of the j-th control parameter. This represents the initial value of the j-th control parameter; is the scaling factor for the j-th control parameter at the current time.

[0035] Secondly, embodiments of this application also provide an unmanned aerial vehicle (UAV) obstacle avoidance detection and control system for high-voltage tower inspection, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described UAV obstacle avoidance detection and control methods for high-voltage tower inspection.

[0036] This application has at least the following beneficial effects:

[0037] This application addresses the issue of low accuracy in obstacle avoidance detection and control of UAVs during high-voltage tower inspection operations due to complex environmental interference. By analyzing the location and distance of obstacles in the UAV's 3D point cloud, the application plans the UAV's obstacle avoidance path at the current moment, thereby acquiring various ideal monitoring data. This provides a reliable basis for subsequent evaluation of UAV flight deviations. The application also improves the analysis of the characteristics of the UAV's flight affected by dynamic environmental interference, constructs flight deviation values, and determines whether adjustments to the UAV's control parameters are necessary. Considering the correlation and coupling characteristics of environmental interference on the UAV's state during actual obstacle avoidance, the application analyzes the changing trends of various monitoring data when the UAV's state changes, constructs a comprehensive state value, and thus assesses the degree of UAV deviation. This allows for precise optimization and adjustment of various control parameters during the UAV's obstacle avoidance process, reducing the impact of dynamic environmental interference factors on the UAV's control response and improving the accuracy and stability of UAV obstacle avoidance detection and control. Attached Figure Description

[0038] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A flowchart illustrating the steps of a drone obstacle avoidance detection and control method for high-voltage tower inspection according to an embodiment of this application;

[0040] Figure 2 A flowchart illustrating the process of obtaining the adjustment coefficient at the current moment, provided as an embodiment of this application. Detailed Implementation

[0041] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the UAV obstacle avoidance detection and control method and system for high-voltage tower inspection proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0043] The following, in conjunction with the accompanying drawings, details the specific scheme of the UAV obstacle avoidance detection and control method and system for high-voltage tower inspection provided in this application.

[0044] Please see Figure 1 The diagram illustrates a flowchart of a UAV obstacle avoidance detection and control method for high-voltage tower inspection according to an embodiment of this application. The method includes the following steps:

[0045] Step 1: Acquire 3D point cloud data and various monitoring data in real time during the drone's flight.

[0046] This application utilizes a drone equipped with a multi-sensor integrated optoelectronic pod to conduct inspections of high-voltage power transmission towers. During the inspection, the multi-sensor integrated optoelectronic pod collects comprehensive detection data. Specifically, the multi-sensor integrated optoelectronic pod integrates at least an E1R series lidar, an IMU inertial measurement unit, and an RTK high-precision positioning module. During obstacle avoidance and detection control, the lidar collects real-time 3D point cloud data of the drone's inspection environment and the surrounding area of ​​the tower for subsequent analysis of the spatial location of obstacles in the environment. The IMU inertial measurement unit collects real-time data on the drone's actual heading angle, flight altitude, and flight speed for subsequent real-time adjustment and control of the drone's attitude and flight speed. The high-precision positioning module collects the drone's latitude and longitude coordinates to determine the relative positional relationship between the drone and obstacles.

[0047] Based on the above data collection, the raw data is transmitted to the data analysis and processing unit of the power inspection cloud platform via a 4G / 5G wireless communication network. During the inspection, the cloud platform can issue control commands such as flight path adjustment and obstacle avoidance strategy optimization to the drone based on the data analysis results, so as to accurately control the drone's obstacle avoidance and detection process and realize two-way data interaction. If communication is interrupted, the drone will trigger the local data storage mechanism and complete the data retransmission after communication is restored.

[0048] Step 2: Cluster the non-ground point clouds in the current 3D point cloud data. Based on the shortest distance between the UAV and each cluster at the current time, as well as the 3D spatial coordinates of the center of each cluster and the 3D raster map it maps, and combined with the target position coordinates of the UAV, obtain the obstacle avoidance path at the current time, and then obtain various ideal monitoring data of the UAV at the current time.

[0049] After receiving the data transmitted by the UAV, the cloud platform's data processing and analysis unit first preprocesses the 3D point cloud data at each sampling time using a voxel filtering algorithm, and then preprocesses the attitude and flight speed data at each sampling time using a Kalman filtering algorithm. This reduces the low data quality issues caused by UAV body shaking, strong wind and airflow disturbances, sensor noise, and transmission delays during actual data acquisition. Both voxel filtering and Kalman filtering algorithms are well-known techniques to those skilled in the art, and their specific processes will not be elaborated upon.

[0050] Furthermore, based on the preprocessed data, the obstacle avoidance detection information during the UAV's inspection process is analyzed and processed. Then, based on the analysis and processing results, the obstacle avoidance control of the UAV is adjusted. The specific analysis and processing process is as follows:

[0051] First, for the preprocessed 3D point cloud data at each sampling time, the 3D point cloud data is used as input, and the ground segmentation algorithm is used to segment and remove the ground point cloud data in the 3D point cloud data to obtain the non-ground point cloud data; the non-ground point cloud is used as input to the Euclidean distance clustering algorithm to obtain the clustering results of the non-ground point cloud data; where the point cloud data in each cluster corresponds to a single obstacle.

[0052] The three-dimensional spatial coordinate data of all cluster centers at the current moment are mapped to a three-dimensional grid space with a preset resolution (set to 20cm resolution in this embodiment) through a grid mapping algorithm. Grids containing cluster center points are marked as danger zones, and grids without cluster center points are marked as safe zones, thus obtaining a three-dimensional grid map at the current moment, which is used to generate obstacle avoidance path planning in the future.

[0053] Furthermore, based on the latitude and longitude coordinates and flight altitude of the UAV at the current moment, the three-dimensional spatial coordinates of the UAV at the current moment are obtained, and the three-dimensional spatial coordinates of the UAV and all point clouds in each cluster at the current moment are converted into the same geographic coordinate system. The Euclidean distance between the UAV and all point clouds in each obstacle is calculated, and the minimum value among all Euclidean distances is recorded as the shortest distance between the UAV and each cluster at the current moment, which is used to determine whether the distance between the current UAV and each obstacle exceeds the preset safety threshold.

[0054] Among them, the ground segmentation algorithm, Euclidean distance clustering algorithm, and grid mapping algorithm are well known to those skilled in the art, and the specific processes will not be described in detail.

[0055] Furthermore, the current 3D grid map, the shortest collision distance between the UAV and each cluster, the 3D spatial coordinates of the cluster centers, and the target position coordinates of the UAV are used as inputs to the A* algorithm. Combined with a fusion path planning algorithm based on the Dynamic Window Method (DWA), the algorithm outputs the obstacle avoidance path of the UAV at the current moment, which is used for obstacle avoidance control of the UAV's flight path. Both the A* algorithm and the fusion path planning algorithm based on the Dynamic Window Method (DWA) are well-known technologies, and their specific processes will not be elaborated further.

[0056] Furthermore, based on the obstacle avoidance driving path and the inspection time set in the actual application process, various ideal monitoring data of the UAV at different times are determined. The monitoring data includes ideal spatial coordinates, ideal flight speed, ideal heading angle and ideal altitude information, which are used to evaluate the difference between the actual flight state of the UAV and the ideal flight state at the current time.

[0057] Step 3: Based on the differences between various monitoring data at the current moment and their ideal monitoring data, construct the driving deviation value of various monitoring data at the current moment, so as to determine whether the control parameters of the UAV need to be adjusted at the current moment; when adjustment is required, construct the comprehensive state value of various monitoring data at the current moment based on the significance and dispersion of the increasing trend of the driving deviation value of various monitoring data at all sampling moments between the current moment and the last adjustment moment.

[0058] Traditional UAV obstacle avoidance methods typically utilize PID control algorithms to feed back various monitoring data based on the difference between ideal and actual flight conditions, generating corresponding heading corrections, speed adjustments, attitude adjustments, and altitude adaptation control commands. However, during real-time control following the obstacle avoidance path, the influence of complex real-world environments can cause deviations in UAV control at different times. Therefore, it is necessary to optimize and adjust the UAV's obstacle avoidance detection and control process based on the characteristics of real-time monitoring data under the influence of actual environmental disturbances.

[0059] During high-voltage tower inspections, drones are prone to deviations due to complex environmental interference, leading to a greater discrepancy between actual and ideal monitoring data. This necessitates adjustments to the drone's monitoring data at any given moment. Therefore, the need for drone control parameter adjustments can be determined by analyzing the differences between current monitoring data and corresponding ideal data.

[0060] Based on the above analysis, various monitoring data of the UAV at the current moment are obtained. These monitoring data include actual spatial coordinates, actual flight speed, actual heading angle, and actual altitude information. The relative error between each type of monitoring data at the current moment and the corresponding ideal data is calculated. The absolute value of the relative error is recorded as the deviation value of each type of monitoring data at the current moment. The larger the value, the greater the deviation of the corresponding category of monitoring data of the UAV due to environmental interference at the current moment. The method for calculating the relative error is a well-known technique, and the specific process will not be elaborated further.

[0061] It should be noted that none of the above monitoring data have physical units; they are only mathematical calculations. Furthermore, this embodiment standardizes all types of monitoring data to prevent inconsistencies in units.

[0062] Because environmental disturbances have a correlated and coupled effect on the attitude and flight speed of the UAV during actual flight attitude control, meaning that under the influence of dynamic disturbances, multiple data points show significant deviations, affecting the overall flight status of the UAV, the mean of the flight deviation values ​​of all types of monitoring data at the current moment is calculated. This mean is then compared with a preset flight deviation threshold. If the mean is greater than or equal to the preset flight deviation threshold, it indicates that the UAV's flight deviation is large under the influence of dynamic environmental disturbances, affecting the accuracy of UAV flight status control. In this case, it is determined that the control parameters of the UAV need to be optimized and adjusted at the current moment; otherwise, it is determined that the control parameters do not need to be optimized and adjusted at the current moment.

[0063] It should be noted that the preset driving deviation threshold setting is related to the driving safety requirements during the inspection process. The smaller the driving deviation threshold setting, the faster the response to the driving deviation of the UAV under the influence of dynamic interference, but the inspection efficiency of the UAV will be affected. Therefore, the implementer can set the driving deviation threshold according to the actual inspection needs. In this embodiment, in order to avoid affecting the inspection efficiency and improve the accuracy of the UAV obstacle avoidance driving control, the preset driving deviation threshold is set to 0.1.

[0064] Furthermore, if the control parameters of the drone need to be optimized and adjusted at the current moment, the following procedure is executed:

[0065] The sampling time at which the control parameters of the UAV need to be adjusted is recorded as the target adjustment time. In order to accurately analyze the dynamic response characteristics of different types of monitoring data of the UAV under the influence of dynamic interference, the driving deviation values ​​of all sampling times between the current time and the previous target adjustment time of various types of monitoring data are obtained. All the driving deviation values ​​are used as input, and the least squares method is used to fit a straight line. The slope of the fitted curve is calculated and recorded as the first state value of various types of monitoring data at the current time. The larger the first state value, the more significant the trend of increasing deviation of the corresponding type of monitoring data caused by dynamic environmental interference between the previous target adjustment time and the current time.

[0066] Calculate the variance of the driving deviation values ​​of various monitoring data at all sampling times between the current time and the previous target adjustment time. This variance is denoted as the second state value of each type of monitoring data at the current time. The larger the second state value, the greater the possibility that the monitoring data of the corresponding category will be affected by dynamic environmental interference and become disordered between the previous target adjustment time and the current time.

[0067] Furthermore, based on the first and second state values ​​of various monitoring data at the current moment, a comprehensive state value for each type of monitoring data at the current moment is constructed to characterize the degree of environmental interference affecting each type of monitoring data at the current moment. The method for constructing the comprehensive state value of each type of monitoring data at the current moment is as follows: The slope of the fitted curve corresponding to the driving deviation values ​​of each type of monitoring data at all sampling moments between the current moment and the previous target adjustment moment is calculated; the variance of the driving deviation values ​​of each type of monitoring data at all sampling moments between the current moment and the previous target adjustment moment is calculated; the comprehensive state value of each type of monitoring data at the current moment is positively correlated with both the slope and the variance. This positive correlation means that the dependent variable increases (decreases) as the independent variable increases (decreases).

[0068] To eliminate the impact of dimensional inconsistencies on the calculation results, the min-max normalization method is used to normalize the first and second state values ​​of various monitoring data at the current moment. The minimum and maximum values ​​are obtained statistically based on the first and second state values ​​of all types of monitoring data at all historical moments. The min-max normalization method is a well-known technique, and its specific process will not be elaborated further.

[0069] In this embodiment, the first The comprehensive status value of the monitoring data at the current moment is recorded as follows: Its specific expression is: In the formula, Indicates the first The comprehensive status value of the monitoring data at the current moment; and They represent the first The normalized results of the first state value and the normalized results of the second state value of the monitoring data at the current moment; , These represent the preset first weight and the preset second weight, respectively, satisfying... In this embodiment, the values ​​are set to 0.5 and 0.5 respectively. The specific values ​​can be set according to the importance of the first state value and the second state value. For example, if the drone vibration amplitude is required to be high during the drone inspection process, the preset second weight should be set to a larger value during the obstacle avoidance process; if the deviation state of the monitoring data is required to be high, the preset first weight should be set to a larger value.

[0070] It should be noted that if there is no target adjustment time before the current time, the initial time of the UAV's obstacle avoidance flight will be used as the previous target adjustment time for analysis, so as to obtain the comprehensive status value of various monitoring data at the current time.

[0071] The larger the overall state value, the greater the degree of interference from the dynamic environment on various monitoring data during the drone's operation at the current moment, and the more unstable the drone's obstacle avoidance process becomes.

[0072] Step 4: Based on the comprehensive state value and driving deviation value of all types of monitoring data at the current moment, construct the adjustment coefficient for the current moment, and adjust the initial values ​​of various control parameters in the PID control algorithm at the current moment.

[0073] Furthermore, based on the calculated comprehensive state value of each type of monitoring data at the current moment, the control parameters in the PID control algorithm at the current moment are optimized. The larger the comprehensive state value of each type of monitoring data relative to other types of monitoring data at the current moment, the more significant the change in the UAV's state reflected by that type of monitoring data at the current moment. Therefore, when evaluating the degree of adjustment of each control parameter at the current moment, that type of monitoring data should be given a greater weight.

[0074] Specifically, the ratio between the comprehensive state value of various monitoring data at the current moment and the sum of the comprehensive state values ​​of all types of monitoring data is calculated and recorded as the state response weight of various types of monitoring data at the current moment. The larger the value, the more important the driving deviation value corresponding to the monitoring data of the corresponding category is in assessing the degree of obstacle avoidance anomaly of the UAV.

[0075] It should be noted that when performing division operations, if the denominator is 0, a preset minimum positive number (0.01 in this embodiment) is added to the denominator as a parameter adjustment factor to prevent the calculation from crashing due to the denominator being 0.

[0076] Furthermore, the mean value of the driving deviation of various monitoring data at all sampling times between the current time and the previous target adjustment time is calculated and recorded as the comprehensive driving deviation value of various monitoring data at the current time. The larger the value, the more significant the deviation of the corresponding category of monitoring data is due to the influence of dynamic environmental interference factors. That is, the UAV may deviate significantly from the driving path during obstacle avoidance. In this case, it is more necessary to restore the UAV to normal as quickly as possible to reduce its obstacle avoidance error.

[0077] As a preferred implementation, an adjustment coefficient is constructed based on the comprehensive state value and the driving deviation value of all types of monitoring data at the current moment. This coefficient characterizes the degree of deviation of the UAV's driving state from the ideal driving state at the current moment. The method for constructing the adjustment coefficient at the current moment is as follows: calculate the ratio between the comprehensive state value of all types of monitoring data at the current moment and the sum of the comprehensive state values ​​of all types of monitoring data; calculate the mean of the driving deviation values ​​of all types of monitoring data at all sampling moments between the current moment and the previous target adjustment moment, and calculate the ratio of the driving deviation value of each type of monitoring data at the current moment to the mean value; the adjustment coefficient at the current moment is positively correlated with both of the above ratios. The flowchart for obtaining the adjustment coefficient at the current moment is shown below. Figure 2 As shown.

[0078] In this embodiment, the adjustment coefficient at the current moment is denoted as... Its specific expression is: In the formula, This is the adjustment factor for the current moment; Indicates the next time step. Status response weights for monitoring data; Indicates the next time step. Driving deviation values ​​from the monitoring data; Indicates the next time step. The comprehensive driving deviation value of the monitoring data; n is the number of monitoring data categories; This is a preset constant, which is set to 0.01 in this embodiment to avoid the denominator being 0.

[0079] The larger the adjustment coefficient value at the current moment, the greater the driving deviation value of each type of monitoring data at the current moment compared to the overall driving deviation value. Under the influence of dynamic environmental interference, the UAV is more likely to exhibit a large control deviation. Therefore, in this case, the control parameters in the PID control algorithm can be appropriately increased to improve the response speed and reduce the static error. Conversely, the smaller the adjustment coefficient value at the current moment, the smaller the driving deviation value of each type of monitoring data at the current moment compared to the overall driving deviation value. The UAV is less affected by dynamic environmental interference during obstacle avoidance. Therefore, the control parameters can be reduced to decrease the system gain and improve the stability of the control response.

[0080] Furthermore, during the inspection of high-voltage towers using drones, sensors onboard the drones collect monitoring data in real time, determining whether adjustments to the drone's control parameters are necessary. If adjustments are required, the adjustment coefficients for the current moment are calculated according to the above steps, and then these adjustment coefficients are used to optimize the initial values ​​of various control parameters in the PID control algorithm. To avoid over-adjusting various control parameters, which could cause them to exceed their normal range and lead to control instability, this application uses a preset control parameter mapping table to adjust the adjustment coefficients for the current moment. These are respectively mapped to the scaling factors of each control parameter at the current time.

[0081] The specific formula for adjusting the initial values ​​of the various control parameters in the PID control algorithm is as follows: In the formula, This represents the adjusted value of the j-th control parameter. This represents the initial value of the j-th control parameter; This is the scaling factor for the j-th control parameter at the current time. The control parameters in the PID control algorithm refer to the proportional coefficient, integral coefficient, and derivative coefficient. The initial values ​​of each control parameter are determined using the decay curve method.

[0082] Furthermore, starting from the current moment, the adjusted control parameters will be used as the control parameters for the PID control algorithm during the UAV's obstacle avoidance maneuver, enabling real-time feedback control and ensuring the UAV travels stably along the obstacle avoidance path. These optimizations reduce the impact of dynamic interference during obstacle avoidance, improve the stability and accuracy of obstacle avoidance detection and control, and enhance the safety of UAV inspections of high-voltage power towers.

[0083] Based on the same inventive concept as the above method, this application also provides an unmanned aerial vehicle (UAV) obstacle avoidance detection and control system for high-voltage tower inspection, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the above-described UAV obstacle avoidance detection and control method for high-voltage tower inspection.

[0084] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0085] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0086] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for obstacle avoidance detection and control of unmanned aerial vehicles (UAVs) used for high-voltage tower inspection, characterized in that, The method includes the following steps: Real-time acquisition of 3D point cloud data and various monitoring data during the drone's flight process; Clustering is performed on the non-ground point cloud in the current 3D point cloud data. Based on the shortest distance between the UAV and each cluster at the current time, as well as the 3D spatial coordinates of the center of each cluster and the 3D grid map it maps, and combined with the target position coordinates of the UAV, the obstacle avoidance driving path at the current time is obtained, and then various ideal monitoring data of the UAV at the current time are obtained. Based on the differences between various monitoring data at the current moment and their ideal monitoring data, the driving deviation values ​​of various monitoring data at the current moment are constructed to determine whether the control parameters of the UAV need to be adjusted at the current moment. When adjustment is required, based on the significance and dispersion of the increasing trend of the driving deviation values ​​of various monitoring data at all sampling moments between the current moment and the last adjustment moment, the comprehensive state value of various monitoring data at the current moment is constructed. Based on the comprehensive state value and driving deviation value of all types of monitoring data at the current moment, the adjustment coefficient for the current moment is constructed, and the initial values ​​of various control parameters in the PID control algorithm at the current moment are adjusted accordingly.

2. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 1, characterized in that, The method for obtaining the obstacle avoidance driving path at the current moment is as follows: Based on the distance between the UAV's current spatial location and the three-dimensional spatial coordinates of each cluster, the shortest distance between the UAV and each cluster at the current moment is obtained. The current 3D grid map, the shortest distance between the UAV and each cluster, the 3D spatial coordinates of the center of each cluster, and the target position coordinates of the UAV are used as inputs to the A* algorithm. Combined with a fusion path planning algorithm based on the dynamic window method, the algorithm outputs the obstacle avoidance path of the UAV at the current moment.

3. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 2, characterized in that, The method for obtaining the shortest distance between the UAV and each cluster at the current moment is as follows: Real-time acquisition of latitude and longitude coordinates and flight altitude of the drone during its flight; Based on the latitude and longitude coordinates and flight altitude of the UAV at the current moment, the three-dimensional spatial coordinates of the UAV at the current moment are obtained, and the three-dimensional spatial coordinates of the UAV and all point clouds in each cluster at the current moment are transformed to the same geographic coordinate system. Calculate the Euclidean distance between the UAV and the transformed 3D spatial coordinates of all point clouds in each obstacle, and record the minimum value among all Euclidean distances as the shortest distance between the UAV and each cluster at the current moment.

4. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 2, characterized in that, The method for constructing the three-dimensional raster map is as follows: The three-dimensional spatial coordinate data of all cluster centers at the current moment are mapped to a three-dimensional raster space of preset resolution using a raster mapping algorithm. Rasters containing cluster center points are marked as danger zones, and rasters not containing cluster center points are marked as safe zones, thus obtaining a three-dimensional raster map at the current moment.

5. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 1, characterized in that, The driving deviation value of various monitoring data at the current moment refers to the absolute value of the relative error between various monitoring data at the current moment and the corresponding ideal data.

6. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 1, characterized in that, The specific process for determining whether the control parameters of the drone need to be adjusted at the current moment is as follows: Calculate the mean value of the driving deviation of all types of monitoring data at the current moment; If the mean value is greater than or equal to the preset driving deviation threshold, it is determined that the control parameters of the drone need to be adjusted at the current moment. Conversely, if the condition is not met, it is determined that no adjustment to the drone's control parameters is needed at the current moment.

7. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 1, characterized in that, The method for constructing the comprehensive state value of the various monitoring data at the current moment is as follows: The sampling time at which the control parameters of the UAV need to be adjusted is recorded as the target adjustment time; The slope of the fitted curve corresponding to the driving deviation values ​​of various monitoring data at all sampling times between the current time and the previous target adjustment time is calculated. The variance of the driving deviation values ​​of various monitoring data at all sampling times between the current time and the previous target adjustment time is statistically analyzed. The overall state value of the various monitoring data at the current moment is positively correlated with the slope and the variance.

8. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 7, characterized in that, The method for constructing the adjustment coefficient at the current moment is as follows: Calculate the ratio between the comprehensive state value of each type of monitoring data at the current moment and the sum of the comprehensive state values ​​of all types of monitoring data; Calculate the mean value of the driving deviation of various monitoring data at all sampling times between the current time and the previous target adjustment time, and calculate the ratio of the driving deviation value of various monitoring data at the current time to the mean value; The adjustment coefficient at the current moment is positively correlated with both of the above ratios.

9. The UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in claim 1, characterized in that, The specific process of adjusting the initial values ​​of each control parameter in the PID control algorithm at the current moment is as follows: Based on the preset control parameter mapping table, the adjustment coefficients at the current time are mapped to the scaling factors of each control parameter at the current time. The specific formula for adjusting the initial values ​​of each control parameter is as follows: In the formula, This represents the adjusted value of the j-th control parameter. This represents the initial value of the j-th control parameter; is the scaling factor for the j-th control parameter at the current time.

10. A drone obstacle avoidance and detection control system for high-voltage tower inspection, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the UAV obstacle avoidance detection and control method for high-voltage tower inspection as described in any one of claims 1-9.

Images